Transfer and Training Effects of Learning to Code

The Idea

Learning to write programs stretches your mind, and helps you think better, creates a way of thinking about things that I think is helpful in all domains.
—Bill Gates (Co-Chairman of the Bill & Melinda Gates Foundation and Co-Founder of Microsoft)

Copyright: Glenn Carstens-Peters,

Technology is everywhere—it determines the ways we communicate, solve problems, and organize our everyday lives. Because of this omnipresence of technology and its inevitable advancement, educational systems around the world have recognized the pressing need for helping students become digitally literate. Over the last decade, one key element of digital literacy has received considerable attention, as it forms the very heart of technological skills: computational thinking.

Among other facets, computational thinking contains the skills to code or, in other words, to program a computer—skills that have been said to help students think, even in domains that have little to do with coding. Recently, claims concerning the positive effects of learning to code have been made, and they all go into the same direction: Learning to code supposedly improves problem solving, reasoning, logical thinking, mathematical, and other skills.

The core idea of this project is to put to test these claims empirically. This project further reviews instructional practices to learn how to code.

The Approach

With the help of meta-analytic methods, this project reviews the existing body of research on the positive effects of learning to code on multiple cognitive skills (transfer effects). This review will result in an aggregated effect that will or will not substantiate the claims surrounding coding. Furthermore, this project reviews the existing body of research on the trainability of coding skills, using coding skills as the outcome variable (training effects).

The initial search and screening for relevant empirical studies resulted in more than 200 studies, which were subjected to data extraction procedures. The overall effects will be reported by differentiating between the types of cognitive skills as outcome variables.


Please find the registered, meta-analytic protocol in the PROSPERO database (external link).

Please find more information on the project’s ResearchGate page (external link).

Published Oct. 11, 2017 2:52 PM - Last modified Oct. 30, 2017 1:20 PM