Develop a research design. First, determine what approach makes the most sense—experimental, quasi-experimental, or nonexperimental. Identify the strengths and weaknesses of the approach that you select. Then identify how and what kinds of data will be collected. Next, identify the unit of observation and the population. Then determine whether or not you can collect population data or if you need to collect sample data. If you need to draw a sample, determine whether or not you can take a random sampling approach. Then identify your sampling strategy and develop a plan that includes the number of observations on which you will collect data and the strengths and weaknesses of the sampling approach.