Computer-Based Language Assessment Software: The Effects of Presentation and Response Format

• Published in: Language, speech & hearing services in schools Vol. 30; no. 1; pp. 68 - 74
• Main Authors: Haaf, Robert, Duncan, Brent, Skarakis-Doyle, Elizabeth, Carew, Maria, Kapitan, Paula
• Format: Journal Article
• Language: English
• Published: United States ASHA 01.01.1999; American Speech-Language-Hearing Association
• Subjects: Automation; Children & youth; Communication; Computer Applications; Computer Assisted Testing; Computer Software; Computer Uses in Education; Computers; Educational Technology; Elementary Education; Evaluation Methods; Experimenter Characteristics; Feedback (Response); Intonation; Language; Language Impairments; Language Tests; Peabody Picture Vocabulary Test; Peabody Picture Vocabulary Test (Revised); Physical Disabilities; Receptive Language; Semantics; Speech; Speech Language Pathology; Speech therapists; Standardization; Standardized tests; Stimuli; Student Evaluation; Test Format; Test Items; Test Reliability; Test Results; Test Validity and Reliability; Validity; Ontario Canada; Canada
• ISSN: 0161-1461; 1558-9129; 1558-9129
• DOI: 10.1044/0161-1461.3001.68

Nonstandard presentation and response formats are often employed by speech-language pathologists when clients are unable to perform the pointing response required of many standardized tests. However, any adaptation to test administration potentially compromises the norming standards of that test. The present study investigated the effects of computerized presentation of the Peabody Picture Vocabulary Test-Revised (PPVT-R, Dunn & Dunn, 1981) Form M, using two computer-based response formats. Seventy-two normally developing children between the ages of 4:0 (years:months) and 8:11 participated in this study. Subjects were randomly assigned to one of three response conditions: (1) standard presentation -direct pointing, (2) computer presentation-trackball, and (3) computer presentation-automated scanning. Statistical analyses revealed that there were no differences in performance across the response conditions, which is consistent with the hypothesis that computerized testforms are equivalent to standard forms. Age was strongly related to increased accuracy regardless of condition. Thus, the adapted response formats of the computerized version constitute statistically equivalent forms of the PPVT-R and can be used with the published norms for this test.