Mobile game development: Improving student engagement and motivation in introductory computing courses – Kurkovsky, 2013
I had much better luck this week finding a quality article. In fact, I may have even found a journal I can stick with through the next review assignment (and maybe for nerdy professional reading). I chose this particular article because it sounded like I’d get some affirmation for my current curriculum choices. A little self-serving, I know, but I wanted to poke around the data behind integrating game development into computer science courses because it’s clearly a trend that is picking up speed and has been a hit in my own classroom.
This article started so hopeful. It included a lengthy lit review to support the use of game development to improve student engagement in intro computer science courses at universities. What many studies noted was that game development rarely makes it into the intro courses because building a full computer game takes high level programming skills. But, the creation of casual mobile games is totally within the capabilities of intro level students. Mobile game development provides an accessible, engaging, and practical application of many computer science concepts. I basically highlighted everything in this section; that’s how excited I was to see all this research affirming my current beliefs about teaching computer science. To sum it up, games are a slick way to teach programming concepts as it allows students to see “the connection between technical material and their everyday lives” (pg.141). It appeals to non-CS majors, women, and minorities (pg.143). Games help students understand that computer science is more than coding; an idea which hopefully gets them in the door and keeps them engaged throughout the semester.
For their study, the researchers created learning modules based on core Java programming concepts with an opportunity to practice and apply that knowledge through the enhancement of a mobile game. Some of these modules included variations on crowd favorites such as Battleship, Connect Four, Space Invaders, and Frogger. Students were not asked to build the games from scratch, but were given an almost functional game so that they could focus on smaller programming objectives while also customizing the interface and/or enhancing the game logic. Honestly, it all sounded awesome; if you have to learn Java then this seems like the way to do it.
The experiment was set up in introductory CS courses at two different universities: one school was more selective and only had STEM majors in the course, the other was less selective and had a wide variety of majors in the course. At each site, professors were given test groups (access to the game development features) and control groups. Researchers would assess the effectiveness of the mobile game modules through student grades/completion, a student survey, and two questionnaires bookending the semester.
And then it all went terribly, terribly wrong. Okay, maybe not wrong, but their findings were severely disappointing after the huge build up for game development at the beginning of the article. The researchers referred to their findings as a “mixed bag” (pg.153). Yikes. In the end, the variation between the two universities kind of hurt the study because nothing could conclusively be said for whether the game development features had a positive effect when one student population was so clearly better prepared from the start. They actually saw negative results in student interest from the more selective school; a suggested explanation being that students were anticipating traditionally taught courses and the new modules were jarring (pg.153). Happily, there was a (limited) positive effect in student engagement overall, and the test group did as well as the control group (pg.154).
Regardless of the findings, the researchers remain stalwart in their belief that game development is a positive teaching tool, and hold that more research on the topic must be done. I’m as baffled as they are as to why the study went awry. I’ll admit I got deeply suckered into the lit review section and now want to forgo these particular bad-to-middling findings, but I think this is a “fail forward” moment as the researchers noted that they would continue testing iterations of their modules. Clearly, there is a plethora of studies supporting game development in CS courses, but the modules that the researchers developed for this study are so similar to those I’m looking at to teach Java next year that I’m still kind of nervous/curious to know why they didn’t see better results. Or maybe we can just blame it on Java…
Kurkovsky, S. (2013). Mobile game development: Improving student engagement and motivation in introductory computing courses. Computer Science Education, 23(2), 138-157. Retrieved from http://dx.doi.org/10.1080/08993408.2013.777236