How To: Select a Good Project Topic

As you transition to the project proposal, you’re tasked with proposing the actual work you’ll do the rest of the semester. The work you’ve done the past several weeks has been aimed at allowing you to choose a good problem or question, but how you actually scope and present it is important as well. This guide is on how you settle on a good actual project to pursue.

Unlike the other guides, which are closer to steps to follow, this guide is more of a set of suggestions curated based on our several semesters teaching this class so far.

Follow Your Passion

This class is a lot of fun if you’re working on something that you care about, but it can be torturous if you settle on something just to complete the coursework. You want to select something that’s going to engage and motivate you to keep pursuing it; I might go so far as to say you want to be disappointed at the end of the term that you didn’t get to do more. (You own your project, though, so you can do more! …just not for a class grade.)

The wonderful thing about education is that it’s ubiquitous to all of us. There are things you’ve encountered that you think could have been done better, or questions you’ve encountered that you want answered. If you’re not excited about the work you’re proposing to do, think about what you would be excited about, and try to move in that direction.

Justify the Need and Justify your Approach

We ask that you spend the first several weeks of the semester exploring the area of your problem rather than just jumping in, and one reason for this is so you can justify that the need exists. If you’re doing a development project, you need to be able to show that you’re addressing a problem that hasn’t been adequately addressed. If you’re doing a research project, you need to demonstrate that the question hasn’t been adequately answered and that it has been adequately justified by related literature. If you’re doing a continent project, you need to show that the material that you’re teaching has not been adequately taught.

Of course, in some ways those are impossible: you can’t prove a negative. But you can document that you’ve done the necessary research to show that if there was a good solution or a fitting theory, you likely would have found it. It still might exist, but it at least hasn’t become well enough known or recognized to turn up with sufficient earnest research.

It’s also important to note that in justifying the need, you don’t then need to create a wholesale competitor or alternative. For example, if you documented a need with second language unmet by apps like DuoLingo, you wouldn’t need to then reimplement everything DuoLingo does as well as the features that address your documented need. You would need to work toward something that complements what is already out there.

There is another side to this as well, which is also justifying that the approach you’re taking has merit. Imagine, for instance, you want to include a leaderboard: why? Your justification should go beyond the idea that it’s a feature that just seems to make sense; you should be able to put your plan for addressing the problem or answering the question in terms of the existing literature on the topic. Why do you believe your method will succeed?

Scope Appropriately

Your project is generally expected to be 100 hours of work, regardless of whether you’re taking this class in a regular or a shortened semester. 100 hours is both a significant chunk of time and an insignificant one. You have time to do something notable and impactful, but probably not time to do absolutely everything you might have in mind.

So, pay attention to what you can scope in a particular semester. If the entire space of what you’d like to do is too large, then find a way to scope it down. Your goal is to select something you can actually complete, even if it’s smaller than what you’d like to do overall; you’d rather do 100% of something small than 50% of something large.

Similarly, there are good ideas that don’t constitute 100 hours of work. Good survey research, for instance, is more time consuming than people think, but a single survey isn’t 100 hours of work either. Consider how you might follow on from the initial work. If you were doing a survey, for example, you might specifically outline follow-up questions or directions that might come from different survey results.

Learn a Little Something New

Depending on your background, you might want to do something that you don’t yet know how to do. Maybe you want to build a web site that better tracks individual contributions to group projects. Maybe you want to perform an analysis to see if grades are biased by visible indicators of race or gender. Maybe you want to teach a field you know, but you don’t really know how to teach.

It’s okay to specifically include in your proposal some time to learn the skills necessary to do your idea, like web design, data analysis, or instructional design. Those tasks should still be specific (what are you learning, from where?), but they can be foundational to executing your actual project.

That said, the majority of your project work should actually be to complete the project, not to prepare for the project. You shouldn’t spend 90 hours learning web design to spend 10 hours creating a web site for this class. We’d generally say no more than 15% of your project work should be spent learning the necessary skills to do the other stuff you have in mind.

That’s not to say that you can’t spend more time learning, but it shouldn’t count toward your project scope. If you don’t mind spending 200 hours on your project with 100 spent learning, that’s fine! Make sure you’re ready for that, though: if you run out of time to learn everything you need to know, you might find yourself in a bad position for the actual project work you proposed.