Re: WASL Fact/Thought of the Day

[Mike Kluznik <mkluznik@HOTMAIL.COM>]

Gerald,
"About once a month I consider forming the National Coalition for the
Aboliktion of Correlation Coefficients."
I think this is a wonderful organization.  I'd like to start a chapter in my
area.
Lately, the ARN list has been revisited by the small school issue.  You have
to wonder what is due to cause & effect and what is due to correlation in
the effectiveness studies that support small schools.
I work in a brand new, large high school.  It really is a beautiful school.
Lately, I've been talking to students and teachers, trying to get their
reaction to the new building.  I haven't heard any negative comments from
students.  But, I digress...
We do need a NCACC chapter in every state.  Please send me sign-up
materials.
Mike




> From: gerald bracey <gbracey@EROLS.COM>
> Reply-To: Assessment Reform Network Mailing List <ARN-L@LISTS.CUA.EDU>
> To: ARN-L@LISTS.CUA.EDU
> Subject: Re: WASL Fact/Thought of the Day
> Date: Sat, 22 Sep 2001 07:23:34 -0400
>
> Mickey:
>
> What two variables do you want to correlate?
>
> In any event, be careful.  About once a month I consider forming the
> National Coalition for the Aboliktion of Correlation Coefficients.  They
> cause so much trouble because human beings have an irresistable impulse to
> infer causality from correlation.
>
> If variables A and B are correlated, you don't know if A causes B, B causes
> A, both A and B are affected by some other variable, or if it's just
> chance.  The cliche is that in education everything is correlated about
> .30.
>
> "Variance explained" or "variance accounted for" is a very slippery concept
> in my opinion because it seems to lead even more to a causal
> interpretation.  I use it rarely and usually do not point to the specific
> value.
>
> Recently, for instance, I wrote that the typical correlation between the
> SAT and freshman college grades is about .45, meaning that the SAT explains
> only 20% of the variance in those grades.  The point is, no matter what the
> precise figures might be, most of what goes into those grades comes from
> other qualities.
>
> Jerry
>
>   ----- Original Message -----
>   From: Mickey VanDerwerker
>   To: ARN-L@LISTS.CUA.EDU
>   Sent: Friday, September 21, 2001 11:13 PM
>   Subject: Re: WASL Fact/Thought of the Day
>
>
>   In a message dated 9/20/2001 6:23:31 PM Eastern Daylight Time,
> aburke@VANSD.ORG writes:
>
>
>
>     A correlation coefficient is a numerical index of the strength of the
> linear relationship between two measures.  The square is sometimes
> interpreted as "proportion of variance explained."
>
>
>   How would I go about asking for this information from our state
>   MIckey


_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp

--------------------------------------------------------------------------
To unsubscribe from the ARN-L list, send command SIGNOFF ARN-L
to LISTSERV@LISTS.CUA.EDU.