Writing items for tests and surveys is often more difficult than it appears. There are professional guidelines for item writing but it is also an art that might be best learned through experience and observing how items perform when they are actually used.

How an item is written can have a substantial impact upon how subjects respond to it and inadvertently skew the results. As an example, I received what was labeled a survey in the mail the other day. Here are a couple of items:

-Do you believe that Barack Obama's nominees for federal courts should be immediately and unquestionably approved for their lifetime appointments by the U.S. Senate?

-Do you believe the best way to increase the quality and effectiveness of public education in the U.S. is to rapidly expand federal funding while eliminating performance standards and accountability?

To be honest, I suspect that the underling intent of the survey is to motivate political supporters (as opposed to actually being an opinion poll) because it was accompanied by a fund raising request. And regardless of one’s political leanings, I suspect that it would be difficult to find anyone who would endorse either of these statements as written. However, it illustrates how an item is constructed can impact the results. In this case, I think that it is fair to conclude this was the intent. Problems do occur, though, when the goal is gather information in an unbiased way but poor item writing gets in the way.

Reid Klion