“If you cannot measure it, you cannot improve it” Lord Kelvin
Despite the recent proliferation of standardized testing in education, there is still a significant number of countries that oppose it. I’ve heard many arguments against standardized testing from policy makers, teachers and school directors, but two of them seem persuasive at first glance. The first one is that the test’s main purpose is to hold teachers and school directors accountable, that is, to reward and punish them based on students’ performance and—per tests’ opponents—this is unfair. The second is that since standardized testing assesses few subject areas, it redistributes attention and resources to these subjects in detriment of other equally important areas of the curriculum. These are valid points, but, as I argue below, they do not justify incurring the very high cost of not testing.
In Latin America, the region that I am most familiar with, standardized testing is often placed in the basket labelled neoliberalism (ipse dixit). The association is not random. Partly based on the “No Child Left Behind” (NCLB) policy in the US, boosted with ideology, for many teachers and school directors, the only purpose of standardized testing is to hold them accountable—euphemism for blaming them—for the poor leaning outcomes of students. Many Latin American teachers and school directors perceive (rightly so) standardized tests as an unfair measure of their true performance, since the test scores levels mainly reflect the socio-economic conditions of students, rather than the teachers’ value added. If not handled properly, the detractors claim, this summative use of standardized tests can lead to confrontation between parents and teachers resulting in even worse conditions for learning.
I fully agree and understand teachers’ fears, but the risks can be addressed with the right test design and guidelines on its uses. As proven by experiences in Mexico and the US, linking test scores with teacher rewards creates incentives for strategic behavior (cheating) among teachers which can end up destroying the credibility of the standardized test. Although providing tools so that citizens can hold public service providers accountable is desirable, I don’t think it should be the only nor the main purpose of testing. Instead, the main purpose of the test should be formative, in other words, the test should help teachers, school directors and parents identify the main challenges and put together a school improvement plan to address them. Governments should provide technical assistance to schools, to help them understand and interpret test results such as score cards and use them as an input in school improvement plans. When presented like this, improving poor learning outcomes is a responsibility shared by schools, parents and the government and standardized tests can align incentives among all three stakeholders. As proven by impact evaluations in the Mexican state of Colima and the Argentine province of La Rioja, this low-stakes accountability or shared responsibility approach is politically less contentious and far more effective to improve learning outcomes than the NCLB approach.
The second argument against standardized tests has its roots in the purpose of education. Although the mission of the school has evolved through time to reflect changes in society’s economic and political context, broadly speaking, education systems have the objective of ensuring that all learners acquire the skills necessary to be productive and participate actively in a democratic society. This purpose goes clearly beyond ensuring numeracy and literacy and includes citizenship and socio-emotional skills.
Basic foundational skills (numeracy and literacy), citizenship and socio-emotional skills are all important to achieve the purpose as stated by many education systems around the world. Per the mission statements of a large sample of US schools, none of these three is more important than the other. Nevertheless, we are currently just assessing and monitoring progress in a few basic skills but not in citizenship and socio-emotional skills.
Is this misalignment between the purpose of education and standardized testing, inadvertently crowding out civic and socio-emotional skills by only assessing numeracy and literacy? The evidence from Mexico and the US suggests that the answer is no. The focus on math and reading comprehension test scores that resulted in a significant improvement in these subjects in Colima had a positive but statistically insignificant impact on a third rotating area (science) included in the Mexican standardized test.
Lack of evidence on the crowding out hypothesis is reassuring, but if testing for formative purposes can improving basic skills, shouldn’t we be following a similar strategy in all—equally important—areas? If the purpose of most education systems is to ensure learners acquire the skills necessary to pursue the life they aspire to, and since this includes socio-emotional skills like empathy and self-efficacy or civic skills nurtured via history and other liberal arts courses, then the metric used to track the system should be perfectly aligned to this objective. There is an ongoing debate on how to define and measure socio-emotional and civic skills, perhaps the lack of a precise and widely accepted definition is part of the problem. This is a challenge we should address soon if our education systems will become the catalyst for inclusive growth in open and democratic societies. If only we had a “thermometer” to measure empathy, tolerance or grit.
In my opinion standardized tests cannot be used for assessing students and it can not be criteria for assessing teachers and other stakeholders of education. There are so many factors which affects the performance of students.