Survey design

PsychData's primary role is to host a survey you've already designed. Survey design involves many elements, and technically precedes your use of PsychData and its tools. In the Library find links to external in-depth references on survey design.

Design can be and often is dynamic. The process of building a survey--formatting, adding and modifying questions, and more--can raise questions that reopen settled design issues. Below, links to some specific items adapted from our library.

Ask only those questions that you truly need answered. Remember: each question adds to total survey time for your participants. If you're not giving them anything in return for taking the survey, shorter is better.

Long surveys turn people off: they're less likely to start and finish a free task they perceive as lengthy. Longer surveys are better for an audience who gets something in return (e.g., class credit, money). If you don't need the data, don't ask for it!

So what's long?

Many factors make up length: number and length of questions, complexity, branching, participant's speed. If you're not giving anything, think 15 minutes as a kind of baseline. Let your own experience guide you: how would you feel if someone asked you to take a survey?

Keep participants informed

Two kinds of time information participants should always know:

How long is this?
Where am I?

If you want your online research to work the right way for all of your participants (who are using a variety of computers and internet browsers), keep your project simple and straightforward. Avoid animated , flashy presentations and complex layouts, and keep your goal in mind—data.

Think from your participant's point of view as you design a survey.

Would you be willing to answer that question online?
Are those choices different and significant?
Is that question clear?

Participants can't ask for clarification when they're taking a survey; ambiguity weakens any survey instrument. If some ambiguity is unavoidable, consider providing explanations within the survey layout (using Add Question or Text>Rich Text Editor).

Define ambiguous, please

Editors earn their pay, because they bring fresh eyes to your work. Have someone who is unfamiliar with your research proof-read your survey for clarity and flow--and simple errors, like typos.

Online research allows you to collect data from anywhere in the country, and many parts of the world.

Expand your contact area. If you are interested in a diverse sample, you can get it online if you plan ahead, make contacts, and network.

Online is not for everyone.

Some populations are not appropriate for online data collection. Don't try to force an unwilling or difficult-to-reach population into an online format. A really inappropriate audience can invalidate your data and cost you time (and money).

Putting your research tool online is only one step in online data collection. Other timing issues:

How long will it take to gather materials and post your survey?
Do you have a deadline to work back from?
Is there a critical window for participants?
What kind of scheduling will you use? (i.e., some participants work better under a deadline, others with an open invitation).
How and when do your participants learn about your survey?
If you participants download materials from a website, when are they likely to visit? If you email them, will they check email and be motivated to respond?
What else is happening in your participants' world?
For a college student population, semester end can be a great time if your survey provides extra credit (because it's a potential grade boost). If it's not for credit, a survey has to compete with final papers and exams.