Checklist for Research Questions
Use the criteria below to evaluate other course participants' submissions as well as your own:

Please provide an example of misunderstanding the question. (Questions should be clear and unambiguous, should have no multiple meanings, and it should be easy to see what exactly do they ask)
Please provide an example of how an answer could look like. (If it's hard to imagine at this point how the answer will look like (not what the answer is but how it roughly might look like) then the question might be problematic.)
Are the questions short and understandable by non-specialists?  (The longer the question, the higher likelihood for ambiguity and obscureness.  Of course not all questions (e.g. P=NP?) can be understood by non-specialists, but it's generally a good thing if yours can.)
Can you think of any obvious answers? (If the answer is obvious, the research question won't probably tell much new.  Every now and then you stumble across a question where the answer is so obvious that you're left wondering why the research was done in the first place.  "Does Moodle enable distance learning?")
Please provide an example of how the questions can be answered through empirical research.  (Is it possible to answer the question through empirical research?  Can data collection and analysis, in principle, provide answers to this question?)
What kind(s) of empirical data collection does the question require?  (In this course all research projects must be empirical, so those that can be answered without any empirical research have to be revised.)
Give examples of different approaches that can be used to answer the questions, leading to different types of answers? (In many types of research study it's a great signal if you can use different methods to address the problem.  This doesn't apply to all research, but it's generally a good thing to have.)
Give an example of a) what the answer looks like if the expected happens, and b) what the answer looks like if the expected results aren't found. (Think of what happens if you don't get what you expected to get.  Does that make your research useless or not?  For instance, any answer to "Which of the eight primary emotions can be automatically extracted from forum posts?" is interesting, including none, some, or all.)
How are the study parts aligned (what sentence(s) in the aim lead to the question, and how do the subquestions link with the objectives?)  (This is one of the most important aspects of excellence in research projects.  The main research question should be well aligned with the aim of the study.  Subquestions should be well aligned with the objectives.  Overall, the research question should address the research problem.  (In many studies objectives aren't required, but in our study here they are.))
To what extent are the questions neutral?  Do they assume something or imply something?  (In most cases, research questions don't need to take sides about "good" and "bad".  Please see the lecture videos for more info on this aspect.)