Insight into Nebraska ACT Scores & Potential Disparities
An estimated 80% of post-secondary schools have stepped away from requiring students to submit their scores from a standardized college readiness exam when applying for admission into their university. Some criticisms for using standardized college readiness test scores in admissions stem from underlying racial biases that may be inherent in the tests alongside findings regarding their weak predictive value for college academic performance and success. While many do not require scores from these tests for admissions, universities may still accept these scores for qualification for merit-based scholarships.
Regardless, since the passage of LB 930 and spring of 2017, Nebraska has used the (nearly tautological) American College Testing (ACT) Test as the primary form of accountability testing in its public school systems. Accountability tests measure and track a student’s progress to assist in evaluating the quality of a school’s (or state’s) pedagogy, curricula, and programs for school improvement purposes. High stakes are often attached to such tests given the prospect of school sanctions placed on under performing schools that have not shown growth.
Unless identified as requiring an alternative assessment, all students in their junior (third) year of high school in Nebraska are administered the ACT to participate in its reading, English, math, science, and writing subject tests. Oftentimes, reading, English, and writing are combined into one English language arts (ELA) score. One’s ACT subject test scores can range from 1 to 36 while an ACT composite score is the calculated average of all four subject test scores which is rounded to the nearest whole number. To categorize student performance on the ACT, subject test and composite scores are binned into one of three groups: developing, on-track, and met benchmark. The table below illustrates how scores are divided into one performance level category.
Table 1
ACT Benchmark Score Intervals
| Performance Level | Math | ELA | Science |
|---|---|---|---|
| ACT Benchmark | 22-36 | 20-36 | 23-36 |
| On Track | 18-21 | 18-19 | 19-22 |
| Developing | 1-17 | 1-17 | 1-18 |
In terms of college readiness, students who score in the “developing” category likely need additional support or remediation before their first year of college courses to succeed. Students whose scores are “on-track” can likely access first-year collegiate course content without remediation. Lastly, students at or above the ACT benchmark are equipped to succeed in first-year courses and have an estimated 50% chance of getting at least a B or 75% chance of attaining a C or above in these courses.
The freely available data used in this post derives from the 2021-22 Nebraska ACT scores that are reported by the Nebraska Department of Education (NDE) (I could not get my hands on the updated 2022-23 AY scores). While we are lucky to have access to such information, the dataset is split into pre-defined strata such as student sex, race/ethnicity, disability status, or unhoused subgroups. Unfortunately, this means that I cannot examine the intersection of multiple groups, identities, or statuses at once. Further, there is no way to distinguish junior and senior student scores.
Three exemplar school districts were selected out of personal curiousity: Grand Island Public Schools, Lincoln Public Schools, and Elkhorn Public Schools. Tab through the information below to see data, figures, and discussion prevalent to each of these districts. Note: some figures require you to scroll to see its entirety.
To illustrate what data is available, Table 2 gives a preview of the (lightly cleaned and transformed) data for the mathematics subject test results from my alma mater, Grand Island Senior High. Specifically, the displayed table shows results for male and female students along with those who had free or reduced-price lunch status and those who had full lunch price status (i.e., a proxy variable for the student’s economic background). The way that the data was given allows me to examine differences across reported student sex, but not across student sex and economic status.
An interesting exercise involves comparing male and female student average ACT scores and benchmark percentages to the state’s averages. Contrast these differences by hovering your mouse over the below horizontal bar charts below.
Table 2
ACT Mathematics Scores for Grand Island Public Schools
| Subgroup | Average Score | Developing | On-Track | Hit Benchmark |
|---|---|---|---|---|
| All Students | 15.10 | 76.1% | 12.11% | 11.78% |
| Female | 14.94 | 80.42% | 10.49% | 9.09% |
| Male | 15.24 | 72.31% | 13.54% | 14.15% |
| Economically Disadvantaged | 13.94 | 83.55% | 10% | 6.45% |
| Non-Economically Disadvantaged | 16.30 | 68.44% | 14.29% | 17.28% |
.
Fifteen percent more students who were considered economically disadvantaged scored in the developing category than those not economically disadvantaged. The percent of non-economically disadvantaged students with math scores that hit the ACT benchmark (17.28%) was more than double the percent of scores that hit the ACT benchmark from economically disadvantaged student group (6.45%). The descriptive table also conveys that more male students are scoring at benchmark in mathematics than female students, though the average math subject test score is not marginally different (~0.30 point gap). Oppositely, a higher percentage of female students are scoring in the developing category than male students.
For the math subject test, the average state of Nebraska ACT scores is:
- 17.74 for all students
- 17.60 for female students
- 17.87 for male students
- 15.10 for economically disadvantaged students
- 19.33 for non-economically disadvantaged students
Personally, I do not find these scores as being enormously different from the average scores from Grand Island Public Schools (~2 to 3 point gap across groups). All Islander average scores exist in the interval of scores that is considered to be developing (scores <18). There are noticeable differences in benchmark scores between average Islander and state ACT math scores. For GIPS, about 15% more male, female, and “all student” groups scored in the developing category at Grand Island Senior High than the overall state. The percent of economically disadvantaged students who scored developing for GIPS closely matched the percentage from the state while 25% more non-economically disadvantaged GIPS students scored in the developing interval than the state.
On average, 27% of students hit the ACT benchmark across all male, female, and “all student” groups for the state of Nebraska while less than 15% of Islander male, female, and “all student” groups hit the ACT benchmark. The percent of the state’s non-economically disadvantaged students met the ACT benchmark was nearly double that of non-economically disadvantaged GIPS students.
It is also astonishing to see that, for the state of Nebraska, economically advantaged students are three-times more likely to have scores that hit the ACT mathematics benchmark than the non-economically advantaged counterparts.
Moving West, the next school district analyzed is Lincoln Public Schools (LPS). In Table 3, we delve into English Language Arts ACT scores between male and female students. Within LPS, a ~10% gap in benchmark and developing scores placed female students above male students. About half as many students who were not economically disadvantaged showed developing ELA scores compared to students who were economically disadvantaged. Meanwhile, 30% more LPS students who were not economically disadvantaged had scores that met the ACT benchmark than LPS students who were economically disadvantaged.
The same comparisons between state and school district can be done by using Table 2 and complementary state average horizontal bar chart.
Table 3
ACT English Language Arts Scores for Lincoln Public Schools
| Subgroup | Average Score | Developing | On-Track | Hit Benchmark |
|---|---|---|---|---|
| All Students | 16.33 | 52.93% | 11.54% | 35.53% |
| Female | 17.22 | 47.63% | 11.34% | 41.03% |
| Male | 15.49 | 57.96% | 11.73% | 30.31% |
| Economically Disadvantaged | 13.07 | 74.13% | 8.84% | 17.03% |
| Non-Economically Disadvantaged | 18.52 | 38.74% | 13.35% | 47.91% |
.
Besides the economically disadvantaged group, the average LPS student scores across all defined groups fell into the developing score category (an average score <18). When comparing LPS to the state, it the Lincoln Public School district closely matches the state percentages and average scores within a 4% margin, meaning that the LPS district is a good example of what is expected of the state for ELA ACT subject test.
The state’s percent of non-economically disadvantaged students who had scores that hit the benchmark was 2.65 times that of the economically disadvantaged students. Almost 32% more economically disadvantaged students scored in the developing range than non-economically disadvantaged students in Nebraska for the ELA ACT subject test.
For the ELA subject test, the state ACT average score is:
- 16.33 for all students
- 17.22 for female students
- 15.49 for male students
- 13.70 for economically disadvantaged students
- 18.43 for non-economically disadvantaged students
Unlike GIPS, LPS had a slightly greater percentage of scores that met the ACT benchmark and slightly fewer scores that were categorized as developing than the state averages.
The final school district inspected is Elkhorn Public Schools (EPS), which is notorious for its wealthier student population. Table 4 gives details about the average ACT science subject test scores and benchmark placements from EPS. In contrast to GIPS (and LPS to a lesser extent LPS), there are striking differences in a) the percent of EPS students who had developing scores, b) the percent of EPD students who scored at or above the ACT benchmark, and c) the average scores for all groups. Similar deviations are also exhibited when comparing EPS ACT scores to the state’s average scores. Male students scored ~10% higher than female students in EPS. Lastly, the average science score for female student falls just below the “Hit benchmark” interval while the average science score for male students fell just within it.
Table 4
ACT Science Scores for Elkhorn Public Schools
| Subgroup | Average Score | Developing | On-Track | Hit Benchmark |
|---|---|---|---|---|
| All Students | 22.64 | 23.06% | 20.78% | 56.17% |
| Female | 22.03 | 25.07% | 24.79% | 50.14% |
| Male | 23.20 | 21.23% | 17.14% | 61.64% |
| Economically Disadvantaged | 19.17 | 51.85% | 20.37% | 27.78% |
| Non-Economically Disadvantaged | 22.92 | 20.81% | 20.81% | 58.38% |
.
Across all groups, the percent of EPS students who scored at or above the ACT benchmark was double that of the state’s average percent of students who hit the same benchmark. Correspondingly, EPS had less than half as many students score in the developing category than the state did.
The state average ACT science subject test scores are:
- 18.06 for all students
- 18.12 for female students
- 18.02 for male students
- 15.31 for economically disadvantaged students
- 19.71 for non-economically disadvantaged students
Across all groups in Table 4, EPS student science scores were higher than the state averages. What I find particularly interesting is the category of economically disadvantaged EPS students. This group had a lower percentage of students scoring in the developing interval, greater percentage of on-track scores, and a higher percentage of students who hit the ACT benchmark than the state. This begs the question: are economically disadvantaged students in other wealthy schools exhibiting higher-than-state-average ACT scores as depicted here? What contextual effects, such as additional teacher benefits, curricular changes, behavioral support systems, school resources, parental involvement, and student characteristics differentiate a wealth-advantaged school from a wealth-disadvantaged school? This type of inquiry will lead future NEED blog posts!
While existing disparities between ACT scores, student sex, and economic background may be worrying for some, the review of gaps in scores across race/ethnicity groups discussed below is a major cause of concern. The following bar charts refer to the state of Nebraska’s ACT scores and are separated by percentage of student scores categorized as developing, on-track, and met benchmark across every self-reported race/ethnicity group. In addition, the average ACT score for every group is printed in bold font at the base of every bar. Within every charts, bars are clustered by ACT subject test (ELA, math, and science).
To keep the general size of each chart consistent, the y-axis changes between every ACT benchmark score category. Please be cognizant of y-axis changes when interpreting. To avoid confusion, percent labels are attached to the top of every bar.
Across the three ACT subject tests, Black or African American students demonstrated the greatest percentage of developing scores, closely followed by Hispanic students. On the other end of the scale, Asian and White students were the groups that demonstrated the smallest percentage of developing scores. The disparity between these groups indicates that the percentage of students who are Black or Hispanic students that show developing ACT scores are twice that of the Asian or White student groups. Per this category’s definition, the percent of Black students in Nebraska who need remediation to be successful in their first year of college is 1.84, 1.90, and 1.99 times greater (for ELA, math, and science, respectively) than the percent of White students who would need the same supports. The percent of Hispanic students in Nebraska who need remediation to be successful in their first year of college is 1.70, 1.71, and 1.80 times greater (for ELA, math, and science, respectively) than the percent of White students who would need the same supports. This is startlingly bad and brings light to existing educational disparities!
The above charts depict that White and Native Hawaiian or Pacific Islander student groups have the greatest percentage of students who scores are considered “on-track” while Black and Hispanic student groups have the lowest percentage of student scores that are considered “on track” for success. This is evident across the three ACT subject tests. The state did not report any on-track ELA scores for the Native Hawaiian or Pacific Islander group, which could mean that there were truly no cases of this in the state or that missing data is present (is that an oxymoron?).
As you can see, the disparities across race/ethnicity subgroups between percentage of student scores that met ACT benchmarks are worse than percent of race/ethnicity subgroup developing scores. Hispanic and Black student groups have the lowest percentage of scores that met the ACT benchmark across all three subject tests. On the opposite end of the scale, White and Asian student subgroups exhibit the greatest percentage of students whose scores met the ACT benchmark. Per this category’s definition, the percent of Asian students in Nebraska whose scores indicate that they are equipped for success in their first year of college is 3.72, 7.34, and 5.49 times greater (for ELA, math, and science, respectively) than the percent of Black students whose scores illustrate college readiness. Similarly, the percent of Asian students in Nebraska whose scores indicate that they are equipped for success in their first year of college is 2.83, 4.21, and 3.34 times greater (for ELA, math, and science, respectively) than the percent of Hispanic students whose scores illustrate college readiness.
Before making conclusions about Nebraska education, it is wise to compare these percents to national percentages. Using the data from the “Percent Meeting Benchmark” from the ACT Tableau Data Viewer, I was able to find the percent of students who met the ACT benchmark standards across each subject tests. Though, unlike the Nebraska ACT data that combined reading and writing tests into an ELA score, the reading and writing subtest scores are separated. Likewise, percentages and scores found below are for students who graduated in 2022, which discludes third-year students.
Nationally, the percent of Black students who graduated in 2022 that met the ACT benchmark scores include:
- 27% in English (average score of \(M_{English}=15.0\))
- 9% in math (average score of \(M_{math}=16.0\))
- 18% in reading (average score of \(M_{reading}=16.6\))
- 10% in science (average score of \(M_{science}=16.5\))
The national percentages for Hispanic students who graduated in 2022 that met the ACT benchmark scores include:
- 38% in English (average score of \(M_{English}=16.6\))
- 18% in math (average score of \(M_{math}=17.5\))
- 28% in reading (average score of \(M_{reading}=18.3\))
- 19% in science (average score of \(M_{science}=18.0\))
Because I am unable to find the total and benchmark sample sizes used to compute the percentages, I cannot conduct significance testing, such as two-proportion z-tests. This would have allowed me to identify if the proportion of Black or Hispanic students had scores that hit the ACT benchmark in Nebraska significantly differed from the national proportion of Black or Hispanic students whose scores hit the ACT benchmark.
Stemming from these findings are three lines of thought:
School resources are inequitable across the state. An economically disadvantaged student in a wealthy school feels the benefits of that school’s resources when compared to economically disadvantaged students in less wealthy schools. For many, this is a “duh” moment, but it still deserves acknowledgement. How does resource accessibility and school funding relate to Nebraska public schools’ racial and ethnic demographics in Nebraska? If you want to further complicate this issue, check out how the Nebraska funds its public school system.
Nebraska high schools may be struggling to support their Black and Hispanic scholar’s college readiness, as exhibited by the gaps in ACT scores between state and national averages. Nebraska ACT scores from students in Black or Hispanic subgroups were lower across all subtests when compared to the United States averages. Similarly, the percents of Black and Hispanic student scores who met the ACT benchmarks in Nebraska were much lower than national percents of Black or Hispanic student scores who met the ACT benchmarks. In most cases, the Nebraska percentages of benchmark scores in Black or Hispanic students were half that of the national percentages. A major caveat to this discussion is that the ACT site specified that percent and average score values were from the graduating class (i.e., fourth-year students) of 2022. One should be cautious when directly comparing scores across the two data sources (NDE and ACT).
The ACT scores do not appropriately reflect the college readiness for Black and Hispanic scholars. This argument is pointed at the validity of the ACT rather than educational practices in Nebraska.
As mentioned before, there are caveats to this description of Nebraska ACT scores that may muddy the brief analysis above. I was unable to examine the intersection of subgroups to see how multiple student statuses and characteristics may play into ACT scores. In addition, it was not possible to identify whether students paid for additional ACT resources (e.g., ACT preparatory books, classes, practice tests), though income is a common indicator of such resources.