[Oz-gifted] tests in the public system

[sol1]

Reliability in psychological tests refers to consistency or stability in 
measurement, not prediction of adult IQ or anything to do with ceilings.

The smaller the measurement error, the greater the reliability or 
repeatability of the test results.

Reliability is measured by test retest reliability (over time reliability) 
where higher reliability means that when a person takes the test on 2 
separate occasions it is more likely the first result will predict the 2nd; 
by internal consistency reliability, where statistical tests check that each 
individual item correlates with each other, and by intertester reliability, 
where the same test given by a number of testers gives the same (or similar) 
results.

Basically the statistical data at age 8, on some tests, suggests the test is 
more reliable (with less measurement error) than at age 5

I am not sure if that is any clearer than mud, sorry

Fin


----- Original Message ----- 
From: "Matthew & Tracey Wood" <mwood at iinet.net.au>
To: <oz-gifted at cobia.ed.qut.edu.au>
Sent: Tuesday, September 19, 2006 7:16 PM
Subject: Re: [Oz-gifted] tests in the public system


>
> Hi Ellen,
>
>>Sorry -- I meant that the reliability of testing instruments for IQ is
> supposedly lower at 5 than at 8.  I don't know >whether that's due to
> ceiling issues, younger children's unfamiliarity with testing, etc, but it
> would be interesting if >anyone here has any definitive answers.
>
> I would have said the opposite was true...that they are more likely to be
> affected by ceilings at an older age. The only problem I could anticipate
> with testing at a young age would be if they were unable to concentrate 
> for
> long periods of time. Our DD was tested at 5.5 and we had no problems at
> all.
>
>>DET's guidelines are for formal assessment to take place before skipping a
> year, and they seem to be sticking to the >script.
>
> I think that's reasonable and necessary and we certainly made sure that
> happened before S was accelerated. I wouldn't consider acceleration 
> without
> a full assessment first.
>
> Tracey
> --
>>Ellen Hrebeniuk
>>Sydney, Australia
>
>
> _______________________________________________
> oz-gifted mailing list
> oz-gifted at rite.ed.qut.edu.au
> http://lists.rite.ed.qut.edu.au/mailman/listinfo/oz-gifted
>
>
> -- 
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.1.405 / Virus Database: 268.12.5/450 - Release Date: 9/18/2006
>
>