Esche, Svana (2024)
Focus on Language in Introductory Programming.
Technische Universität Darmstadt
doi: 10.26083/tuprints-00027043
Ph.D. Thesis, Primary publication, Publisher's Version
Text
Dissertation_SvanaEsche.pdf Copyright Information: In Copyright. Download (4MB) |
Item Type: | Ph.D. Thesis | ||||
---|---|---|---|---|---|
Type of entry: | Primary publication | ||||
Title: | Focus on Language in Introductory Programming | ||||
Language: | English | ||||
Referees: | Weihe, Prof. Dr. Karsten ; Schroeder, Prof. Dr. Ulrik | ||||
Date: | 26 April 2024 | ||||
Place of Publication: | Darmstadt | ||||
Collation: | xiv, 129 Seiten | ||||
Date of oral examination: | 17 April 2024 | ||||
DOI: | 10.26083/tuprints-00027043 | ||||
Abstract: | This thesis analyzes the relationship between natural language and reasoning when introducing programming in postsecondary education. The particular focus is on the insights that emerge when this relationship is made explicit in teaching-learning situations. As a major result, natural language does indeed play an important role in these teaching-learning situations but deserves further research attention. Making the described relationship explicit is itself a new meta-perspective for thinking about research and teaching in computing education. Developing this new meta-perspective is an important contribution of this work. As far as we know, this meta-perspective has not yet been synthesized as such and at the same time supported by empirical evidence in postsecondary education. Another contribution is the insight that diagnosing and using students' terms as part of language are particularly fruitful for research and teaching. Further research should pursue this new approach in order to shed light on well-known problems from a new perspective. For the empirical evidence of these insights, this thesis used a mixed methods research design in which four studies were conducted. The mixed methods research design combined qualitative approaches such as qualitative content analysis with quantitative approaches such as experiments with students as participants. This approach enabled a comprehensive and multi-layered analysis of the results. We included all actors in the teaching-learning situations in order to obtain a more comprehensive analysis. All four studies were conducted according to the respective gold standard as far as possible. This includes rigorous measurements and large sample sizes. The four studies were divided into an exploratory basis study and three subsequent application studies. The Basis Study focused on students, uncovered the terms they used and indicated a possible relationship between programming language and natural language. In addition, it led to new insights about correct conceptions and misconceptions of programming languages. Furthermore, the methodological approach of this study is itself a contribution, as it shows how a qualitative analysis of terms can be carried out with large samples. Application Study I examined instructional videos as a teaching method used by instructors. As a basis for instructional videos to support students, the implementation of language focus across multiple representations turned out not to be convincing. Thus, the hypothesis about the positive impact of these language-based videos on students' code writing and self-efficacy could be rejected with reasonable certainty. In the Application Study II, we turn our attention to teaching staff in general and their competences in supporting students. We developed and evaluated a rubric, which is a guide that lists specific criteria for grading by category and level. The rubric serves as an assessment scheme for the quality in their responses, focusing on language. Finally, in the Application Study III, a test of testing programming aptitude was developed and validated based on converting programming thinking into natural language. The resulting Natural Language Computing Test, or NLCT, has proven to be highly accurate, valid, and reliable. As a minor weakness, the NLCT did not prove to be a predictor of success at the end of the course. To summarize, the four studies conducted have shown that the potential of this new meta-perspective lies in the way in which the relationship described is carried out. The use of terms as a link between programming language and natural language seems to be particularly promising. In contrast, the use of multiple representations of programming language, natural language and visualization has shown to be less promising. |
||||
Alternative Abstract: |
|
||||
Uncontrolled Keywords: | Informatikdidaktik | ||||
Status: | Publisher's Version | ||||
URN: | urn:nbn:de:tuda-tuprints-270435 | ||||
Classification DDC: | 000 Generalities, computers, information > 004 Computer science 300 Social sciences > 370 Education |
||||
Divisions: | 20 Department of Computer Science > Algorithmics 20 Department of Computer Science > Didactics of Informatic |
||||
Date Deposited: | 26 Apr 2024 12:15 | ||||
Last Modified: | 30 Apr 2024 06:39 | ||||
URI: | https://tuprints.ulb.tu-darmstadt.de/id/eprint/27043 | ||||
PPN: | 517570742 | ||||
Export: |
View Item |