Questionnaire design - Agree/Disagree scale I recently constructed a questionnaire to carry out a market survey on soap brands used. One of the questions which was asked was to rate few statements which influence the respondent to buy a certain brand. For each of the statements(Eg: price, advertisements etc) I gave four options strongly agree, agree, disagree, strongly disagree. At that time I decided not to add a "moderate" option. However, I am halfway through collecting the data an I am seriously worried if I had made a big mistake by going with a rating system with even number of responses. Is this okay? At the end of the survey I will have to justify why I did so hence I am in need for some good answers to justify my decision. I would really appreciate if someone could help me with this. Thanks in advance
 A: Even number of choices in Likert's scale is not unusual and you didn't create any monster. So don't worry. Books on how to conduct survey often show scales with odd and even numbers of choice.
The problem (or strength) with even-number scale is that it forces people to take a stand. Usually, this scheme does not flow well with controversial issues cause respondents may just want to hand wave it away by replying with a neutral answer. A tell-tale sign that your respondents being unhappy is high rate of missing. If you see an unusual amount of missing answers, chance is the respondents wanted a neutral item and couldn't find one, thus frustrated and decided to just skip. You can check your interim responses and see if that's the case.
In your situation, whether the decision is justifiable or not depends on if "neutral" makes sense to your questions. If it doesn't, then even number options are fine. For a list of pros and cons about the two approaches, Losby and Wetmore has put a good introduction to both types on line.
