tag:blogger.com,1999:blog-5841910768079015534.post1075850933542071800..comments2024-03-18T08:28:01.624+00:00Comments on BishopBlog: Great Expectations: Our early assessments of schoolchildren are misleading and damaging deevybeehttp://www.blogger.com/profile/15118040887173718391noreply@blogger.comBlogger4125tag:blogger.com,1999:blog-5841910768079015534.post-5795598357845876292015-05-15T16:43:12.729+01:002015-05-15T16:43:12.729+01:00The "misleading and damaging" consequenc...The "misleading and damaging" consequences related to "age" is only part of the "expectations" etiology. Not only is the "measurement" information faulty, it mis-directs attention from the instruction/schooling to accomplish the "expectation"--if any formal effort is at all necessary. "Norming" and other statistical adjustments skate over this fatal flaw in the logic, rather than removing it.<br /><br /><b>follow instructions involving several ideas or action...express themselves effectively, showing awareness of listeners’ needs</b><br /><br />On the one hand, all children have been doing these things from birth. On the other hand, every adult is "emerging" in the expectations. <br /><br />The antidote is simple in principle: Specify an observable instructional consequence of interest. Specify minimal prerequisites for instruction to accomplish the intent. Devise products and protocols for reliably attaining the expectation. That's standard science/technology practice, but it has yet to be applied to schooling.<br /><br /><br /> Dick Schutzhttp://ssrn.com/author=1199505noreply@blogger.comtag:blogger.com,1999:blog-5841910768079015534.post-53384850657893674142015-05-06T11:20:26.515+01:002015-05-06T11:20:26.515+01:00Dear Franck and Tim
Thanks for clarification.
As ...Dear Franck and Tim<br />Thanks for clarification. <br />As you note, yes, of course, we'd expect tests to reflect age changes. Indeed, if they didn't we'd worry whether they were sensitive and valid measures. The point I was making was that those devising EYFS don't seem to have realised that!<br />They therefore have a measure that is confounded with age.<br />Ddeevybeehttps://www.blogger.com/profile/15118040887173718391noreply@blogger.comtag:blogger.com,1999:blog-5841910768079015534.post-60602507117606575212015-05-05T14:13:57.764+01:002015-05-05T14:13:57.764+01:00Makes sense that scores are age-dependent. But ess...Makes sense that scores are age-dependent. But essential then to use age-linked norms for decision making. And in research to include age-in-days as a covariate, probably along a growth curve (age^2 age^3), and nest children in class to account for young child in advanced class, versus older child in a less-advanced classroom.timhttps://www.blogger.com/profile/12707381996365946983noreply@blogger.comtag:blogger.com,1999:blog-5841910768079015534.post-71197861638407242592015-05-05T12:34:44.649+01:002015-05-05T12:34:44.649+01:00I understand what you mean and I agree, but there ...I understand what you mean and I agree, but there is a risk that you will misunderstood. <br />A test is not flawed just because it shows an important age (or month of birth) effect. Indeed, there will always be such an effect, in a system where there is one grade per year, so children in the same class have (at least) a 12-month age range.<br />The main conclusion that should be drawn from this observation is that test scores can only be interpreted relatively to very precise norms, that provide centiles trimester by trimester, if not month by month (and, ideally, taking into account both age and grade). This should be feasible if the EYFS has been administered nationwide. And of course, standards of achievement should be defined realistically, based on these very norms, rather than a priori. Franck Ramushttps://www.blogger.com/profile/02656240693713885894noreply@blogger.com