Clinical Strategies for Sampling Word Recognition Performance.

Purpose: Computer simulation was used to estimate the statistical properties of searches for maximum word recognition ability (PB max). These involve presenting multiple lists and discarding all scores but that of the 1 list that produced the highest score. The simulations, which model limitations i...

Descripción completa

Detalles Bibliográficos
Publicado en:Journal of Speech, Language & Hearing Research Vol. 61; no. 4; pp. 936 - 945
Autores principales: Schlauch, Robert S., Carney, Edward
Formato: Artículo
Publicado: American Speech-Language-Hearing Association Apr2018
Materias:
Acceso en línea:Ver este registro en EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=ssf&AN=129160615&site=ehost-live
header:
  @attributes:
    shortDbName: ssf
    uiTerm: 129160615
    longDbName: Social Sciences Full Text (H.W. Wilson)
    uiTag: AN
  controlInfo:
    bkinfo:
    jinfo:
      jid:
        10924388
        1SM
      jtl: Journal of Speech, Language & Hearing Research
      issn: 10924388
      maglogo: N
    pubinfo:
      dt: Apr2018
      vid: 61
      iid: 4
      pid: 42
      pub: American Speech-Language-Hearing Association
    artinfo:
      ui:
        129160615
        10.1044/2017_JSLHR-H-17-0236
      ppf: 936
      ppct: 9
      formats:
        fmt:
          @attributes:
            type: P
            size: 688KB
      tig:
        atl: Clinical Strategies for Sampling Word Recognition Performance.
      aug:
        au:
          Schlauch, Robert S.
          Carney, Edward
        affil: Department of Speech-Language-Hearing Sciences, University of Minnesota, Minneapolis
      su:
        Phonemics
        Decision making
        Word recognition
        Sampling (Process)
        Performance evaluation
      sug:
        subj:
          Phonemics
          Decision making
          Word recognition
          Sampling (Process)
          Performance evaluation
      ab: Purpose: Computer simulation was used to estimate the statistical properties of searches for maximum word recognition ability (PB max). These involve presenting multiple lists and discarding all scores but that of the 1 list that produced the highest score. The simulations, which model limitations inherent in the precision of word recognition scores, were done to inform clinical protocols. A secondary consideration was a derivation of 95% confidence intervals for significant changes in score from phonemic scoring of a 50-word list. Method: The PB max simulations were conducted on a "client" with flat performance intensity functions. The client's performance was assumed to be 60% initially and 40% for a second assessment. Thousands of estimates were obtained to examine the precision of (a) single lists and (b) multiple lists using a PB max procedure. This method permitted summarizing the precision for assessing a 20% drop in performance. Results: A single 25-word list could identify only 58.4% of the cases in which performance fell from 60% to 40%. A single 125-word list identified 99.8% of the declines correctly. Presenting 3 or 5 lists to find PB max produced an undesirable finding: an increase in the word recognition score. Conclusions: A 25-word list produces unacceptably low precision for making clinical decisions. This finding holds in both single and multiple 25-word lists, as in a search for PB max. A table is provided, giving estimates of 95% critical ranges for successive presentations of a 50-word list analyzed by the number of phonemes correctly identified.
      pubtype: Academic Journal
      doctype: Article
      src: R
    language: English
    refInfo:
    copyright:
      @attributes:
        flag: N
    holdings:
      @attributes:
        islocal: N