Effects of autocorrelation on single-subject single-facet crossed-design generalizability assessment

Hoi K. Suen, Patrick S.C. Lee, Steven V. Owen

Research output: Contribution to journalArticlepeer-review

4 Scopus citations

Abstract

The effects of first-order serial dependence on estimates of variance components and common reliability indices in single-subject single-facet crosseddesign generalizability analyses were investigated by adding autoregressive and moving-average components to four sets of direct observation data known not to contain serial dependence. Results indicated that, whereas universe score, relative error, and expected observed score variance estimates were systematically biased, the amount of bias with signal/noise ratios, generalizability coefficients, and the D-study analog of Cronbach alpha were negligible.

Original languageEnglish (US)
Pages (from-to)305-315
Number of pages11
JournalBehavioral Assessment
Volume12
Issue number3
StatePublished - Dec 1 1990

All Science Journal Classification (ASJC) codes

  • Psychiatry and Mental health

Fingerprint Dive into the research topics of 'Effects of autocorrelation on single-subject single-facet crossed-design generalizability assessment'. Together they form a unique fingerprint.

Cite this