Social and Philosophical Foundations of Education

A lesson and a Question
Page 1 of 1

Author:  Stella_Brewer [ Tue Jun 06, 2006 9:00 pm ]
Post subject:  A lesson and a Question

On page 32 in The Big Test, Lemann explains the difference b/w reliablility (how one person scores on administation of same test/usually multiple choice) and validity (the tests' ability to predict future outcome/like student's grades for next year). This was a good lesson for my statistically challenged mind. However, it developed another question...If the SAT did not show true validity and was weak on reliability; how in the heck did it become the be all end all TEST?

Author:  Jen Davis [ Tue Jun 06, 2006 9:53 pm ]
Post subject: 

Maybe this is where our discussion from last night comes in. Perhaps all the evidence that they really needed or ever saw was the "correlation" that existed between the test takers and their freshman years on campus (and please remember how extremely unreliable and unrelated correlations can be at times). The Ivy League schools had a purpose for the test- all they really needed were some indicative results and a group to administer to. You know I actually paused when reading that part because it was the first time that those 'terms' were used explicitly in this text. Ironically, those same explicit terms are those things which are plaguing us under the NCLB act right now. Scientific based research (SBR) demands that results be both valid and reliable (among other things) and so we in turn are being forced to look for products and programs that have landed themselves within this noteworthy category because they accurately 'correspond' to the prescribed formulas we need to reach our goal. So what is the big deal- who knows? But it seems to be a vicious circle. The SAT itself seemed to have encouraged and even gave birth to countless other exams and tests within the educational field. And yet, one must also acknowledge the lasting hand and impressions made by the SBR movement spurned to life by the NCLB act.

Author:  gayle [ Wed Jun 07, 2006 2:19 pm ]
Post subject: 

This is EXACTLY why I brought up the issue of correlation v. causality on Monday. Great connection, Jen!

There are other aspects of this that we will talk about more tonight, for example, the correlation between parental income and student test scores on the SAT. Historically, the strongest correlation has been this one: parental income goes up, child's SAT score is higher, and vice versa.

This is also a disturbing (but unsurprising) correlation on EOG scores - along with the correlation between lighter skin color and higher EOGs. Again - correlation, but not causality!

Author:  suzanne averett [ Thu Jun 08, 2006 2:33 pm ]
Post subject: 

How does an IQ test in the third grade determine your placement in reading and math for the rest of your formal schooling? Our principal was saying just the other day that they're probably going to stop giving the IQ test to any grade but third. Apparently they(whoever they is) think that third grade testing is so valid and reliable that it will only be given that year. How does that test carry so much weight?

Page 1 of 1 All times are UTC - 5 hours
Powered by phpBB® Forum Software © phpBB Group