“It wasn’t what he said…”
Have you ever found yourself responding flippantly to a question — or not responding at all — because of the way the question was phrased? “It wasn’t what she said, it was the way she said it…” is a pretty common expression, and it’s particularly appropriate if the question being posed seems kind of personal.
A touchy subject
So how does this apply to online surveys? Because we frequently need to collect personal information. Some items, like the respondent’s age, education level, income, household type and employment status are common (and are available as AYTM built-in screening items). But there are others that maybe unique to your topic, and you may be planning on adding them to your questionnaire design.
Why go there in online surveys?
There are a couple of reasons we might need to get personal in an online survey. Sometimes it’s to qualify people. Is your target market people who enjoy vodka? Then you need to find out — up front — whether or not this is the case.
Too, you may also need it for analysis. If your research goal includes determining if specific social activities correlate to interest in a new product, you have to ask for that data.
But be really careful how you do that. Recently I saw an online survey which asked about net worth as one of the screening questions. That’s right, actual net worth. It even specified, “include any equity in your home should you own a home.” That’s really personal information. Household income is commonly asked about, but total net worth? To me, that crossed the line. It felt rude.
It’s all in how you ask
Social behaviors, personal hygiene and religious views are just a few examples of topics that are all pretty personal to be asking a stranger about, so we have to be very careful about how we ask. First, because we don’t want to turn people off so much that they won’t complete the survey. And second, because we want them to be truthful. So how can we ask for personal information in a way that won’t leave survey participants feeling embarrassed or in any way uncomfortable? How do we to make it easy for them to give us honest, candid information?
Once you’ve determined what you really need to know, there are strategies you can use in your online surveys to collect needed personal information.
• Use ranges. “How many servings of alcohol do you consume in a typical week?” A lot of people aren’t comfortable writing in a specific number. But give them a choice of ranges (none, under 3, 3-5, 6-9, 10 or more) and you’ve made it comfortable and easy for them to respond.
• Give choices. Asking “Do you attend church regularly?” can feel judgmental, a little bit intrusive and uncomfortable. Even if that is what you mean, phrasing it to “feel” more positive is always a good idea. “Please select the statement which best describes your religious activity”, with choices of “no formal services attended”, “attend religious services for holidays and special events”, etc., puts a different spin on the issue, but still gets you the information you want.
• Offer an opt-out. With personal questions, give the respondent an option to decline to answer. It’s a polite way for them to say, “I’m not comfortable answering this question. I’ll finish your survey but I’m not going to give you that one piece of information.” Sometimes just offering that “out” will convince them to go ahead and answer. Typically I find that fewer than 10 percent of people will actually check that option, so you’re still likely getting complete data from a good 90 percent in most cases.
Just don’t go there
And finally, consider: do you really need that personal information? If you’re not going to use it for your analysis and it’s not a qualifying parameter of your target market, why collect it? Rather than collecting a lot of “nice to know” but unnecessary items, select only the ones actually needed.
“…it was the way he said it.”
In the end, only you can decide what information is required to get the job done. If you’re actually going to use specific pieces of personal information for screening or analysis, then ask—nicely. Taking the time to phrase your request politely can make the difference between good data and bad, or maybe no data at all. At the end, the participants may even think, “How polite.” And even children know, if they want something, it is best to ask nice.