Online surveys are ubiquitous. Most of us get requests to take one almost every day. That’s one reason why, if you plan to do a survey for your research or service offering, you’ll want to think it through carefully.
Conducting an effective survey takes time and effort. Tasks include:
- Preparing the survey instrument with clear questions that will maximize useful results
- Determining the list of invitees
- Writing an email invitation
- Getting any needed approvals in advance
- Analyzing and reporting your results
That said, there’s a great resource on campus that can help you prepare, conduct and analyze a survey. The Office of the Provost, Institutional Research (or IR for short), has posted Survey Guidelines [PDF] that cover everything from the survey process, to considerations such as permissions, to reporting and analyzing results.
This article will give a brief summary — based on the IR Guidelines — of some of the options and tools involved in conducting surveys at MIT.
There are three ways you can conduct surveys at MIT.
- Administer the survey yourself. Commercial tools, such as the popular SurveyMonkey, let you create and host a survey.
- Have IR administer and host the survey for you. IR may be able to assist with the design of your survey and then host and administer it for you. This option is subject to the availability of IR staff and the group’s existing survey schedule, and the service may be subject to a fee.
- Include your questions on an already scheduled survey. In some instances, constituencies on campus can append a limited number of targeted questions to surveys IR administers to various MIT populations. IR maintains editorial rights and the right to refuse questions.
Regardless of which option you choose, IR staff can help you form your survey questions. For examples, see IR’s Surveys page.
Tools for analysis
After you’ve collected the responses to your survey, you’ll need to analyze the results. Simple analyses can be done using Excel with Pivot Tables, histograms, and graphs, but more advanced statistical analyses will require the use of a statistical software package such as SPSS, SAS, Stata, or R. Stata and SAS are available to the MIT community on Athena, and R is open‐source free software, but all three systems have a learning curve.
IR primarily uses SPSS and may be able to analyze survey results for you, depending on its schedule and staff availability. This service may be subject to a fee.
Once your results have been analyzed, it’s a good idea to share them in a way that’s accessible to the community you’ve surveyed. By being transparent with results, the populations you survey can see that their contributions count. For examples of overall summaries and results for each question, see IR’s Surveys page.Need help?
Successful surveys require extensive preparation and access to the right tools for conducting and analyzing them. Before you undertake a survey, be sure to check out IR’s online resources. Also feel free to contact IR staff with questions or to ask for help: you can reach them by sending email to email@example.com.