[etni] Fw: relialbility

----- Original Message ----- From: "Motti" <motti46@xxxxxxxxxxxx>
Subject: re: relialbility

I am sorry, my previous answer did not come through like planned.

Dear Ziona,
If you cared to read what I wrote in my original note you could
possibly have saved yourself the trouble of commenting that
"this is not reliability".
In that note I referred readers to a quotation from an academic source:

http://www.georgetown.edu/departments/psychology/researchmethods/researchanddesign/validityandreliability.htmTest/RetestTest/RetestTest/retest
 is the more conservative method to estimateRELIABILITY.Simply put, the idea 
behind test/retest is that you should get the samescore on test 1 as you do on 
test 2. The three main components to thismethod are as follows:1) implement 
your measurement instrument at two separate times for eachsubject;2) compute 
the correlation between the two separate measurements; and3) assume there is no 
change in the underlying condition (or trait you aretrying to measure) between 
test 1 and test 2.This means that if a studenttakes Moed A and gets 70 and then 
Moed B a month later and gets 100, themeaning of these scores is that the test 
is NOT RELIABLEMotti Mendelkern
--------------------------------------------
Visit the ETNI site - http://www.etni.org
Send a message to the ETNI list: etni@xxxxxxxxxxxxx
Send queries and questions to: ask@xxxxxxxx
--------------------------------------------

Other related posts: