Hedy: A Gradual Language for Programming Education

Sverrir Thorgeirsson Z. Su

10 Oktober 2021

Empirical research suggests that programming language syntax is a common impediment for beginners, a concern that is mitigated to a varying degree by visual programming. In this paper, we introduce a novel visual programming language that is founded on program synthesis and the programming-by-demonstration paradigm. By using an intuitive visual syntax, we show how we can meet our primary goal of providing support to computer programming novices in exploring foundational programming concepts. We present the language's current and planned use in computer science education, provide preliminary evidence for its effectiveness, and discuss its future possibilities.

Sumit Gulwani Tung Phung Tobias Kohn + 4 lainnya

24 Januari 2023

Large language models (LLMs), such as Codex, hold great promise in enhancing programming education by automatically generating feedback for students. We investigate using LLMs to generate feedback for fixing syntax errors in Python programs, a key scenario in introductory programming. More concretely, given a student's buggy program, our goal is to generate feedback comprising a fixed program along with a natural language explanation describing the errors/fixes, inspired by how a human tutor would give feedback. While using LLMs is promising, the critical challenge is to ensure high precision in the generated feedback, which is imperative before deploying such technology in classrooms. The main research question we study is: Can we develop LLMs-based feedback generation techniques with a tunable precision parameter, giving educators quality control over the feedback that students receive? To this end, we introduce PyFiXV, our technique to generate high-precision feedback powered by Codex. The key idea behind PyFiXV is to use a novel run-time validation mechanism to decide whether the generated feedback is suitable for sharing with the student; notably, this validation mechanism also provides a precision knob to educators. We perform an extensive evaluation using two real-world datasets of Python programs with syntax errors and show the efficacy of PyFiXV in generating high-precision feedback.

Pierre Weill-Tessier A. Costache Maksymilian Sekula + 2 lainnya

1 Agustus 2022

Objectives. Java is a popular programming language for use in computing education, but it is difficult to get a wide picture of the issues that it presents for novices; most studies look only at the types or frequency of errors. In this observational study, we aim to learn how novices use different features of the Java language. Participants. Users of the BlueJ development environment have been invited to opt in to anonymously record their activity data for the past 8 years. This dataset is called Blackbox, which was used as the basis for this study. BlueJ users are mostly novice programmers, predominantly male, with a median age of 16 years. Our data subset featured approximately 225,000 participants from around the world. Study Methods. We performed a secondary data analysis that used data from the Blackbox dataset. We examined over 320,000 Java projects collected over the course of 8 years and used source code analysis to investigate the prevalence of various specifically selected Java programming usage patterns. As this was an observational study without specific hypotheses, we did not use significance tests. Instead, we present the results themselves with commentary, having applied seasonal trend decomposition to the data. Findings. We found many long-term trends in the data over the course of the 8 years, most of which were monotonic. There was a notable reduction in the use of the main method (common in Java but unnecessary in BlueJ) and a general reduction in the complexity of the projects. We find that there is only a small number of frequently used types: int, String, double, and Boolean, but also a wide range of other infrequently used types. Conclusions. We find that programming usage patterns gradually change over a long period of time (a period in which the Java language was not seeing major changes) once seasonal patterns are accounted for. Any changes are likely driven by instructors and the changing demographics of programming novices. The novices use a relatively restricted subset of Java, which implies that designers of languages specifically targeted at novices can satisfy their needs with a smaller set of language constructs and features. We provide detailed recommendations for the designers of educational programming languages and supporting development tools.

J. Prather Brett A. Becker Nigel Bosch + 3 lainnya

22 Februari 2022

As computing students progress through their studies they become proficient with multiple programming languages. Prior work investigating language transitions for novices has tended to analyze program artifacts rather than explore the benefits and difficulties as perceived by students in their own words, and has often overlooked problems that may arise in switching paradigms or where familiar syntax has a different meaning in the new language. In this paper, we ask students to reflect on the transition from an interpreted language and environment (MATLAB) to a compiled language (C), prompting comments on the aspects of learning the new language that they found both easier and harder. Analysis of over 70,000 words written by 771 students revealed that the highest-performing students expressed more negative sentiments towards the language transition -- a surprising result that we hypothesize is explained by their generally stronger metacognitive skills. We also report the most common difficulties described by students, which include challenges with syntax, error messages, and the process of compilation, and suggest teaching practices that might help students as they transition to a new programming language.

Fionnuala Johnson John O'Donnell Stephen McQuistin

9 Januari 2020

Python has become a popular language for the delivery of introductory programming courses. Two reasons for this are Python's convenience and syntactic simplicity, giving a low entry barrier for beginners and the ability to solve complex problems with short snippets of code. However, students exhibit widespread misconceptions about the meaning of basic language constructs, inhibiting their ability to solve problems and damaging their understanding of fundamental concepts. In this paper, we document our observations of level 1 university students over several years, as well as surveys probing the nature of their misconceptions. We analyze the misconceptions in relation to a notional machine model for Python, and show that many students form inadequate and brittle mental models of the language. Our results indicate that one of the major sources of misunderstanding is the heavy use of overloading in Python. Overloading hides the complexity of algorithms and data structures, often leading students to write code that involves mutability, sharing, copying, side effects, coroutines, concurrency, and lazy evaluation -- and none of those topics are accessible to students who haven't yet mastered basic assignments, conditionals, and looping. We suggest that Python, when taught alone, is insufficient as an introductory language: students can gain a firmer grasp of programming fundamentals when Python is presented alongside a complementary low level language that makes a notional machine clear and explicit.