Survey questions

Continuing our discussion of surveys:

4. You know how you will use the answers.

You need to have a clear idea as to how you will use the survey questions to address your research questions. Keep in mind that short surveys are generally much better than long ones in keeping participants focused and finishing the survey.

Each of your research questions typically relate to a number of survey questions. It is a good idea to match up your research and interview questions to make sure that you will actually be able to answer your research questions. If you have survey questions without a matching research question, you probably do not need the survey questions. If you have a research question without a survey question, you need to add more survey questions. Here is an example of matching a research question and survey questions. 

Research Questions	Survey Questions
1. How do patients' searches, for health related information on the internet, relate to visits to their primary health provider?	1. In the past month, have you searched for health related information on the internet? 2. Before your most recent visit to your primary health provider, did you search for any health related information on the internet? 3. After your most recent visit to your primary health provider, did you search for any health related information on the internet?

You should also have a plan as to how you will analyze the data you collect. Without such a plan, you may come to realize that you have collected hundreds of surveys with a missing variable that is needed for your analysis. As an example, if you would like to analyze your data using gender as an independent variable, make sure you have included it in the survey.

5. You can expect an adequate response rate.

An otherwise excellent survey study can be ruined by a low response rate. The difficulty is that if only a small proportion of the people you ask to be in the survey actually complete it, you can no longer generalize to the population of interest. For this reason, if you suspect a low response rate is likely, you would be better off having fewer people and then interview them.

What is a good response rate? There is not a clear answer to this. You will need to review the literature and see what others report. There are a few things you can do to improve your response rate. Using a university's participant pool, generally results in a high rate of response (keep in mind they are not representative of the general population). If you can guarantee respondents that their answers will be anonymous (the researcher does not know who responded), it generally helps. The second best choice is guaranteeing confidentiality (only the researcher knows who responded).

Next time we will continue our discussion of surveys. Do you have an issue or a question that you would like me to discuss in a future post? Would you like to be a guest writer? Send me your ideas! leann.stadtlander@waldenu.edu