Methods to Evaluate Scales and Sample Size for Stable Task Inventory Information
Abstract
Methods were developed to determine: (1) the stability and redundancy of responses to two job scales--the continuous Relative Time-Spent scale and the dichotomous Task-Performed scale, (2) the stability of 'job types' (i.e., clusters of job incumbents) derived from scale responses, and (3) the change in stability when sample size is reduced. Results indicated that the Task-Performed scale yields stable, meaningful task information (i.e., percentages of personnel performing tasks) from responses by job incumbents, but no practical gain in information is achieved from the Relative Time-Spent scale. A better way to collect time-spent data is proposed. Findings also demonstrate that highly stable scale data and cluster solutions are obtainable from samples substantially smaller than those presently administered. The study's empirically developed relationship between sample size and stability can be usefully employed to determine cost-effective sampling for task inventory surveys.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 01, 1980
- Accession Number
- ADA085600
Entities
People
- David W. Robertson
- John J. Pass