Questionnaire design in survey research happens to be one of the key areas in the process from conceptualization to survey analysis. Questionnaire design is the step that follows the list of required information. This list of required information is generated from the client brief and understanding of brief.

This article doesn't try to educate you on how to design a questionnaire, but on how to avoid some of the common mistakes in a survey research questionnaire. One of the most common mistakes committed by the questionnaire designer is to have overlap of response categories in a close ended question. Consider this for example – there is a question about age of the respondent. It is common practice to offer the response categories like these – upto 25, 25 to 35, 35 to 45 and above 45 years. I think, you may have got the idea. Field interviewer will have trouble placing a respondent of 35 years - in 26 to 35 or 35 to 45. The correct way of asking this question will be - upto 25, 26 to 35, 36 to 45 and above 45 years.

Another area of problems is using number of response categories in a question. Suppose you want to have response to this question - how much are you satisfied with the services offered by the service center? A researcher could use three response categories - very satisfied, neither satisfied nor dissatisfied and very dissatisfied. Another researcher could use five response categories. What is the 'right' number of categories? There is no definite answer to this question. It depends an important aspect of the attribute - wisdom or intelligence of the respondent and differentiability of the response categories. For example, if your respondent is not 'literate' or 'intelligent' enough, she may not be able to articulate her satisfaction level in a precise manner.

However, if one is 'intelligent' enough, using more number of response categories may help. Let me illustrate it this way - your respondents are senior maintenance engineers with more than 10 years of experience. If you are asking a question about a particular brand of furnace or cooling tower, they can precisely articulate their responses even on a nine point scale. However, if the audience is junior types. limiting the number of response categories to a minimum level will be advisable. Last but not the least, language makes a huge difference. The tendency of the questionnaire designer (most of the times) is to impress the reader - the objective should be acquire truthful response. There is never ever any need for flowery or poetic language. Keep it straight, simple and easy to comprehend.

For more information about survey research and survey analysis, visit here: http://www.migindia.biz

Source: www.articlesbase.com