CCP 05-14-1990 - Laserfiche WebLink

Home Browse Search

15 . analyses have ranged from .87 to .96, indicating an extremely high level of reliability for these questionnaires. In addition, we have conducted test-retest measures of the reliability of our questionnaires. The results of these tests have also indicated that these instruments are reliable measurement devices. As one assessment of the validity of our surveys, we verify that all respondents have the same interpretation of terms, phrases, and questions included in the survey. This enables us to determine whether an instrument is measuring what it purports to be measuring. Other means of checking the validity of a questionnaire may be considered (i.e., correlation, factor analysis, etc.); however, because the number of respondents is relatively small for most pretests, this type of check is seldom conducted prior to data collection for the main study. . c. Questionnaire Format For the resident questionnaire, primariiy closed questions having a predetermined set of response options that are exhaustive and mutually exclusive will be used, This means that the response options will include all possible answers and that no one can legitimately respond with more than one answer to a given question, Such wording avoids confusing respondents and prevents problems of interpreting multiple responses to a single question. The questionnaire will be constructed so the layout and design prevent confusion and are compatible with the data analysis strategy. After Anderson, Niebuhr & Associates, Inc. has prepared a draft of the survey questionnaire, copies will be provided for review and critique by City of Arden Hills representatives. Suggestions. for additions, deietlons, or other changes will be incorporated into a revised version of the survey instrument. The revised questionnaire will then be pretested. . -----