A comparative test of magnitude estimation and pair-comparison treatment of complete ranks for scaling a small number of equalinterval frequency response anchors

Chester Schriesheim, Claudia C. Gardiner

Research output: Contribution to journalArticle

9 Scopus citations

Abstract

This study investigated whether previously-found differences in two sets of recommended five-point equal-interval response anchors could have been caused by scaling too many stimuli at one time. One set of recommended anchors, produced by magnitude estimation, was compared with a set produced by Thurstone Case III pair-comparison treatment of complete ranks. Subjects (N =110) completed magnitude estimations and rankings of 8 frequency expressions. Few of the scale values produced by magnitude estimation differed significantly from the means obtained in previous studies or from the “ideal” values expected of exactly equal-interval anchors. However, this outcome was not true of the Case III results (they were seriously discrepant). Implications for future research are discussed.

Original languageEnglish (US)
Pages (from-to)867-877
Number of pages11
JournalEducational and Psychological Measurement
Volume52
Issue number4
DOIs
StatePublished - 1992

ASJC Scopus subject areas

  • Developmental and Educational Psychology
  • Applied Psychology
  • Applied Mathematics
  • Education

Fingerprint Dive into the research topics of 'A comparative test of magnitude estimation and pair-comparison treatment of complete ranks for scaling a small number of equalinterval frequency response anchors'. Together they form a unique fingerprint.

  • Cite this