Atkins court decision: Farad Roland v USA (NJ; 2018)

Today the opinion regarding the Atkins ID decision for Farad Roland was issued.  As per my policy, having served as an expert witness in this particular case, I offer no comments.  The opinion can be found here.

Dr. Doug Detterman's bytes: Psychometric validity

I have been remiss (busy) in my posting of Dr. Doug Detterman's bytes. Here is a new one on validity

Validity is the extent to which a test measures what it is supposed to measure and predicts what it is supposed to predict. When Binet developed his intelligence test, his goal was to identify children who would not do well in school so they could be given help. To the extent that Binet's test identified such children, it was valid. In Binet's case, proving the validity of the test amounted to showing that the test predicted or correlated with school performance. (Binet was handicapped, though, since the correlation coefficient was not widely known at the time of his first test.) Note that there is no requirement to provide an explanation of why the test predicts what it was designed to predict, only that it do it. Validity provides an empirical relationship that may be absent of any theoretical meaning. Theoretical meaning is given to the relationship when people attempt to explain why the test works to produce this validity relationship.

Tests designed to predict one thing may be found to predict other things. This is
certainly the case with intelligence tests. Relationships between intelligence and many other variables have been found. Such relationships help to build a theory about how and why the test works and ultimately about the relationship of the variables studied.

- iPost using BlogPress from Kevin McGrew's iPad

intelligence IQ tests IQ testing IQ scores CHC intelligence theory CHC theory Cattell-Horn-Carroll human cognitive abilities psychology school psychology individual differences cognitive psychology neuropsychology neuroscience psychology special education educational psychology psychometrics psychological assessment psychological measurement IQs Corner general intelligence intelligent IQ testing validity Doug Detterman

Generated by: Tag Generator

Intelligent IQ testing: Joel Schneider on proper interpretation of composite/cluster scores

Dr. Joel Schneider has (again) posted an amazing and elegant video tutorial to help individuals who engage in intelligence test interpretation understand whether composite/cluster scores should be interpreted as valid when the individual subtests comprising the composite are significantly different or discrepant (according to Dr. Schneider--"short answer: not very often"). It is simply AWESOME...and makes me envious that I don't have the time or skills to develop similar media content.

His prior and related video can be found here.

Clearly the message is that the interpretation of test scores is not simple and is clearly a mixture of art and science. As Tim Keith once said in a journal article title (1997)...."Intelligence is important, intelligence is complex." This should be modified to read "intelligence is important, intelligence is complex, and intelligent intelligence test interpretation is also complex."


- iPost using BlogPress from my Kevin McGrew's iPad

intelligence IQ tests IQ testing IQ scores CHC intelligence theory CHC theory Cattell-Horn-Carroll human cognitive abilities psychology school psychology individual differences cognitive psychology neuropsychology neuroscience psychology special education educational psychology psychometrics psychological assessment psychological measurement IQs Corner general intelligence intelligent IQ testing validity composite IQ scores

Generated by: Tag Generator

Visual-graphic of how to develop psychological measures of constructs

I found this figure, which I had developed a few years ago for a specific grant process (thus the scratched out box that is not relevant to this post), which summarizes in a single figure the accepted/recommended approach to developing and validating tests. In simple terms, one starts with the specification of the theoretical domain construct(s) of interest, then examines the measurement domain for possible types of tests to operationalize the constructs, and then one develops and scales the test items (optimally using IRT scaling methods) Very basic. Thought I would share---I love visual-graphic explanations.

Double click on image to enlarge





- iPost using BlogPress from my Kevin McGrew's iPad

intelligence IQ tests IQ scores CHC theory Cattell-Horn-Carroll human cognitive abilities psychology school psychology individual differences cognitive psychology neuropsychology special education educational psychology psychometrics psychological assessment psychological measurement IQs Corner neuroscience neurocognitive cognitive abilities cognition