As a State Representative in Massachusetts, I have received numerous emails and phone calls from parents, teachers and education organizations, Republican and Democrat parents, political minded citizens and those who are apolitical asking for my support on reconsidering the use of standardized tests in Massachusetts. It didn't take much convincing; I don't support standardized testing, also called high-stakes testing.
There is an argument that standardized testing is useful to develop national standards as with Common Core, to help assess how a student is doing, how a well a teacher is teaching, and which school districts may be in better need of funding. All of these are fair questions to ask. However, not only is standardized testing not answering these questions, there are other problems that arise.
Reasons to Oppose Standardized Testing
Testing takes the fun out of learning, but testing is necessary. However, a standardized test not only takes the fun out of learning, it attempts to create a standard that all students are subjected to and bases progress and competency on a one-size-fits-all approach. A one-size-fits-all approach does not reflect the range of talents and abilities in our students or nation.
Moreover, a one-size-fits-all approach falls apart when we take into consideration special needs students. Special education students learn differently and as such should be assessed differently. Special education teachers have told me that they spend an inordinate amount of time preparing their students to the test compared to non-special education students.
I understand that there is a need for testing to pass subjects and grades. However, there is no need to have a one-size-fits-all approach. Teaching and learning are very personal and should be custom. Assessment should be as well.
Parents and teachers alike report that many students become needlessly pressured and stressed by the test. A little stress and pressure may be good for children to prepare them for the real world, but in this case, standardized tests have the perverse effect of limiting a child's possibilities with a poor test grade.
Additionally, the PARCC test is an online computer test assessing the Common Core standard. However, not all school districts are prepared for this test; many school districts have neither the internet capability nor the computer terminals to administer the test. The result would be that student would be taking the test at different times throughout the day; this is problematic. The time of a day a test is taken affects the test scores. This raises questions of reliability and validity.
Standardized Tests Are Not Even Measuring Student Achievement
Standardized tests are supposed to measure student achievement as a reflection of the teacher's instruction. However, most standardized tests do not rule out third variables as being responsible for the outcome of the test. For example, socioeconomic status, parent involvement, health, sleep, the natural cognitive abilities of the child and many more variables are all are responsible factors when determining a test grade. Standardized tests as they are administered now do not rule out these and other variables.
Measuring students on a continuum against one another invariably creates a bell curve distribution where students at the higher end of scores are the beneficiaries of more than just a good education and the students at the lower end are the by-product of risk factors that contribute to weaker educational achievement. The standardized test scores reflect these extra-educational differences.
In my opinion the proper way to measure whether or not we are doing a good job educating students is far more complex than what we are doing. Proper education testing would involve multiple groups, one of which receive the education curriculum and one of which did not receive the curriculum, where the two groups are basically the same so that we are comparing apples to apples. What standardized tests do right now is just compares students to students without taking into consideration all of the confounding third variables. Standardized test are measuring students against each other on a continuum, which is a descriptive statistic, as opposed to looking at the actual outcome educational instruction with a control group, which is an analytical statistic. Analytical statistics are far harder to produce then descriptive statistics.
Reliability places an upper limit on validity and, as mentioned above, the time of day a test is taken affects students test scores. With this being the case the time a test is taken undermines the reliability of test retest accuracy. If test-retest reliability is weak or non-existent we don't really have a valid test.
Our students need time learning and educators teaching, not preparing for a test that is missing the mark.