03 September 2008

Standardized Tests and Behavior Driven Metrics

One problem with public education and standardized tests is the way bureaucracies subdivide test results in order to gain some meaningful measurements from the results. Every year in Texas after tests are taken, many schools sit on pins and needles waiting to see if their “sub-pops” make the grade. “Sub-pop” is a bit of a euphemism for racial group or economic background. Students are measured at grade level, and then by racial/ethnic/ background within that grade level. Students are also divided by “economic situation”, meaning poorer students get grouped.

The trick is that if one of a school’s “sub-pops” doesn’t make the grade for a given year, the school is rated “unacceptable”. After all, no child can be left behind, right?

But it is my opinion that these “sub-pops” attempt to measure school success by dividing students in precisely the wrong way. It is arguably a racist position to think that a child’s educational success ought to be calculated with his or her skin color as a factor. Similarly, factoring in the student’s economic situation erroneously places the focus on something the kid can’t change. And while there may be a correlation between economic status and school achievement, one hardly causes the other any more than color does. Yet these are the primary methods for dividing and measuring student success today.

Instead, students should be divided by other objective criteria which probably have far greater impact on their educational success. Students might be “sub-popped” based on the student’s social promotion or grade retention history. Students’ criminal background might create another meaningful “sub-pop”. Students with a history of absenteeism might create a meaningful “sub-pop” as well.

These criteria might not only look at the at-risk end of the spectrum, either. Students with high attendance histories or records of high achievement could form “sub-pops” as well. However, the bulk of students at most schools would simply fall into a middle ground “sub-pop”; most students are, after all, just regular students working to graduate.

The thing of note in these alternative criteria is that they are driven by student behavior – something the student can control – and not race or economic background – things the student cannot control. After all, if the point of these testing metrics is to ensure student success, doesn’t it make greater sense to focus on student actions when creating metrics?

(Note: I realize that the privacy issue with regard to student records may well make these kinds of metrics impossible. However, those privacy policies are created by government, and government rules can always be changed.)

5 comments:

Anonymous said...

How about the "sub Pop" of students who for year after year are assigned inexperienced, poorly certified (if any) in content, and little to no training in the classroom.
How about a "sub-pop" of students not given textbooks and only State mandated Test booklets or worksheets?
How about a "sub pop" of students in over crowded and run down facilities with one of the highest paid administrations in the nation?
Wait, I think all DISD would become a sub pop and that defeats the purpose.

Bob M. said...

Anonymous,

Thanks for reading and your input. I appreciate your frustration with these issues. You mention come very specific problems which are quite important, but ones which I did not intend to address in this post. I really wanted to focus on the erroneous sub-pops that students are lumped into. Other things in other posts.

Cheers.
Flapjack

Anonymous said...

Flabjack,

I loved the article. I have often wondered why it is important to divide everyone by race or ethnicity? I even suggested once that we all bubble in "other" on one of the standardized test...of course the administration told me we could not do that...they never gave me an explanation as to why we all couldn't be classified as "other". So, I started doing it on my own. While I'm not ashamed of my ethnicity. I am a "Black-American" I don't feel it has anything to do with my job, my education, my socioeconomic status, etc. If there just has to be a descriptor then I chose to be "Black-American". However, I notice often that it is not listed as a choice...so, I become an "other". I do not use the "African-American" label simply because I was born in America just as my parents and their parents. We are of African descent...but, since the government must "describe" my heritage...I am an other. Is the US the only country that labels ethnicity in this manner?

It's amazing how over the years there have been so many "ethnicity" labels...octoroon,quadroon, mulatto...what difference does it make? None. These type divisions have driven families apart in our culture and now it's tearing our school systems apart.
Even in the presidential race Barack Obama's mixed heritage is often mentioned...but we hear nothing of McCain, Palin, or Biden's ancestry...why is that?

Bob M. said...

Anonymous II,

Thanks for the insightful commentary. One might think that in what is supposed to be a country moving toward "racial neutrality" or "beyond racism", bureaucracies could move beyond these groupings as well. However, it seems that in all bureaucracies, thinking in easily definable categories (with which one can "calculate" success or failure) is a paradigm that will never die. It's simply too difficult to calculate with some other metric; better to rest on the "easy" metric of "race/ethnicity".

By extension, it's probably too much to ask for subjective qualitative analysis of students and teachers. That would require a level of professionalism (or training, or institutional honesty) that may be lacking. But perhaps I go to far here...

Cheers!
Flapjack

Anonymous said...

While there are some privacy issues of concern here--you're right, however, those are available for change--I really like the fact that your version of metric focuses on what we as individuals are capable of altering. I understand the need for accountability of students, teachers, and families in measuring student success, but if our measurements are focused more on individual initiative rather than often irrelevant factors such as race and SE status, wouldn't that empower each and every one of us? And wouldn't that be very "American" and humane?

A measurement of results based largely on individual, historical choice is somewhat analogous to "local control" of schools, something NCLB does not encourage. Institutions and bureaucracies are also accountable for "professional" behavior; by encouraging and rewarding forward-thinking, responsible behaviors, we can assure their continuance. (Success breeds success.) Isn't this preferable to the mandated, mediocrity-inducing effects of NCLB?

Thanks, Flapjack, for some interesting fodder. Bookwurm1010.