Volltextdatei(en) vorhanden
Titel: Student Performance and Collaboration in Introductory Courses to Theory of Computation
Sonstige Titel: Studierendenperformance und Kollaboration in Einführungskursen der Theoretischen Informatik
Sprache: Englisch
Autor*in: Frede, Christiane
Schlagwörter: Computer science education; Theory of computation; Videography; Explorative Data Analysis; Performance Analysis
GND-Schlagwörter: InformatikGND
HochschuldidaktikGND
Theoretische InformatikGND
VideoGND
Empirische ForschungGND
Erscheinungsdatum: 2021
Tag der mündlichen Prüfung: 2022-02-10
Zusammenfassung: 
In tertiary computer science education, computer science undergraduate programs usually include one or two compulsory courses in theory of computation. Although computer science curriculum recommendations indicate that theory of computation courses are a highly relevant part of computer science undergraduate programs, the courses suffer from high failure rates, and only a minority of students perform well. Several pedagogical approaches have been introduced in the last decade to address the problem and improve the situation in theory of computation courses. These approaches offer elaborated pedagogical solutions for engaging computer science students for theory of computation and lowering attrition and failure rates. Most of the existing approaches were developed with the assumption that students’ difficulties with theory of computation are mainly caused by a lack of interest, motivation, or ability to understand the relevant concepts and theorems due to the abstract and formal nature of computation. Thereby, the assumptions are often based on occasional oral feedback or surveys with given answers which were conducted after the courses. This leads to the fact that none of the assumed student difficulties have been empirically validated in ways that would inform pedagogical considerations by detailed insights about the nature of students’ actual difficulties.

In the present work, I have undertaken a detailed investigation of the difficulties of students with theory of computation. Thereby, I provide more sustained information than the general assumptions on which current pedagogy has been based. In more detail, I conducted two studies: (1) a quantitative study within an introductory course about Formal Languages and Automata to investigate the student performance in all assignments and topics covered, and (2) a qualitative study to explore students’ difficulties in assignments selected based on the results of the quantitative study.

Using an exploratory data analysis approach and a one-way analysis of variance, I analyzed the final exam and homework performance of about 1500 students over three consecutive years. The results show that all students perform low on almost all proof assignments, regardless of their final exam grades. While students performed worst in the final exam on an assignment that required a formal proof using the pumping lemma, performance on a similar homework assignment was not as low. Furthermore, I detected how one assignment of the first year of analysis had a significantly lower performance in the following years after a sub-task was added that required proof development. The results underline that students have most difficulties with formal proof assignments and add that this can affect students regardless of their final exam grade.

Based on the performance discrepancy between pumping lemma final exams and homework assignments, I conducted a qualitative study. Using a videography and a video interaction analysis, I observed three student groups working on two pumping lemma homework assignments. Thereby, I came to the following conclusions: Students have the same difficulties on the pumping lemma assignments in final exams and homework. However, when it comes to homework, students usually solve the problems while working together, so performance on homework solutions tends to be higher than on individual final exams. Nevertheless, through an analysis of student interactions, I found that there is a particular distribution of roles in the groups. Generally, one student acts as an explaining teacher, one as a questioning student, while all other students hardly participate in the group work, regardless of the group size. One possible explanation for this type of distribution lies in how students externalize and internalize their knowledge. They focus heavily on the tutor session and their tutor’s explanations and use the sample solutions they receive online for various assignments as patterns for their own solutions. The overall study gives the impression that students in group work are trying to achieve a result that will earn them as many points as possible with the tutor, rather than really internalizing and learning the topics.
Through an extensive quantitative study and a detailed qualitative study, the present work offers new insights and explanations for the low performance and high failure rates in theory of computation courses. The findings offer starting points for changing the pedagogical design to improve the poor situation in theory of computation courses. In addition to teaching proof skills, special attention needs to be paid to collaborative teaching-learning situations.
URL: https://ediss.sub.uni-hamburg.de/handle/ediss/9483
URN: urn:nbn:de:gbv:18-ediss-98947
Dokumenttyp: Dissertation
Betreuer*in: Knobelsdorf, Maria
Enthalten in den Sammlungen:Elektronische Dissertationen und Habilitationen

Dateien zu dieser Ressource:
Datei Beschreibung Prüfsumme GrößeFormat  
Frede_Dissertation_final.pdf2ac84c1c63252ce785532f46cb8486807.14 MBAdobe PDFÖffnen/Anzeigen
Zur Langanzeige

Diese Publikation steht in elektronischer Form im Internet bereit und kann gelesen werden. Über den freien Zugang hinaus wurden durch die Urheberin / den Urheber keine weiteren Rechte eingeräumt. Nutzungshandlungen (wie zum Beispiel der Download, das Bearbeiten, das Weiterverbreiten) sind daher nur im Rahmen der gesetzlichen Erlaubnisse des Urheberrechtsgesetzes (UrhG) erlaubt. Dies gilt für die Publikation sowie für ihre einzelnen Bestandteile, soweit nichts Anderes ausgewiesen ist.

Info

Seitenansichten

80
Letzte Woche
Letzten Monat
geprüft am 17.05.2022

Download(s)

84
Letzte Woche
Letzten Monat
geprüft am 17.05.2022
Werkzeuge

Google ScholarTM

Prüfe