I have a copy. The .99 correlation seems to be coming from this:
> Paired sample t-tests showed that children diagnosed as gifted in this study (N = 44) achieved equal scores on the HAWIK-IV (M = 126.9, SD = 7.1) and the IDS (M = 128.9, SD = 8.2), t(43) = –1.30, p = .20. Average nongifted children (N = 69) scored equal as well on the HAWIK-IV (M = 101.2, SD = 8.5) and the IDS (M = 99.86, SD = 0.03), t(68) = 1.42, p = .16.
(IDS is this new line/weight-based test, the HAWIK-IV the usual IQ battery.)
Thanks for the reference and for the quotation (to the parent and grandparent of this reply). Yes, the HAWIK would be the usual child IQ battery in a German-speaking country. I'm not following the statistics shown there completely, but I note the sample size. Did the same group of test-givers give both tests? Were they "blind" as to the results of each test when giving the other?
I appreciate the references. I'm still doubtful that the general finding would be that the five-minute test would be strongly correlated with full scale child IQ batteries, e.g. the WPPSI. The way to find out would be for other groups of test-givers to attempt to replicate the result.
> Paired sample t-tests showed that children diagnosed as gifted in this study (N = 44) achieved equal scores on the HAWIK-IV (M = 126.9, SD = 7.1) and the IDS (M = 128.9, SD = 8.2), t(43) = –1.30, p = .20. Average nongifted children (N = 69) scored equal as well on the HAWIK-IV (M = 101.2, SD = 8.5) and the IDS (M = 99.86, SD = 0.03), t(68) = 1.42, p = .16.
(IDS is this new line/weight-based test, the HAWIK-IV the usual IQ battery.)