I’m about to share the secret sauce of designing rigorous and compelling research projects. I call it a secret sauce, but it’s actually a tool that I was taught as an undergrad research assistant in psychology. Thus, it’s just one of those skills I learned through apprenticeship rather than formal coursework. The tool is using your hypotheses to iteratively improve your research design.
First, you need to make hypotheses about your research questions. Because your research questions should include the independent and dependent variables you will measure, for each level of your independent variable(IVs), what results do you expect on the dependent variable(DVs)? Use a table like the one below to help you plan it out.
In the table below, the hypothesis is that the control group performs average-ly on the process DV (measures the process of learning) and the product DV (measures the outcome of learning). This is what makes them the control group. For the first level of the IV, we expected the learning process to improve (e.g., take less time) with similar outcomes as the control group. For the second level of the IV, we expect the learning process to be worse but with better outcomes (e.g., retention).
|DV Process||DV Product|
|IV level 1||↑||=|
|IV level 2||↓||↑|
Making a table like this should be difficult if you have an interesting research question (i.e., if the result isn’t obvious). When deciding what to place in each box, take note of any tradeoffs/decision-making factors. For example, on IV level 1, maybe you expect the product DV to be equivalent to the control ONLY IF learners take less time to complete the instruction. If learners take the same amount of time, then you might expect the product DV to improve.
You want to keep note of these deciding factors because they will help you anticipate alternative results and possible explanations for those results. When you have possible alternative results and likely explanations, make sure you are measuring something about that likely explanation (e.g., time to complete instruction). Then if you find alternative results, you already have the data you need to examine likely explanations. This forethought makes null results much more compelling.
Another strategy to create a better research design is to pre-write your limitations section, either for the results you expect or for null results. Then analyze that limitations section for anything that you can improve in the research design before collecting data.
This is the final post in the Research Design Series. View the Research Design: Series Introduction for more posts. To access the worksheet that will help apply these concepts to a research project, choose from the editable PDF or Word versions below.
One thought on “Research Design: Improve Your Design Before Collecting Data”
Pingback: Research Design: Series Introduction | Lauren Margulieux