The effects of first programming language on college students’ computing attitude and achievement: a comparison of graphical and textual languages

ABSTRACT

Background and Context: The relationship between novices’ first programming language and their future achievement has drawn increasing interest owing to recent efforts to expand K–12 computing education. This article contributes to this topic by analyzing data from a retrospective study of more than 10,000 undergraduates enrolled in introductory computer science courses at 118 U.S. institutions of higher education.

Objective: We explored the relationship between students’ first programming languages and both their final grades in an introductory computer science course and their attitudes about programming.

Method: Multiple matching techniques compared those whose first language was graphical (e.g., Scratch), textual (e.g., Java), or absent prior to college.

Findings: Having any prior programming experience had positive effects on both attitudes about programming and grades in introductory computer science courses. Importantly, students whose first language was graphical had higher grades than did students whose first language was textual, when the languages were introduced in or before early adolescent years.

Implications: Learning any computer language is better than learning none. If programming is to be taught to students before early adolescence, it is advised to start with a graphical language. Future work should investigate the transition between different types of programming languages.

KEYWORDS:

• Programming and programming languages
• first programming language
• pedagogical issues
• secondary education
• teaching and learning strategies
• postsecondary education

Acknowledgments

This work was supported by the National Science Foundation (grant number 1339200). Any opinions, findings, and conclusions in this article are the authors’ and do not necessarily reflect the views of the National Science Foundation. Without the excellent contributions of many people, the FICSIT project would not have been possible. We thank the members of the FICSIT team: Wendy Berland, Hal Coyle, Zahra Hazari, Annette Trenga, and Bruce Ward. We would also like to thank several STEM educators and researchers who provided advice or counsel on this project as members of our Advisory Board: Hal Abelson; Lecia Barker, Chair of Advisory Board; Randy Battat; Joanne Cohoon; Maria Litvin; Clayton Lewis; Irene Porro; Kelly Powers; Lucy Sanders; Susanne Steiger–Escobar; and Jane Stout. Last but not least, we are grateful to the many college computer science professors and their students who gave up a portion of a class to provide data.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the National Science Foundation [1339200]; Scratch Foundation.

Notes on contributors

Chen Chen

Chen Chen is a postdoctoral fellow at the Science Education Department of the Harvard-Smithsonian Center for Astrophysics.