Tuesday, September 3, 2019

Test Should Discriminate

Tests should Discriminate

Tests are designed to discriminate and that is good.  The purpose of a test is to discriminate between people who are good at something and those people who are not.   The key issue is not that tests discriminate.  The key issue is whether the test uses relevant factors in making its discrimination.

For example, driving tests are designed to determine who should be allowed to drive a car and who should not be allowed to have a license to operate a vehicle.   Almost all people would agree that knowledge of traffic regulations is a relevant factor in driving a car.   Vision would also be a relevant factor.   And it is.  But vision can be modified by eye glasses or contact lenses to the point where visual acuity is sufficient to operate a motor vehicle.   If vision cannot be sufficiently modified then the individual cannot be allowed to drive a car.   However, most people would agree that eye color or hair color are not relevant variables and that they should not be used.   On the other hand, some variables such as the ability to parallel park a car have now been determined to be used so infrequently that it is no longer tested in driving tests.

This same paradigm can be used in analyzing math test scores.  There has been a great deal of consternation of late about the decline in math scores on standardized tests.  Some have suggested that the issue is elementary school teachers, who are considered to be generalists, and not sufficiently skilled in math to teach it.  That argument begs the question of why are high school students also having declining math scores when most high school math teachers are specialists in the field.  Next up for concern, is the curriculum itself.   Local school districts with declining scores are rushing to change up the curriculum and the teaching methodology that goes with it.  The theory is that a different and, of course, better curriculum will yield better test scores.   Changing the curriculum has been the “go to” for school systems needing to improve test score since the millennium and yet test scores do not vary that much.   We could blame the students and suggest that climate change is lowering kids’ abilities to do math.  But no one with good sense would go there so we won’t.

To me the most obvious culprit in the declining test scores is the measuring tool being used to take the measure.   Almost 20 years ago a curriculum known as Common Core was developed by educators across the country (NO, it was not imposed by the feds) to create a common standard among states for educating kids so that the skills of a fifth grader in Alabama would be similar to the expectations for a fifth grader in Oregon.   In order to measure how kids were doing in meeting these Common Core standards companies were paid large sums of money to develop the tests that would make the measure.  Maryland chose the group of tests known as PARCC.   From the beginning everyone rushed to teach to the test so their scores would be great.  Teachers were not involved in developing the Common Core curriculum.  It was done by state level educators.   If someone had deigned to ask those same generalist elementary educators, they would have  quickly learned that the children would not be at the developmental levels appropriate to the skills being demanded of them.   In order to worship at the throne of the God of Rigor, skills were pushed down from higher grades to lower ones.   If the kids’ developmental levels are not there yet, no amount of good teaching is going to make the plants bloom before their time.   There is also the issue of relevancy because scores decline as kids get older.   Try to convince an adolescent that the tasks of algebra II are useful in their daily lives, or even will be useful when they become adults.   Good luck with that.

Last school year was the last time Maryland gave PARCC tests.   Now Maryland will give the MCAP ( Maryland Comprehensive Assessment Program) test.   This set of tests will be shorter than PARCC and we are assured that it will be just as rigorous- all bend to rigor.  But the big question is – will the scores be better.   If not, we will again blame the usual suspects:  teachers, curriculum, poor schools.  There is a big elephant in the room but we will not see it.   It will have the letters MCAP written big and bold on its side. And it will discriminate, but will it be discriminating on the correct variables.

No comments:

Post a Comment