A poorly designed questionnaire may not obtain the results the firm or organization is looking for. There are four categories in which numbers are generally grouped. In increasing order of sophistication, they are (1) nominal numbers, (2) ordinal numbers, (3) intervally scaled numbers, and (4) ratio-scaled numbers. This paper will examine each briefly and how they can be used effectively in the design of a survey questionnaire. Qualitative Data
Data that can be categorized into one of several non-numerical categories is qualitative. You are recording some quality that the data possesses. Care needs to be taken to avoid assigning numbers to the categories then computing with them. This type of data is either nominal or ordinal. Nominal
The number we assign to some object, idea, or behavior is entirely arbitrary, although in some cases a tradition may establish the rules of assignment. If measurements are assigned arbitrary numbers, they are called nominal numbers, and their sole purpose in the analysis is to differentiate an item possessing one characteristic from an item possessing a different characteristic. Nominal data is a type of categorical data in which objects do not have a natural, meaningful order. You can count but not order or measure nominal data. Only calculations based on the frequencies of occurrence are valid. Nominal scales have no numeric properties. Qualitative information is obtained from a nominal scale. This means objects are classified by name only. Counting is the only operation that can be performed on a nominal scale. Examples of nominal questions that may be used are: state of residence; gender; or hair color: blonde, brown, red, black. Ordinal
Ordinal data is a type of categorical data in which objects have a natural and meaningful order but no magnitude. You can count and order, but not measure, ordinal data. Calculations based on an ordering process...