Grade 11 ACT Scores

Last week, we shared that the grade 11 ACT scores from last spring were essentially the same as the ACT scores from the prior year.  When we examined the school-level, however, there were some schools that had serious increases from prior ACT scores!

We’ve posted the data on our website and included change in scores so you can dig into it yourself.

Three high schools really caught our eye, and we want to celebrate Scranton High, Shirley High, and Eureka Springs High for the improvement made by their students since Grade 11 ACT testing began in the 2015-16 school year. We present the ACT data from these schools below.

These schools have demonstrated large increases in the percentage of students meeting ACT college readiness benchmarks in all four subject areas (Math, English, Reading, and Science).  A student who meets the ACT College Readiness Benchmarks has at least a 50% chance of getting a “B” or better in the corresponding college course. Statewide, 14% of students meet this criteria, and the rates in these three districts are higher than in similar districts.

Scranton High, Shirley High, and Eureka Springs High Schools haven’t only shown improvements in the percentage of students meeting all four ACT readiness benchmarks, however.  Students from these schools also increased the average ACT score in each content area over prior performance and are outscoring districts serving similar populations.

You may notice that these schools serve a small number of students- meaning that the scores are more likely to fluctuate than those in larger districts. We agree, but many small schools have had big decreases in scores over time, and we think the three-year trend is a good start.

Scranton High, Shirley High, and Eureka Springs High schools also had strong ESSA growth scores last year, and were in the top 10% for growth overall, and the top 5% of high school for growth in ELA.  This got us to wondering about how ESSA growth (which doesn’t involve 11th graders at all) is related to ACT score improvement?  We were also wondering how 11th grade ACT scores are related to school poverty rates, and to the 10th grade ACT Aspire scores from the prior year.

ACT improvement and ESSA growth:

We might anticipate that ESSA growth values and ACT improvement would be related, because if a school is seeing growth from the students who complete the ACT Aspire (given in grades 3-10), then it would make sense that students who complete the ACT would also demonstrate growth. As you can see in the scatter plot below, however, there is essentially no relationship between the values at the school level (r=0.09).

While we presented English scores, the relationship was consistent for math growth as well. We thought maybe it was because some of the schools we are considering as ‘high schools’ serve a wider grade range than just 9-12, and so maybe the growth values was based on growth in lower grades, but when we limited the sample to only schools serving grades 9-12 the correlation remained as low as with the whole sample. We think this lack of relationship highlights the difference between the longitudinal growth model and a simple change value. The ESSA growth score includes more sensitive information about how students are performing longitudinally, while the ACT information which is just snapshots of the scores of two different groups of students in subsequent years.

ACT scores and 10th grade ACT Aspire scores:

This got us wondering about how correlated were these schools’ 10th grade ACT Aspire scores from 2016-17 with the 11th grade ACT  scores from 2017-18.  We would expect they would be correlated, since the ACT Aspire is meant to predict performance on the ACT!  Although the state-level values didn’t change for either assessment, we wanted to examine the relationship at a school level.  We use the OEP’s weighted GPA to examine the relationship between the scores, and find the values are strongly correlated (r=0.84).  Math scores were more correlated (r=0.92), while ACT science scores were less correlated with the 10th grade performance (r=0.70).

Given that the ACT scores are so correlated with ACT Aspire scores, students, parents, teachers, and other education stakeholders should take action on ACT Aspire results in earlier grades to address any areas of academic weakness well before the ACT.

ACT scores school poverty rates:

As we would expect with assessments measuring academic performance, we find that the ACT is negatively correlated with school poverty rates (r=-0.58), so it is important to compare your school’s performance to those serving similar student populations.

We are so glad that the state is investing in our students and providing them with all the opportunity to take the ACT in 11th grade, but just testing students doesn’t help them learn.  We need to work hard to effectively use information from both formative and summative assessments to support students in their learning.