School Effectiveness: Test Scores Versus Learning Growth


For decades we have tracked, rated, and ranked schools based on their standardized test scores. The narrative has been that schools with the highest test scores are the best, most effective schools. The connection has seemed so obvious that few voices have questioned the practice.

The high correlation between test scores and the socioeconomic status of students’ families has consistently fed the perception that families with means choose or demand schools that are effective. Again, the rationale seems obvious.

A recent study at Stanford University has called these assumptions and the system that has perpetuated this thinking into question (Spector, 2017). Rather than focusing solely on test scores to measure school effectiveness, Professor Sean Reardon studied the learning growth demonstrated by students over a period of years. Specifically, Reardon collected and analyzed achievement data for students in grades three through eight from 11,000 school districts.

Professor Reardon discovered that despite where students started on the performance continuum relative to other students, their academic growth over time was not predicted by their third-grade scores. In fact, he discovered that students in many schools and school districts in high poverty areas demonstrated more growth than their counterparts in wealthier neighborhoods and communities.

Reardon’s analysis points to early childhood and initial elementary grade performance as heavily influenced by community socioeconomic-related factors such as income and family education levels. Yet, growth between third and eighth grade appears to have minimal relationship to these factors. The perception that students in poverty are not capable of learning at a rate comparable to more economically advantaged students appears not to be based in reality.

So, what should educators, parents, and communities take from this study? First, we cannot assume that students from families that are challenged socioeconomically are not capable of learning growth comparable to students from other backgrounds and environments. Rather, we need to encourage, support, and expect learning success.

Second, we need to do all we can to provide young children from more challenged and less supportive environments with every learning opportunity we can to prevent them from having to start behind and catch up with their more advantaged schoolmates. It is not fair that they must learn more and progress faster than other students to catch up with them.

Third, we can analyze data on the growth of our students’ learning rather than be satisfied with their current performance as an indicator of our effectiveness and support. We cannot determine what happens in the lives of students before they reach us, but we can focus our attention and efforts on ensuring that they are making progress each day, week, and year they are with us.

Fourth, we can help students focus on their learning growth rather than place disproportionate attention on test scores alone. When students see their progress and understand their power to determine this progress, we can help them counter the confidence-sapping impact of comparisons with other students who may be at different points on the academic performance continuum.

Fifth, we can advocate for accountability systems that consider the trajectory of student learning, not solely reflect the advantages that families and communities provide before students start formal education. After all, doesn’t it make more sense to pay attention to how students are progressing rather than assume current performance is all that matters?

Are there other factors that continue to have an influence on the learning of students beyond third grade? Of course, and we need to be aware and do all that we can to mitigate and prevent these forces from derailing or diminishing student learning. Meanwhile, we cannot be satisfied with or take too much pride in how our students are doing unless they also are growing.

Spector, C. (2017, December 5). Students’ early test scores do not predict academic growth over time, Stanford research finds. Stanford News. Retrieved from

Leave a Reply

Your email address will not be published. Required fields are marked *