Can someone neither agree nor disagree?
One of the recurring debates in the research community is about using "even" or "odd" scales. This refers to scales such as:
Where the labels go from one extreme to the other (this Likert scale is a common label set). An odd number of scale points has a neutral point in the middle, while an even scale requires the respondent go to one side or the other.
The argument for even scales (4 or 6 levels) is that everyone has an opinion and a neutral point is essentially a cop-out.
The argument for odd scales (5 or 7 levels) is that a respondent's experience with an element may have been limited or mixed, and they may not have a firm opinion one way or the other.
Last time this cropped up on the American Marketing Association Market Research discussion list, the vast majority were in favor of providing a mid-point.
In my experience, I've also found it important to provide a Not Applicable, N/A, or No Opinion option. If you can, it's best to use skip patterns to bypass sections which may not apply to a respondent, such as rating e-mail assistance when they've only called customer service. However, there are times within a category of ratings when a respondent doesn't feel they can rate a question or two. On paper they can just skip the item; on the Web they can also skip non-required responses, but there's a catch on radio button scales like the one pictured here. Once you select an option on a set of radio buttons, you can change your response, but you can't clear it—hence the value of an N/A.
I was keen on getting all the juice out of our survey findings. Your consultation helped us do that and I would do it again!