Assessing Engineering Student Perceptions of Introductory CS Courses in an Indian Context
Extracted source text · arXiv 2508.06563
\documentclass[lettersize,journal]{IEEEtran}
\usepackage[colorlinks,urlcolor=black,linkcolor=blue,citecolor=blue]{hyperref}
% \IEEEoverridecommandlockouts
\def\BibTeX{{\rm B\kern-.05em{\sc i\kern-.025em b}\kern-.08em
T\kern-.1667em\lower.7ex\hbox{E}\kern-.125emX}}
%\iffalse
%\documentclass[conference]{IEEEtran}
%\IEEEoverridecommandlockouts
%\pdfoutput=1
\ifCLASSOPTIONcompsoc
\usepackage[nocompress]{cite}
\else
\usepackage{cite}
\fi
\usepackage{algorithm}
\usepackage{pgffor}
\usepackage{algpseudocode}
\usepackage{epsfig}
\usepackage{graphicx}
\usepackage{amsfonts}
\usepackage{comment}
\usepackage{color}
\usepackage{amssymb}
\usepackage{amsmath}
% \usepackage{mathtools}
\usepackage{amsthm}
% \usepackage{multirow}
% \usepackage{enumitem}
% \usepackage[shortlabel]{enumitem}
\usepackage{etoolbox}
%\usepackage{balance}
% \usepackage{booktabs}
\usepackage{color}
\usepackage{flushend}
% \usepackage{lipsum}
\usepackage{babel,blindtext}
\usepackage{caption}
\usepackage{subcaption}
%===========CHANDRA============
\usepackage{csquotes}
\usepackage{rotating}
\usepackage{amssymb}
\usepackage{algorithm}
\usepackage{algpseudocode}
\usepackage{color}
\usepackage{epsf}
\usepackage{psfrag}
\usepackage{epsfig}
\usepackage{graphicx}
\usepackage{amsfonts}
\usepackage{textcomp}
\usepackage{comment}
\usepackage{footnote}
%\usepackage{hyperref}
\usepackage{tablefootnote}
\usepackage[flushleft]{threeparttable}
\usepackage{paralist}
\usepackage{mdwlist}
\usepackage{amssymb}
% \usepackage{amsmath}
\usepackage{url}
\usepackage{verbatim}
\usepackage{alltt}
\usepackage{balance}
%\usepackage{hyperref}
\usepackage{multirow}
\usepackage{epstopdf}
\usepackage{lipsum}% http://ctan.org/pkg/lipsum
\usepackage{float}
\usepackage{booktabs}
% \usepackage{lineno}
\usepackage{slashbox}
\usepackage[table]{xcolor}
\usepackage{array}
\usepackage{boldline}
\usepackage{caption,setspace}
\usepackage{mathtools}
% \usepackage[table]{xcolor}
%\usepackage{amsmath}
%\usepackage{algorithm}
%\usepackage{algorithmic}
\usepackage{csquotes}
\usepackage{stackengine}[2013-10-15]
\newcommand\textss[1]{\stackengine{.9ex}{}{\scriptsize#1}{O}{l}{F}{F}{L}}
\usepackage[T1]{fontenc}
\usepackage{frcursive}
\usepackage{calligra}
%\usepackage{emerald}
\usepackage{wedn}
\usepackage{bm}
%\usepackage{fontspec}
%\setmainfont{QTCoronation}
%\usepackage[default]{frcursive}
\usepackage{aurical}
%\usepackage{amsmath}
\usepackage{upgreek}
\usepackage{pgffor}
\newcommand{\setfont}[2]{{\fontfamily{#1}\selectfont #2}}
\definecolor{green}{RGB}{0,204,0}
%\definecolor{blue1}{RGB}{0,150,200}
%\definecolor{orange1}{RGB}{236,112,10}
\DeclarePairedDelimiter\floor{\lfloor}{\rfloor}
%\usepackage[colorlinks,urlcolor=blue,linkcolor=blue,citecolor=blue]{hyperref}
%
% \iffalse
% \usepackage[colorlinks = true,
% linkcolor = red,
% urlcolor = black,
% citecolor = blue,
% anchorcolor = magenta]{hyperref}
%\usepackage{hyperref}
% \fi
\usepackage{cite}
%\usepackage{amsmath}
%\usepackage{algorithmic}
\usepackage{graphicx}
\usepackage{adjustbox}
\usepackage{graphicx}
\usepackage{subcaption}
\usepackage{esint}
\usepackage{tipa}
% \usepackage{rotating}
% \usepackage{multirow}
% \usepackage{longtable}
\usepackage{pgfplots}
% \usepackage{pgfplots}
\usepackage{pgfplotstable}
\pgfplotsset{compat=1.18}
% \pgfplotsset{compat=1.17}
\usetikzlibrary{patterns}
\newcommand{\mhyphen}[1]{{\operatorname{#1}}}
\DeclareMathOperator{\E}{\mathbb{E}}
\newcolumntype{?}{!{\vrule width 1pt}}
\newcolumntype{+}{!{\vrule width 1.25pt}}
\def\hlineb#1{%
\noalign{\ifnum0=`}\fi\hrule \@height #1 %
\futurelet\reserved@a\@xhline}
\newcommand{\argmax}{\arg\!\max}
\usepackage{cancel}
\newcommand{\highlight}[1]{\textbf{\textcolor{blue}{#1}}} % Define highlight command
\definecolor{darkgreen}{RGB}{0,153,0}
\definecolor{darkred}{RGB}{192,0,0}
%==========CHANDRA============
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% \captionsetup[table]{format=plain,labelformat=simple,labelsep=period}%
% \usepackage[top=0.8in,bottom=0.7in,left=0.6in,right=0.6in]{geometry}
% \setlength{\textfloatsep}{10pt plus 1pt minus 2pt} % set space between float and text
% \setlength{\intextsep}{4pt plus 1pt minus 2pt} % set space between text and float
% \setlength{\columnsep}{20pt} % set space between columns
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\ifCLASSINFOpdf
\else
\fi
\hyphenation{}
\begin{document}
%===================
% \title{Engineering Students’ Experiences of Assessment in Introductory Computer Science Course in Indian Colleges}
% \title{Perceptions of Assessment in Introductory Computer Science Course: Insights from Indian Engineering Students}
% \title{Perceptions of Assessment in Introductory Programming Course: Insights from Indian Engineering Students}
% \title{Perceptions of Assessment in Introductory CS Course: Insights from Indian Engineering Students}
% \title{Engineering Student Perceptions of Introductory CS Course Assessment in an Indian Context}
\title{Assessing Engineering Student Perceptions of Introductory CS Courses in an Indian Context}
\author{Utsav Kumar Nareti,~\IEEEmembership{Student Member,~IEEE},
Divyansh Gupta,
Chandranath Adak,~\IEEEmembership{Senior Member,~IEEE},
Soumi Chattopadhyay,~\IEEEmembership{Senior Member,~IEEE},
Emma Riese,
Tanujit Chakraborty,
Mayank Agarwal,
Satendra Kumar
\thanks{
% U.K. Nareti, C. Adak, M. Agarwal, and S. Kumar are with the Dept. of CSE, Indian Institute of Technology Patna, Bihar 801106, India.
% \\
% D. Gupta is with the Dept. of EE, Indian Institute of Technology Patna, Bihar 801106, India.
U.K. Nareti, D. Gupta, C. Adak, M. Agarwal, and S. Kumar are with the Indian Institute of Technology Patna, Bihar 801106, India.
\\
S. Chattopadhyay is with the Dept. of CSE, Indian Institute of Technology Indore, Madhya Pradesh 453552, India.
\\
E. Riese is with the School of EECS, KTH Royal Institute of Technology, SE-100 44 Stockholm, Sweden.
\\
T. Chakraborty is with SAFIR, Sorbonne University Abu Dhabi, and SCAI, Sorbonne University, Paris, 75006, France.
\\
\emph{Corresponding author: C. Adak (email: chandranath@iitp.ac.in)}
}
}
\markboth{U. K. N\MakeLowercase{areti \textit{et al.}}}{}
%\IEEEpubid{0000--0000/00\$00.00~\copyright~2021 IEEE}
% Remember, if you use this you must call \IEEEpubidadjcol in the second
% column for its text to clear the IEEEpubid mark.
\maketitle
% \begin{abstract}
% \textit{Contribution:}
% This study provides a comprehensive analysis of \textcolor{blue}{engineering students' perceptions of assessment} practices in an introductory course on programming and data structures (CS102) and its associated lab \textcolor{blue}{course} (CS112) in an Indian engineering college context. It \textcolor{blue}{also} explores how students' assessment experiences and perceptions vary by prior programming experience, technology familiarity, gender, academic branch, and grade.
% \textit{Background:}
% Basic programming skills \textcolor{blue}{have} become essential for all engineering students \textcolor{blue}{to master.} Large and diverse student cohorts present challenges in ensuring fair, supportive, and meaningful assessment experiences. \textcolor{blue}{Understanding students' perceptions of the assessments and their} effectiveness and equity is increasingly important.
% \textit{Research Questions:} 1) How do engineering students perceive and engage with different types of assessments, such as lab assignments,
% midterm exams, and projects? 2) How do students view the roles and contributions of teaching assistants (TAs) and instructors during
% assessment situations? \textcolor{green}{3) Do the performance of students in the introductory CS course follow a Gaussian distribution?} 4) What differences exist based on prior programming and technology experience? and how do gender, grades, and academic branch influence perceptions and responses? 5) Are there any differences in perceptions or factors between
% a European and Indian educational context?
% \textit{Methodologies:} Based on previously identified themes from an interview study \textcolor{blue}{and a follow-up quantitative study in a different educational context}, this study employed a survey of 25 statements weekly to 318 first-year \textcolor{blue}{Bachelor of Technology} students over 9 weeks. Quantitative analysis was conducted using descriptive statistics \textcolor{green}{and statistical tests (e.g.,} the Mann–Whitney U test and the Kruskal–Wallis test) to identify significant differences and trends.
% \textit{Findings:} Lab assignments are widely seen as valuable learning opportunities. Students perceive exams and projects \textcolor{blue}{as} authentic and important for improving skills and knowledge. They see instructors as shaping course content and receiving helpful feedback from TAs, with some reported variations in TA practices.
% \end{abstract}
\begin{abstract}
Understanding student perceptions of assessment is vital for designing inclusive and effective learning environments, especially in technical education. This study explores engineering students’ perceptions of assessment practices in an introductory computer science/ programming course, and its associated laboratory within an Indian engineering institute context. A total of 318 first-year Bachelor of Technology students participated in a weekly 25-statement Likert-scale survey conducted over nine weeks. Using descriptive statistics and non-parametric tests (Mann-Whitney U and Kruskal-Wallis), the analysis reveals that students largely perceive lab assignments as effective learning activities and view exams and projects as authentic and skill-enhancing. Students appreciated the role of instructors in shaping course content and found teaching assistants to be approachable and helpful, despite some inconsistencies. The study also finds significant variations in students’ academic performance and assessment perceptions based on prior programming experience, technology familiarity, gender, and academic branch.
Notably, the performance data did not follow a Gaussian distribution, challenging common assumptions in grade modeling. A comparative analysis with European cohorts highlights both universal patterns and contextual differences, offering valuable insights for designing inclusive and equitable assessment strategies in programming education.
\end{abstract}
\begin{IEEEkeywords}
% Assessment, Student feedback, Higher education, Student perceptions, Programming education.
Programming education, Student perceptions, Assessment practices, Cross-cultural higher education.
\end{IEEEkeywords}
\input{1intro}
\input{2related}
\input{3method}
\input{4_result}
\input{5conclusion}
\input{7app_QQ}
\balance
\bibliographystyle{IEEEtran}
\bibliography{ref.bib}
\clearpage
% \balance
\input{7appendix}
\end{document}
\section{Introduction}\label{sec:intro}
\IEEEPARstart{I}{ntroductory} programming, offered as the first Computer Science (CS) course along with its associated laboratory course, are mandatory requirements for many first-year {engineering} students, including both CS and non-CS majors. Introductory programming courses have been the focus of numerous studies, largely due to the common challenges students face in acquiring programming skills \cite{10.1145/3293881.3295779}. Difficulties in mastering the course are attributed not only to the inherent complexity of programming \cite{article1}, but also to social factors like a competitive classroom environment \cite{article}. Historically, the performance of students on course exams is often assumed to follow a normal distribution, reflecting that most are clustered around the mean and a small number either overachieve or underachieve \cite{gaultney2001grade}. Although the Gaussian assumption is the ubiquitous choice for modeling exam scores, there is empirical evidence of an asymmetric logit-normal assumption at the item level \cite{arthurs2019grades}. Computer programming is considered an essential requirement for all engineering students, providing key skills for academic and professional growth \cite{10036401}.
To manage the workload of large mandatory courses, course coordinators/ main instructors often work with teaching assistants (TAs), generally Master's or PhD students, to coordinate labs, lead tutorials, and assist with grading \cite{10036401}. While extensive research exists on CS courses, students' experiences with assessments remain underexplored \cite{10.1145/3293881.3295779}. This study addresses the gap by investigating how engineering students experience assessments in their introductory CS course and how they perceive the roles of course coordinators and TAs in these contexts. Furthermore, this study contributes with an Indian perspective to contrast the previous study results in a European setting \cite{10036401}. This group is particularly relevant, as many engineering programs require foundational programming skills, which are valuable even outside a CS major \cite{10.1145/3517134}.
% Non-CS majors make up a significant portion of CS102/ CS112 enrollment and often do not view themselves as future programmers or computer scientists. Nevertheless, the skills acquired in this course are considered valuable for their broader engineering education and future job applications.
Non-CS majors form a significant portion of the course enrollment and often do not view themselves as future programmers or computer scientists. Nevertheless, the skills acquired in this course are considered essential for broader engineering education and future career opportunities.
Building on prior European research \cite{qualitative_step,10036401}, this study examines how Indian students' perceptions differ based on prior experience, gender, academic branch, and grades, and investigates how these patterns align with findings from international contexts.
The central research questions (\emph{RQ}s) explored in this study are:
% \begin{enumerate}[(RQ1)]
% \item How do engineering students perceive and engage with different types of assessments, such as lab assignments, mid/end-term exams, and projects?
% \item How do students view the roles and contributions of TAs and instructors during assessment?
% \item {Do the performance of students in the introductory CS course follow a Gaussian distribution? }
% \item What differences exist based on prior programming and technology experience {and how do factors such as gender, grades, and academic branch influence perceptions and responses?}
% \item {Are there any differences in perceptions or factors between a European and Indian educational context}?
% \end{enumerate}
\emph{RQ1}. How do engineering students perceive and engage with different types of assessments, such as lab assignments, mid/end-term exams, and projects?
\emph{RQ2}. How do students view the roles and contributions of TAs and instructors during assessments?
\emph{RQ3}. Does the performance of students in the introductory CS course follow a Gaussian distribution?
\emph{RQ4}. What differences exist based on prior programming and technology experience, and how do factors such as gender, grades, and academic branch influence perceptions?
\emph{RQ5}. Are there any differences in perceptions or influencing factors between European and Indian educational contexts?
% The rest of the paper is organized as follows:
% Section \ref{sec:related_work} discusses background,
% Section \ref{sec:method} describes the methodology,
% Section \ref{4sec:result} presents the results, and
% Section \ref{5sec:discussion_conclusion} concludes the study with discussions and directions for future research.
The rest of the paper is organized as follows:
Section~\ref{sec:related_work} reviews background, Section~\ref{sec:method} details the methodology, Section~\ref{4sec:result} presents the results, and Section~\ref{5sec:discussion_conclusion}
concludes the study with discussions and future research directions.
% offers discussion and conclusions.
% \section{Related Work}
\section{Background}
\label{sec:related_work}
% This section presents the foundation for this research with key concepts.
\subsection{Foundation and Context}
% \noindent
This study adopts a constructive learning approach \cite{cambridge}, wherein the course is developed from the ground up by integrating theoretical instruction with hands-on practice/ lab. Course coordinators emphasize foundational concepts to equip students for subsequent, related courses \cite{10.1145/3517134}. The assessment structure consists of two assignments, two quizzes, and two main examinations: midterm and endterm that hold greater weight in the overall evaluation. A course project assigned early in the term requires students to demonstrate running code with a brief presentation and Q\&A session \cite{10.1145/1565799.1565820}. The lab component is closely aligned with the main course, involving weekly structured three-hour sessions and assessments through weekly evaluations, as well as midterm and end-term lab exams comprising practical tasks and a viva. This format, common across many engineering institutions, fosters interest in programming among non-CS majors and underscores its relevance across engineering disciplines \cite{10.1145/961290.961519}. By addressing structural gaps identified through student feedback, the study aims to promote student engagement and reposition the course as a meaningful component of academic and professional development rather than a mere graduation requirement.
\subsection{Assessment Pattern in the Studied Courses}
% \noindent
The courses follow a continuous evaluation model, incorporating a combination of assignments, quizzes, projects, and examinations to assess student learning comprehensively. Examinations are conducted in both pen-and-paper and computer-based formats, while assignments are designed to foster both practical application and theoretical understanding. Quizzes, typically comprising multiple-choice questions, are administered before major exams to reinforce recently taught concepts and enhance retention \cite{articleMCQ}. Both multiple-choice and open-ended questions form key components of the assessment strategy and have proven effective in evaluating students’ programming knowledge. However, pen-and-paper exams have been criticized for their limited authenticity in assessing hands-on programming skills \cite{Bennedsen01102007}. To address this, the use of real-world assignments is increasingly recommended in computer science curricula \cite{articleL}. In the context of this study, project- and portfolio-based grading is also employed, where students compile a portfolio of work throughout the semester, which is evaluated at the end. This approach offers a holistic perspective on student progress and allows room for improvement, enabling recovery from occasional underperformance \cite{7062585}. Continuous evaluation extends to lab sessions as well, with weekly assessments and culminating in a comprehensive lab examination and viva that rigorously test their problem-solving skills. A persistent challenge for instructors is designing assignments that fall within students’ Zone of Proximal Development (ZPD), defined as the distance between the {actual developmental level} determined by independent problem solving and the {level of potential development} achievable through guided support or collaboration with more capable peers \cite{ZPD_1}.
\subsection{Role of TAs}
% \noindent
Teaching assistants (TAs) play a crucial role in handling increasing enrollment in CS courses by leading tutorials, conducting office hours, supervising lab sessions, and assisting with grading tasks \cite{TAs1}. Typically, master's or PhD students serve as TAs; however, many receive minimal formal training, which can pose significant challenges, particularly in assessment-related responsibilities \cite{10.1145/3291279.3339418}. TAs often face difficulties in addressing the diverse abilities and expectations of students in large, heterogeneous classrooms \cite{TAs1}. While collaborative grading among TAs, through peer calibration, has been shown to improve grading reliability, students may still perceive inconsistencies in evaluation practices \cite{TA_gradning}.
\subsection{Student Success Factors}
% \noindent
% Student comfort and willingness to seek help are strong predictors of academic success in CS courses. Anxiety in CS education arises not only from programming challenges but also from test-related stress and collaborative tasks like pair programming \cite{10.1145/2999541.2999557}. The CS classroom is often seen as evaluative and competitive rather than supportive \cite{article}. Persistent stereotypes portraying CS as male-dominated can hinder students’ sense of belonging, particularly among female students.
% Males often enter with prior CS experience, contributing to higher achievement \cite{10.1145/364447.364581}, while females often report lower confidence, limited exposure, and a reduced sense of belonging, contributing to lower grades.
% Moreover, many students prefer seeking help from peers with stronger programming skills rather than TAs, indicating trust in peer networks \cite{10.1145/2729094.2742611}. To foster an inclusive learning environment, institutions must counteract stereotypes, encourage collaboration, connect content to student interests, and promote positive engagement with TAs and instructors \cite{inproceedingsF}. Key factors such as confidence, prior experience, digital literacy, and sense of belonging significantly influence student retention and can help reduce dropout rates in introductory CS courses \cite{10.1145/2729094.2742611}.
Student comfort and willingness to seek help are strong predictors of academic success in CS courses. Anxiety in CS education stems not only from programming challenges but also from test-related stress and collaborative tasks like pair programming~\cite{10.1145/2999541.2999557}. The CS classroom is often perceived as evaluative and competitive rather than supportive~\cite{article}. Persistent stereotypes portraying CS as male-dominated can hinder students’ sense of belonging, particularly among female students. Males often enter with prior CS experience, contributing to higher achievement, while females often report lower confidence, limited exposure, and a reduced sense of belonging, leading to lower grades \cite{female_3}. Many students prefer seeking help from peers with stronger programming skills rather than TAs, indicating greater trust in peer networks~\cite{10.1145/2729094.2742611}. To foster an inclusive environment, institutions must counter stereotypes, encourage collaboration, connect content to student interests, and promote engagement with TAs and instructors~\cite{inproceedingsF}. Confidence, prior experience, digital literacy, and sense of belonging significantly influence retention and can help reduce dropout rates in introductory CS courses~\cite{10.1145/2729094.2742611}.
% This study adopts a constructive learning approach \cite{cambridge}, in which the course is built from the ground up, integrating theoretical instruction and hands-on practice. Course coordinators focus on foundational concepts to prepare students for related future courses \cite{10.1145/3517134}. Assessment includes two assignments and two quizzes, one before and after the midterm, along with two major exams: the midterm and the \textcolor{purple}{endterm}, which carry greater weight in overall grading. In the second or third week, a course project is introduced, to be submitted near the end of the term, followed by a brief presentation and Q\&A session \cite{10.1145/1565799.1565820}. The associated lab course is closely aligned with the main curriculum. Each week, students attend a structured three-hour lab session and are assessed through midterm and endterm lab exams, which include performing tasks and a viva. This format, common in many engineering colleges, elevates programming interest among non-CS majors and highlights its relevance across engineering disciplines \cite{10.1145/961290.961519}. This study addresses gaps in the learning structure based on student feedback, aiming to help students see the course as a valuable part of their academic and professional growth, rather than just a mere requirement to pass.
% The course\textcolor{blue}{s} employs a continuous evaluation model, utilizing a combination of assignments, quizzes, projects, and examinations to assess student learning comprehensively. Examinations are based on both pen \& paper and computer-based formats, while assignments are designed to emphasize both practical and theoretical problem-solving abilities. Quizzes, typically consisting of multiple-choice questions, are scheduled prior to major examinations to reinforce recently covered material and enhance retention \cite{articleMCQ}. Multiple-choice and \textcolor{blue}{open-ended}\textcolor{green}{} questions form a core component of the assessment strategy and have demonstrated effectiveness in evaluating students' programming knowledge. Nevertheless, pen \& paper exams have been critiqued for their limited authenticity in assessing practical programming skills \textcolor{blue}{\cite{Bennedsen01102007}}. Along with these, real-world assignments are \textcolor{blue}{recommended to be} used in the computer science \textcolor{purple}{curriculum} \cite{articleL}. \textcolor{blue}{In the studied courses}, project and portfolio-based grading is also employed, with students developing a portfolio of work throughout the semester that is evaluated at the end. This approach provides a holistic view of student progress and offers opportunities for improvement, enabling students to recover from isolated instances of poor performance \cite{7062585}.
% Continuous evaluation extends to laboratory sessions, where students are assessed weekly, culminating in a comprehensive lab examination and viva that rigorously test their problem-solving skills.
% \textcolor{blue}{Another challenge for teachers can be to develop assignments that are in within the students Zone of Proximal Development (ZPD).} \textcolor{green}{Vygotsky \cite{ZPD_1} defined the ZPD as the distance between the actual developmental level as determined by independent problem solving and the level of potential development as determined through problem solving under adult guidance or in collaboration with more capable peers \cite{learning_1}.}
% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% % Divyansh
% The course follows a continuous evaluation model, assessing students through assignments, quizzes, a project, and exams. Exams include both pen-and-paper and computer-based formats, while assignments emphasize practical and theoretical problem-solving. Quizzes, usually multiple-choice, are taken before major exams to reinforce recent topics \cite{articleMCQ}. Multiple-choice and theoretical questions are central to the course and have proven effective in assessing programming knowledge. However, pen-and-paper exams are often criticized for lacking authenticity in evaluating real programming skills \cite{10.5555/1563334.1563345}. Research on assessment authenticity reflects both student and faculty perspectives, leading to the development of "real-world" assignments in CS curricula \cite{articleL}. The course also uses project and portfolio-based grading, with students building a portfolio throughout the semester, assessed at the end. This approach offers a comprehensive view of student progress and allows for improvement over time, helping students recover from poor performance in individual assessments \cite{7062585}. Continuous evaluation also applies to lab sessions, with students assessed weekly, culminating in a comprehensive lab exam and viva that tests problem-solving skills.
%While CS research often overlooks the diversity of student assessment experiences, prior phenomenographic studies identified five key categories: (1) grading as important to the teacher, (2) grading as important to the student, (3) assessment as guidance, (4) assessment as a learning opportunity, and (5) assessment as communication. The first three are viewed as the most beneficial \cite{8064428}.
% Teaching assistants (TAs) play a key role in managing rising enrollment in CS courses by leading tutorials, holding office hours, supervising labs, and grading assignments \cite{10.1109/FIE44824.2020.9274245}. Usually, master's or PhD students are selected as TAs \cite{10.1145/3291279.3339418}. However, many receive limited training, leading to challenges, especially with assessment tasks \cite{8659243}. \textcolor{blue}{It is also observed that TAs often struggle to manage the wide range of student abilities and expectations, especially in large, diverse classes \cite{TAs1}. Research shows that collaborative grading activities among TAs improves reliability through peer calibration \cite{TA_gradning}. Nonetheless, the grading can still be perceived as inconsistent by the students \cite{TA_gradning}}.
%However, time and resource constraints often limit such collaboration.
%These challenges highlight the need for structured training and support systems to enhance TA effectiveness and improve student engagement in increasingly crowded CS courses \cite{inproceedings2}.
% Student comfort and their willingness to seek assistance are recognized as strong predictors of academic success in computer science courses. Research indicates that anxiety in CS education stems not only from the challenges of learning to program but also from test-related pressures and collaborative activities such as pair programming \cite{10.1145/2999541.2999557},\cite{4271597}. The computer science classroom is frequently perceived as evaluative and competitive, rather than as a supportive environment \cite{article}. Persistent stereotypes portraying CS as a male-dominated discipline can hinder students’ sense of belonging, particularly for female students. Male students are more likely to enter with prior CS experience, which is associated with higher achievement rates \cite{10.1145/364447.364581}. In contrast, female students often report a diminished sense of belonging and tend to earn lower grades, influenced by factors including confidence, previous exposure to CS, and the overall learning environment \cite{femalecite}.
% Additionally, many students prefer to seek help from peers with stronger programming backgrounds rather than from TAs, suggesting a greater trust in peer support networks \cite{10.1145/2729094.2742611}. To improve an inclusive and effective learning environment, it is essential for institutions to actively counteract stereotypes, encourage collaborative learning, connect course content to students’ interests, and promote positive, supportive interactions with TAs and course coordinators \cite{inproceedingsF}. Critical factors such as confidence, prior CS experience, computer literacy, interest in technology, and a sense of belonging play a significant role in student retention. Addressing these elements can contribute to reducing dropout rates in introductory CS courses \cite{10.1145/2729094.2742611}.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Divyansh
% Student comfort, especially their willingness to seek help, is a strong predictor of success in computer science courses. Research shows that anxiety in CS education arises not only from learning to program but also from test-related and collaborative activities like pair programming \cite{10.1145/2999541.2999557},\cite{4271597}. The CS classroom is often seen as evaluative and competitive rather than supportive \cite{article}. Persistent stereotypes of CS as a male-dominated field can hinder students’ sense of belonging, particularly among females. Male students also tend to have more prior CS experience, which is linked to higher success rates \cite{10.1145/364447.364581}. Female students often report a lower sense of belonging and tend to earn lower grades in CS courses, influenced by factors such as confidence, prior experience, and the learning environment \cite{femalecite}. Many students also prefer seeking help from peers with stronger programming skills rather than relying on teaching assistants, indicating a greater trust in peer support \cite{10.1145/2729094.2742611}. To improve an inclusive and effective learning environment, the institute should avoid reinforcing stereotypes, promote collaborative learning, relate content to student interests, and encourage supportive interactions with TAs and course coordinators \cite{inproceedingsF}. Key factors such as confidence, prior CS experience, computer literacy, interest in technology, and a sense of belonging significantly influence student retention. Addressing these can help reduce dropout rates in introductory CS courses \cite{10.1145/2729094.2742611}.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\section{Method}
\label{sec:method}
% This section outlines the study framework, research setting, data collection process, and data analysis methods.
\subsection{Study Framework}
% \noindent
This study builds on prior research conducted in a European educational setting and employs a comprehensive research design to explore key phenomena while validating the generalizability of previous findings in an Indian context. It began with qualitative data collection \cite{qualitative_step}, followed by quantitative data collection, preprocessing, and non-parametric statistical analysis to uncover underlying patterns \cite{10036401}. These insights informed the final structured quantitative phase presented in this study, which aims to validate and compare earlier findings within a different educational and cultural setting. Feedback statements adapted from \cite{10036401}, as listed in Table \ref{tab:themes_statements}, are analyzed across factors such as gender, academic branch, grades, and prior experience (refer to Table \ref{tab:pvalues_last_week_1}).
\subsection{Research Settings}
% \noindent
The study was conducted with first-year B.Tech. students from the 2022–2026 batch at IIT Patna during Spring 2023. All first-year students were required to enroll in an introductory programming and data structures course (CS102) and its associated laboratory component (CS112). Of the 516 enrolled students, 318 consistently participated in the weekly feedback surveys. A summary of participant information is presented in Table \ref{tab:metadata_analysis}.
The course curriculum focused on C programming, covering topics such as basic syntax, data types, control flow, functions, pointers, fundamental data structures (e.g., linked lists, stacks, queues), file handling, and error management. The course was coordinated by three faculty members, resulting in a faculty-to-student ratio of 1:172, with additional support from TAs at a ratio of about 1:20.
The assessment structure of CS102 included two assignments, two quizzes, a midterm, and an end-term examination, comprising multiple-choice, fill-in-the-blank, and justification-based questions.
Weekly 3-hour lab sessions were evaluated through hands-on tasks, and midterm/ end-term lab exams of CS112 additionally included a viva component.
A group project was assigned early in the semester and assessed near the end, just before the end-term exam. Students formed self-selected groups of three to five for collaborative tasks such as pair programming~\cite{4271597}. For project evaluation, they demonstrated running code and gave a brief presentation with Q\&A.
While final grading was the responsibility of course instructors, TAs evaluated lab work, assignments, and quizzes \cite{10.1145/3488042.3488052}. This structure was designed to build foundational programming skills within a supportive and collaborative learning environment.
\input{tables/tab_metadata}
% \subsection{Data Collection}
% An online Google Form survey was distributed via email to all enrolled students at the end of each week for \textcolor{blue}{the} 12 weeks \textcolor{green}{period}, following the completion of course activities. Participation was encouraged both during class and through reminder emails. The survey, conducted in English, included four optional questions regarding prior experience: C programming experience (PCE), any programming experience (PAE), computer experience (PCoE), and smartphone experience (PSE), each rated on an ordinal scale: nil, medium, or a lot. Additionally, students could optionally provide demographic information such as gender (male/female) and academic branch.
% Participation was entirely voluntary, and no incentives were offered to increase response rates. Students were informed about the purpose of the study, and informed consent was obtained through a final question on the survey. Ethical approval for data collection was secured from both the students and the institute. The survey consisted of 25 statements (S1–S25) \cite{10036401}, developed based on previously identified themes (Table \ref{tab:themes_statements}). Students rated each statement using a 7-point Likert scale, ranging from 1 (strongly disagree) to 7 (strongly agree), with 4 representing a neutral response.
% Ethical clearance for this study was obtained from IIT Patna (ITP/EC/2022-24/09, dated 10 July 2023).
% \subsection{Data Collection}
% \noindent
% An online portal was developed to conduct the survey, which included 25 statements (S1–S25) derived from previously identified themes~\cite{10036401}, as detailed in Table~\ref{tab:themes_statements}.
% To ensure respondent authenticity, the portal required login through the institute-issued official email ID.
% Students rated each statement on a 7-point Likert scale, ranging from 1 (strongly disagree) to 7 (strongly agree), with 4 indicating a neutral response.
% Survey links were emailed to all enrolled students at the end of each week over a 9-week period, following the completion of course activities. Participation was encouraged during class and through reminder emails. The survey was administered in English, which was also the medium of instruction.
% A total of 318 students voluntarily responded to the survey. At the beginning of the course, background information was collected on students’ prior experience with C programming (PCE), any programming language (PAE), computer usage (PCoE), and smartphone usage (PSE), each rated on an ordinal scale: \textit{nil}, \textit{medium}, or \textit{high}. Additionally, gender and academic branch were also recorded (refer to Table~\ref{tab:metadata_analysis}).
% Participation was entirely voluntary, with no incentives offered. Students were informed of the study's objectives, and informed consent was obtained.
% Ethical approval was obtained from the Institute Ethics Committee at IIT Patna (ITP/EC/2022-24/09, dated 10 July 2023).
\subsection{Data Collection}
% \noindent
An online portal was developed to administer the survey, comprising 25 mandatory statements (S1--S25) derived from previously established themes~\cite{10036401}, as listed in Table~\ref{tab:themes_statements}.
To ensure respondent authenticity, the portal required login through the institute-issued official email ID.
Students rated each statement on a 7-point Likert scale, where 1 indicated \emph{strongly disagree}, 7 indicated \emph{strongly agree}, and 4 represented a \emph{neutral} response.
Survey links were distributed via email to all enrolled students at the end of each week over a span of nine weeks, following the completion of that week’s course activities. Participation was further encouraged through in-class announcements and follow-up reminders.
The survey was conducted in English, which also served as the medium of instruction for the course.
A total of 318 students voluntarily participated in the survey. At the beginning of the course, demographic and experiential background data were collected, including prior experience in C programming (PCE), general programming (PAE), computer usage (PCoE), and smartphone usage (PSE), each rated on a three-level ordinal scale: \textit{nil}, \textit{medium}, and \textit{high}. Gender and academic branch information were also recorded (see Table~\ref{tab:metadata_analysis}).
Participation was completely voluntary and uncompensated. Students were informed of the study objectives, and informed consent was obtained. Ethical clearance was granted by the Institute Ethics Committee at IIT Patna (Ref: ITP/EC/2022-24/09, dated 10 July 2023).
% \subsection{Data Analysis}
% The study employed a combination of descriptive \textcolor{green}{and inferential statistics methods to address the research questions}. Students were stratified into subgroups based on demographic and experiential variables, including gender, academic branch, grade, and self-reported prior experience in C programming, general programming, computer use, and smartphone use. \textcolor{purple}{We begin by examining QQ plots to assess the distribution of the data and find that it does not conform to a normal distribution. Appendix \ref{App:qq_plots} presents QQ plots for all 25 questions during the 12$^{th}$ week.} Therefore, for gender-based comparisons, \textcolor{green}{a non-parametric hypothesis test, namely} the Mann-Whitney U test was applied, a non-parametric method appropriate for comparing distributions between two independent groups using ranked ordinal data. To analyze differences across multiple subgroups (e.g., academic branches, prior experience categories), the Kruskal–Wallis test was utilized, which extends this approach to three or more independent groups. Initial hypotheses posited no significant differences between subgroups. While course grades were inherently ordinal (e.g., AA–CD), they were treated as continuous variables for calculating grade point averages (GPAs). All analyses were conducted using Python’s statistical libraries, with significance assessed at the 95\% confidence level across subgroups based on branch (9 groups), gender (2 groups), grade (5 groups), and prior experience (3 groups), yielding a comprehensive understanding of patterns and variations within the dataset.
% \subsection{Data Analysis}
% \noindent
% The study employed a combination of descriptive and inferential statistical methods to address the research questions. Students were stratified into subgroups based on demographic and experiential variables, including gender, academic branch, course grade, and self-reported prior experience in C programming, general programming, computer use, and smartphone use.
% We began by examining Q-Q plots to assess the distribution of responses and found that the data did not follow a normal distribution. Appendix \ref{App:qq_plots} presents Q-Q plots for all 25 statements from Week-9.
% As a result, the Mann-Whitney U test, a non-parametric hypothesis test suitable for comparing distributions between two independent groups using ranked ordinal data, was applied for gender-based comparisons. For comparisons across multiple subgroups (e.g., academic branches, prior experience levels), the Kruskal-Wallis test was used, as it is suitable for analyzing three or more independent groups.
% Initial hypotheses assumed no significant differences between subgroups. Although course grades were inherently ordinal (e.g., AA–CD), they were treated as continuous variables for calculating grade point averages (GPAs).
% All analyses were conducted using Python \emph{scipy.stats 1.16.1} statistical library, with significance tested at the 95\% confidence level among subgroups based on academic branch (9 groups), gender (2 groups), grade (5 groups), and prior experience (3 levels), yielding a comprehensive understanding of patterns and variations within the dataset.
\subsection{Data Analysis}
% \noindent
To address the research questions, the study employed a combination of descriptive and inferential statistical methods. Students were stratified into subgroups based on demographic and experiential variables, including gender, academic branch, course grade, and self-reported prior experience in C programming (PCE), general programming (PAE), computer use (PCoE), and smartphone use (PSE).
The distribution of responses was first examined using Q-Q plots, which revealed that the data did not follow a normal distribution.
% Appendix B
Appendix \ref{App:qq_plots}
provides the Q-Q plots for all 25 statements collected in Week-9.
Given the non-normality of the data, the Mann-Whitney U test, a non-parametric test suitable for comparing two independent groups using ranked ordinal data, was applied to analyze gender-based differences. For comparisons across multiple subgroups (e.g., academic branches or prior experience levels), the Kruskal-Wallis test was employed, which is appropriate for assessing differences among three or more independent groups.
Although course grades (AA–CD) are inherently ordinal, they were treated as continuous variables for the purpose of computing grade point averages (GPAs). All analyses were performed using the \emph{scipy.stats 1.16.1} library in Python. Statistical significance was evaluated at the 95\% confidence level across subgroups defined by
academic branch (9 groups),
gender (2 groups),
grade (6 groups), and
prior experience (3 levels), enabling a comprehensive understanding of response patterns and subgroup variations within the dataset.
% This study builds on prior research and employs a comprehensive research design to explore key phenomena while validating the generalizability of previous findings. It began with qualitative data collection \cite{qualitative_step}, followed by quantitative data collection, preprocessing, and non-parametric statistical analysis to uncover underlying patterns \cite{10036401}. These insights informed the final structured quantitative phase presented in this study, which aims to validate and compare earlier findings within a new educational context. Statements adapted from \cite{10036401}, as listed in Table \ref{tab:themes_statements}, are analyzed across factors such as gender, academic branch, grades, and prior experience (see Table \ref{tab:pvalues_last_week_1}).
% This study \textcolor{blue}{builds on prior research} and employs a comprehensive research design that explores key phenomena while validating the generalizability of findings. It began with qualitative data collection \cite{qualitative_step}, followed by \textcolor{blue}{a quantitative data collection}, preprocessing, and non-parametric statistical analysis to uncover underlying patterns \cite{10036401}. These insights informed \textcolor{blue}{a final} structured quantitative phase \textcolor{blue}{, presented in this study}, aimed at validating \textcolor{blue}{and comparing the previous findings in a new educational context}. Statements drawn from \cite{10036401}, as shown in Table \ref{tab:themes_statements}, are analyzed across factors such as gender, academic branch, grades, and prior experience (refer to Table \ref{tab:pvalues_last_week_1}).
%The main course was conducted alongside the lab component, offering students practical experience to reinforce classroom learning. The study also gathered data on students’ prior exposure to C and other programming languages. Given the importance of foundational skills in the Indian context, students’ familiarity with computers and smartphones was also assessed, as these are essential for success in CS courses.
% \textcolor{red}{The participants belonged to the 2022–2026 B.Tech batch at IIT Patna.}
% All first-year bachelor of technology students were required to enroll in the mandatory CS course along with its associated laboratory course. A total of 516 students registered, and 318 consistently participated in the weekly course feedback surveys. Table~\ref{tab:metadata_analysis} provides a detailed breakdown of these active participants. While the courses did not mandate prior programming experience, a basic familiarity with smartphones and computers was expected. The 12-week course\textcolor{blue}{s, running in parallel, were} coordinated and taught by three faculty members, with one serving as the lead coordinator responsible for the overall course structure. The curriculum focused on C programming, covering topics such as basic syntax, data types, control flow (conditions and loops), functions, pointers, fundamental data structures \textcolor{purple}{(linked list, stack, queue, etc.)}, file handling, and error handling. The faculty-to-student ratio was approximately 1:180, and the TA-to-student ratio was about 1:20.
% The assessment structure included an initial assignment, Labs 1–5, a pre-midterm quiz, and a midterm exam featuring multiple-choice, fill-in-the-blank, and justification questions. This was followed by a second assignment, Labs 6–10, and a lab \textcolor{purple}{endterm} exam comprising on-computer programming tasks and a viva, conducted prior to the \textcolor{purple}{endterm} exam \cite{10.1145/3209635.3209639}. A group project, assigned early in the term, was evaluated before the \textcolor{purple}{endterm} exam. The \textcolor{purple}{endterm} exam carried the highest weight in the overall grading scheme. Laboratory sessions were conducted in parallel with lectures, and assignments were graded out of 10. To fulfill pair programming requirements and promote collaborative learning, students formed self-selected groups of three \cite{4271597}. TAs were not officially responsible for final grading, \textcolor{green}{but assessed students'} work, and provided feedback to the course coordinator. Each TA conducted weekly tutorials to reinforce lecture material and evaluated lab assignments during three-hour sessions, as well as course assignments and quizzes. This structure was designed to create a supportive learning environment and strengthen students’ programming foundations \cite{10.1145/3488042.3488052}.
%%%%%%%%%%%%%%%%%%%%%%%%%%
% Assignments were graded out of 10, and students formed self-selected groups of three to five for collaborative activities such as pair programming \cite{4271597}. While final grading was the responsibility of course instructors, TAs evaluated lab work, assignments, and quizzes, and conducted weekly tutorials to reinforce lecture content \cite{10.1145/3488042.3488052}. This structure was designed to build foundational programming skills within a supportive and collaborative learning environment.
% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% % Divyansh
% All first-year B.Tech students were enrolled in the mandatory CS course and its associated lab, with all eligible for data collection. Out of 516 enrolled students, 318 consistently participated in the weekly course feedback. Table~\ref{tab:metadata} presents an overview of these active participants, detailing their distribution by branch, gender, grades, and prior experience. \textbf{THIS PART TO BE CHECKED}
% Although the course does not require prior programming experience, basic familiarity with smartphones and computers is expected. Three faculty members coordinated and taught the 12-week course, with the lead coordinator overseeing its structure, which followed a traditional C programming curriculum. Topics included data types, basic data structures, control flow (conditions and loops), functions, classes, memory storage, and methods. The faculty-to-student ratio was approximately 1:180, while the TA-to-student ratio was around 1:20.
% The course sequence included an initial assignment, Labs 1–5, a pre-midterm quiz, and a midterm exam comprising multiple-choice, fill-in-the-blank, and justification questions. This was followed by a second assignment, Labs 6–10, and a lab end-semester exam with on-computer programming tasks and a viva, conducted before the final exam \cite{10.1145/3209635.3209639}. A group project, assigned early in the term, culminated in a presentation. The final exam carried the highest weight. Labs ran parallel to lectures, and assignments were graded out of 10. Students worked in self-selected groups of three to meet pair programming requirements and foster collaborative learning \cite{4271597}. A minimum of 30\% was required to pass both the course and lab components.
% Teaching Assistants (TAs), though not officially responsible for grading, assessed student work and shared feedback with the course coordinator. Each TA led weekly tutorials to reinforce lecture content and evaluated lab assignments during three-hour sessions, as well as course assignments and quizzes. This structure aimed to foster a supportive environment and strengthen students' programming foundations \cite{10.1145/3488042.3488052}.
% ''''''''ADD EXAMPLE OF SOMETHING RELATED TO THE PROJECT, LAB ASSIGNMENT, COURSE ASSIGNMENT, QUIZ TYPE, ENDSEM EXAM PATTERN, TUTORIAL, ETC'''''''
% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% % Divyansh
% A Google Form survey was emailed to all enrolled students at the end of each week for 12 weeks, following the final course activity. Participation was encouraged in class and through reminder emails. The survey, conducted in English, included three optional background questions on prior experience with C programming, any programming language, smartphones, and computers—rated from "Nil" to "Lot." Students also provided optional details such as gender (female/male), branch (PH, CSE, MC, ME, EEE, CE, CB, AI, MM), hometown, name, and roll number. All participation was voluntary, and students could choose which questions to answer.
% Survey participation varied weekly. While 318 students consistently responded during weeks 0, 3, 6–12, responses dropped below 300 in weeks 1, 2, and 5. No week had full participation from all 516 students, and no incentives were offered. Students were informed of the study's purpose, and consent was obtained. As no sensitive data were collected and no interventions introduced, ethical approval was not required per national and institutional guidelines.The survey included 25 statements (S1–S25), based on previously identified themes (Table \ref{tab:themes_statements}. Students rated each statement on a 7-point Likert scale: 1 (strongly disagree) to 7 (strongly agree), with 4 as neutral.
% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%
% % Divyansh
% To address the research questions, both descriptive statistics and non-parametric statistical tests were applied. Students were categorized into subgroups based on factors such as gender, grade, branch, and self-reported prior experience in C programming, computer usage, smartphone usage, and any programming experience. Hypotheses tested assumed no significant difference between subgroups.
% To compare responses by gender, Mann–Whitney U tests were conducted. This non-parametric test is effective for comparing distributions between two independent groups based on ranks, making it well-suited for ordinal data . To explore differences across multiple groups (such as branch, gender, and prior experience), Kruskal–Wallis tests were utilized, which is a non-parametric test used to evaluate distributions across more than two independent groups.
% While grade was generally treated as ordinal data, it was analyzed here as continuous data for GPA calculations. All statistical analyses were performed using Python libraries with significant differences at 95\% confidence level across subgroups based on branch (9 groups), gender (2 groups), grade (5 groups), and prior experience (3 groups), yielding a comprehensive understanding of patterns and variations within the dataset.
% %%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Version -5
\begin{table}[]
\centering
% \caption{Academic and background information of survey participants}
\caption{Basic information of survey participants}
\begin{adjustbox}{width=1\linewidth}
\begin{tabular}{c|c|cc|cc|cc|cc|cc|cc|cc|cc|cc|ccc}
\hline
\multicolumn{2}{c|}{\textbf{Branch (B)}} & \multicolumn{2}{c|}{\textbf{AI}} & \multicolumn{2}{c|}{\textbf{CB}} & \multicolumn{2}{c|}{\textbf{CE}} & \multicolumn{2}{c|}{\textbf{CSE}} & \multicolumn{2}{c|}{\textbf{EE}} & \multicolumn{2}{c|}{\textbf{MC}} & \multicolumn{2}{c|}{\textbf{ME}} & \multicolumn{2}{c|}{\textbf{MM}} & \multicolumn{2}{c|}{\textbf{PH}} & \multicolumn{3}{c}{\textbf{Total}} \\ \hline
\multicolumn{2}{c|}{\textbf{Gender (G)}} & \textbf{M} & \textbf{F} & \textbf{M} & \textbf{F} & \textbf{M} & \textbf{F} & \textbf{M} & \textbf{F} & \textbf{M} & \textbf{F} & \textbf{M} & \textbf{F} & \textbf{M} & \textbf{F} & \textbf{M} & \textbf{F} & \textbf{M} & \textbf{F} & \textbf{M} & \textbf{F} & \textbf{M + F} \\ \hline %\hline
\multicolumn{2}{c|}{\textbf{Student count}} & 28 & 6 & 30 & 10 & 25 & 11 & 40 & 12 & 35 & 7 & 19 & 5 & 39 & 10 & 19 & 3 & 16 & 3 & 251 & 67 & 318 \\ \hline %\hline
\textbf{} & Nil & 19 & 5 & 25 & 7 & 22 & 11 & 31 & 9 & 28 & 7 & 16 & 5 & 32 & 6 & 14 & 3 & 9 & 3 & 196 & 56 & 252 \\ %\cline{2-23}
\textbf{PCE} & Medium & 9 & 1 & 5 & 3 & 2 & 0 & 9 & 3 & 7 & 0 & 3 & 0 & 7 & 3 & 5 & 0 & 6 & 0 & 53 & 10 & 63 \\ %\cline{2-23}
\textbf{} & High & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 1 & 0 & 2 & 1 & 3 \\ \hline %\hline
\textbf{} & Nil & 9 & 4 & 22 & 3 & 13 & 8 & 12 & 5 & 15 & 5 & 9 & 2 & 13 & 1 & 6 & 2 & 7 & 3 & 106 & 33 & 139 \\ %\cline{2-23}
\textbf{PAE} & Medium & 19 & 2 & 8 & 6 & 11 & 3 & 21 & 6 & 20 & 2 & 9 & 3 & 23 & 9 & 13 & 1 & 8 & 0 & 132 & 32 & 164 \\ %\cline{2-23}
\textbf{} & High & 0 & 0 & 0 & 1 & 1 & 0 & 7 & 1 & 0 & 0 & 1 & 0 & 3 & 0 & 0 & 0 & 1 & 0 & 13 & 2 & 15 \\ \hline %\hline
\textbf{} & Nil & 2 & 3 & 6 & 1 & 3 & 1 & 2 & 1 & 4 & 1 & 3 & 0 & 4 & 0 & 1 & 0 & 0 & 0 & 25 & 7 & 32 \\ %\cline{2-23}
\textbf{PCoE} & Medium & 14 & 2 & 10 & 6 & 16 & 10 & 25 & 8 & 20 & 4 & 11 & 5 & 25 & 5 & 10 & 3 & 8 & 2 & 139 & 45 & 184 \\ %\cline{2-23}
\textbf{} & High & 12 & 1 & 14 & 3 & 6 & 0 & 13 & 3 & 11 & 2 & 5 & 0 & 10 & 5 & 8 & 0 & 8 & 1 & 87 & 15 & 102 \\ \hline %\hline
\textbf{} & Nil & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 3 & 0 & 0 & 0 & 0 & 0 & 4 & 0 & 4 \\ %\cline{2-23}
\textbf{PSE} & Medium & 12 & 1 & 5 & 6 & 9 & 8 & 13 & 4 & 12 & 5 & 4 & 2 & 15 & 3 & 1 & 1 & 0 & 1 & 71 & 31 & 102 \\ %\cline{2-23}
\textbf{} & High & 16 & 5 & 25 & 4 & 15 & 3 & 27 & 8 & 23 & 2 & 15 & 3 & 21 & 7 & 18 & 2 & 16 & 2 & 176 & 36 & 212 \\ \hline %\hline
& AA & 9 & 2 & 1 & 2 & 0 & 1 & 17 & 2 & 9 & 0 & 9 & 2 & 11 & 2 & 6 & 0 & 3 & 0 & 65 & 11 & 76 \\ %\cline{2-23}
& AB & 7 & 1 & 3 & 0 & 6 & 2 & 11 & 4 & 10 & 3 & 8 & 0 & 15 & 3 & 3 & 0 & 3 & 0 & 66 & 13 & 79 \\ %\cline{2-23}
\textbf{Gd} & BB & 8 & 2 & 10 & 3 & 12 & 4 & 9 & 4 & 13 & 2 & 1 & 3 & 10 & 4 & 5 & 2 & 6 & 1 & 74 & 25 & 99 \\ %\cline{2-23}
& BC & 0 & 1 & 12 & 4 & 4 & 3 & 1 & 0 & 2 & 0 & 1 & 0 & 1 & 1 & 4 & 0 & 4 & 0 & 29 & 9 & 38 \\ %\cline{2-23}
& CC & 3 & 0 & 3 & 1 & 2 & 1 & 2 & 2 & 1 & 0 & 0 & 0 & 2 & 0 & 1 & 0 & 0 & 2 & 14 & 6 & 20 \\ %\cline{2-23}
& CD & 1 & 0 & 1 & 0 & 1 & 0 & 0 & 0 & 0 & 2 & 0 & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 3 & 3 & 6 \\ \hline
\multicolumn{23}{r}{\textbf{PCE}: Prior C programming experience,
\textbf{PAE}: Prior any programming experience,
\textbf{PCoE}: Prior computer usage experience,} \\
\multicolumn{23}{r}{\textbf{PSE}: Prior smartphone usage experience,
% \textbf{Gd}: Grade,
\textbf{AI}: Artificial Intelligence,
\textbf{CB}: Chemical Engineering,
\textbf{CE}: Civil Engineering,} \\
\multicolumn{23}{r}{\textbf{CSE}: Computer Science \& Engineering,
\textbf{EE}: Electrical \& Electronics Engineering,
\textbf{MC}: Mathematics \& Computing, } \\
\multicolumn{23}{r}{\textbf{ME}: Mechanical Engineering,
\textbf{MM}: Metallurgy Engineering,
\textbf{PH}: Engineering Physics,
\textbf{Gd}: Grade,
\textbf{M}: Male,
\textbf{F}: Female}
\end{tabular}
\end{adjustbox}
\label{tab:metadata_analysis}
\end{table}
\input{tables/tab_theme_statements}
% \begin{figure}[!t]
% \centering
% \includegraphics[width=0.45\textwidth]{figs/likert_plot.pdf}
% \caption{Distribution of Likert-scale responses for all 25 survey statements, based on feedback from all 318 students}
% \label{fig:likert_plot_old}
% \end{figure}
\input{tables/likert_chart}
\section{Results} \label{4sec:result}
% \noindent
This section presents results derived from the survey data and their subsequent analysis. Prior experience variables were encoded on an ordinal scale: \emph{nil} = 0, \emph{medium} = 1, and \emph{high} = 2. Academic grades were mapped to numerical values as follows: AA = 10, AB = 9, BB = 8, BC = 7, CC = 6, and CD = 5, with AA representing the highest possible grade on a 10-point scale.
While the survey spanned nine weeks, the main paper focuses on Week-9 responses, which offer the most comprehensive feedback following the completion of course activities. Results from earlier weeks are included in the supplementary material.
\subsection{Student Perceptions of Assessment}
The survey findings related to lab assignments (S1–S5), exams (S6–S9), and projects (S10–S13) are presented in the top half of Fig.~\ref{fig:likert_plot}. While many students selected neutral responses across these items, a substantial number perceived lab assignments as meaningful learning activities rather than obligatory tasks. These assignments were generally not seen as stressful, with several students citing peer collaboration as a key factor in managing deadlines.
Student responses regarding awareness of exam content and question types were varied, with no clear consensus. However, many acknowledged that the skills and knowledge assessed in the exams were both important and beneficial, and viewed the exams as authentic assessments.
Regarding projects, a significant proportion of students remained neutral on whether there was a notable increase in difficulty compared to lab assignments. Nevertheless, some students noted variations in the workload required to attain similar grades across different project topics. Despite this, the majority found the projects to be authentic learning experiences and reported substantial knowledge gains from completing them.
\subsection{Student Perceptions of Course Coordinators and TAs}
% From the lower part of Fig.~\ref{fig:likert_plot}, it can be observed that
The lower part of Fig.~\ref{fig:likert_plot} shows that for statements S14 and S15, which relate to the course coordinator's effectiveness, students generally acknowledged the coordinator’s active engagement with various course components beyond just lectures. Overall, students perceived the coordinator as well-informed and interested in learning activities.
Regarding TAs (statements S16–S25 in Fig.~\ref{fig:likert_plot}), student responses showed considerable variation in perceptions of TA involvement in assessing lab assignments and projects. Many students reported being treated professionally by TAs, appreciated the feedback and guidance they received, and found it easier to seek assistance from TAs than from course coordinators.
However, most students remained neutral regarding inconsistencies in how different TAs assessed lab and project work, the level of help provided, and the content delivered during tutorials. Interestingly, and contrary to conventional expectations, a notable proportion of students preferred consulting TAs over searching for solutions online. While opinions on TA availability varied, a significant number of students indicated that adequate teaching support was available during course activities.
\input{tables/p_value_meta_data}
\subsection{Differences}
Table~\ref{tab:meta_data_pvalues} summarizes the results of statistical tests examining differences across various subgroups, including
gender (G), academic performance grade (Gd), branch (B), and
prior experiences in C programming (PCE)/ any programming language (PAE)/ computer usage (PCoE)/ smartphone usage (PSE).
Key observations are discussed below.
\subsubsection{\textbf{Gender (G)}}
In this study, female students were underrepresented compared to their male counterparts (see Table~\ref{tab:metadata_analysis}). As shown in Table~\ref{tab:meta_data_pvalues}, a statistically significant difference was observed in PSE (Mann-Whitney U test, $p$-value = 0.049), with male students reporting slightly higher engagement level. This may reflect a tendency for greater male participation in technology-oriented activities.
Additionally, female students achieved lower average grades than male students (Mann-Whitney U test, $p$-value = 0.017), with a mean cumulative GPA of 8.07 compared to 8.52 for males.
% \subsubsection{\textbf{Prior C programming Experience (PCE)}}
% As detailed in Table \ref{tab:meta_data_pvalues}, significant differences emerged when analyzing students grouped by PCE. The Kruskal-Wallis test revealed substantial disparities in PAE across PCE groups ($p$ < 0.001). Post hoc Dunn-Bonferroni pairwise comparisons demonstrated that students with medium PCE differed significantly from those with no PCE (adjusted $p$ < 0.001). Similar statistically significant variations were observed in PCoE and PSE when stratified by PCE (Kruskal-Wallis test, $p$ < 0.001 for both).
% It is worth noting that PCE exhibited strong correlations with PAE, PCoE, and PSE. These findings support the hypothesis that foundational knowledge in one programming language, such as C, often facilitates easier learning of additional \textcolor{green}{coding} languages, potentially due to the transfer of logical problem-solving frameworks and algorithmic thinking. Furthermore, prior C programming experience correlated with the familiarity \textcolor{green}{with} computers or smartphones. A possible explanation is that individuals inclined toward programming often engage more deeply with digital tools, fostering greater technical adeptness across platforms. This aligns with observed patterns where programming-experienced students demonstrated more exploratory and efficient use of smartphones, computers, and related technologies.
\subsubsection{\textbf{Prior C Programming Experience (PCE)}}
As shown in Table~\ref{tab:meta_data_pvalues}, significant differences were observed among student groups based on PCE. The Kruskal-Wallis test indicated strong disparities in PAE across PCE levels ($p$~<~0.001). Post hoc Dunn–Bonferroni pairwise comparisons further revealed that students with medium PCE significantly differed from those with no PCE (adjusted $p$~<~0.001). Similar significant variations were found in PCoE and PSE, with $p$-values less than 0.001 for both.
These results suggest that PCE is strongly associated with broader technological familiarity. Specifically, students with PCE also tended to report higher levels of programming experience in general (PAE), as well as more frequent and confident use of computers and smartphones (PCoE and PSE).
These findings suggest that early exposure to programming, particularly in C, supports the development of coding skills through enhanced logical thinking. Students with PCE also tend to engage more with digital tools, likely contributing to greater technical proficiency across platforms.
% This supports the hypothesis that early exposure to programming, particularly in foundational languages like C, can facilitate the acquisition of additional coding skills by reinforcing logical reasoning and algorithmic thinking.
% Moreover, the correlation between PCE and digital tool usage may reflect a broader inclination: students interested in programming are more likely to explore and engage with technology, thereby developing greater proficiency across platforms. This pattern aligns with observations that programming-experienced students often demonstrate more exploratory, efficient, and confident use of smartphones, computers, and related technologies.
% \subsubsection{\textbf{Prior Any programming Experience (PAE)}}
% We notice a substantial difference in grades when students were grouped by PAE (Kruskal-Wallis test, $p$-value < 0.001). Students with no PAE received lower grades (mean grade = 7.91) compared to those with moderate PAE (mean grade = 8.79) and high PAE (mean grade = 9.13). This underscores a strong positive association between PAE and grade, suggesting that foundational programming skills enhance learning outcomes in subsequent courses. Additionally, the academic branch varied significantly across PAE groups (Kruskal-Wallis test, $p$-value = 0.015). Post hoc pairwise comparisons with Bonferroni corrections revealed a statistically significant \textcolor{blue}{difference} between students with no PAE and high PAE (adjusted $p$ = 0.024). As illustrated in Table \ref{tab:metadata_analysis}, high PAE students were predominantly enrolled in the Computer Science and Engineering (CSE) branch, aligning with the expectations of self-selection into technical disciplines. These findings highlight the critical role of prior programming exposure in shaping both academic performance and disciplinary trajectories \textcolor{green}{among engineering students}.
\subsubsection{\textbf{Prior Any programming Experience (PAE)}}
A substantial difference in academic performance was observed when students were grouped by PAE, as indicated by the Kruskal-Wallis test ($p$~<~0.001). Students with no PAE had lower average grades (mean = 7.91) compared to those with moderate (mean = 8.79) and high (mean = 9.13) PAE. This suggests a strong positive association between PAE and academic achievement (grade), indicating that foundational programming skills contribute to improved learning outcomes in subsequent courses.
In addition, academic branch distribution varied significantly across PAE groups (Kruskal-Wallis test, $p$ = 0.015). Post hoc Dunn-Bonferroni comparisons revealed a statistically significant difference between students with no PAE and those with high PAE (adjusted $p$ = 0.024).
% As shown in Table~\ref{tab:metadata_analysis}, students with high PAE were predominantly from the Computer Science and Engineering (CSE) branch, reflecting a likely self-selection into technically focused disciplines.
%
These findings underscore the importance of prior programming experience in shaping both academic success and disciplinary pathways among engineering students.
% \subsubsection{\textbf{Prior Computer usage Experience (PCoE)}}
% Similar to PAE, the Kruskal-Wallis test with $p$-value \textcolor{green}{of} 0.051 was observed in the grade when grouped by PCoE. There is a notable difference between no PCoE (mean grade = 7.91) and high PCoE (mean grade = 8.56, adjusted $p$-value = 0.047) and moderate PCoE (mean grade = 8.44). \textcolor{green}{These findings indicate a clear, positive relationship between prior computer familiarity and academic success: the more experience students have with computers, the better their academic performance \textcolor{blue}{in the study course} tends to be.}
\subsubsection{\textbf{Prior Computer usage Experience (PCoE)}}
A near-significant difference in grades was observed across PCoE groups (Kruskal–Wallis test, $p$ = 0.051). Students with no prior computer experience had a lower average grade (mean = 7.91) compared to those with moderate (mean = 8.44) and high (mean = 8.56) PCoE. Post hoc analysis showed a statistically significant difference between students with no and high PCoE (adjusted $p$ = 0.047).
These findings suggest a positive association between computer familiarity and academic performance; students with greater experience using computers tend to perform better in the course.
% \subsubsection{\textbf{Prior Smartphone usage Experience (PSE)}}
% We grouped students by PSE, resulting in $p$-value \textcolor{green}{of} 0.014 \textcolor{green}{using} the Kruskal-Wallis test \textcolor{green}{for} gender. Notably, students with no PSE were exclusively male. However, there is a considerable difference between moderate and high PSE (adjusted $p$-value = 0.019). As shown in Table \ref{tab:metadata_analysis}, 70\% of male students reported high PSE, compared to 53\% of female students. This suggests a potential gender-related pattern in \textcolor{blue}{reported} smartphone usage, differences in technology adoption trends between genders, possibly influenced by societal or personal preferences.
\subsubsection{\textbf{Prior Smartphone usage Experience (PSE)}}
Grouping students by PSE revealed a significant gender-related difference (Kruskal-Wallis test, $p$ = 0.014). Notably, all students with no reported PSE were male. A statistically significant difference was also observed between moderate and high PSE levels (adjusted $p$ = 0.019). As shown in Table~\ref{tab:metadata_analysis}, 70.11\% of male students reported high smartphone usage, compared to 53.73\% of female students.
This suggests a potential gender-related pattern in reported smartphone usage, possibly reflecting differences in technology adoption trends influenced by societal norms or personal preferences.
% \subsubsection{\textbf{Grade (Gd)}}
% We observe a significant difference in PAE while grouping with grade (Kruskal-Wallies test, $p$-value < 0.001), having differences among multiple grade pairs: AA - CD (adjusted $p$-value = 0.004), AA - CC (adjusted $p$-value = 0.001), BC - AA (adjusted $p$-value < 0.001), BB - AA (adjusted $p$-value < 0.001), AB - CD (adjusted $p$-value = 0.024), AB - CC (adjusted $p$-value = 0.017), BC - AB (adjusted $p$-value = 0.026). Students with higher grades, such as AA (mean PAE = 0.87) and AB (mean PAE = 0.75), tend to have more PAE. In contrast to students with lower grades, particularly CD (mean PAE = 0), CC (mean PAE = 0.3), and BC (mean PAE = 0.39) have less PAE. This suggests that foundational experience in programming may enhance performance in subsequent technical coursework.
\subsubsection{\textbf{Grade (Gd)}}
A significant difference in PAE was observed when students were grouped by academic grade (Kruskal-Wallis test, $p$~<~0.001), with multiple grade pairs showing statistically significant differences:
AA–CD (adjusted $p$ = 0.004),
AA–CC (adjusted $p$ = 0.001),
BC–AA (adjusted $p$ < 0.001),
BB–AA (adjusted $p$ < 0.001),
AB–CD (adjusted $p$ = 0.024),
AB–CC (adjusted $p$ = 0.017), and
BC–AB (adjusted $p$ = 0.026).
Students with higher grades, AA (mean PAE = 0.87) and AB (mean PAE = 0.75), tended to report more PAE. In contrast, students with lower grades, particularly those receiving CD (mean PAE = 0), CC (mean PAE = 0.3), and BC (mean PAE = 0.39), had less PAE.
This suggests that foundational programming exposure may contribute to improved performance in subsequent technical coursework.
% \subsubsection{\textbf{Branch (B)}}
% As detailed in Table \ref{tab:metadata_analysis}, CSE students have the highest PAE (mean = 0.83), while CB students exhibited the lowest (mean PAE = 0.4). CE students demonstrated the least PSE with a mean of 1.47. Statistically \textcolor{green}{significant} differences in academic grade were observed across branches (Kruskal-Wallis test, $p$-value < 0.001), post hoc pairwise comparisons revealed notable differences: CB \textcolor{green}{with} CSE (adjusted $p$-value < 0.001), CE \textcolor{green}{with} CSE (adjusted $p$-value < 0.001), CB \textcolor{green}{with} MC (adjusted $p$-value < 0.001), CE \textcolor{green}{with} MC (adjusted $p$-value < 0.001). Students in CB (mean grade < 8) and CE (mean grade < 8) underperformed relative to their peers in MC (mean grade = 9.20) and CSE (mean grade = 8.84). Other branches achieved intermediate results, with average grades ranging from 8.0 to 8.8.
\subsubsection{\textbf{Branch (B)}}
CSE students reported the highest prior programming experience (PAE) with a mean of 0.83, while the CB students had the lowest (mean PAE = 0.40), as comprehended from Table~\ref{tab:metadata_analysis}. CE students reported the lowest PSE, with a mean of 1.47.
Statistically significant differences in academic grades were observed across branches (Kruskal-Wallis test, $p$ < 0.001). Post hoc pairwise comparisons revealed notable differences:
CB vs. CSE (adjusted $p$ < 0.001),
CE vs. CSE (adjusted $p$ < 0.001),
CB vs. MC (adjusted $p$ < 0.001), and
CE vs. MC (adjusted $p$ < 0.001).
Students in the CB and CE branches had lower average grades (both < 8.0), whereas students in MC (mean = 9.20) and CSE (mean = 8.84) performed better. Other branches fell in between, with average grades ranging from 8.0 to 8.8.
These results suggest that academic performance and prior technical experience vary notably by discipline, with students from computing-focused branches (CSE, MC) generally exhibiting stronger preparation and outcomes.
% Table \ref{tab:pvalues_last_week_1} provides an in-depth analysis of Week-9 responses. Analyses of other weeks are presented in Appendix \ref{App:weeks_analysis}.
Table~\ref{tab:pvalues_last_week_1} provides a detailed Week-9 analysis; while results from initial eight weeks are available in
% Appendix A.
\ref{App:weeks_analysis}.
% in preamble
\input{tables/p_value_last_week}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Version 3
\begin{table}[!hbt]
\centering
% \caption{Themes identified in the qualitative step \cite{qualitative_step} and mapping from themes to statements in the quantitative step}
\caption{Survey statements}
\scriptsize % Reduces font size to fit content
\begin{adjustbox}{width=0.48\textwidth} % Adjust to half page width
\begin{tabular}{c |c p{7cm}} % Adjust column width for statements
\hline
\textbf{Theme} & \textbf{SN.} & \textbf{Statement} \\
\hline
\multirow{8}{*}{\rotatebox{90}{\shortstack{Lab\\assignments}}}
& S1: & I am experiencing the lab assignments foremost as an activity for me to learn the course content. \\
& S2: & I am experiencing the lab assignments as a necessary evil. \\
& S3: & I found it stressful to complete the lab assignments to the deadline. \\
& S4: & I have gotten a lot of help from my classmates when I solved the lab assignments. \\
& S5: & I have often helped other classmates when they solved the lab assignments. \\
\hline
\multirow{8}{*}{\rotatebox{90}{Examination}}
& S6: & I experienced that I knew in advance what types of questions would be on the exam. \\
& S7: & I experienced that I knew in advance what would be assessed on the written exam. \\
& S8: & I experienced that the type of knowledge and skills that were assessed on the written exam, are important to know/master when programming. \\
& S9: & I experienced the written exam as authentic, meaning that it consisted of questions I could encounter when developing programs. \\
\hline
\multirow{8}{*}{\rotatebox{90}{Project}}
& S10: & I experienced that there was a big leap in difficulty between the lab assignments and the project. \\
& S11: & I experienced that I learnt a lot from doing the project assignment. \\
& S12: & I experienced the project assignments as authentic, meaning that they consisted of questions I could encounter when developing programs. \\
& S13: & I experienced it to be a huge difference between different project assignments when it comes to the amount of work that was required to achieve the same grade. \\
\hline
\multirow{6}{*}{\rotatebox{90}{\shortstack{Course\\Coordinator}}}
& S14: & I experienced that the course coordinator has been able to put their own touch on the course, control the course structure and learning activities. \\
& S15: & I experienced that the course coordinator has had too little insight into, and shown too little interest in other course elements than lectures, such as tutorials and assessment of lab assignments. \\
\hline
\multirow{19}{*}{\rotatebox{90}{Teaching Assistants (TAs)}}
& S16: & I experienced that assessment of the lab assignments and projects is mainly done by the TAs. \\
& S17: & I experienced that there was a huge difference between how different TAs assessed the lab assignments during this course. \\
& S18: & I experienced that there was a huge difference between how different TAs assessed the project assignments during this course. \\
& S19: & I experienced that there was a huge difference between how much help you received from different TAs. \\
& S20: & I experienced that the structure and content of the tutorials differed depending on which TA you were assigned/chose to go to. \\
& S21: & I felt that the TAs in this course treated me professionally. \\
& S22: & I experienced that I got useful feedback and guidance from the TAs during the course. \\
& S23: & I felt it was easier to ask a TA for help, rather than the course coordinator. \\
& S24: & I preferred to search for answers on the internet, rather than asking a TA for help. \\
& S25: & I experienced that there often were too few TAs during the lab sessions, which made it difficult to get help or present my solution on time. \\
\hline
% \multicolumn{3}{r}{\textbf{SN:} Statement No}
\end{tabular}
\end{adjustbox}
\label{tab:themes_statements}
\end{table}
% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% % Version 2
% \begin{table}[!hbt]
% \centering
% \caption{Themes identified in the qualitative step \cite{qualitative_step} and mapping from themes to statements in the quantitative step}
% \scriptsize % Reduces font size to fit content
% \begin{adjustbox}{width=0.48\textwidth} % Adjust to half page width
% \begin{tabular}{l|p{6.2cm}} % Adjust column width for statements
% \hline
% \textbf{Regarding} & \textbf{Statement No: Full Statement} \\
% \hline
% \multirow{9}{*}{Lab Assignments}
% & S1: I am experiencing the lab assignments foremost as an activity for me to learn the course content. \\
% & S2: I am experiencing the lab assignments as a necessary evil. \\
% & S3: I found it stressful to complete the lab assignments to the deadline. \\
% & S4: I have gotten a lot of help from my classmates when I solved the lab assignments. \\
% & S5: I have often helped other classmates when they solved the lab assignments. \\
% \hline
% \multirow{4}{*}{About Exams}
% & S6: I experienced that I knew in advance what types of questions would be on the exam. \\
% & S7: I experienced that I knew in advance what would be assessed on the written exam. \\
% & S8: I experienced that the type of knowledge and skills that were assessed on the written exam, is important to know/master when programming. \\
% & S9: I experienced the written exam as authentic, meaning that it consisted of questions I could encounter when developing programs. \\
% \hline
% \multirow{4}{*}{Course Project}
% & S10: I experienced that there was a big leap in difficulty between the lab assignments and the project. \\
% & S11: I experienced that I learnt a lot from doing the project assignment. \\
% & S12: I experienced the project assignments as authentic, meaning that it consisted of questions I could encounter when developing programs. \\
% & S13: I experienced it to be a huge difference between different project assignments when it comes to the amount of work that was required to achieve the same grade. \\
% \hline
% \multirow{2}{*}{Course Coordinator}
% & S14: I experienced that the course coordinator has been able to put their own touch on the course, control the course structure and learning activities. \\
% & S15: I experienced that the course coordinator has had too little insight into, and shown too little interest in other course elements than lectures, such as tutorials and assessment of lab assignments. \\
% \hline
% \multirow{10}{*}{Teaching Assistants (TAs)}
% & S16: I experienced that assessment of the lab assignments and projects are mainly done by the teaching assistants. \\
% & S17: I experienced that there was huge difference between how different TAs assessed the lab assignments during this course. \\
% & S18: I experienced that there was huge difference between how different TAs assessed the project assignments during this course. \\
% & S19: I experienced that there was huge difference between how much help you received from different TAs. \\
% & S20: I experienced that the structure and content of the tutorials differed depending on which TA you were assigned/chose to go to. \\
% & S21: I felt that the TAs in this course treated me professionally. \\
% & S22: I experienced that I got useful feedback and guidance by the TAs during the course. \\
% & S23: I felt it was easier to ask a TA for help, rather than the course coordinator. \\
% & S24: I preferred to search for answers on the internet, rather than asking a TA for help. \\
% & S25: I experienced that it often were too few TAs during the lab sessions, which made it difficult to get help or present my solution on time. \\
% \hline
% \end{tabular}
% \end{adjustbox}
% \label{tab:themes_statements}
% \end{table}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Version 1
% \begin{table}[h]
% \centering
% \caption{Themes identified in the qualitative step \cite{qualitative_step} and mapping from themes to statements in the quantitative step}%Themes and Statements}
% \begin{adjustbox}{width=0.40\textwidth} % Adjusts table to fit page width
% \begin{tabular}{l|l} % Slightly smaller column widths
% \hline
% \textbf{Regarding} & \textbf{Statement No: Theme} \\
% \hline
% \multirow{5}{*}{Lab Assignments}
% & S1: Learning Activity \\
% % \cline{2-2}
% & S2: Necessary Evil \\
% % \cline{2-2}
% & S3: Stressful \\
% % \cline{2-2}
% & S4: Took help from peers \\
% % \cline{2-2}
% & S5: Helped peers \\
% \hline
% \multirow{4}{*}{About Exams}
% & S6: Predictable exam questions \\
% % \cline{2-2}
% & S7: Predictable assessment \\
% % \cline{2-2}
% & S8: Important skills and knowledge \\
% % \cline{2-2}
% & S9: Authentic \\
% \hline
% \multirow{4}{*}{Course Project}
% & S10: A big leap from it \\
% % \cline{2-2}
% & S11: Learnt a lot \\
% % \cline{2-2}
% & S12: Fair and authentic \\
% % \cline{2-2}
% & S13: Difference between projects \\
% \hline
% \multirow{2}{*}{Course Coordinator}
% & S14: On course structure \\
% % \cline{2-2}
% & S15: Too little insight \\
% \hline
% \multirow{10}{*}{Teaching Assistants (TAs)}
% & S16: Helping hand during labs \\
% % \cline{2-2}
% & S17: Variation in lab assessment \\
% % \cline{2-2}
% & S18: Variation in project assessment \\
% % \cline{2-2}
% & S19: Seeking help \\
% % \cline{2-2}
% & S20: Variation in teaching \\
% % \cline{2-2}
% & S21: Professionalism \\
% % \cline{2-2}
% & S22: Useful feedback \\
% % \cline{2-2}
% & S23: Vital presence \\
% % \cline{2-2}
% & S24: Prefer internet for help \\
% % \cline{2-2}
% & S25: Too few during labs \\
% \hline
% \end{tabular}
% \end{adjustbox}
% \label{tab:themes_statements}
% \end{table}
\begin{figure}
\centering
\begin{tikzpicture}
\begin{axis}[
xbar stacked,
bar width=6pt,
width=0.48\textwidth,
height=12cm,
enlargelimits=0.05,
xmin=-100, xmax=100,
xmajorgrids = true, % ← Enables vertical grid lines
grid style = {dashed, black!20},
xtick={-100, -75, -50, -25, 0, 25, 50, 75, 100},
xlabel={Likert scale response (\%)},
symbolic y coords={S1,S2,S3,S4,S5,S6,S7,S8,S9,S10,S11,S12,S13,S14,S15,S16,S17,S18,S19,S20,S21,S22,S23,S24,S25},
ytick=data,
y dir=reverse,
xticklabel style={/pgf/number format/fixed},
tick label style={font=\footnotesize},
label style={font=\footnotesize},
legend style={at={(0.5,-0.1)}, anchor=north,legend columns=3, draw=none}
]
% Disagree (1-3 + half of 4)
\addplot+[xbar, fill=gray!60, draw=black!100] coordinates {
(-20.91,S1) (-20.12,S2) (-21.22,S3) (-21.22,S4) (-22.01,S5)
(-23.11,S6) (-24.21,S7) (-22.48,S8) (-21.54,S9) (-26.10,S10)
(-16.66,S11) (-20.75,S12) (-25.94,S13) (-23.89,S14) (-21.22,S15)
(-25,S16) (-24.52,S17) (-28.30,S18) (-25,S19) (-26.10,S20)
(-22.95,S21) (-21.06,S22) (-24.68,S23) (-24.68,S24) (-25.94,S25)
};
\addplot+[xbar, fill=red!20, draw=black!100] coordinates {
(-4.40,S1) (-12.57,S2) (-12.89,S3) (-5.03,S4) (-7.86,S5)
(-11.32,S6) (-11.32,S7) (-4.40,S8) (-5.97,S9) (-10.37,S10)
(-4.40,S11) (-4.40,S12) (-11.32,S13) (-2.51,S14) (-13.83,S15)
(-7.86,S16) (-11.01,S17) (-11.01,S18) (-12.26,S19) (-9.11,S20)
(-8.49,S21) (-5.66,S22) (-8.17,S23) (-10.06,S24) (-11.32,S25)
};
\addplot+[xbar, fill=red!50, draw=black!100] coordinates {
(-2.83,S1) (-13.52,S2) (-15.09,S3) (-6.60,S4) (-3.14,S5)
(-8.80,S6) (-8.80,S7) (-2.20,S8) (-2.20,S9) (-10.37,S10)
(-0.94,S11) (-2.20,S12) (-6.60,S13) (-1.25,S14) (-15.09,S15)
(-6.28,S16) (-4.71,S17) (-4.08,S18) (-7.54,S19) (-7.86,S20)
(-3.14,S21) (-2.83,S22) (-2.51,S23) (-9.11,S24) (-10.69,S25)
};
\addplot+[xbar, fill=darkred, draw=black!100] coordinates {
(-0.62,S1) (-13.20,S2) (-11.01,S3) (-5.03,S4) (-1.25,S5)
(-4.08,S6) (-3.77,S7) (-0.62,S8) (-1.25,S9) (-4.71,S10)
(-0.31,S11) (-0.31,S12) (-4.08,S13) (-0.62,S14) (-11.94,S15)
(-4.71,S16) (-5.34,S17) (-5.66,S18) (-4.08,S19) (-5.34,S20)
(-2.20,S21) (-1.25,S22) (-2.20,S23) (-8.49,S24) (-8.80,S25)
};
% Agree (5-7 + half of 4)
\addplot+[xbar, fill=gray!60, draw=black!100, forget plot] coordinates {
(20.91,S1) (20.12,S2) (21.22,S3) (21.22,S4) (22.01,S5)
(23.11,S6) (24.21,S7) (22.48,S8) (21.54,S9) (26.10,S10)
(16.66,S11) (20.75,S12) (25.94,S13) (23.89,S14) (21.22,S15)
(25,S16) (24.52,S17) (28.30,S18) (25,S19) (26.10,S20)
(22.95,S21) (21.06,S22) (24.68,S23) (24.68,S24) (25.94,S25)
};
% No new legend entry here — already labeled as 4
\addplot+[xbar, fill=green!20, draw=black!100] coordinates {
(18.55,S1) (13.83,S2) (13.52,S3) (23.89,S4) (23.58,S5)
(21.06,S6) (17.61,S7) (20.75,S8) (21.06,S9) (16.35,S10)
(28.61,S11) (22.64,S12) (16.98,S13) (24.21,S14) (13.52,S15)
(14.15,S16) (17.92,S17) (12.26,S18) (15.40,S19) (15.09,S20)
(19.18,S21) (21.06,S22) (19.18,S23) (13.20,S24) (11.63,S25)
};
\addplot+[xbar, fill=green!50, draw=black!100] coordinates {
(15.09,S1) (5.03,S2) (3.45,S3) (11.32,S4) (12.89,S5)
(5.03,S6) (6.60,S7) (17.29,S8) (15.40,S9) (4.08,S10)
(20.44,S11) (18.86,S12) (6.28,S13) (14.77,S14) (2.51,S15)
(11.32,S16) (6.60,S17) (5.66,S18) (7.54,S19) (7.23,S20)
(12.57,S21) (16.03,S22) (12.26,S23) (4.40,S24) (3.45,S25)
};
\addplot+[xbar, fill=darkgreen, draw=black!100] coordinates {
(16.66,S1) (1.57,S2) (1.57,S3) (5.66,S4) (7.23,S5)
(3.45,S6) (3.45,S7) (9.74,S8) (11.01,S9) (1.88,S10)
(11.94,S11) (10.01,S12) (2.83,S13) (8.80,S14) (0.62,S15)
(5.66,S16) (5.34,S17) (4.74,S18) (3.14,S19) (3.14,S20)
(8.49,S21) (11.01,S22) (6.28,S23) (5.34,S24) (2.20,S25)
};
% Manually define the legend with specific colors
\addlegendimage{xbar, fill=gray!60, draw=none}
\addlegendentry{\textcolor{gray!100}{neutral}}
\addlegendimage{xbar, fill=red!20, draw=none}
\addlegendentry{\textcolor{red!80}{weakly disagree}}
\addlegendimage{xbar, fill=red!50, draw=none}
\addlegendentry{\textcolor{red!90}{disagree}}
\addlegendimage{xbar, fill=darkred, draw=none}
\addlegendentry{\textcolor{darkred}{strongly disagree}}
\addlegendimage{xbar, fill=green!20, draw=none}
\addlegendentry{\textcolor{green!90}{weakly agree}}
\addlegendimage{xbar, fill=green!50, draw=none}
\addlegendentry{\textcolor{green!100}{agree}}
\addlegendimage{xbar, fill=darkgreen, draw=none}
\addlegendentry{\textcolor{darkgreen}{strongly agree}}
\end{axis}
\end{tikzpicture}
\caption{Distribution of Likert-scale responses for all 25 survey statements, based on feedback from all 318 students}
\label{fig:likert_plot}
\end{figure}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% VERSION -2
\begin{table}[!ht]
% \caption{Parameters Correlation P-values: RESULTS FROM THE KRUSKAL–WALLIS TESTS COMPARING SUBGROUPS BASED ON Prior Experiences, Grade, Branch RESPECTIVELY, AND THE MANN–WHITNEY U TEST COMPARING SUBGROUPS BASED ON GENDER}
\caption{Subgroup comparison of $p$-values using Kruskal-Wallis test/ Mann-Whitney U test}
\centering
\begin{adjustbox}{width=0.4\textwidth}
\begin{tabular}{l| cccc ccc}
\cline{2-8}
\multicolumn{1}{c}{} & \textbf{G} & \textbf{PCE} & \textbf{PAE} & \textbf{PCoE} & \textbf{PSE} & \textbf{Gd} & \textbf{B} \\
\hline
% \cline{2-8}
\textbf{G} & 0 & 0.506 & 0.32 & 0.153 & \textbf{0.049} & \textbf{0.017} & 0.782 \\
\textbf{PCE} & 0.475 & 0 & \textbf{<0.001} & \textbf{<0.001} & \textbf{<0.001} & 0.448 & 0.217 \\
\textbf{PAE} & 0.504 & \textbf{<0.001} & 0 & \textbf{<0.001} & \textbf{<0.001} & \textbf{<0.001} & \textbf{0.015} \\
\textbf{PCoE} & 0.153 & \textbf{<0.001} & \textbf{<0.001} & 0 & \textbf{<0.001} & \textbf{0.051} & 0.993 \\
\textbf{PSE} & \textbf{0.014} & \textbf{0.002} & \textbf{<0.001} & \textbf{<0.001} & 0 & 0.658 & 0.361 \\
\textbf{Gd} & 0.148 & 0.762 & \textbf{<0.001} & 0.096 & 0.057 & 0 & \textbf{<0.001} \\
\textbf{B} & 0.836 & 0.296 & \textbf{0.01} & 0.398 & \textbf{0.004} & \textbf{0.001} & 0 \\
\hline
\end{tabular}
\end{adjustbox}
\label{tab:meta_data_pvalues}
\end{table}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%0
% VERSION 1
% \begin{table}[h]
% % \caption{Parameters Correlation P-values: RESULTS FROM THE KRUSKAL–WALLIS TESTS COMPARING SUBGROUPS BASED ON Prior Experiences, Grade, Branch RESPECTIVELY, AND THE MANN–WHITNEY U TEST COMPARING SUBGROUPS BASED ON GENDER}
% \caption{Results from the Kruskal-Wallis test and Mann-Whitney U test comparing subgroups among factor parameters.}
% %\caption{Subgroup comparison using Kruskal-Wallis test and Mann-Whitney U test on various factors.}
% \centering
% \begin{adjustbox}{width=0.4\textwidth}
% \begin{tabular}{rlccccccc}
% \toprule
% \textbf{} & & \textbf{G} & \textbf{PCE} & \textbf{PAE} & \textbf{PCoE} & \textbf{PSE} & \textbf{Gd} & \textbf{B} \\
% \hline
% \textbf{G} & & 0 & 0.506 & 0.32 & 0.153 & \textcolor{blue}{\textbf{0.049}} & \textcolor{blue}{\textbf{0.017}} & 0.782 \\
% \textbf{PCE} & & 0.475 & 0 & \textcolor{blue}{\textbf{<0.001}} & \textcolor{blue}{\textbf{<0.001}} & \textcolor{blue}{\textbf{<0.001}} & 0.448 & 0.217 \\
% \textbf{PAE} & & 0.504 & \textcolor{blue}{\textbf{<0.001}} & 0 & \textcolor{blue}{\textbf{<0.001}} & \textcolor{blue}{\textbf{<0.001}} & \textcolor{blue}{\textbf{<0.001}} & \textcolor{blue}{\textbf{0.015***}} \\
% \textbf{PCoE} & & 0.153 & \textcolor{blue}{\textbf{<0.001}} & \textcolor{blue}{\textbf{<0.001}} & 0 & \textcolor{blue}{\textbf{<0.001}} & \textcolor{blue}{\textbf{0.051}} & 0.993 \\
% \textbf{PSE} & & \textcolor{blue}{\textbf{0.014}} & \textcolor{blue}{\textbf{0.002}} & \textcolor{blue}{\textbf{<0.001}} & \textcolor{blue}{\textbf{<0.001}} & 0 & 0.658 & 0.361 \\
% \textbf{Gd} & & 0.148 & 0.762 & \textcolor{blue}{\textbf{<0.001}} & 0.096 & 0.057 & 0 & \textcolor{blue}{\textbf{<0.001***}} \\
% \textbf{B} & & 0.836 & 0.296 & \textcolor{blue}{\textbf{0.01}} & 0.398 & \textcolor{blue}{\textbf{0.004}} & \textcolor{blue}{\textbf{0.001}} & 0 \\
% \bottomrule
% \end{tabular}
% \end{adjustbox}
% \label{tab:meta_data_pvalues}
% \end{table}
\begin{table*}[]
% \scriptsize
% \footnotesize
% \small
% \begin{rotate}{90}
% \captionof{table}{Detailed Week-9 analysis across various subgroups: Mann-Whitney U test for gender (G), and Kruskal-Wallis test based on academic performance grade (Gd), branch (B), and prior experience in C programming (PCE)/ any programming language (PAE)/ computer usage (PCoE)/ smartphone usage (PSE)}
% \end{rotate}
% \caption{\footnotesize Detailed Week-9 analysis across various subgroups: Mann-Whitney U test for gender (G), and Kruskal-Wallis test based on academic performance grade (Gd), branch (B), and prior experience in C programming (PCE)/ any programming language (PAE)/ computer usage (PCoE)/ smartphone usage (PSE)}
\caption{\footnotesize{Week-9 analysis on $p$-values across subgroups: Mann-Whitney U test for gender (G), and Kruskal-Wallis test based on academic grade (Gd), branch (B), and prior experiences in C programming (PCE)/ any programming language (PAE)/ computer usage (PCoE)/ smartphone usage (PSE)}}
\begin{adjustbox}{rotate=90, width=1\textwidth}
\begin{tabular}{c|lp{2.1cm}|ccccccc|p{20cm}}
\hline
\textbf{Theme} & \textbf{SN.} & \textbf{Statement} & \textbf{G} & \textbf{PCE} & \textbf{PAE} & \textbf{PCoE} & \textbf{PSE} & \textbf{Gd} & \textbf{B} & \multicolumn{1}{c}{\textbf{Post-hoc test results explanations}} \\ \hline \hline
\multirow{12}{*}{\rotatebox{270}{\shortstack{Lab\\Assignments}}} & S1: & Learning activity & 0.903 & 0.582 & \textcolor{darkred}{ \textbf{0.018}} & 0.313 & \textcolor{darkred}{\textbf{0.048}} & 0.400 & 0.763 & \textcolor{darkred}{\textbf{PAE}}: Students with \emph{high} PAE disagreed to a greater extent, significantly differing from those with \emph{medium} PAE (adj. $p$ = 0.047).
\textcolor{darkred}{\textbf{PSE}}: Students with \emph{nil} PSE disagreed to a greater extent, significantly differing from those with \emph{high} (adj. $p$ = 0.041) and \emph{medium} PSE (adj. $p$ = 0.041). \\
& S2: & Necessary evil & 0.968 & \textcolor{darkgreen}{\textbf{0.154}} & 0.242 & 0.431 & 0.469 & 0.697 & \textcolor{darkgreen}{\textbf{0.139}} & \textcolor{darkgreen}{\textbf{B}}: CB students (mean = 3.95) and EE students (mean = 3.83) showed the highest level of agreement compared to other branches, while PH students disagreed the most, with a mean of 2.95. \\
& S3: & Stressful & 0.486 & 0.425 & 0.483 & 0.276 & 0.382 & \textcolor{darkgreen}{\textbf{0.107}} & \textcolor{darkgreen}{\textbf{0.144}} & \textcolor{darkgreen}{\textbf{B}}: CB students showed the highest agreement (mean = 4.03), while PH students had the highest level of disagreement (mean = 3.11). \textcolor{darkgreen}{\textbf{Gd}}: Students with a BC grade exhibited the highest agreement (mean = 3.79), whereas those with a CD grade showed the most disagreement (mean = 2.95). \\
& S4: & \shortstack{Help from peers} & 0.469 & 0.400 & 0.791 & 0.320 & 0.308 & 0.234 & \textcolor{blue}{\textbf{0.094}} & \textcolor{blue}{\textbf{B}}: Students from different branches generally agreed, except for those in CSE and CE, who expressed disagreement. \\
& S5: & Help to peers & 0.887 & \textcolor{darkgreen}{\textbf{0.175}} & \textcolor{darkgreen}{\textbf{0.158}} & 0.448 & 0.599 & 0.998 & 0.634 & \textcolor{darkgreen}{\textbf{PCE}}: Students with \emph{nil} (mean = 4.58) and \emph{medium} (mean = 4.40) PCE strongly agreed, whereas those with \emph{high} PCE disagreed the most (mean = 3.67). \textcolor{darkgreen}{\textbf{PAE}}: Students with \emph{high} PAE showed more disagreement (mean = 4.14) than \emph{medium} (mean = 4.59) and \emph{nil} (mean = 4.51) PAE. \\ \hline
\multirow{9}{*}{\rotatebox{270}{\shortstack{Examination}}} & S6: & Predictable question types & 0.497 & \textcolor{darkgreen}{\textbf{0.117}} & 0.237 & \textcolor{blue}{\textbf{0.093}} & \textcolor{darkgreen}{\textbf{0.121}} & 0.591 & 0.418 & \textcolor{blue}{\textbf{PCoE}}: Students with \emph{medium} PCoE (mean = 3.87) disagreed more strongly than those with a \emph{high} (mean = 4.17) or \emph{nil} (mean = 4.25) PCoE. \textcolor{darkgreen}{\textbf{PCE}}: Students with \emph{nil} PCE (mean = 3.94) disagreed more strongly than those with a lot (mean = 4.33) or \emph{medium} (mean = 4.24) PCE. \textcolor{darkgreen}{\textbf{PSE}}: Students with \emph{high} PSE (mean = 4.11) disagreed more strongly than those with \emph{medium} PSE (mean = 3.77). \\
& S7: & Predictable assessment & 0.799 & 0.555 & 0.462 & \textcolor{darkgreen}{\textbf{0.179}} & 0.648 & 0.236 & 0.683 & \textcolor{darkgreen}{\textbf{PCoE}}: Students with \emph{nil} PCoE showed the highest level of agreement (mean = 4.34), followed by those with \emph{high} (mean = 4.14) and \emph{medium} PCoE (mean = 3.88). \\
& S8: & Skills \& knowledge & 0.973 & 0.631 & \textcolor{darkgreen}{\textbf{0.141}} & 0.630 & \textcolor{darkgreen}{\textbf{0.114}} & \textcolor{darkgreen}{\textbf{0.170}} & 0.497 & \textcolor{darkgreen}{\textbf{PSE}}: Students with \emph{nil} PSE tended to report stronger disagreement (mean = 3.75) than those with \emph{medium} (mean = 4.83) and \emph{high} (mean = 4.71) PSE. \textcolor{darkgreen}{\textbf{Gd}}: Students with AA grade showed the strongest agreement (mean = 4.97), followed by those with BC (mean = 4.89), CC (mean = 4.90), CD (mean = 4.83), AB (mean = 4.68), and BB (mean = 4.51). \\
& S9: & Authentic & 0.462 & 0.977 & \textcolor{blue}{\textbf{0.075}} & 0.973 & 0.359 & 0.373 & 0.379 & \textcolor{blue}{\textbf{PAE}}: Moderate significance was observed. \\ \hline
\multirow{7}{*}{\rotatebox{270}{\shortstack{Project}}} &
S10: & Big leap & 0.282 & 0.339 & 0.628 & 0.368 & 0.502 & 0.618 & 0.415 & --- \\
& S11: & Learnt a lot & 0.442 & 0.460 & \textcolor{darkred}{\textbf{0.017}} & 0.443 & \textcolor{blue}{\textbf{0.069}} & 0.263 & 0.824 & \textcolor{darkred}{\textbf{PAE}}: Students with \emph{medium} PAE showed the highest agreement (mean = 5.13), significantly differing from those with \emph{nil} PAE (mean = 4.84, adj. $p$ = 0.028). Students with \emph{high} PAE had the lowest agreement (mean = 4.67). \textcolor{blue}{\textbf{PSE}}: Students with \emph{nil} PSE showed the lowest agreement (mean = 3.5), significantly differing from those with \emph{medium} (mean = 5.1, adj. $p$ = 0.07) and \emph{high} PSE (mean = 4.95, adj. $p$ = 0.107). \\
& S12: & Authentic & 0.690 & 0.267 & \textcolor{darkred}{\textbf{0.006}} & 0.743 & 0.823 & 0.736 & 0.657 & \textcolor{darkred}{\textbf{PAE}}: Students with \emph{medium} prior programming experience showed the highest agreement (mean = 4.98), significantly differing from those with a lot of prior experience (mean = 4.13, adj. $p$ = 0.024) and those with \emph{nil} prior experience (mean = 4.68, adj. $p$ = 0.069). \\
& S13: & Difference &
0.604 & 0.229 & 0.606 & 0.520 & 0.727 & 0.433 & 0.873 & --- \\ \hline
\multirow{6}{*}{\rotatebox{270}{\shortstack{Course\\Coordinator}}} & S14: & \small{Course structure} &
0.568 & 0.402 & \textcolor{darkgreen}{\textbf{0.142}} & \textcolor{blue}{\textbf{0.082}} & 0.368 & 0.470 & 0.842 & \textcolor{darkgreen}{\textbf{PAE}}: Students with \emph{high} PAE showed stronger agreement. \textcolor{blue}{\textbf{PCoE}}: All students strongly agreed. \\
& S15: & Little insight & \textcolor{darkgreen}{\textbf{0.122}} & 0.701 & \textcolor{darkred}{\textbf{0.001}} & \textcolor{blue}{\textbf{0.074}} & 0.381 & 0.516 & 0.214 & \textcolor{darkgreen}{\textbf{G}}: The male group had equally \emph{high} values (median = 4) as the female group (median = 4). Male students showed \emph{high}er agreement (mean = 3.46) compared to female students (mean = 3.19). \textcolor{darkred}{\textbf{PAE}}: Pairwise group comparisons between \emph{nil} - \emph{medium} and \emph{high} - \emph{medium} showed adj. $p$-values less than 0.05. Students with \emph{medium} PAE (mean = 3.09) strongly disagreed, while students with \emph{high} (mean = 4.2) and \emph{nil} PAE (mean = 3.69). \textcolor{blue}{\textbf{PCoE}}: Students with \emph{high} PCoE (mean = 3.16) strongly disagreed, while students with \emph{medium} (mean = 3.49) and \emph{nil} PCoE (mean = 3.69). \\ \hline
\multirow{26}{*}{\rotatebox{270}{\shortstack{Teaching Assistants (TAs)}}} & S16: & Assessment & 0.956 & \textcolor{darkgreen}{\textbf{0.133}} & \textcolor{darkgreen}{\textbf{0.158}} & 0.384 & 0.338 & 0.354 & 0.618 & \textcolor{darkgreen}{\textbf{PCE}}: Students with \emph{medium} PCE (mean = 3.89) somewhat disagreed compared to students with \emph{nil} PCE (mean = 4.26). \textcolor{darkgreen}{\textbf{PAE}}: All students were neutral in their response, with an average mean of 4.08. \\
& S17: & Variation in lab assessment & 0.694 & \textcolor{darkgreen}{\textbf{0.119}} & 0.681 & \textcolor{blue}{\textbf{0.097}} & 0.261 & 0.876 & 0.337 & \textcolor{darkgreen}{\textbf{PCE}}: Students with \emph{medium} PCE (mean = 3.83) showed stronger disagreement compared to those with \emph{high} PCE (mean = 4.67). \textcolor{blue}{\textbf{PCoE}}: Overall, students were neutral in their responses, with an average mean of 4.21. \\
& S18: & Variation for projects & 0.870 & \textcolor{blue}{\textbf{0.073}} & 0.710 & \textcolor{blue}{\textbf{0.068}} & \textcolor{darkgreen}{\textbf{0.141}} & 0.845 & \textcolor{darkgreen}{\textbf{0.174}} & \textcolor{blue}{\textbf{PCoE}}: Overall, students were neutral in their responses, with an average mean of 4.15. \textcolor{darkgreen}{\textbf{B}}: CB students showed stronger agreement (mean = 4.5), while AI (mean = 3.79), ME (mean = 3.78), and PH (mean = 3.68) students showed comparatively lower agreement. \\
& S19: & Variation for help & 0.906 & 0.244 & 0.697 & 0.732 & \textcolor{darkgreen}{\textbf{0.158}} & 0.577 & 0.305 & \textcolor{darkgreen}{\textbf{PSE}}: Students with \emph{nil} PSE strongly disagreed (mean = 3.25), while those with \emph{medium} to \emph{high} PSE were generally neutral (mean = 3.99). \\
& S20: & Tutorials & 0.739 & 0.532 & 0.593 & 0.371 & 0.834 & 0.873 & 0.817 & --- \\
& S21: & Professionalism & 0.674 & 0.543 & \textcolor{darkred}{\textbf{0.018}} & 0.909 & \textcolor{darkgreen}{\textbf{0.141}} & 0.655 & 0.863 & \textcolor{darkred}{\textbf{PAE}}: All adj. $p$-values were greater than 0.05, indicating no significant pairwise differences. However, students with \emph{medium} PAE showed stronger agreement (mean = 4.66) compared to those with \emph{nil} or \emph{high} PAE (weighted mean = 4.29). \textcolor{darkgreen}{\textbf{PSE}}: Students with \emph{medium} (mean = 4.56) and \emph{high} PSE (mean = 4.47) strongly agreed, whereas those with \emph{nil} PSE strongly disagreed (mean = 3.5). \\
& S22: & Feedback & 0.650 & 0.913 & \textcolor{darkgreen}{\textbf{0.197}} & 0.922 & 0.947 & \textcolor{darkred}{ \textbf{0.010}} & 0.931 & \textcolor{darkgreen}{\textbf{PAE}}: All students strongly agreed, with a weighted mean of 4.71. \textcolor{darkred}{\textbf{Gd}}: A significant difference was observed between students with BC and BB grades (adj. $p$ = 0.009). Students with grades BC (mean = 5.24), CC (mean = 5.10), CD (mean = 4.83), and AA (mean = 4.83) showed strong agreement. In contrast, students with BB grade (mean = 4.37) showed comparatively lower agreement. \\
& S23: & Asking TA help & 0.505 & 0.411 & 0.894 & 0.744 & \textcolor{blue}{\textbf{0.100}} & 0.269 & 0.571 & \textcolor{blue}{\textbf{PSE}}: Students with \emph{medium} to \emph{high} PSE strongly agreed, with a weighted mean of 4.44. In contrast, students with \emph{nil} PSE strongly disagreed (mean = 3.5). \\
& S24: & Preferred internet & 0.382 & 0.763 & \textcolor{blue}{\textbf{0.091}} & \textcolor{blue}{\textbf{0.091}} & \textcolor{darkred}{\textbf{0.027}} & 0.895 & 0.882 & \textcolor{blue}{\textbf{PAE}}: Students with \emph{high} PAE agreed (mean = 4.33), while those with \emph{nil} PAE disagreed (mean = 3.68). Students with \emph{medium} PAE were nearly neutral (mean = 3.94). \textcolor{blue}{\textbf{PCoE}}: Students with \emph{nil} (mean = 3.59) and \emph{medium} (mean = 3.73) PCoE disagreed, with a weighted mean of 3.70. In contrast, students with \emph{high} PCoE were somewhat neutral (mean = 4.12). \textcolor{darkred}{\textbf{PSE}}: A significant difference was observed between students with \emph{high} and \emph{medium} PSE (adj. $p$ = 0.025). Students with \emph{medium} PSE strongly disagreed (mean = 3.52), while those with \emph{high} PSE were more agreeable (mean = 4.00). \\
& S25: & Few TAs & 0.890 & \textcolor{darkred}{ \textbf{0.008}} & 0.572 & 0.488 & \textcolor{blue}{\textbf{0.098}} & 0.780 & 0.449 & \textcolor{darkred}{\textbf{PCE}}: Significant differences were found in pairwise comparisons between \emph{medium} - \emph{high} (adj. $p$ = 0.007) and \emph{high} - \emph{nil} (adj. $p$ = 0.014). Students with \emph{medium} (mean = 3.46) and \emph{nil} (mean = 3.68) PCE (weighted mean = 3.66) strongly disagreed, whereas students with \emph{high} PCE (mean = 6.00) strongly agreed. \textcolor{blue}{\textbf{PSE}}: However, students with \emph{high} (mean = 3.73) and \emph{medium} PSE (mean = 3.49, weighted mean = 3.65) strongly disagreed, while those with \emph{nil} PSE (mean = 4.50) were more agreeable. \\ \hline
% \multicolumn{11}{r}{Note: \textcolor{darkred}{\textbf{Red-colored}} $p$-values indicate strong significance ($p \leq 0.05$), \textcolor{blue}{\textbf{blue-colored}} $p$-values indicate moderate significance ($0.05 < p \leq 0.1$),} \\
% \multicolumn{11}{r}{and \textcolor{darkgreen}{\textbf{red-colored}} $p$-values also \emph{high}light values with $0.1 < p \leq 0.2$, potentially signaling directionally important results.}
\multicolumn{11}{r}{
\textcolor{darkred}{\textbf{Red}} $p$-values indicate strong significance ($p \leq 0.05$),
\textcolor{blue}{\textbf{blue}} indicate moderate significance ($0.05 < p \leq 0.1$), and
\textcolor{darkgreen}{\textbf{green}} denote marginal trends ($0.1 < p \leq 0.2$).
}
\label{tab:pvalues_last_week_1}
\end{tabular}
\end{adjustbox}
\end{table*}
\section{Discussion and Conclusion}
\label{5sec:discussion_conclusion}
% This section interprets the results using the theoretical perspectives and the \emph{RQ}s.
This section interprets the results through theoretical perspectives, discusses practical implications, outlines the study’s limitations, and suggests directions for future research.
% \subsection{Student Perceptions of Assessment:}
% The findings indicated that the students perception of lab assignments as a valuable learning activity (S1) \textcolor{green}{(refer to Table \ref{tab:pvalues_last_week_1} for details)} that aligns with formative assessment principles \cite{theory_perspective} and do not find it as a necessary evil (S2) \cite{qualitative_step}, implying that assessments are designed to enhance learning rather than merely certify outcomes. The collaborative efforts reported by students resonate with social constructivist theories \cite{collab_perspective_1}, where peer interactions and shared problem-solving foster knowledge construction \cite{collab_perspective}, \cite{student_opinion}. The lower stress levels (S3) linked to lab sessions may be attributed to constructive alignment \cite{constructive_alignment}, where a clear connection between learning objectives, activities, and assessments, along with students collaborating and supporting one another (S4 and S5), helps reduce anxiety \cite{10.1145/2999541.2999557}.
% Students demonstrated varied levels of awareness regarding the content of exams (S6 and S7), a trend also reflected in the qualitative findings \textcolor{blue}{(\cite{qualitative_step}}. The view that exams were both authentic and valuable to assess skill and knowledge (S8 and S9) aligns with the principles of authentic assessment \cite{authentic_assessment}, which emphasizes tasks that replicate real-world scenarios, \textcolor{blue}{and previous results \cite{10036401}.} Although the exams primarily utilized traditional formats such as multiple-choice questions, students’ perception of their authenticity may be attributed to the transparency of expectations, a key factor in reducing test anxiety \cite{authentic_assessment_1}.
% Projects were widely regarded as authentic learning opportunities (S12), in line with experiential learning theory \cite{learning}, which highlights the educational value of practical, hands-on experiences. Students reported significant learning gains from their project work (S11). However, the inconsistency in the effort required to achieve comparable grades raises issues of assessment fairness, potentially disadvantaging some students (S13). The neutral responses concerning the increased difficulty from lab assignments to projects (S10) may indicate \textcolor{blue}{a too difficult assignment, not within the students' ZPD \cite{ZPD_1}}, suggesting that a more gradual increase in task complexity is necessary to effectively support student learning progression.
%=============
\subsection{Student Perceptions of Assessment}
The findings indicate that students perceived lab assignments (S1) as valuable learning activities (refer to Table \ref{tab:themes_statements} for details), aligning well with the principles of formative assessment \cite{theory_perspective}. They did not view them as a necessary evil (S2) \cite{qualitative_step}, suggesting that the assessments were seen as tools to enhance learning rather than merely to certify achievement. Students’ emphasis on collaborative efforts during labs resonates with social constructivist theories \cite{collab_perspective_1}, where peer interaction and shared problem-solving foster knowledge construction \cite{collab_perspective,student_opinion}. Lower stress levels associated with lab sessions (S3) may be attributed to constructive alignment \cite{constructive_alignment}, where coherence between learning objectives, activities, and assessments, combined with peer support (S4 and S5), helps reduce anxiety \cite{10.1145/2999541.2999557}.
Students showed varying levels of awareness about the exam content (S6 and S7), a trend also reflected in the qualitative findings \cite{qualitative_step}. However, many perceived exams as authentic and meaningful assessments of skills and knowledge (S8 and S9), aligning with the theory of authentic assessment \cite{authentic_assessment} and previous findings \cite{10036401}. Although the exams primarily relied on conventional formats such as multiple-choice questions, students’ perception of authenticity may stem from clear and transparent expectations, an essential factor in reducing assessment-related anxiety \cite{authentic_assessment_1}.
Projects were generally seen as authentic learning experiences (S12), consistent with experiential learning theory \cite{learning}, which emphasizes the educational value of hands-on engagement. Students also reported substantial learning gains from completing projects (S11). However, some noted inconsistencies in the effort required to achieve similar grades (S13), raising concerns about perceived fairness in assessment. Neutral responses regarding the increased difficulty from lab assignments to projects (S10) may suggest that the jump in complexity was too steep and possibly beyond some students' ZPD \cite{ZPD_1}. This highlights to the need for a more gradual increase in task difficulty to better scaffold student learning and support their progression.
% \subsection{Student Perceptions of Course Coordinator and TAs}
% The largely neutral responses regarding course coordinator effectiveness, with a notable subset recognizing the coordinator’s engagement (S14) beyond lectures, suggest that the coordinator’s role was perceived as both accessible and involved. This finding aligns with the concept of distributed leadership in educational settings, where instructional leadership extends beyond direct instruction to encompass broader course oversight and student engagement in various learning contexts. The absence of negative perceptions regarding the coordinator’s insight into non-lecture activities (S15) may reflect the benefits of transparent communication and visible involvement, which encourage student trust and satisfaction.
% In relation to TAs (S16-S25), students generally valued the professional conduct, feedback, and guidance provided by TAs, and found them more approachable than course coordinators \cite{wilson2002study}. This supports previous research indicating that approachability and social presence of TAs can enhance students’ comfort and willingness to seek help, which is a predictor of academic success \cite{10.1145/3291279.3339418}. Many students prefer consulting TAs over searching online resources. This may indicate a shift in student attitudes toward valuing personalized, context-specific support within the classroom environment or due to the availability of sufficient number of TAs (S25), in contrast to the qualitative results\cite{qualitative_step}. However, the wide variation in student responses regarding TA consistency in assessment, assistance, and tutorial content raises concerns. Inconsistent experiences with TAs can undermine students’ perceptions of fairness and equal opportunity, echoing findings from previous studies that highlight the importance of standardized training and clear guidelines for TAs to ensure equitable learning experiences \cite{TAs1, TAS2}. The presence of adequate TAs, as reported by many students, is a positive indicator; however, the variability in perceived availability suggests that support distribution may not always align with student needs.
% These findings underscore the importance of ensuring that all students receive consistent feedback and guidance, and that support is available, especially for tasks within their ZPD \cite{ZPD_1}, \textcolor{blue}{assignments that students cannot solve on their own}.
%%%%%%%%%%%%%%%%%
\subsection{Student Perceptions of Course Coordinator and TAs}
The largely neutral responses of students to the effectiveness of the course coordinator, with a notable subset recognizing their active participation beyond lectures (S14), suggest that the coordinator was perceived as accessible and involved in multiple aspects of the course. This perception aligns with the concept of distributed leadership in higher education, where instructional responsibilities are shared across various roles to promote collective responsibility for student learning \cite{harris2022distributed}. In this model, leadership extends beyond direct instruction to include coordination of assessments, learning resources, and support mechanisms. The absence of negative perceptions about the coordinator’s insight into non-lecture activities (S15) may reflect the benefits of visible involvement and transparent communication, both of which foster student trust and satisfaction.
Regarding TAs (S16-S25), students generally reported positive experiences with professional behavior, constructive feedback, and approachability. Notably, students found TAs more accessible than the course coordinator, a finding consistent with prior research emphasizing the role of TA approachability and social presence in fostering help-seeking behavior and academic engagement \cite{wilson2002study, 10.1145/3291279.3339418}. Interestingly, many students indicated a preference for consulting TAs over online resources when encountering challenges. This shift may highlight a growing appreciation for personalized, context-aware support within the classroom, possibly enabled by the availability of a sufficient number of TAs (S25). This contrasts with earlier qualitative findings \cite{qualitative_step}, suggesting that improvements in TA accessibility may have occurred during the studied course.
However, notable variability was observed in students' perceptions of TA consistency regarding assessment, assistance, and tutorial content. Such inconsistencies can negatively impact students’ perceptions of fairness and uniformity, aligning with concerns raised in prior studies that emphasize the need for standardized training and assessment guidelines for TAs \cite{TAs1, TAS2}. Although the majority of students acknowledged the presence of adequate TA support, the uneven perceptions of their availability suggest potential mismatches between support distribution and student needs during high-demand periods.
These findings underscore the importance of structured TA coordination to ensure consistent and equitable support. All students must receive timely guidance and formative feedback, especially for tasks situated within their ZPD, i.e., tasks they cannot yet complete independently but can accomplish with appropriate scaffolding.
%%%%%%%%%%%%%%%%%%%%%%%%%
% \subsection{Differences \textcolor{blue}{between subgroups of students}}
% The observed differences in academic grade and prior experience across gender, programming exposure, and academic branch can be contextualized through several theories. The lower grades and low prior programming experience among female students align with stereotype threat theory \cite{diff_theory_1}, where societal stereotypes about gender roles in STEM fields can undermine confidence and performance. The gender gap in prior smartphone and computer experience (e.g., 70\% of males reporting high PSE vs. 53\% of females) reflects gender socialization \textcolor{green}{in an Indian setting} \cite{diff_theory_2}, where societal norms often steer males toward technology-centric activities like gaming or troubleshooting. This disparity in prior programming experience may contribute to differential self-efficacy \cite{diff_theory_3}, influencing both engagement and outcomes. As shown in previous studies \cite{inproceedingsF, female_1, female_2, female_3},
% The underrepresentation of women in technical branches exacerbates these issues, as isolation in male-dominated environments can reduce the sense of belonging \cite{diff_theory_4}, further impacting performance.
% The strong correlation among prior experience (PAE, PCE, PCoE, PSE) and academic success supports transfer of learning \cite{transfer_learning}, where foundational skills in one domain (e.g., C programming) enhance problem-solving efficiency in new contexts. Students with prior experience benefit from reduced cognitive load \cite{cognitive_load}, enabling them to focus on advanced tasks rather than \textcolor{green}{learning} basic syntax. The strong relationship between PCoE and grades further underscores the role of cultural capital \cite{cultural_capital}, where students from \textcolor{green}{technology-rich} backgrounds enter courses with implicit advantages. However, the significant performance gap between students with no prior experience and their peers raises concerns about equitable course design. Courses assuming prior knowledge may violate principles of constructive alignment \cite{constructive_alignment}, disadvantaging true beginners. The lack of scaffolding for novices \textcolor{green}{is evident} in the difficulty leap between labs and projects fails to align with students’ ZPD \cite{ZPD_1}, where incremental challenges and guided support are critical for progression.
% The digital divide in prior computer and smartphone experience (PCoE, PSE) highlights systemic inequities in technology access \cite{digital_divide}, which disproportionately affect students from \textcolor{purple}{underprivileged} backgrounds. These differences are compounded by inconsistent TA support and feedback, undermining assessment reliability \cite{assessment_1}. Variability in TA training and workload allocation evident in courses can lead to inequitable learning experiences.
% The dominance of CSE and MC students \textcolor{green}{(mainly Math and Computer Science students)} in high-performance cohorts illustrates self-selection bias, where students with prior technical exposure gravitate toward disciplines perceived as aligned with their skills. This creates a feedback loop, enriching these branches with peer networks that foster communities of practice \cite{diff_theory_5}. Conversely, non-technical branches (e.g., CB, CE) often attract students with less prior experience, perpetuating performance gaps. The correlation between high PAE and CSE enrollment also reflects expectancy-value theory \cite{exp_theory}, where students invest in fields they perceive as valuable or attainable based on prior success.
%%%%%%%%%%%%%%%%%%%%%%%%%
\subsection{Differences between Subgroups of Students}
The observed differences in academic performance and prior experience across gender, programming exposure, and academic branches can be contextualized through multiple theoretical lenses. The relatively lower grades and limited prior programming experience reported by female students align with {stereotype threat theory} \cite{diff_theory_1}, which suggests that societal stereotypes about gender roles in STEM can undermine confidence and hinder performance. The gender gap in smartphone and computer usage, e.g., 70.11\% of males reporting high PSE compared to 53.73\% of females, may reflect patterns of {gender socialization} within the Indian context \cite{diff_theory_2}. This disparity in early exposure can affect {self-efficacy} \cite{diff_theory_3}, influencing both academic engagement and learning outcomes. As previous studies have shown \cite{inproceedingsF, female_3}, the under-representation of women in technical branches can further exacerbate these challenges. Isolation in male-dominated environments may reduce the sense of belonging \cite{diff_theory_4}, thereby negatively affecting performance.
The strong correlation among prior experiences (PCE, PAE, PCoE, PSE) and academic success supports the theory of {transfer of learning} \cite{transfer_learning}, whereby foundational skills in one domain (e.g., C programming) enhance students' efficiency in new contexts. Prior experience can lower {cognitive load} \cite{cognitive_load}, allowing students to focus on higher-order problem-solving rather than struggling with basic syntax. The positive relationship between PCoE and academic grades also highlights the influence of {cultural capital} \cite{cultural_capital}, wherein students from technology-rich environments begin courses with built-in advantages.
However, these advantages expose equity concerns. The substantial performance gap between students with and without prior experience suggests a misalignment with the principles of {constructive alignment} \cite{constructive_alignment}, particularly when courses implicitly assume prior knowledge. The abrupt difficulty progression between labs and projects may also fail to respect learners' ZPD \cite{ZPD_1}, where incremental challenge and guided support are essential for skill acquisition. Lack of scaffolding disproportionately affects true beginners, leading to potential disengagement or underperformance.
The digital divide in computer and smartphone experience (PCoE, PSE) further underscores systemic inequities in access to technology \cite{digital_divide}, disproportionately impacting students from underprivileged backgrounds. These disparities are compounded by inconsistencies in TA feedback and support, undermining {assessment reliability} \cite{assessment_1}. Variations in TA training, workload, or availability may result in inequitable academic experiences across students within the same course.
The dominance of CSE (Computer Science \& Engg.) and MC (Math \& Computing) students in higher performance bands illustrates a form of {self-selection bias}, wherein students with greater prior exposure naturally gravitate toward technical disciplines. This trend creates a reinforcing feedback loop, nurturing {communities of practice} \cite{diff_theory_5} within these branches and further supporting academic success.
In contrast, students from non-CS branches (e.g., CB, CE) tend to enter with less exposure and continue to lag in performance. The correlation between high PAE and CSE enrollment also supports the {expectancy-value theory} \cite{exp_theory}, where students are more likely to invest effort in fields they perceive as valuable and attainable based on past success.
%=========================
% \subsection{Comparative Study}
% Comparing the results in the previously studied Swedish context to the results \cite{10036401} with the results in this Indian data collection, we find some similarities and differences focusing on engineering student perceptions of assessment, course coordinators, and teaching assistants (TAs), and the influence of demographic and experiential factors on academic outcomes. In both contexts, students recognize labs as more than mere requirements and value projects as authentic learning experiences with variation in workload across projects. Positive coordinator engagement, while concerns about TA consistency and availability are present in both settings. However, students from both settings still found TAs as more approachable than coordinators and valued their feedback and guidance. In both countries, prior programming and technical experience are correlated with academic success, and females are underrepresented, performing slightly lower on average. The Swedish students did perceive a sharper increase in difficulty from labs to projects, but stated to have clearer expectations about exams, while the Indian students' responses are more varied on these questions. Another interesting finding is that Indian students preferred TAs over online resources, while Swedish students preferred online resources over TAs. Swedish analysis focuses more on programming experience and gender, while the analysis on the Indian data collection explores a broader range of prior experiences (C programming, general programming, computer, and smartphone usage) and academic branches. Overall, the results from both studies emphasize the need for greater standardization, improved TA training, and enhanced student support systems within the Indian educational context. Some challenges seem to be universal, but it also suggests that we as educators can learn from each other's strengths.
%77777777777777777777777777
\subsection{Comparative Study}
This subsection compares findings from the previously studied Swedish context \cite{10036401} with those from the current Indian study, focusing on engineering students’ perceptions of assessment, the roles of course coordinators and teaching assistants (TAs), and the impact of demographic and experiential factors on academic outcomes.
In both settings, students recognized lab assignments as meaningful learning activities rather than formal requirements and viewed projects as authentic learning experiences, though they reported variations in workload across projects. Course coordinators were generally perceived as positively engaged beyond lectures, while concerns about TA consistency and availability emerged as common issues. Despite these concerns, students in both countries
found TAs more approachable and helpful than coordinators.
% appreciated the approachability and feedback of TAs more than that of coordinators.
A consistent trend observed across both contexts is the positive association between prior technical experience and academic performance, alongside the under-representation of female students, who, on average, performed slightly lower. However, while Swedish students reported a clearer understanding of exam expectations, Indian students' responses were more varied, indicating potential differences in communication or transparency. Swedish students also perceived a sharper difficulty increase from labs to projects.
An interesting divergence lies in support-seeking behavior: Indian students preferred consulting TAs, whereas Swedish students relied more on online resources. Additionally, the Indian study examined a broader spectrum of prior experiences, including C programming, general programming, computer, and smartphone use, and analyzed differences across academic branches, offering a more granular understanding of student backgrounds.
Overall, the results from both studies emphasize the need for greater standardization, improved TA training, and enhanced student support systems, particularly within the Indian educational context. While some challenges appear to be universal, the findings also suggest that we, as educators, can learn from each other’s strengths.
% Overall, while some challenges, such as TA variability and gender disparities, appear universal, others are context-specific. These findings underscore the importance of improved standardization, TA training, and equitable support systems, particularly within the Indian setting, and suggest opportunities for cross-contextual learning to enhance engineering education globally.
%===============
% \textcolor{purple}{
% \subsection{Comparative Study}
% This comparative study highlights similarities and differences in the findings from the Swedish and Indian educational settings, focusing on engineering student perceptions of assessment, course coordinators, and teaching assistants (TAs), and the influence of demographic and experiential factors on academic outcomes.
% }
% \textcolor{purple}{
% \textcolor{blue}{Comparing the results in the previously studied Swedish context to the results \cite{10036401} with the results in this Indian data collection, we find some similarities and differences}. In both contexts, students recognize labs as more than mere requirements and value projects as authentic learning experiences with variation in workload across projects. Positive coordinator engagement, while concerns about TA consistency and availability are present in both settings. However, students from both settings still found TAs as more approachable than coordinators and valued their feedback and guidance. In both countries, prior programming and technical experience are correlated with academic success, and females are underrepresented, performing slightly lower on average. The Swedish students did perceive a sharper increase in difficulty from labs to projects, but stated to have clearer expectations about exams, while the Indian students' responses are more varied on these questions. Another interesting finding is that Indian students preferred TAs over online resources, while Swedish students preferred online resources over TAs. Swedish analysis focuses more on programming experience and gender, while the analysis on the Indian data collection explores a broader range of prior experiences (C programming, general programming, computer, and smartphone usage) and academic branches. Overall, the results from both studies emphasize the need for greater standardization, improved TA training, and enhanced student support systems within the Indian educational context. \textcolor{blue}{Some challenges seem to be universal, but it also suggests that we as educators can learn from each other's strengths.}
% }
% \subsection{Implications for Practice}
% The findings from this study highlight actionable strategies for course coordinators, TAs, program designers, and institutions to improve equitable, effective, and inclusive learning environments. First, implementing pre-course modules or bridging workshops is essential for students with limited prior programming or computer experience. These initiatives should introduce foundational concepts and computational thinking to establish a baseline skill set before the course begins. Assignments and projects must be designed with incremental complexity, aligning with students’ ZPD \cite{ZPD_1}, to ensure that true beginners are not overwhelmed and can build confidence as they progress.
% Ensuring equity and consistency in assessment is also critical. Comprehensive and regular training for TAs \cite{TAs1}, can help minimize discrepancies in evaluation and feedback. Such measures would promote fairness and reliability across course sections and TA groups. Additionally, cultivating an inclusive and supportive classroom environment is essential. Instructors should adopt teaching practices that counteract stereotype threat, such as featuring diverse role models, encouraging a growth mindset, and elevating a sense of belonging for all students. Structured peer collaborations, such as pair programming and study groups, can strengthen support networks, particularly benefiting students from underrepresented groups. Technology access remains a pivotal determinant of success. Institutions should ensure that all students have the necessary resources, such as computers and smartphones, either through campus facilities or loan programs. Supplementary digital literacy workshops can help level the playing field for students from less technology-rich backgrounds. When designing projects, course coordinators should prioritize authenticity and equity \cite{constructive_alignment}, ensuring tasks reflect real-world challenges but are also manageable for students regardless of their prior experience. Providing options within projects can allow all students to participate meaningfully.
% Finally, the institute needs to adopt a data-driven approach to continuous improvement. Regularly monitoring student performance and engagement, disaggregated by variables such as gender, prior experience, and academic branch, can identify differences and inform targeted interventions. Systematically collecting and acting on student feedback regarding assessment fairness, TA support, and course difficulty will further refine instructional practices. By integrating these strategies, the institute can reduce achievement gaps, support diverse student populations, and ensure equitable opportunities for success.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\subsection{Implications for Practice}
The findings of this study highlight several actionable strategies for course coordinators, TAs, program designers, and institutions to foster equitable, effective, and inclusive learning environments.
%
First, offering {pre-course modules or bridging workshops} is vital for students with limited programming or computer experience. These initiatives should introduce foundational concepts and computational thinking to establish a common baseline before course commencement. Assignments and projects must be structured with {gradual complexity}, aligned with students’ ZPD \cite{ZPD_1}, allowing beginners to build confidence and competence over time.
To ensure {equity and consistency in assessment}, institutions should implement {regular and comprehensive TA training} \cite{TAs1}, minimizing variability in grading and feedback across different groups. Cultivating a {supportive and inclusive classroom culture} is equally important. Instructors can help mitigate stereotype threat by showcasing diverse role models, promoting a growth mindset, and fostering a sense of belonging among all students.
{Structured peer-learning activities}, such as pair programming and study groups, can enhance engagement, especially for students from underrepresented groups. Addressing {technology access gaps} is also critical. Institutions should offer {digital literacy workshops} to support students from less technology-rich backgrounds.
%
Project work should be {authentic yet adaptable}, reflecting real-world problems while being accessible to students regardless of their prior experience. Offering{project options or scaffolded challenges} can help ensure meaningful participation for all.
Finally, institutions should adopt a {data-informed approach} to continuous improvement. Regular analysis of student performance and engagement, disaggregated by gender, prior experience, and academic branch, can guide targeted interventions. Collecting and acting on student feedback regarding assessment fairness, TA support, and course difficulty will further refine instructional practices.
%
By implementing these strategies, institutions can help close achievement gaps, support diverse learners, and promote equitable academic success.
% \subsection{Limitations}
% A key limitation of this study lies in its confinement to a single Indian engineering institution, which constrains the generalizability of findings to other educational contexts with differing curricular structures, student demographics, or institutional policies. Additionally, students’ prior programming, computer, and smartphone experience were self-reported, a methodology that may introduce recall bias or inaccuracies in self-assessment of skill levels, potentially compromising the reliability of the outcomes.
% Further, only 318 of 516 eligible students (61.6\%) participated consistently in the survey, introducing potential non-response bias. Volunteers may exhibit systematic differences from non-participants (e.g., higher motivation or survey comfort), skewing results and limiting the generalizability of findings to the broader student population. These factors collectively raise concerns about the external validity of the study’s conclusions.
%======================
\subsection{Limitations}
A key limitation of this study lies in its confinement to a single Indian engineering institution, which constrains the generalizability of findings to other educational contexts with differing curricular structures, student demographics, or institutional policies. Additionally, students’ prior programming, computer, and smartphone experience were self-reported, a methodology that may introduce recall bias or inaccuracies in self-assessment of skill levels, potentially compromising the reliability of the outcomes.
Only 318 of the 516 eligible students (61.6\%) volunteered consistently in the survey, introducing potential non-response bias. Volunteers may differ systematically from non-respondents, such as being more motivated or comfortable with surveys, thus skewing the results.
Finally, while subgroup analyses were conducted (e.g., by gender, academic branch, and prior experience), some subgroups, such as female students or students from less-represented branches, were relatively small.
% \textcolor{red}{\subsection{Comparison Study}
% The comparative analysis highlights both similarities and differences in how engineering students from Indian and international institutions perceive assessments in CS1 courses. Both the Indian and Swedish studies underscore the value of lab assignments and project-based work; however, they differ significantly in terms of support structures and pedagogical approaches. Indian students frequently reported stress and uncertainty related to assessment timelines and expectations, whereas their Swedish counterparts found examinations to be more predictable and better aligned with learning objectives—reflecting stronger constructive alignment in course design. Challenges faced by Indian students include inconsistency in TA support, elevated assessment-related stress, and insufficient project scaffolding. In contrast, Swedish students benefited from structured TA training, consistent assessment practices, and a more inclusive learning environment. While prior programming experience influenced outcomes in both contexts, gender-based disparities were more pronounced in the international study, with female students reporting less effective feedback and higher levels of stress. Overall, these findings emphasize the need for greater standardization, improved TA training, and enhanced student support systems within the Indian educational context.
% }
% \subsection{Future Research Directions}
% Future research could explore how variations in TA practices influence student learning outcomes and perceptions, particularly across different institutional or regional contexts. It would also be valuable to investigate the specific characteristics of lab assignments, exams, and projects that contribute to their perceived authenticity and effectiveness. Additionally, examining the experiences of subgroups such as first-generation college students or non-CS majors may uncover unique challenges or expectations in an Indian context. Another important direction is to study the growing use of large language models (LLMs) by students and how these tools are integrated into coursework, their impact on learning and assessment integrity, and how students and instructors perceive their value and limitations in an Indian educational context.
%%%%%%%%%%%%%%%%%%5
\subsection{Future Research Directions}
Future research could explore how variations in TA practices influence student learning outcomes and perceptions, particularly across different institutional or regional contexts. It would also be valuable to investigate the specific characteristics of lab assignments, exams, and projects that contribute to their perceived authenticity and effectiveness.
Additionally, examining the experiences of subgroups such as first-generation college students or non-CS majors may reveal unique challenges or expectations. Another important direction involves studying the growing use of large language models (LLMs) by students, investigating how these tools are integrated into coursework, their impact on learning and assessment integrity, and how students and instructors perceive their value and limitations in an Indian educational context.
% \section{Conclusion}
% This study provides a comprehensive analysis of engineering students’ perceptions of assessment practices, instructional support, and the influence of prior experiences on academic outcomes within an Indian educational context. Key findings underscore the importance of equitable assessment design, consistent TA support, and the need to address disparities rooted in gender and prior technical exposure. Comparative insights with the Swedish context further highlight both universal and context-specific challenges. By implementing targeted, evidence-based interventions, institutions can foster more inclusive, effective, and learner-centered environments that support diverse student populations.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Divyansh
% \subsubsection{Lab Assignments (S1--S5)}
% For S1, significant differences were observed in students' perceptions based on prior programming experience (PAE) and prior smartphone experience (PSE). Students with substantial prior programming experience disagreed more than those with medium experience (adj. p-value: 0.047), suggesting that lab assignments were more beneficial for students with medium experience but less engaging for highly experienced students. Similarly, students with no prior smartphone experience disagreed significantly more than those with medium or high experience (adj. p-value: 0.041), indicating that students unfamiliar with smartphone usage found lab assignments less effective.
% For S2, students from the Chemical and Biochemical Engineering (CB, mean: 3.95) and Electrical Engineering (EE, mean: 3.83) branches strongly agreed that lab assignments were a necessary burden. In contrast, Physics (PH) students disagreed the most (mean: 2.95), indicating that PH students found lab assignments more beneficial for learning rather than burdensome. These differences suggest that discipline-specific curriculum structures may affect students’ perceptions of lab assignments.
% For S3, CB students reported the highest stress levels in completing lab assignments before deadlines (mean: 4.03), while PH students reported the least stress (mean: 3.11). Additionally, students with a BC grade perceived lab assignments as stressful (mean: 3.79), while those with a CD grade disagreed the most (mean: 2.95). This pattern suggests that students who struggle academically might either feel overwhelmed by lab assignments or benefit from a more flexible deadline structure.
% For S4, students across different branches generally agreed that they received significant help from classmates when solving lab assignments. However, students in Computer Science and Engineering (CSE) and Civil Engineering (CE) were exceptions, as they disagreed with this statement. This may indicate that these students relied more on individual problem-solving or alternative resources rather than peer collaboration.
% For S5, students with nil (mean: 4.58) and medium (mean: 4.4) prior experience in C programming strongly agreed that they often helped classmates in solving lab assignments. In contrast, students with substantial prior experience in C programming strongly disagreed (mean: 3.67). A similar trend was observed for general programming experience, where highly experienced students (mean: 4.14) were less inclined to assist their peers, whereas students with medium (mean: 4.59) and nil (mean: 4.51) prior experience strongly agreed that they frequently helped classmates. This finding suggests that students with more experience may either feel less obligated to assist or find lab assignments too simple to require collaborative learning.
% \subsubsection{Exam (S6--S9)}
% For S6, students with medium prior computer experience disagreed more (mean: 3.87) about predicting exam questions based on the course and lab content than students with high (mean: 4.17) and nil (mean: 4.25) experience. This suggests that students with no prior experience relied more on course materials, while experienced students were better at predicting exam patterns.
% For S7, students with no prior computer experience strongly agreed (mean: 4.34) that they knew in advance what would be assessed in the written exam, compared to students with medium (mean: 3.88) and high (mean: 4.14) experience. This suggests that less experienced students depended more on structured learning resources.
% For S8, students with AA grades (mean: 4.97) strongly agreed that the knowledge and skills assessed in the exam were important for programming compared to students with grades AB (mean: 4.68), BB (mean: 4.51), BC (mean: 4.89), CC (mean: 4.90), and CD (mean: 4.83). This indicates that students with higher academic performance recognized the exam's relevance more than others.
% \subsubsection{Project Assignments (S10--S13)}
% For S11, students with medium prior experience in programming (\(\text{mean} = 5.13\)) reported significantly higher agreement that they learned a lot from the project assignments compared to students with nil prior experience (\(\text{mean} = 4.84\), \(\text{adj. p-value} = 0.028\)). This suggests that students with moderate programming experience benefited the most, likely because they had sufficient foundational knowledge to engage with the projects effectively, whereas those with no prior experience may have struggled with fundamental concepts.
% Regarding smartphone experience, students with nil prior experience (\(\text{mean} = 3.5\)) were significantly less likely to agree that they learned a lot from the project assignments compared to those with medium (\(\text{mean} = 5.1\), \(\text{adj. p-value} = 0.07\)) and a lot of experience (\(\text{mean} = 4.95\), \(\text{adj. p-value} = 0.107\)). This indicates that familiarity with smartphone-based development environments or related technologies might have influenced students' ability to grasp project-related concepts effectively.
% For S12, students with medium prior programming experience (\(\text{mean} = 4.98\)) strongly agreed that the project assignments were authentic, meaning that they closely resembled real-world programming tasks. They significantly differed from students with a lot of prior programming experience (\(\text{mean} = 4.13\), \(\text{adj. p-value} = 0.024\)) and those with nil prior experience (\(\text{mean} = 4.68\), \(\text{adj. p-value} = 0.069\)), who exhibited comparatively lower agreement. This suggests that students with extensive prior experience may have found the assignments too simplistic, while those with no experience may have found them too challenging, reducing their perceived authenticity.
% \subsubsection{Course Coordinators (S14-S15)}
% For S14, students with any level of prior experience in programming strongly agreed that the course coordinator effectively managed the course structure and learning activities, bringing a personal touch to the delivery. Similarly, all students with any prior experience in computer usage also strongly agreed with this statement.
% For S15, male students reported slightly higher agreement (Mdn = 4, mean = 3.46) than female students (Mdn = 4, mean = 3.19), indicating a perception that the course coordinator showed limited insight into or interest in course components beyond lectures, such as tutorials and lab assessments.
% Students with medium prior experience in programming (mean = 3.09) strongly disagreed with S15 compared to those with a lot (mean = 4.20) and nil (mean = 3.69) prior experience. Additionally, students with extensive prior computer usage (mean = 3.16) disagreed more with S15 than those with medium (mean = 3.49) or no (mean = 3.69) prior experience.
% \subsubsection{Teaching Assistants (S16--S25)}
% The students' perceptions of Teaching Assistants (TAs) regarding lab assignments and projects were analyzed based on their prior experience in programming, smartphone usage, and computer literacy. The following discusses key insights for each statement.
% For S16, students with medium prior experience (mean: 3.89) somewhat disagreed that TAs primarily assessed the lab assignments and projects, whereas students with nil prior experience in C programming (mean: 4.26) agreed. The average response across students with any programming experience was neutral (avg. mean: 4.08). This suggests that students with more programming exposure may have had different expectations about assessment roles, possibly perceiving TAs as facilitators rather than primary evaluators.
% For S17, students with medium prior experience in C programming (mean: 3.83) strongly disagreed that there was a large variation in TA grading, whereas students with a lot of prior experience (mean: 4.67) strongly agreed. The overall average for students with any prior computer experience was neutral (avg. mean: 4.21). This indicates that students with greater programming proficiency may have been more sensitive to inconsistencies in TA grading.
% For S18, students with any prior computer experience were neutral (avg. mean: 4.15) regarding differences in TA assessment of project assignments. However, CB students strongly agreed (mean: 4.5), while AI (mean: 3.79), ME (mean: 3.78), and PH (mean: 3.68) students somewhat disagreed. This suggests variations in how students from different disciplines perceived TA grading fairness.
% For S19, students with nil prior smartphone experience (mean: 3.25) strongly disagreed that TA assistance varied, while students with medium-to-lot experience remained neutral (mean: 3.99). This implies that students unfamiliar with smartphone-based learning tools may have had lower expectations for TA support.
% For S21, students with medium prior programming experience (mean: 4.66) strongly agreed that TAs treated them professionally, whereas students with nil-to-lot experience had a slightly neutral stance (weighted avg. mean: 4.29). This suggests that students with programming exposure valued structured TA interactions more.
% For S22, all students with programming experience strongly agreed (weighted avg. mean: 4.71) that they received useful feedback and guidance from TAs. However, students with BC (mean: 5.24), CC (mean: 5.1), and CD (mean: 4.83) grades agreed more than those with BB (mean: 4.37), who somewhat disagreed. AA-grade students (mean: 4.83) strongly agreed. This highlights that students struggling academically may have found TA guidance more beneficial.
% For S23, students with medium-to-lot smartphone experience strongly agreed (avg. mean: 4.44) that it was easier to ask a TA for help than the course coordinator, while students with nil experience (mean: 3.5) strongly disagreed. This implies that familiarity with digital tools may have influenced their willingness to approach TAs.
% For S24, students with a lot of programming experience (mean: 4.33) strongly agreed that they preferred searching online for answers rather than asking a TA, while students with nil (mean: 3.68) disagreed. Medium-experience students were neutral (mean: 3.94). Students with nil (mean: 3.59) and medium (mean: 3.73) prior computer experience disagreed with this preference, whereas students with a lot of experience (mean: 4.12) were neutral. Those with medium mobile experience (mean: 3.52) strongly disagreed, while those with a lot of experience (mean: 4.0) agreed (adj. p-value: 0.025). This indicates that students confident in independent problem-solving preferred online resources.
% For S25, students with medium (mean: 3.46) and nil (mean: 3.68) prior experience in C programming strongly disagreed that there were too few TAs during lab sessions, while students with a lot of experience (mean: 6.0) strongly agreed (adj. p-value: 0.007). Similarly, students with nil prior smartphone experience (mean: 4.5) agreed, whereas students with medium (mean: 3.49) and a lot (mean: 3.73) disagreed (weighted mean: 3.65). These results suggest that while many students felt the number of TAs was adequate, more experienced students felt otherwise, possibly due to higher expectations for personalized guidance.
\section*{Acknowledgment}
The authors gratefully acknowledge the support of all students and teaching assistants in the studied course.
\begin{figure*}
\centering
\footnotesize
\begin{tabular}{c | c | c | c | c}
\hline
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q1_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q2_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q3_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q4_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q5_W12.pdf} \\
S1 & S2 & S3 & S4 & S5 \\ \hline
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q6_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q7_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q8_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q9_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q10_W12.pdf} \\
S6 & S7 & S8 & S9 & S10 \\ \hline
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q11_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q12_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q13_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q14_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q15_W12.pdf} \\
S11 & S12 & S13 & S14 & S15 \\ \hline
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q16_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q17_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q18_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q19_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q20_W12.pdf} \\
S16 & S17 & S18 & S19 & S20 \\ \hline
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q21_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q22_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q23_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q24_W12.pdf} &
\includegraphics[width=0.18\linewidth]{figs/QQPlot_Q25_W12.pdf} \\
S21 & S22 & S23 & S24 & S25 \\
\hline
\end{tabular}
\caption{Q-Q plots for survey questions S1 to S25 from Week-9. The plots indicate that response distributions deviate from normality, thereby justifying the application of non-parametric statistical tests in the analysis.}
\label{fig:QQ}
\end{figure*}
\iffalse
\begin{figure*}
\centering
\footnotesize
%%%%%%%%%%%%% ROW 1
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q1_W12.pdf}
\caption*{S1}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q2_W12.pdf}
\caption*{S2}
\end{subfigure}
% \vspace{12pt}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q3_W12.pdf}
\caption*{S3}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q4_W12.pdf}
\caption*{S4}
\end{subfigure}
% \vspace{12pt}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q5_W12.pdf}
\caption*{S5}
\end{subfigure}
\hfill
%%%%%%%%%% ROW 2
% \vspace{12pt}
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q6_W12.pdf}
\caption*{S6}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q7_W12.pdf}
\caption*{S7}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q8_W12.pdf}
\caption*{S8}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q9_W12.pdf}
\caption*{S9}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q10_W12.pdf}
\caption*{S10}
\end{subfigure}
\hfill
%%%%% ROW 3
% \vspace{12pt}
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q11_W12.pdf}
\caption*{S11}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q12_W12.pdf}
\caption*{S12}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q13_W12.pdf}
\caption*{S13}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q14_W12.pdf}
\caption*{S14}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q15_W12.pdf}
\caption*{S15}
\end{subfigure}
\caption{Q-Q plots for questions S1 to S15 for Week-9. These plots show that none of the question responses follow a normal distribution, validating the use of non-parametric statistical tests.}
\label{fig:qqplots_week12_1}
\end{figure*}
\begin{figure*}
\centering
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q16_W12.pdf}
\caption*{S16}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q17_W12.pdf}
\caption*{S17}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q18_W12.pdf}
\caption*{S18}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q19_W12.pdf}
\caption*{S19}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q20_W12.pdf}
\caption*{S20}
\end{subfigure}
\hfill
%%%%%%% ROW 5
% \vspace{12pt}
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q21_W12.pdf}
\caption*{S21}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q22_W12.pdf}
\caption*{S22}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q23_W12.pdf}
\caption*{S23}
\end{subfigure}
\hfill
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q24_W12.pdf}
\caption*{S24}
\end{subfigure}
\hfill
% Last plot alone
\begin{subfigure}[t]{0.30\textwidth}
\centering
\includegraphics[width=\linewidth]{figs/QQPlot_Q25_W12.pdf}
\caption*{S25}
\end{subfigure}
\caption{Q-Q plots for questions S16 to S25 for Week-9. These plots show that none of the question responses follow a normal distribution, validating the use of non-parametric statistical tests.}
\label{fig:qqplots_week12_2}
\end{figure*}
\fi
\appendices
% \balance
% \counterwithin{figure}{section}
% \counterwithin{table}{section}
% =========================================
\section{Q-Q Plots for Week-9} \label{App:qq_plots}
Analysis of the Q-Q plots presented in Fig. \ref{fig:QQ} for student responses to all 25 Week-9 statements indicates that the data for each question deviates from a normal distribution. Due to this non-normality, non-parametric statistical methods, specifically the Kruskal-Wallis and Mann-Whitney U tests, were employed for the Week-9 analysis, as they are well-suited for such data.
% \input{7app_QQ}
\section{Analyses of Initial Weeks} \label{App:weeks_analysis}
For the initial eight weeks, Tables~\ref{tab:week0_analysis}--\ref{tab:week11_analysis} present the week-wise $p$-value analysis using the Mann-Whitney U test for comparing subgroups based on gender (G), and the Kruskal-Wallis test for comparing subgroups based on
academic performance grade (Gd), branch (B), and
prior experiences in C programming (PCE)/
any programming language (PAE)/
computer usage (PCoE)/
smartphone usage (PSE).
Here, $p$-values indicating strong significance ($p \leq 0.05$) are shown in \textcolor{darkred}{\textbf{red}} color.
\input{supp_tables/p_value_0_week}
\input{supp_tables/p_value_3_week}
\input{supp_tables/p_value_6_week}
\input{supp_tables/p_value_7_week}
\input{supp_tables/p_value_8_week}
\input{supp_tables/p_value_9_week}
\input{supp_tables/p_value_10_week}
% \newpage
\input{supp_tables/p_value_11_week}
% \input{7app_QQ}
% =========================================
% \newcommand{\highlight}[1]{\textbf{\textcolor{blue}{#1}}} % Define highlight command
\begin{table}[!b]
\centering
\caption{Analysis of $p$-values for Week-1}
\begin{adjustbox}{width=0.45\textwidth}
\begin{tabular}{ll|c|c|c|c|c|c|c}
\hline
\textbf{SN.} & \textbf{Statement} & \textbf{G} & \textbf{PCE} & \textbf{PAE} & \textbf{PCoE} & \textbf{PSE} & \textbf{Gd} & \textbf{B} \\ \hline
S1: & Learning activity & 0.491 & \textcolor{darkred}{\textbf{0.022}} & 0.335 & 0.170 & 0.751 & 0.947 & 0.974 \\
S2: & Necessary evil & 0.543 & 0.860 & 0.838 & 0.749 & 0.571 & 0.930 & 0.322 \\
S3: & Stressful & 0.829 & \textcolor{darkred}{\textbf{0.036}} & 0.244 & \textcolor{darkred}{\textbf{0.016}} & 0.988 & 0.299 & 0.693 \\
S4: & Help from peers & 0.458 & 0.690 & \textcolor{darkred}{\textbf{0.029}} & 0.279 & 0.221 & \textcolor{darkred}{\textbf{0.004}} & 0.342 \\
S5: & Help to peers & 0.342 & 0.119 & 0.597 & \textcolor{darkred}{\textbf{0.001}} & 0.267 & 0.112 & 0.475 \\ \hline
S6: & Predictable question types & 0.432 & 0.244 & \textcolor{darkred}{\textbf{0.017}} & \textcolor{darkred}{\textbf{0.007}} & 0.565 & 0.339 & 0.134 \\
S7: & Predictable assessment & 0.110 & 0.288 & 0.093 & 0.056 & 0.135 & 0.529 & 0.058 \\
S8: & Skills and knowledge & 0.378 & 0.135 & 0.232 & 0.388 & 0.340 & 0.704 & 0.783 \\
S9: & Authentic & 0.893 & 0.179 & \textcolor{darkred}{\textbf{0.038}} & 0.439 & 0.362 & 0.305 & 0.581 \\ \hline
S10: & Big leap & 0.951 & 0.208 & 0.370 & 0.600 & 0.821 & 0.079 & \textcolor{darkred}{\textbf{0.047}} \\
S11: & Learnt a lot & 0.742 & 0.135 & 0.173 & 0.211 & 0.963 & 0.864 & 0.221 \\
S12: & Authentic & 0.886 & 0.093 & \textcolor{darkred}{\textbf{0.018}} & 0.112 & 0.465 & 0.372 & 0.444 \\
S13: & Difference & 0.589 & 0.123 & \textcolor{darkred}{\textbf{0.023}} & 0.205 & 0.130 & 0.464 & 0.131 \\ \hline
S14: & Course structure & 0.300 & 0.057 & \textcolor{darkred}{\textbf{0.013}} & 0.427 & 0.251 & 0.329 & 0.394 \\
S15: & Little insight & 0.370 & 0.065 & 0.756 & 0.052 & 0.240 & 0.520 & 0.602 \\ \hline
S16: & Assessment & 0.507 & 0.728
& 0.179 & 0.142 & 0.223 & 0.560 & 0.111 \\
S17: & Variation in lab assessment & 0.597 & 0.353 & 0.468 & \textcolor{darkred}{\textbf{0.021}} & 0.668 & 0.909 & 0.229 \\
S18: & Variation for projects & 0.803 & 0.774 & 0.503 & \textcolor{darkred}{\textbf{0.012}} & 0.737 & 0.735 & 0.417 \\
S19: & Variation for help & \textcolor{darkred}{\textbf{0.012}} & 0.113 & 0.554 & \textcolor{darkred}{\textbf{0.004}} & 0.809 & 0.952 & 0.314 \\
S20: & Tutorials & 0.232 & 0.500 & 0.678 & 0.065 & 0.600 & 0.106 & 0.113 \\
S21: & Professionalism & 0.571 & 0.737 & 0.124 & 0.062 & \textcolor{darkred}{\textbf{0.024}} & 0.711 & 0.589 \\
S22: & Feedback & 0.768 & 0.561 & 0.261 & 0.376 & 0.683 & 0.801 & 0.171 \\
S23: & Asking TA help & 0.098 & 0.118 & 0.220 & 0.095 & 0.472 & 0.823 & \textcolor{darkred}{\textbf{0.008}} \\
S24: & Preferred internet & \textcolor{darkred}{\textbf{0.036}} & 0.688 & \textcolor{darkred}{\textbf{0.043}} & 0.183 & 0.281 & 0.152 & 0.563 \\
S25: & Few TAs & \textcolor{darkred}{\textbf{0.023}} & 0.307 & 0.587 & \textcolor{darkred}{\textbf{0.004}} & 0.520 & 0.696 & 0.519 \\ \hline
\end{tabular}
\end{adjustbox}
\label{tab:week0_analysis}
\end{table}
% \newcommand{\highlight}[1]{\textbf{\textcolor{blue}{#1}}} % Define highlight command
\begin{table}[!b]
\centering
\caption{Analysis of $p$-values for Week-2}
\begin{adjustbox}{width=0.45\textwidth}
\begin{tabular}{ll|c|c|c|c|c|c|c}
\hline
\textbf{SN.} & \textbf{Statement} & \textbf{G} & \textbf{PCE} & \textbf{PAE} & \textbf{PCoE} & \textbf{PSE} & \textbf{Gd} & \textbf{B} \\ \hline
S1: & Learning activity & 0.084 & 0.136 & \textcolor{darkred}{\textbf{0.050}} & 0.570 & 0.184 & 0.858 & 0.494 \\
S2: & Necessary evil & 0.871 & \textcolor{darkred}{\textbf{0.023}} & 0.654 & 0.536 & 0.866 & 0.659 & 0.399 \\
S3: & Stressful & 0.715 & 0.239 & \textcolor{darkred}{\textbf{0.070}} & 0.853 & 0.935 & 0.339 & \textcolor{darkred}{\textbf{0.022}} \\
S4: & Help from peers & 0.279 & 0.336 & 0.066 & 0.165 & \textcolor{darkred}{\textbf{0.042}} & 0.070 & 0.734 \\
S5: & Help to peers & 0.505 & 0.486 & 0.331 & \textcolor{darkred}{\textbf{0.006}} & 0.474 & 0.762 & 0.171 \\ \hline
S6: & Predictable question types & 0.279 & 0.466 & \textcolor{darkred}{\textbf{0.001}} & \textcolor{darkred}{\textbf{0.009}} & 0.080 & 0.507 & \textcolor{darkred}{\textbf{0.022}} \\
S7: & Predictable assessment & 0.453 & 0.492 & \textcolor{darkred}{\textbf{0.009}} & 0.150 & 0.184 & 0.481 & 0.087 \\
S8: & Skills and knowledge & 0.105 & 0.766 & 0.548 & 0.782 & 0.557 & 0.774 & 0.986 \\
S9: & Authentic & 0.183 & 0.090 & \textcolor{darkred}{\textbf{0.006}} & 0.623 & \textcolor{darkred}{\textbf{0.026}} & 0.301 & 0.600 \\ \hline
S10: & Big leap & 0.913 & 0.058 & 0.829 & 0.939 & 0.192 & 0.262 & \textcolor{darkred}{\textbf{0.043}} \\
S11: & Learnt a lot & \textcolor{darkred}{\textbf{0.030}} & 0.644 & 0.524 & 0.931 & 0.684 & 0.480 & 0.149 \\
S12: & Authentic & 0.807 & 0.398 & 0.427 & 0.636 & 0.894 & 0.195 & 0.467 \\
S13: & Difference & 0.738 & 0.415 & 0.447 & 0.201 & 0.590 & 0.065 & 0.077 \\ \hline
S14: & Course structure & 0.508 & \textcolor{darkred}{\textbf{0.040}} & 0.744 & 0.123 & 0.468 & 0.367 & 0.825 \\
S15: & Little insight & 0.955 & 0.338 & 0.060 & 0.323 & 0.715 & \textcolor{darkred}{\textbf{0.003}} & \textcolor{darkred}{\textbf{0.038}} \\ \hline
S16: & Assessment & 0.633 & 0.770 & 0.944 & 0.904 & 0.239 & 0.716 & \textcolor{darkred}{\textbf{0.039}} \\
S17: & Variation in lab assessment & 0.263 & 0.417 & 0.871 & 0.465 & 0.519 & 0.871 & 0.141 \\
S18: & Variation for project & 0.272 & 0.143 & 0.569 & 0.260 & 0.171 & 0.636 & 0.092 \\
S19: & Variation for help & 0.873 & 0.142 & 0.715 & 0.367 & 0.822 & 0.532 & 0.139 \\
S20: & Tutorials & 0.624 & 0.224 & 0.115 & 0.880 & 0.827 & 0.558 & 0.110 \\
S21: & Professionalism & 0.755 & 0.516 & 0.428 & 0.062 & \textcolor{darkred}{\textbf{0.003}} & 0.551 & 0.142 \\
S22: & Feedback & 0.745 & 0.400 & 0.691 & 0.364 & 0.084 & 0.846 & 0.676 \\
S23: & Asking TA help & 0.151 & 0.470 & 0.776 & 0.425 &0.055 & 0.854 & 0.979 \\
S24: & Preferred internet& 0.987 & 0.512 & 0.051 & \textcolor{darkred}{\textbf{0.046}} & 0.124 & 0.421 & 0.350 \\
S25: & Few TAs & 0.564 & 0.506 & 0.885 & 0.216 & 0.897 & \textcolor{darkred}{\textbf{0.026}} & 0.087 \\ \hline
\end{tabular}
\label{tab:week3_analysis}
\end{adjustbox}
\end{table}
\begin{table}[]
\centering
\caption{Analysis of $p$-values for Week-3}
\begin{adjustbox}{width=0.45\textwidth}
\begin{tabular}{ll|c|c|c|c|c|c|c}
\hline
\textbf{SN.} & \textbf{Statement} & \textbf{G} & \textbf{PCE} & \textbf{PAE} & \textbf{PCoE} & \textbf{PSE} & \textbf{Gd} & \textbf{B} \\
\hline
S1: & Learning activity & 0.591 & \textcolor{darkred}{\textbf{0.033}} & \textcolor{darkred}{\textbf{0.003}} & 0.131 & 0.579 & 0.261 & 0.615 \\
S2: & Necessary evil & 0.197 & 0.069 & 0.599 & \textcolor{darkred}{\textbf{0.010}} & 0.177 & 0.490 & \textcolor{darkred}{\textbf{0.028}} \\
S3: & Stressful & 0.449 & 0.092 & \textcolor{darkred}{\textbf{<0.001}} & 0.057 & 0.162 & 0.159 & 0.497 \\
S4: & Help from peers & 0.570 & 0.246 & \textcolor{darkred}{\textbf{0.010}} & 0.195 & 0.249 & 0.128 & \textcolor{darkred}{\textbf{0.050}} \\
S5: & Help to peers & 0.886 & 0.658 & 0.144 & 0.406 & 0.372 & \textcolor{darkred}{\textbf{0.009}} & 0.943 \\
\hline
S6: & Predictable question types & 0.618 & 0.214 & 0.280 & 0.513 & 0.178 & 0.840 & 0.114 \\
S7: & Predictable assessment & 0.547 & 0.316 & 0.433 & 0.326 & 0.094 & 0.309 & 0.064 \\
S8: & Skills and knowledge & 0.487 & 0.172 & \textcolor{darkred}{\textbf{0.010}} & 0.895 & 0.437 & 0.780 & 0.571 \\
S9: & Authentic & 0.812 & 0.232 & \textcolor{darkred}{\textbf{<0.001}} & 0.241 & 0.214 & 0.455 & 0.183 \\
\hline
S10: & Big leap & 0.550 & \textcolor{darkred}{\textbf{0.026}} & 0.483 & 0.197 & 0.326 & 0.769 & 0.286 \\
S11: & Learnt a lot & 0.420 & \textcolor{darkred}{\textbf{0.013}} & \textcolor{darkred}{\textbf{0.020}} & \textcolor{darkred}{\textbf{0.048}} & 0.373 & 0.696 & 0.453 \\
S12: & Authentic & 0.835 & \textcolor{darkred}{\textbf{0.037}} & \textcolor{darkred}{\textbf{0.028}} & \textcolor{darkred}{\textbf{0.048}} & 0.307 & 0.832 & 0.076 \\
S13: & Difference & 0.852 & 0.080 & 0.340 & 0.757 & 0.623 & 0.149 & 0.077 \\
\hline
S14: & Course structure & 0.913 & 0.308 & \textcolor{darkred}{\textbf{<0.001}} & \textcolor{darkred}{\textbf{0.009}} & 0.260 & 0.179 & 0.629 \\
S15: & Little insight & 0.483 & 0.374 & \textcolor{darkred}{\textbf{0.004}} & \textcolor{darkred}{\textbf{0.032}} & 0.380 & 0.166 & \textcolor{darkred}{\textbf{0.044}} \\
\hline
S16: & Assessment & 0.115 & 0.108 & 0.071 & 0.274 & 0.107 & 0.295 & 0.651 \\
S17: & Variation in lab assessment & 0.675 & 0.436 & 0.769 & 0.949 & 0.677 & 0.416 & 0.292 \\
S18: & Variation for project & 0.746 & 0.325 & 1.000 & 0.666 & 0.781 & 0.546 & 0.084 \\
S19: & Variation for help & 0.674 & 0.297 & 0.608 & 0.699 & 0.726 & 0.813 & 0.897 \\
S20: & Tutorials & 0.746 & 0.941 & 0.789 & \textcolor{darkred}{\textbf{0.027}} & 0.696 & 0.257 & 0.802 \\
S21: & Professionalism & 0.946 & 0.676 & 0.226 & 0.209 & \textcolor{darkred}{\textbf{0.013}} & 0.194 & 0.230 \\
S22: & Feedback & 0.595 & 0.185 & 0.055 & 0.100 & 0.412 & 0.321 & 0.706 \\
S23: & Asking TA help & 0.251 & 0.852 & 0.360 & 0.737 & 0.743 & 0.454 & 0.412 \\
S24: & Preferred internet & 0.311 & 0.964 & 0.213 & 0.226 & 0.125 & 0.760 & 0.197 \\
S25: & Few TAs & 0.516 & 0.570 & 0.149 & \textcolor{darkred}{\textbf{0.020}} & 0.689 & 0.292 & 0.189 \\
\hline
\end{tabular}
\label{tab:week6_analysis}
\end{adjustbox}
\end{table}
\begin{table}[]
\centering
\caption{Analysis of $p$-values for Week-4}
\begin{adjustbox}{width=0.45\textwidth}
\begin{tabular}{ll|c|c|c|c|c|c|c}
\hline
\textbf{SN.} & \textbf{Statement} & \textbf{G} & \textbf{PCE} & \textbf{PAE} & \textbf{PCoE} & \textbf{PSE} & \textbf{Gd} & \textbf{B} \\
\hline
S1: & Learning activity & 0.424 & 0.285 & \textcolor{darkred}{\textbf{0.031}} & 0.171 & 0.491 & 0.734 & 0.644 \\
S2: & Necessary evil & 0.987 & 0.117 & 0.052 & 0.559 & 0.710 & 0.520 & 0.228 \\
S3: & Stressful & 0.440 & 0.141 & \textcolor{darkred}{\textbf{<0.001}} & 0.883 & 0.144 & 0.844 & 0.515 \\
S4: & Help from peers & \textcolor{darkred}{\textbf{0.005}} & 0.270 & \textcolor{darkred}{\textbf{0.005}} & 0.374 & 0.280 & 0.167 & 0.630 \\
S5: & Help to peers & 0.673 & 0.718 & \textcolor{darkred}{\textbf{0.047}} & 0.060 & 0.135 & 0.271 & 0.710 \\
\hline
S6: & Predictable question types & 0.245 & 0.571 & 0.751 & 0.541 & 0.131 & 0.764 & 0.542 \\
S7: & Predictable assessment & 0.098 & 0.331 & 0.312 & 0.639 & 0.103 & 0.971 & 0.218 \\
S8: & Skills and knowledge & 0.496 & 0.441 & \textcolor{darkred}{\textbf{0.015}} & 0.131 & 0.778 & 0.277 & \textcolor{darkred}{\textbf{0.006}} \\
S9: & Authentic & 0.923 & 0.842 & 0.088 & 0.262 & 0.906 & \textcolor{darkred}{\textbf{0.016}} & 0.141 \\
\hline
S10: & Big leap & 0.342 & 0.385 & 0.939 & 0.489 & 0.676 & 0.808 & 0.276 \\
S11: & Learnt a lot & 0.759 & 0.648 & 0.169 & 0.473 & 0.764 & 0.378 & 0.824 \\
S12: & Authentic & 0.469 & \textcolor{darkred}{\textbf{0.049}} & \textcolor{darkred}{\textbf{0.009}} & 0.347 & 0.698 & \textcolor{darkred}{\textbf{0.026}} & 0.396 \\
S13: & Difference & 0.930 & \textcolor{darkred}{\textbf{0.036}} & 0.327 & 0.791 & 0.167 & 0.533 & 0.759 \\
\hline
S14: & Course structure & 0.860 & 0.549 & 0.122 & 0.452 & 0.805 & 0.213 & 0.497 \\
S15: & Little insight & 0.318 & 0.632 & \textcolor{darkred}{\textbf{0.008}} & 0.312 & 0.916 & 0.089 & 0.597 \\
\hline
S16: & Assessment & 0.628 & \textcolor{darkred}{\textbf{0.023}} & 0.674 & 0.487 & 0.718 & 0.399 & 0.323 \\
S17: & Variation in lab assessment & 0.213 & 0.131 & 0.759 & 0.487 & 0.853 & 0.790 & 0.695 \\
S18: & Variation for project & 0.072 & 0.448 & 0.583 & 0.703 & 0.666 & 0.833 & 0.864 \\
S19: & Variation for help & 0.377 & 0.087 & 0.374 & 0.415 & 0.601 & 0.509 & \textcolor{darkred}{\textbf{0.047}} \\
S20: & Tutorials & 0.539 & 0.053 & 0.691 & 0.206 & 0.425 & 0.950 & 0.400 \\
S21: & Professionalism & 0.531 & 0.667 & \textcolor{darkred}{\textbf{0.006}} & \textcolor{darkred}{\textbf{0.028}} & \textcolor{darkred}{\textbf{0.029}} & 0.194 & \textcolor{darkred}{\textbf{0.013}} \\
S22: & Feedback & \textcolor{darkred}{\textbf{0.018}} & 0.365 & 0.218 & 0.449 & 0.815 & 0.839 & 0.663 \\
S23: & Asking TA help & 0.503 & 0.712 & 0.645 & 0.691 & 0.719 & 0.239 & 0.095 \\
S24: & Preferred internet & \textcolor{darkred}{\textbf{0.024}} & 0.980 & 0.138 & 0.244 & 0.137 & 0.880 & 0.881 \\
S25: & Few TAs & 0.257 & 0.867 & \textcolor{darkred}{\textbf{0.030}} & 0.059 & 0.373 & 0.789 & 0.900 \\
\hline
\end{tabular}
\label{tab:week7_analysis}
\end{adjustbox}
\end{table}
\begin{table}[]
\centering
\caption{Analysis of $p$-values for Week-5}
\begin{adjustbox}{width=0.45\textwidth}
\begin{tabular}{ll|c|c|c|c|c|c|c}
\hline
\textbf{SN.} & \textbf{Statement} & \textbf{G} & \textbf{PCE} & \textbf{PAE} & \textbf{PCoE} & \textbf{PSE} & \textbf{Gd} & \textbf{B} \\ \hline
S1: & Learning activity & 0.415 & 0.380 & \textcolor{darkred}{\textbf{0.002}} & 0.061 & 0.503 & 0.777 & 0.395 \\
S2: & Necessary evil & 0.142 & 0.881 & 0.615 & 0.900 & 0.641 & 0.157 & 0.067 \\
S3: & Stressful & 0.866 & 0.608 & 0.488 & 0.898 & 0.610 & 0.504 & 0.709 \\
S4: & Help from peers & 0.113 & 0.578 & 0.108 & 0.118 & 0.549 & 0.054 & 0.345 \\
S5: & Help to peers & 0.197 & 0.350 & 0.217 & 0.190 & 0.230 & 0.160 & 0.602 \\ \hline
S6: & Predictable question types & 0.059 & 0.313 & \textcolor{darkred}{\textbf{0.029}} & \textcolor{darkred}{\textbf{0.035}} & \textcolor{darkred}{\textbf{0.011}} & 0.591 & 0.531 \\
S7: & Predictable assessment & 0.313 & 0.648 & 0.689 & 0.849 & 0.074 & 0.374 & 0.474 \\
S8: & Skills and knowledge & 0.583 & 0.560 & 0.302 & 0.127 & 0.126 & 0.521 & 0.495 \\
S9: & Authentic & 0.621 & 0.321 & 0.125 & 0.745 & 0.626 & 0.533 & 0.087 \\ \hline
S10: & Big leap & 0.785 & 0.150 & 0.406 & 0.543 & 0.941 & 0.902 & 0.104 \\
S11: & Learnt a lot & 0.677 & 0.676 & 0.273 & 0.281 & 0.358 & 0.808 & 0.069 \\
S12: & Authentic & 0.867 & 0.348 & \textcolor{darkred}{\textbf{0.003}} & \textcolor{darkred}{\textbf{0.020}} & 0.867 & 0.621 & 0.116 \\
S13: & Difference & 0.338 & 0.791 & 0.452 & 0.197 & 0.281 & \textcolor{darkred}{\textbf{0.033}} & \textcolor{darkred}{\textbf{0.021}} \\ \hline
S14: & Course structure & 0.857 & 0.386 & 0.142 & 0.059 & 0.383 & 0.582 & 0.670 \\
S15: & Little insight & 0.391 & 0.498 & \textcolor{darkred}{\textbf{0.017}} & 0.092 & 0.201 & \textcolor{darkred}{\textbf{0.005}} & 0.147 \\ \hline
S16: & Assessment & 0.131 & \textcolor{darkred}{\textbf{0.030}} & 0.736 & 0.438 & 0.158 & 0.734 & 0.313 \\
S17: & Variation in lab assessment & 0.187 & 0.062 & 0.919 & 0.784 & 0.999 & 0.985 & 0.344 \\
S18: & Variation for project & 0.690 & 0.639 & 0.655 & 0.635 & 0.683 & 0.895 & 0.628 \\
S19: & Variation for help & 0.463 & 0.872 & 0.940 & 0.785 & 0.283 & 0.843 & 0.263 \\
S20: & Tutorials & 0.791 & 0.104 & 0.184 & 0.305 & 0.122 & 0.822 & \textcolor{darkred}{\textbf{0.018}} \\
S21: & Professionalism & 0.853 & 0.526 & 0.206 & \textcolor{darkred}{\textbf{0.026}} & \textcolor{darkred}{\textbf{0.037}} & 0.324 & \textcolor{darkred}{\textbf{0.040}} \\
S22: & Feedback & 0.905 & 0.845 & 0.438 & 0.288 & 0.933 & 0.836 & 0.162 \\
S23: & Asking TA help & 0.932 & 0.920 & 0.360 & 0.346 & 0.240 & 0.366 & 0.276 \\
S24: & Preferred internet & 0.487 & 0.698 & 0.327 & 0.155 & 0.105 & 0.641 & 0.813 \\
S25: & Few TAs & 0.431 & 0.944 & 0.599 & 0.194 & 0.924 & 0.424 & 0.619 \\ \hline
\end{tabular}
\label{tab:week8_analysis}
\end{adjustbox}
\end{table}
\begin{table}[]
\centering
\caption{Analysis of $p$-values for Week-6}
\begin{adjustbox}{width=0.45\textwidth}
\begin{tabular}{ll|c|c|c|c|c|c|c}
\hline
\textbf{SN.} & \textbf{Statement} & \textbf{G} & \textbf{PCE} & \textbf{PAE} & \textbf{PCoE} & \textbf{PSE} & \textbf{Gd} & \textbf{B} \\
\hline
S1: & Learning activity & 0.078 & 0.575 & 0.704 & 0.397 & 0.238 & 0.619 & 0.813 \\
S2: & Necessary evil & 0.240 & 0.236 & 0.511 & 0.403 & 0.856 & 0.781 & 0.283 \\
S3: & Stressful & 0.807 & 0.843 & 0.824 & \textcolor{darkred}{\textbf{0.010}} & 0.662 & 0.959 & 0.611 \\
S4: & Help from peers & 0.170 & 0.300 & 0.388 & 0.295 & 0.712 & \textcolor{darkred}{\textbf{0.003}} & 0.301 \\
S5: & Help to peers & 0.649 & 0.948 & 0.678 & 0.750 & 0.760 & 0.586 & 0.296 \\ \hline
S6: & Predictable question types & \textcolor{darkred}{\textbf{0.020}} & 0.299 & 0.085 & 0.938 & 0.689 & 0.179 & 0.090 \\
S7: & Predictable assessment & 0.075 & 0.307 & 0.082 & 0.847 & 0.219 & 0.614 & 0.531 \\
S8: & Skills and knowledge & 0.517 & 0.060 & 0.396 & 0.452 & \textcolor{darkred}{\textbf{0.019}} & 0.353 & 0.526 \\
S9: & Authentic & 0.405 & 0.309 & 0.125 & 0.765 & \textcolor{darkred}{\textbf{0.009}} & 0.415 & 0.901 \\ \hline
S10: & Big leap & \textcolor{darkred}{\textbf{0.018}} & 0.803 & 0.427 & 0.851 & 0.484 & 0.517 & 0.630 \\
S11: & Learnt a lot & 0.784 & 0.565 & 0.866 & 0.314 & 0.104 & 0.520 & 0.695 \\
S12: & Authentic & 0.858 & 0.063 & 0.158 & 0.212 & 0.393 & 0.573 & 0.672 \\
S13: & Difference & 0.295 & 0.588 & 0.929 & 0.127 & 0.140 & 0.370 & 0.604 \\ \hline
S14: & Course structure & 0.175 & 0.239 & 0.501 & 0.366 & 0.562 & 0.801 & 0.636 \\
S15: & Little insight & 0.075 & 0.683 & 0.203 & 0.377 & 0.270 & 0.239 & 0.797 \\ \hline
S16: & Assessment & 0.471 & 0.179 & 0.272 & 0.344 & 0.795 & 0.180 & 0.592 \\
S17: & Variation in lab assessment & 0.152 & 0.778 & \textcolor{darkred}{\textbf{0.033}} & 0.221 & \textcolor{darkred}{\textbf{0.046}} & 0.760 & 0.263 \\
S18: & Variation for project & \textcolor{darkred}{\textbf{0.012}} & 0.130 & 0.726 & 0.427 & 0.134 & 0.925 & 0.696 \\
S19: & Variation for help & 0.410 & 0.330 & 0.118 & 0.546 & 0.569 & 0.860 & 0.622 \\
S20: & Tutorials & \textcolor{darkred}{\textbf{0.038}} & 0.622 & 0.437 & 0.062 & 0.058 & 0.812 & 0.556 \\
S21: & Professionalism & 0.251 & 0.492 & 0.356 & 0.880 & 0.812 & 0.962 & 0.100 \\
S22: & Feedback & 0.304 & 0.265 & \textcolor{darkred}{\textbf{0.020}} & 0.258 & \textcolor{darkred}{\textbf{0.046}} & 0.887 & 0.899 \\
S23: & Asking TA help & 0.057 & \textcolor{darkred}{\textbf{0.002}} & 0.916 & 0.764 & \textcolor{darkred}{\textbf{0.045}} & 0.715 & 0.684 \\
S24: & Preferred internet & 0.639 & 0.578 & \textcolor{darkred}{\textbf{0.005}} & 0.118 & 0.063 & 0.884 & 0.226 \\
S25: & Few TAs & 0.099 & 0.120 & 0.184 & 0.753 & 0.686 & 0.962 & 0.912 \\ \hline
\end{tabular}
\label{tab:week9_analysis}
\end{adjustbox}
\end{table}
\begin{table}[]
\centering
\caption{Analysis of $p$-values for Week-7}
\begin{adjustbox}{width=0.45\textwidth}
\begin{tabular}{ll|c|c|c|c|c|c|c}
\hline
\textbf{SN.} & \textbf{Statement} & \textbf{G} & \textbf{PCE} & \textbf{PAE} & \textbf{PCoE} & \textbf{PSE} & \textbf{Gd} & \textbf{B} \\ \hline
S1: & Learning activity & 0.788 & 0.231 & 0.568 & \textcolor{darkred}{\textbf{0.003}} & 0.086 & 0.149 & 0.984 \\
S2: & Necessary evil & 0.714 & \textcolor{darkred}{\textbf{0.039}} & 0.537 & 0.207 & 0.581 & 0.199 & 0.246 \\
S3: & Stressful & 0.438 & 0.561 & 0.253 & 0.944 & 0.712 & 0.399 & 0.878 \\
S4: & Help from peers & 0.347 & 0.125 & 0.358 & 0.118 & 0.707 & 0.118 & 0.183 \\
S5: & Help to peers & 0.799 & 0.777 & 0.084 & 0.436 & 0.893 & 0.077 & 0.348 \\ \hline
S6: & Predictable question types & 0.562 & 0.691 & 0.663 & 0.575 & 0.126 & 0.060 & 0.813 \\
S7: & Predictable assessment & 0.687 & 0.780 & 0.323 & 0.834 & 0.132 & 0.826 & 0.679 \\
S8: & Skills and knowledge & 0.471 & 0.892 & \textcolor{darkred}{\textbf{0.047}} & 0.728 & 0.484 & 0.596 & 0.592 \\
S9: & Authentic & 0.313 & 0.574 & 0.097 & 0.506 & 0.397 & 0.083 & 0.543 \\ \hline
S10: & Big leap & 0.942 & 0.154 & 0.971 & 0.506 & 0.714 & 0.638 & 0.291 \\
S11: & Learnt a lot & 0.231 & \textcolor{darkred}{\textbf{0.023}} & 0.214 & 0.675 & \textcolor{darkred}{\textbf{0.014}} & \textcolor{darkred}{\textbf{0.027}} & 0.435 \\
S12: & Authentic & 0.544 & \textcolor{darkred}{\textbf{0.041}} & 0.281 & 0.096 & 0.980 & 0.273 & 0.681 \\
S13: & Difference & 0.897 & 0.391 & 0.605 & 0.781 & 0.334 & 0.837 & 0.098 \\ \hline
S14: & Course structure & 0.874 & 0.170 & 0.456 & \textcolor{darkred}{\textbf{0.003}} & 0.196 & 0.531 & 0.413 \\
S15: & Little insight & 0.752 & 0.319 & \textcolor{darkred}{\textbf{0.014}} & \textcolor{darkred}{\textbf{0.008}} & \textcolor{darkred}{\textbf{0.041}} & \textcolor{darkred}{\textbf{0.008}} & 0.512 \\ \hline
S16: & Assessment & 0.485 & 0.167 & 0.799 & 0.657 & 0.487 & 0.623 & 0.596 \\
S17: & Variation in lab assessment & 0.304 & 0.190 & 0.124 & 0.384 & 0.530 & 0.701 & 0.790 \\
S18: & Variation for project & 0.941 & 0.115 & 0.925 & 0.840 & 0.922 & 0.665 & 0.625 \\
S19: & Variation for help & 0.851 & 0.228 & 0.961 & 0.256 & 0.205 & 0.322 & 0.095 \\
S20: & Tutorials & 0.325 & 0.164 & 0.930 & 0.788 & 0.926 & 0.075 & 0.058 \\
S21: & Professionalism & 0.949 & 0.978 & 0.361 & 0.197 & 0.898 & 0.614 & 0.812 \\
S22: & Feedback & 0.351 & 0.215 & 0.152 & 0.193 & 0.070 & 0.241 & 0.950 \\
S23: & Asking TA help & 0.432 & \textcolor{darkred}{\textbf{0.039}} & 0.746 & 0.875 & 0.298 & 0.747 & \textcolor{darkred}{\textbf{0.044}} \\
S24: & Preferred internet & 0.686 & 0.070 & 0.358 & 0.175 & 0.263 & 0.780 & 0.974 \\
S25: & Few TAs & 0.663 & 0.078 & 0.252 & 0.136 & 0.367 & 0.443 & 0.175 \\ \hline
\end{tabular}
\label{tab:week10_analysis}
\end{adjustbox}
\end{table}
% \newpage
\begin{table}[]
\centering
\caption{Analysis of $p$-values for Week-8}
\begin{adjustbox}{width=0.45\textwidth}
\begin{tabular}{ll|c|c|c|c|c|c|c}
\hline
\textbf{SN.} & \textbf{Statement} & \textbf{G} & \textbf{PCE} & \textbf{PAE} & \textbf{PCoE} & \textbf{PSE} & \textbf{Gd} & \textbf{B} \\ \hline
S1: & Learning activity & 0.927 & 0.947 & 0.230 & 0.118 & 0.087 & 0.081 & 0.938 \\
S2: & Necessary evil & 0.635 & 0.305 & 0.820 & 0.347 & 0.166 & 0.534 & 0.120 \\
S3: & Stressful & 0.054 & 0.105 & 0.746 & 0.078 & 0.554 & 0.547 & 0.655 \\
S4: & Help from peers & 0.254 & \textcolor{darkred}{\textbf{0.040}} & 0.595 & 0.678 & 0.366 & \textcolor{darkred}{\textbf{0.022}} & 0.877 \\
S5: & Help to peers & 0.572 & 0.958 & 0.630 & 0.144 & 0.294 & 0.519 & 0.983 \\ \hline
S6: & Predictable question types & 0.640 & 0.655 & 0.568 & 0.654 & 0.673 & \textcolor{darkred}{\textbf{0.045}} & 0.251 \\
S7: & Predictable assessment & 0.299 & 0.564 & 0.836 & 0.783 & 0.938 & 0.113 & 0.563 \\
S8: & Skills and knowledge & 0.244 & 0.444 & 0.739 & 0.385 & 0.407 & 0.456 & 0.712 \\
S9: & Authentic & 0.578 & 0.677 & 0.463 & 0.172 & 0.237 & 0.224 & \textcolor{darkred}{\textbf{0.027}} \\ \hline
S10: & Big leap & 0.109 & 0.259 & 0.332 & 0.415 & \textcolor{darkred}{\textbf{0.037}} & 0.362 & 0.737 \\
S11: & Learnt a lot & 0.661 & 0.610 & 0.342 & 0.593 & \textcolor{darkred}{\textbf{0.034}} & 0.409 & 0.375 \\
S12: & Authentic & 0.880 & 0.156 & 0.590 & 0.113 & 0.341 & 0.691 & 0.504 \\
S13: & Difference & 0.071 & 0.508 & 0.615 & 0.223 & 0.574 & 0.178 & 0.850 \\ \hline
S14: & Course structure & 0.955 & 0.285 & 0.101 & 0.177 & 0.100 & 0.818 & 0.525 \\
S15: & Little insight & 0.236 & 0.973 & \textcolor{darkred}{\textbf{0.023}} & 0.157 & 0.654 & \textcolor{darkred}{\textbf{0.014}} & \textcolor{darkred}{\textbf{0.048}} \\ \hline
S16: & Assessment & 0.760 & 0.325 & 0.632 & 0.199 & 0.510 & 0.681 & 0.980 \\
S17: & Variation in lab assessment & 0.893 & 0.424 & 0.671 & 0.968 & 0.571 & 0.863 & 0.567 \\
S18: & Variation for project & 0.617 & 0.129 & 0.267 & 0.220 & 0.751 & 0.255 & 0.527 \\
S19: & Variation for help & 0.847 & 0.313 & 0.449 & 0.450 & 0.423 & 0.162 & 0.362 \\
S20: & Tutorials & 0.332 & 0.561 & 0.420 & 0.628 & 0.474 & 0.055 & \textcolor{darkred}{\textbf{0.043}} \\
S21: & Professionalism & 0.950 & 0.842 & 0.933 & 0.852 & 0.560 & 0.273 & 0.245 \\
S22: & Feedback & 0.391 & 0.958 & 0.420 & 0.940 & 0.129 & 0.573 & 0.598 \\
S23: & Asking TA help & 0.195 & 0.063 & 0.769 & 0.634 & 0.531 & 0.700 & 0.507 \\
S24: & Preferred internet & 0.120 & 0.374 & 0.062 & 0.257 & 0.093 & 0.811 & 0.767 \\
S25: & Few TAs & 0.952 & \textcolor{darkred}{\textbf{0.035}} & 0.379 & 0.851 & 0.114 & 0.429 & 0.308 \\ \hline
\end{tabular}
\label{tab:week11_analysis}
\end{adjustbox}
\end{table}
% Generated by IEEEtran.bst, version: 1.12 (2007/01/11)
\begin{thebibliography}{10}
\providecommand{\url}[1]{#1}
\csname url@samestyle\endcsname
\providecommand{\newblock}{\relax}
\providecommand{\bibinfo}[2]{#2}
\providecommand{\BIBentrySTDinterwordspacing}{\spaceskip=0pt\relax}
\providecommand{\BIBentryALTinterwordstretchfactor}{4}
\providecommand{\BIBentryALTinterwordspacing}{\spaceskip=\fontdimen2\font plus
\BIBentryALTinterwordstretchfactor\fontdimen3\font minus \fontdimen4\font\relax}
\providecommand{\BIBforeignlanguage}[2]{{%
\expandafter\ifx\csname l@#1\endcsname\relax
\typeout{** WARNING: IEEEtran.bst: No hyphenation pattern has been}%
\typeout{** loaded for the language `#1'. Using the pattern for}%
\typeout{** the default language instead.}%
\else
\language=\csname l@#1\endcsname
\fi
#2}}
\providecommand{\BIBdecl}{\relax}
\BIBdecl
\bibitem{10.1145/3293881.3295779}
A.~Luxton-Reilly \emph{et~al.}, ``Introductory programming: a systematic literature review,'' in \emph{ITiCSE}, 2018, p. 55–106.
\bibitem{article1}
R.~Medeiros \emph{et~al.}, ``A systematic literature review on teaching and learning introductory programming in higher education,'' \emph{IEEE ToE}, pp. 1--14, 2018.
\bibitem{article}
L.~Barker \emph{et~al.}, ``Making visible the behaviors that influence learning environment: A qualitative exploration of computer science classrooms,'' \emph{Computer Science Education}, vol.~14, pp. 119--145, 2004.
\bibitem{gaultney2001grade}
J.~F. Gaultney and A.~Cann, ``Grade expectations,'' \emph{Teaching of Psychology}, vol.~28, no.~2, pp. 84--87, 2001.
\bibitem{arthurs2019grades}
N.~Arthurs \emph{et~al.}, ``Grades are not normal: Improving exam score models using the logit-normal distribution.'' in \emph{EDM}, 2019, pp. 1--6.
\bibitem{10036401}
E.~Riese and S.~Stenbom, ``Engineering students’ experiences of assessment in introductory computer science courses,'' \emph{IEEE ToE}, vol.~66, no.~4, pp. 350--359, 2023.
\bibitem{10.1145/3517134}
E.~Riese \emph{et~al.}, ``A qualitative study of experienced course coordinators’ perspectives on assessment in introductory programming courses for non-cs majors,'' \emph{ACM TOCE}, vol.~22, no.~4, 2022.
\bibitem{qualitative_step}
------, ``Experiences of assessment in introductory programming from the perspective of noncomputer science majors,'' in \emph{FIE}, 2020, pp. 1--9.
\bibitem{cambridge}
S.~A. Fincher \emph{et~al.}, \emph{The Cambridge handbook of computing education research}.\hskip 1em plus 0.5em minus 0.4em\relax Cambridge University Press, 2019.
\bibitem{10.1145/1565799.1565820}
J.~Hundley \emph{et~al.}, ``Engaging students in software development course projects,'' in \emph{Tapia}, 2009, p. 87–92.
\bibitem{10.1145/961290.961519}
J.~P. Barros \emph{et~al.}, ``Using lab exams to ensure programming practice in an introductory programming course,'' \emph{ACM SIGCSE}, vol.~35, no.~3, pp. 16--20, 2003.
\bibitem{articleMCQ}
P.~Henriques~Abreu \emph{et~al.}, ``Multiple-choice questions in programming courses: Can we use them and are students motivated by them?'' \emph{ACM TOCE}, vol.~19, pp. 1--16, 11 2018.
\bibitem{Bennedsen01102007}
J.~Bennedsen \emph{et~al.}, ``Assessing process and product: A practical lab exam for an introductory programming course,'' \emph{ITALICS}, vol.~6, no.~4, pp. 183--202, 2007.
\bibitem{articleL}
A.~Bellino \emph{et~al.}, ``A real-world approach to motivate students on the first class of a computer science course,'' \emph{ACM TOCE}, vol.~21, pp. 1--23, 2021.
\bibitem{7062585}
A.~Cain, ``Factors influencing student learning in portfolio assessed introductory programming,'' in \emph{TALE}, 2014, pp. 55--62.
\bibitem{ZPD_1}
L.~S. Vygotsky, \emph{Mind in society: The development of higher psychological processes}.\hskip 1em plus 0.5em minus 0.4em\relax Harvard university press, 1978, vol.~86.
\bibitem{TAs1}
E.~Riese, ``Teaching assistants’ experiences of lab sessions in introductory computer science courses,'' in \emph{FIE}, 2018, pp. 1--5.
\bibitem{10.1145/3291279.3339418}
Y.~Ren \emph{et~al.}, ``What help do students seek in ta office hours?'' in \emph{ACM ICER}, 2019, p. 41–49.
\bibitem{TA_gradning}
K.~Willey and A.~Gardner, ``Perceived differences in tutor grading in large classes: Fact or fiction?'' in \emph{FIE}, 2010, pp. S2G--1--S2G--6.
\bibitem{10.1145/2999541.2999557}
K.~Nolan \emph{et~al.}, ``The role of anxiety when learning to program: a systematic review of the literature,'' in \emph{Koli Calling}, 2016, p. 61–70.
\bibitem{female_3}
V.~Barr, ``Different denominators, different results: reanalyzing {CS} degrees by gender, race, and ethnicity,'' \emph{ACM Inroads}, vol.~9, no.~3, pp. 40--47, 2018.
\bibitem{10.1145/2729094.2742611}
J.~Wang \emph{et~al.}, ``Gender differences in factors influencing pursuit of computer science and related fields,'' in \emph{ITiCSE}, 2015, p. 117–122.
\bibitem{inproceedingsF}
E.~Winter \emph{et~al.}, ``{'It's a Bit Weird, but it's OK'? How Female Computer Science Students Navigate being a Minority},'' in \emph{ITiCSE}, 2021, pp. 436--442.
\bibitem{4271597}
J.~C. Carver \emph{et~al.}, ``Increased retention of early computer science and software engineering students using pair programming,'' in \emph{CSEET}, 2007, pp. 115--122.
\bibitem{10.1145/3488042.3488052}
A.~Doebling \emph{et~al.}, ``Patterns of academic help-seeking in undergraduate computing students,'' in \emph{Koli Calling}, 2021.
\bibitem{theory_perspective}
M.~Taras, ``Assessment--summative and formative--some theoretical reflections,'' \emph{British journal of educational studies}, vol.~53, no.~4, pp. 466--478, 2005.
\bibitem{collab_perspective_1}
A.~S. Palincsar, ``Social constructivist perspectives on teaching and learning,'' \emph{Annual review of psychology}, vol.~49, no.~1, pp. 345--375, 1998.
\bibitem{collab_perspective}
D.~H. Schunk, \emph{Learning Theories: An Educational Perspective}, 6th~ed.\hskip 1em plus 0.5em minus 0.4em\relax Boston, MA: Pearson Education Inc., 2012.
\bibitem{student_opinion}
C.~B. Nicol \emph{et~al.}, ``Students' opinions, views, and perceptions of science laboratory learning: A systematic review of the literature.'' \emph{Eurasia Journal of Maths., Sc. and Tech. Edu.}, vol.~18, no.~3, 2022.
\bibitem{constructive_alignment}
J.~Biggs, ``Enhancing teaching through constructive alignment,'' \emph{Higher education}, vol.~32, no.~3, pp. 347--364, 1996.
\bibitem{authentic_assessment}
J.~T. Gulikers \emph{et~al.}, ``A five-dimensional framework for authentic assessment,'' \emph{Educational technology research and development}, vol.~52, no.~3, pp. 67--86, 2004.
\bibitem{authentic_assessment_1}
R.~McDermott \emph{et~al.}, ``First year computing students' perceptions of authenticity in assessment,'' in \emph{ITiCSE}, 2017, pp. 10--15.
\bibitem{learning}
E.~Cell, \emph{Learning to learn from experience}.\hskip 1em plus 0.5em minus 0.4em\relax SUNY Press, 1984.
\bibitem{harris2022distributed}
A.~Harris \emph{et~al.}, ``Distributed leadership: taking a retrospective and contemporary view of the evidence base,'' \emph{School Leadership \& Management}, vol.~42, no.~5, pp. 438--456, 2022.
\bibitem{wilson2002study}
B.~C. Wilson, ``A study of factors promoting success in computer science including gender differences,'' \emph{Computer Science Education}, vol.~12, no. 1-2, pp. 141--164, 2002.
\bibitem{TAS2}
E.~Riese \emph{et~al.}, ``Challenges faced by teaching assistants in computer science education across europe,'' in \emph{ITiCSE}, 2021, pp. 547--553.
\bibitem{diff_theory_1}
C.~M. Steele, ``Stereotyping and its threat are real.'' 1998.
\bibitem{diff_theory_2}
C.~Leaper \emph{et~al.}, ``The socialization of gender,'' \emph{Handbook of socialization: Theory and research}, vol. 561, p. 587, 2007.
\bibitem{diff_theory_3}
A.~Bandura \emph{et~al.}, ``Analysis of self-efficacy theory of behavioral change,'' \emph{Cognitive therapy and research}, vol.~1, no.~4, pp. 287--310, 1977.
\bibitem{diff_theory_4}
G.~M. Walton \emph{et~al.}, ``A question of belonging: race, social fit, and achievement.'' \emph{Journal of personality and social psychology}, vol.~92, no.~1, p.~82, 2007.
\bibitem{transfer_learning}
J.~M. Royer, ``Theories of the transfer of learning,'' \emph{Educational psychologist}, vol.~14, no.~1, pp. 53--69, 1979.
\bibitem{cognitive_load}
J.~Sweller, ``Cognitive load during problem solving: Effects on learning,'' \emph{Cognitive science}, vol.~12, no.~2, pp. 257--285, 1988.
\bibitem{cultural_capital}
D.~Throsby, ``Cultural capital,'' \emph{Journal of cultural economics}, vol.~23, pp. 3--12, 1999.
\bibitem{digital_divide}
J.~A. Van~Dijk, ``Digital divide research, achievements and shortcomings,'' \emph{Poetics}, vol.~34, no. 4-5, pp. 221--235, 2006.
\bibitem{assessment_1}
S.~Messick, ``Meaning and values in test validation: The science and ethics of assessment,'' \emph{Educational researcher}, vol.~18, no.~2, pp. 5--11, 1989.
\bibitem{diff_theory_5}
J.~Lave, ``Situating learning in communities of practice,'' in \emph{Perspectives on Socially Shared Cognition}.\hskip 1em plus 0.5em minus 0.4em\relax APA, 1991, pp. 63--82.
\bibitem{exp_theory}
A.~Wigfield \emph{et~al.}, ``Expectancy--value theory of achievement motivation,'' \emph{Contemporary educational psychology}, vol.~25, no.~1, pp. 68--81, 2000.
\end{thebibliography}