Psychometric Properties of Three Addition Tasks with Different Response Requirements
Abstract
Technical advances in personal computers make them very useful for automated testing of questionnaires, personality inventories, and performance tasks. When traditional assessment instruments are adapted for administration by computer, changes in task characteristics output responses are sometimes inevitable. For example, writing numbers on a paper and pencil task versus entering them with number keys on a task administered by computer. Hence, implementing tasks on the computer may change their psychometric properties. Testing tradition implies that the computerized instrument should be validated. Indeed, recent evaluations with systematic criteria suggest such caution is warranted (Bittner, Carter, Kennedy, Harbeson, Krause, 1986; Bittner, Smith, Kennedy, Staley, & Harbeson, 1985; Farrell, 1983, and Smith, Krause, Kennedy, Bittner, & Harbeson, 1983). Recently, we adapted a paper and pencil Addition Task for administration on a portable computer. We found that it seemed insensitive to experimental effects and its response rate was 65% of that for the paper and pencil task. Therefore, we developed another Addition Task to overcome these shortcomings. The purpose of this investigation was to determine the psychometric properties of three Addition Tasks with different response requirements. Reprints.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1988
- Accession Number
- ADA206939
Entities
People
- Alvah C. Bittner Jr.
- Barbara L. Shukitt
- Gary G. Kay
- Louis E. Banderet
- Michael A. Walthers
- Robert S. Kennedy
Organizations
- United States Army Research Institute of Environmental Medicine