Samiha Marwan

Samiha Marwan

Auto-generated profile, based on publicly available information

Samiha Marwan

The University of Virginia, Charlottesville, VA, USA

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 December 2022
Verified
Conference Paper
Open access
JJJohan Jeuring
Johan Jeuring
HKHieke Keuning
Hieke Keuning
SMSamiha Marwan
Samiha Marwan
DBDennis Bouvier
Dennis Bouvier
CICruz Izu
Cruz Izu
SSSami Sarsa
Sami Sarsa
Every year, millions of students learn how to write programs. Learning activities for beginners almost always include programming tasks that require a student to write a program to solve a particular problem. When learning how to solve such a task, many students need feedback on their previous actions, and hints on how to proceed. For tasks such as programming, which are most often solved stepwise, the feedback should take the steps a student has taken towards implementing a solution into account, and the hints should help a student to complete or improve a possibly partial solution. This paper investigates how previous research on feedback is translated to when and how to give feedback and hints on steps a student takes when solving a programming task. We have selected datasets consisting of sequences of steps students take when working on a programming problem, and annotated these datasets at those places at which experts would intervene, and how they would intervene. We have used these datasets to compare expert feedback and hints to feedback and hints given by learning environments for programming. Although we have constructed extensive guidelines on when and how to give feedback, we observed plenty of disagreement between experts. We also found several differences between feedback given by experts and learning environments. Experts intervene at specific moments, while in learning environments students have to ask for feedback themselves. The contents of feedback is also different; experts often give (positive) feedback on subgoals, which is not supported by most environments.

8 June 2022 in IEEE Transactions on Learning Technologies
Verified
Journal Article
Closed access
SMSamiha Marwan
Samiha Marwan
BABita Akram
Bita Akram
TBTiffany Barnes
Tiffany Barnes
TPThomas W. Price
Thomas W. Price
Theories on learning show that formative feedback that is immediate, specific, corrective, and positive is essential to improve novice students’ motivation and learning. However, most prior work on programming feedback focuses on highlighting student's mistakes, or detecting failed test cases after they submit a solution. In this work, we present our Adaptive Immediate Feedback (AIF) system, which uses a hybrid data-driven feedback generation algorithm to provide students with information on their progress, code correctness, and potential errors, as well as encouragement in the middle of programming. We also present an empirical controlled study using the AIF system across several programming tasks in a CS0 classroom. Our results show that the AIF system improved students’ performance, and the proportion of students who fully completed the programming assignments, indicating increased persistence. Our results suggest that the AIF system has potential to scalably support students by giving them real-time formative feedback and the encouragement they need to complete assignments.

21 November 2022 in IEEE Transactions on Learning Technologies
Verified
Journal Article
Closed access
SMSamiha Marwan
Samiha Marwan
TPThomas W. Price
Thomas W. Price
Novice programmers often struggle on assignments, and timely help, such as a hint on what to do next, can help students continue to progress and learn, rather than giving up. However, in large programming classrooms, it is hard for instructors to provide such real-time support for every student. Researchers have therefore put tremendous effort into developing algorithms to generate automated, data-driven hints to help students at scale. Despite this, few controlled studies have directly evaluated the impact of such hints on students' performance, and learning. It is also unclear what specific design features make hints more or less effective. In this work, we present iSnap, a block-based programming environment that provides novices with data-driven, next-step hints in real-time. This paper describes our improvements to iSnap over 4 years, including its “enhanced” next-step hints with three design features: textual explanations, self-explanation prompts and an adaptive hint display. Moreover, we conducted a controlled study in an authentic classroom setting over several weeks to evaluate the impact of iSnap's enhanced hints on students' performance and learning. We found students who received the enhanced hints perform better on in-class assignments and have higher programming efficiency in homework assignments than those who did not receive hints, but that hints did not significantly impact students' learning. We also discuss the challenges of classroom studies and the implications of enhanced hints compared to prior evaluations in laboratory settings, which is essential to validate the efficacy of next-step hints' impact in a real classroom experience.

12 November 2024
Verified
Conference Paper
Open access
SMSamiha Marwan
Samiha Marwan
NCNicki Choquette
Nicki Choquette
VCVeronica Cateté
Veronica Cateté
BMBriana B. Morrison
Briana B. Morrison


18 February 2025
Verified
Conference Paper
Open access
SMSamiha Marwan
Samiha Marwan
MIMohamed Ibrahim
Mohamed Ibrahim
BMBriana Morrison
Briana Morrison
The use of subgoal labels in introduction to programming classrooms has been shown to improve student performance, learning, retention, and reduce students' drop out rates. However, creating and adding subgoal labels to programming assignments is often hard to articulate and very time-intensive for instructors. In Computing Education Research, Large Language Models (LLMs) have been widely used to generate human-like outputs such as worked examples and source code. In this work, we explore whether ChatGPT could be used to generate high-quality and appropriate subgoal labels in two programming curricula. Our qualitative data analysis suggests that LLMs can assist instructors in creating subgoal labels in their classrooms, opening up directions to empower students' learning experience in programming classrooms.