<html>
  <head>
    <meta content="text/html; charset=windows-1252"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">Trevor:<br>
      <br>
      Q3 is the correlation between the raw residuals of pairs of items.
      These are listed, for instance, in Winsteps Table 23.99 with
      PRCOMP=R<br>
      <br>
      Summarizing the Q3's across all items with PCA makes the
      dependency pattern clearer. See, for instance, Winsteps Table 23.1<br>
      <br>
      Hope this help.<br>
      <br>
      Mike L.<br>
      <br>
      On 3/27/2018 10:38 AM, Bond, Trevor wrote:<br>
    </div>
    <blockquote cite="mid:D6DFCF97.1786F%25trevor.bond@jcu.edu.au"
      type="cite">
      <div style="color: rgb(0, 0, 0);"><font face="Times"><span
            style="font-size: 16px;">Dear Colleagues,</span></font></div>
      <div style="color: rgb(0, 0, 0);"><font face="Times"><span
            style="font-size: 16px;">It has been claimed that <span
              lang="EN-US">Yen’s Q3 index </span></span></font><span
          style="font-family: Times; font-size: 16px;"><span
            lang="EN-US">(1984) </span></span><span style="font-family:
          Times; font-size: 16px;"><span lang="EN-US">can be used to
            identify local dependence in Rasch analyses.</span></span></div>
      <p class="MsoCommentText"><span lang="EN-US"><font style="color:
            rgb(0, 0, 0);" face="Times">e.g., Chou, Y.-T., & Wang,
            W.-C. (2010). Checking dimensionality in item response
            models with principal component analysis on standardized
            residuals. Educational and Psychological Measurement, 70(5),
            717-731.</font></span></p>
      <p class="MsoCommentText"><span style="font-size: 12pt;
          font-family: Times;">Christensen, K. B., Makransky, G., &
          Horton, M. (2017). Critical values for Yen’s Q3:
          Identification of local dependence in the Rasch model using
          residual correlations. Applied Psychological Measurement,
          41(3), 178-194.</span></p>
      <p class="MsoCommentText"><span style="font-size: 12pt;
          font-family: Times;">Have you used Q3 for that purpose? Do you
          know of papers that have adopted it? Is it helpful/effective?
          Is it available in the software that you use, or know about?</span></p>
      <p class="MsoCommentText"><span style="font-size: 12pt;
          font-family: Times;">Looking forward to your advice and
          guidance.</span></p>
      <p class="MsoCommentText"><span style="font-size: 12pt;
          font-family: Times;">Collegially</span></p>
      <p class="MsoCommentText"><span style="font-size: 12pt;
          font-family: Times;">Trevor</span></p>
    </blockquote>
    <br>
    <br>
    <div class="moz-signature">-- <br>
      Mike Linacre, <a class="moz-txt-link-abbreviated" href="mailto:mike@winsteps.com">mike@winsteps.com</a> or <a class="moz-txt-link-abbreviated" href="mailto:winsteps1234@gmail.com">winsteps1234@gmail.com</a><br>
      Winsteps 4.0.1 and Facets 3.80.4 - <a
        href="http://www.winsteps.com/index.htm"><a class="moz-txt-link-abbreviated" href="http://www.winsteps.com">www.winsteps.com</a></a><br>
      Concerned about aches, pains, youthfulness? Mike suggests
      <a href="https://www.modere.com/liquidbiocell?referralCode=790842">Liquid
        Biocell</a></div>
  </body>
</html>