This program is used to summarize item data from questionnaires for which there are no known correct answers. It produces a distribution of the proportion of people answering each response and an optional frequency distribution, weighted mean, and standard deviation for each item. Unless the user indicates otherwise, blank or invalid responses are ignored if the mean is computed. The weight values that are applied to each response for computing item means may be positive or negative numbers or decimal fractions. The "best" alternative for each item, the one receiving the highest weight in computing the mean, also is indicated. The program optionally may print the items to the left of the response distributions.
The QUEST analysis is produced for the entire sample. It may also be broken down by different independent variables, such as discussion sections within a course.
A separate proportion distribution with labeling may be provided for up to 15 "group items," such as sex, grade, or status, which usually cannot be weighted reasonably. Means and standard deviations may be provided for these demographic items.
Interpretation of output
The printout from the QUEST program has three parts. The "group item" printed on the top of the output is optional; it contains descriptive information about the group which usually cannot be weighted reasonably. In the example, sex is the group information variable; 13% of the sample are female, 63% are male, and 24% did not indicate their sex.
*Abstracted with modifications from MERMAC Manual, Test and Questionnaire Analysis, Programs Written for the IBM System/360, University of Illinois Press: Urbana, 1971.
The center portion of the printout summarizes the responses to the questionnaire items and also may print the items to the left of the responses. The questionnaire itself consists of statements to which the students could respond Strongly Agree (SA) Agree (A) Disagree (D) Strongly Disagree (SD). The proportion and frequency responding to each alternative and the proportion and frequency omitting are indicated for each item. The most desirable or favorable response is indicated under Best. This Best response indicates the weighting scheme for computing item means applied to the alternatives. In this case, when SA is the best response, the weighting scheme is SA = 4, A = 3, D = 2, E = 1; and when SD is the most favorable response, SA = 1, A = 2, D = 3, SD = 4. That is, the higher the mean, the more favorable are the respondents to that item.