Another look at the behaviors of novice programmers

Fenwick, James B.; Norris, Cindy; Barry, Frank E.; Rountree, Josh; Spicer, Cole J.; Cheek, Scott D.

doi:10.1145/1508865.1508973

Cited by 28 publications

(15 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Jadud's results indicate that syntactical errors, such as missing semicolons or misspellings of variable names, are the most common type of beginners' Java errors. This study was repeated by Fenwick et al (2009), who obtained similar results. Fenwick et al also presented higher-level patterns of student behavior, confirming, for example, the benefits of starting projects early and working incrementally.…”

Section: Related Worksupporting

confidence: 66%

“…The closest reference to the present study is the one by Chambers et al (2012), where student errors were collected using an observational method. Our study complements Chambers' study as we apply a computer-based analysis as the first step, similar to the studies by Jadud (2005), Fenwick et al (2009), and Denny et al (2012). Moreover, our method complements the work of Jadud, Fenwick, and Denny by incorporating a manual analysis step.…”

Section: Related Workmentioning

confidence: 85%

See 1 more Smart Citation

Understanding beginners' mistakes with Haskell

Tirronen¹,

Uusi-Mäkelä²,

Isomöttönen³

2015

J. Funct. Prog.

View full text Add to dashboard Cite

This article presents an overview of student difficulties in an introductory functional programming (FP) course taught in Haskell. The motivation for this study stems from our belief that many student difficulties can be alleviated by understanding the underlying causes of errors and by modifying the educational approach and, possibly, the teaching language accordingly. We analyze students' exercise submissions and categorize student errors according to compiler error messages and then manually according to the observed underlying cause. Our study complements earlier studies on the topic by applying computer and manual analysis while focusing on providing descriptive statistics of difficulties specific to FP languages. We conclude that the majority of student errors, regardless of cause, are reported by three different compiler error messages that are not well understood by students. In addition, syntactic features, such as precedence, the syntax of function application, and deeply nested statements, cause difficulties throughout the course.

show abstract

Section: Related Worksupporting

confidence: 66%

Section: Related Workmentioning

confidence: 85%

Understanding beginners' mistakes with Haskell

Tirronen¹,

Uusi-Mäkelä²,

Isomöttönen³

2015

J. Funct. Prog.

View full text Add to dashboard Cite

show abstract

“…Examples of these patterns that can be extracted are programming behavioral patterns such as debugging (Ahmadzadeh et al 2005), compilation, and coding (Vee et al 2006a;Spacco et al 2006); patterns depicting practices common to successful students but not to struggling students (Norris et al 2008); problem-solving strategies (Kiesmueller et al 2010;Hosseini et al 2014) and anomalies (Helminen et al 2012); submission patterns (Falkner and Falkner 2012;Allevato et al 2008); and code evolutions (Piech et al 2012;Blikstein 2011). The possibility of cheating and the effect of starting late on projects can also be traced from these patterns (Fenwick et al 2009) as well as distinguishing experienced students from poorly performing students (Annamaa et al 2015;Leinonen et al 2016). The data collected can also be explored to predict student performance identifying at-risk students in programming (Tabanao et al 2008;Falkner and Falkner 2012;Watson et al 2013;Koprinska et al 2015) with the aim of providing them tailored remedial instructions.…”

Section: A Process-oriented Approach In Programmingmentioning

confidence: 99%

A review on process-oriented approaches for analyzing novice solutions to programming problems

Villamor

2020

RPTEL

View full text Add to dashboard Cite

High attrition and dropout rates are common in introductory programming courses. One of the reasons students drop out is loss of motivation due to the lack of feedback and proper assessment of their progress. Hence, a process-oriented approach is needed in assessing programming progress, which entails examining and measuring students' compilation behaviors and source codes. This paper reviews the elements of a processoriented approach including previous studies that have used this approach. Specific metrics covered are Jadud's Error Quotient, the Watwin Score, Probabilistic Distance to Solution, Normalized Programming State Model, and the Repeated Error Density.

show abstract

“…While these type of systems offer the flexibility to submit assignments at any time or location, there are downsides for using such systems. The lack of human guidance may lead to undesired study practices such as trial and error problem solving [46,90] or returning assignments at the last minute [48,52,55,117].…”

Section: Detailed Research Questionsmentioning

confidence: 99%

QR code programming tasks with automated assessment

Hakulinen

Malmi

2014

Proceedings of the 2014 Conference on Innovation &Amp; Technology in Computer Science Education - ITiCSE '14

View full text Add to dashboard Cite

The main objective of this thesis is to enrich computer science education by studying two gameful teaching interventions: 1) achievement badges and 2) alternate reality games.Gamification is the use of game design elements in non-game contexts, and achievement badges are one commonly used gamification method that can be used to provide optional challenges for students. We studied the use of achievement badges in a Data Structures and Algorithms course and found out that badges can be used to steer students' behavior even if they are not tied to tangible rewards such as course grades. Specifically, students used more time in the online learning environment, had more sessions in the system, and spent more time between exercise submissions when the badges were used. Furthermore, we found out that students' attitudes towards the badges varied. We also studied the use of badges in relation to achievement goal orientations that characterize students' preferences to different goals and outcomes. We found out that students who had high avoidance orientation were less motivated by the badges than other students. On the other hand, students who were the most motivated by the badges had higher mastery-intrinsic, mastery-extrinsic, and performance-approach orientations, and lower avoidance orientation. Furthermore, we compared the badge intervention with the use of heatmaps that provided feedback on one's behavior in a nongameful way. An interest towards the badges correlated with mastery-extrinsic and performance-approach, whereas interest towards the heatmaps correlated with masteryextrinsic and performance-avoidance orientations.Alternate Reality Games (ARG), are games that blur the line between reality and fiction and typically they involve puzzles and an interactive narrative. We explored the use of ARGs for teaching computer science by organizing an ARG covering a wide range of computer science concepts. In the feedback collected after the game, participants reported learning several computer science concepts during the game. We also evaluated how authenticity, intrinsic motivation, and replayability were taken into account in the game design. Furthermore, we present an example of an educational ARG that was not part of any official curriculum and relied on voluntary participation. Based on the results, alternate reality games seem like a promising approach to be used in computer science education.In addition to evaluating the two approaches, we provide support for applying them in education by describing the implementation of our badge systems and introducing a method for producing automatically assessed programming tasks that are suitable for ARGs. Tiivistelmä Väitöskirjassa tarkastellaan kahta pelillistä tapaa monimuotoistaa tietotekniikan opetusta: 1) suoritusmerkit ja 2) vaihtoehtotodellisuuspelit.Pelillistämisellä tarkoitetaan pelielementtien soveltamista eri ympäristöissä. Suoritusmerkit ovat eräs yleinen pelillistämisen keino, jolla voidaan tarjota opiskelijoille vapaaehtoisia lisä-haasteita. Väitöskir...

show abstract

Another look at the behaviors of novice programmers

Cited by 28 publications

References 8 publications

Understanding beginners' mistakes with Haskell

Understanding beginners' mistakes with Haskell

A review on process-oriented approaches for analyzing novice solutions to programming problems

QR code programming tasks with automated assessment

Contact Info

Product

Resources

About