Research - Bio 260

Although there are no universal criteria for a good research question, this LibGuide will help you write a better one. In general, good research questions should clearly address a specific issue or problem that is within your scope of expertise and can be answered through research (either in the literature or through data collection) using an analytical approach. 

Here are the key points to consider when formulating scientific research questions. It is important to note that these categories are not mutually exclusive, i.e. they can overlap.

Your question should clearly state what you hope to answer with all the necessary details but without any extra details. 

• Example: How can drunk driving be prevented?
  • Although the topic of this question is clear (drunk driving), it is an open-ended question (i.e. “how?”) without a focused hypothesis. It needs more detail on what factors are believed to cause drunk driving (i.e. the exposure). For example: “Will lowering the legal blood alcohol content (BAC) to 0.5% reduce drunk driving by at least 50% nationally?” is a more clear and focused question. Also, this question can be answered simply: don’t drive drunk. Therefore, it is not a focused question. 

Avoid being too broad or too narrow - Make sure your question is within your scope and can be reasonably addressed or answered. 

• Broad topics may be too complicated to answer given your timeframe and resources or result in a description of existing knowledge rather than data collection and synthesis of new knowledge. 
  • Example: Which is more important, nature or nurture? 
    • Given our current understanding of this topic, this question is too broad because it cannot be reasonably answered. Furthermore, “importance” is ambiguous–important in what sense? 
• Narrow topics, in contrast, can be too easily answered. Also, there may be a ton of existing information that can overwhelmingly answer the question. 
  • Example: Does eating 100 grams of sugar per day increase weight by 2 pounds in one month? 
    • This question is too narrow because 2 pounds and 100 grams are very specific amounts. It is a testable question but could be a bit broader to be more interesting and useful: “For every 100 grams of sugar consumed per day, how much weight is gained per month?”
• Other Examples
  • Does brushing teeth once per day prevent cavities?
    • Not too broad and not too narrow. You could conduct a study where you have one group of people brushing their teeth daily and compare it to those who brush more or less frequently. You could then collect data on how often they get cavities. It is a testable hypothesis. 
  • Does exercising twice daily lower systolic blood pressure by 20 mmHg?
    • Too broad and too narrow. Exercise is not well-defined (i.e. walking for 10 minutes, cycling, HIIT?” 20 mmHg is too specific because blood pressure is highly variable.

Questions should be researchable - meaning it is possible to find an answer by analyzing quality research materials such as academic books and refereed journal articles or collected data from a study.

• Example: Is addiction biological or environmental?

It is unlikely that you will be able to conduct a study that can sufficiently answer this question. Although you may be able to find literature on this subject and propose an argument one way or another, this question cannot ultimately be answered given our current level of knowledge.

Aim for an analytical question rather than a descriptive one - It should allow you to produce an analysis of an issue or problem rather than a simple description of it. 

• Example: What are the causes of diabetes?
  • This is a question that can be easily answered using multiple sources. All that is required is a simple description of what is currently known about diabetes. There is no testable hypothesis. 
• Example: Is TikTok or Instagram better at increasing voter turnout among voters under the age of 30?
  • If you can find recent information to answer this question then it is purely descriptive. However, assuming that this has not yet been tested, answering this question would require gathering data and analyzing it. Therefore, it is a reasonable research question. It is a reminder that you should always do your “homework” and research your question before you start working on it, because there may already be a lot of information on the topic you are interested in! Please see a librarian so we can help you further.