ERIC Number: ED097956
Record Type: Non-Journal
Publication Date: 1973
Reference Count: N/A
Determining a Sufficient Sample Size for Analysis of Mail Questionnaire Data. Center for Human Services Development, Report No. 33.
Cohen, Allan S.; And Others
This paper presents the description and several applications of a model which can be used to determine how long a researcher must wait for the return of completed mail questionnaires to be sure the data collected reflect the true values of the parameters of interest. This model proposes to fit two linear time trend lines to a set of mailed questionnaire data. Application of the model provides an estimate of the intersection point during the time period at which the estimate of the parameter has stabilized. It is at the stabilization point that the researcher can be relatively confident that his estimates do not differ significantly from those which he would make were he to wait for additional returns to be received. To illustrate the application of this model, questionnaires were sent to 1,120 child care centers in Pennsylvania. Nine weekly subsamples of the 545 returned questionnaires were analyzed according to the model to estimate the average maximum allowed enrollment, the proportion of centers either licensed or approved, and the proportion of centers reporting Black children in attendance. Also, the appendices contain analyses of the data from each of twenty categories of information contained in the questionnaire. (SDH)
Publication Type: Reports - Research
Education Level: N/A
Sponsor: Pennsylvania State Dept. of Public Welfare, Harrisburg.
Authoring Institution: Pennsylvania State Univ., University Park. Coll. of Human Development.
Identifiers - Location: Pennsylvania