Ask an elementary teacher how much science she taught last week, and you will often get an apologetic answer. Twenty minutes on a Friday. Something after the reading block if it ends on time. In many schools, science becomes a real subject in the grade where the state test arrives.
That is not a failure of teachers. It is arithmetic. When reading and math carry the accountability weight, science is what gets cut, and it is cut hardest in the schools under the most pressure. Which means the students who arrive with the least science knowledge from home are the ones most likely to get the least at school.
Over the past decade, the science of reading has changed more than literacy instruction. It changed how educators argue. The conversation moved from competing philosophies to harder questions: What does the research show? What does good instruction look like? Which students are falling behind, and what do we do before the gap widens?
Science instruction never got that treatment. It should.
The cumulative nature of learning science
Scientific knowledge is cumulative in ways that are easy to underestimate. A student cannot make sense of heredity without cells, or chemical reactions without matter, or Earth systems without energy.
When a fifth grader misses the foundation, she does not simply score lower that year. She carries the hole forward into every science course that follows, and by high school it looks like a student who is bad at science rather than one who was never taught it.
We already know what to do when this happens in reading and math. We screen, we intervene, we check results and we adjust. Science almost never gets that machinery.
In most districts, there is no science equivalent of a reading intervention block, no routine check on whether a misconception from third grade is still sitting there in seventh.
Identifying misconceptions early and giving students structured opportunities to revisit foundational concepts can prevent unfinished learning from compounding from one grade to the next.
Evidence on science interventions
That gap led me to the evidence on science interventions. I want to be careful about what I found and what I did not.
I synthesized 13 independent school-based datasets covering more than 33,000 students in grades three through eight, comparing students using Penda Learning, a science intervention platform, with peers receiving their district’s usual instruction.
In the largest of those datasets, covering 46 schools and roughly 19,500 students, outcomes were measured on assessments I did not control: the nationally normed NWEA MAP Science test and the Florida state science assessment.
The results were consistent. Students using the intervention outperformed peers by a statistically significant margin at every grade examined and were meaningfully more likely to reach proficiency. The gains held for students in Title I schools, English learners and students with disabilities, which matters more than the headline number.
Interventions that work only for students already close to the line are not much use to the schools that need them.
One finding surprised me. Time on the program was not the strongest predictor of growth. Mastery was. Students who completed activities with 80% accuracy or better gained substantially more than those who simply logged more minutes. That is uncomfortable for anyone who buys a program and then measures success by usage dashboards, a common practice in education.
Here is what I am not claiming. This is one intervention, studied across many sites and several years. It is a replication, not a synthesis of a field, and no single study should drive a purchase or a policy. The science of reading did not reshape literacy because of one paper. It gained authority because findings piled up, converged and withstood attempts to knock them down.
Science education deserves that same accumulation of evidence, and we need far more of it.
Key questions for district leaders
So the useful question for district leaders is not which program to buy. It is whether they are asking the questions they already ask about reading. Is science taught in every grade, or only where a test is waiting? When a student falls behind, does anyone find out before the following spring?
When a vendor claims results, what was the comparison group, who was included, was the outcome measured on an assessment the vendor did not design and is the reported gain large enough to be worth the money?
Districts ask those questions about reading curricula. Very few ask them about science. None of this requires treating science and literacy as rivals.
Students who investigate a phenomenon, argue from evidence and explain their reasoning in writing are doing literacy work, and strong readers learn science more easily. The trade-off districts think they are making, protecting reading time by shortening science, may be costing them some of the reading gains they are trying to protect.
The science of reading worked because a field decided to stop arguing from conviction and start arguing from evidence. Science education has not yet had that argument. It is overdue.