An Endless Summer: How COVID Has Reversed Academic Achievement

There is no doubt that the coronavirus pandemic slammed educators and policy makers with a herculean task of pivoting from classrooms to virtual instruction. At the same time, there is no dispute that learning gains that have occurred since school buildings were closed have been negatively affected. The only open question has been: how bad is it?

 

The Center for Research on Education Outcomes (CREDO) at Stanford University calculated the magnitude of student-level learning losses on behalf of nineteen of our state research partners for the 2019-2020 school year. The results are not good.

 

Every summer, students forget some of what they learned during the prior school year. We call this learning decay.

 

We can see the same phenomenon with COVID. As schools sent their students home in mid-March, students were effectively on an extended summer break. Every day they were out of the classroom, they fell further behind. 

 

We modeled the learning decay that students faced during the extended break. The results are tragic.  

 

Of the nineteen states we reviewed, the largest learning loss in reading was about 183 days of schooling, equal to one school year. The smallest effect was about 57 school days. That’s still approximately one-third of the school year. 

 

Losses in math achievement were even larger. The average loss was 136 days of schooling and the largest was 232 days.

 

These findings lead to four important considerations.

 

First, recovering from the 2019-2020 losses will take years. The significant differences in lost learning between states and particular cohorts of students reflect the fact that school closures had highly unequal impacts. Disadvantaged students generally suffered much more than students from advantaged families. And as the pandemic continues, we can assume learning loss will continue, so recovery will be extended.

 

Second, the wide variation within states (and often within schools) means that conventional models of classroom-based instruction – a one-to-many, fixed-pace approach – will not meet the needs of students. New approaches must be allowed to ensure high-quality instruction is available in different settings, recognizing that different skills may be needed for the different channels. It is worth asking who is preventing those approaches from being adopted.

 

Third, the need for rigorous student-level learning assessments has never been higher. In particular, this crisis means we must adopt strong diagnostic assessments and frequent progress checks. Both should be aligned with historical assessment trends to plot a recovery course. 

 

Fourth, the measures of average loss and the range around it immediately call into question the existing practice of allowing communities to plot their own paths forward. Communities suffering the largest estimated learning losses generally do not have the means and capacity to create and implement school improvement plans on their own - we should anticipate similar limitations as they build and deploy their recovery plans. Insistence on local autonomy in this case will not yield equitable responses.

 

The sad reality is that the shutdowns have had an unequal effect on many students. Subsequent shutdowns through the 2020-2021 school year will only magnify those effects. Recovering from the shutdown requires us to transform the way we educate our children, and that we do it soon.