University of Arkansas Office for Education Policy

Arkansas’ Achievement Gaps: Analysis By Race and Gender

In The View from the OEP on April 24, 2019 at 2:00 pm

This week we continue our examination of long-term achievement for different student groups on the state assessments. Today we are discussing differences in average achievement in math and literacy by gender and race. We think this is important descriptive work for education stakeholders throughout the state, because identifying and acknowledging the differences in achievement is a first step to ameliorating them. 


As we mentioned in our earlier blog about the achievement gap between students participating the Free and Reduced Lunch program (an indicator of low family income), for this analysis we use ten years of the annual Arkansas state assessments, which have changed over time. For the first six years of our analysis we use data from the Benchmark and End-of-Course exams. In the 2014-15 school year, students in Arkansas completed the PARCC exam, before switching to the ACT Aspire exam for 2016-17 and beyond.

All of these test changes have made it essentially impossible for Arkansas schools to examine achievement gaps over time, due to unique test scoring and reporting of each assessment. Although the assessments have varied over time, each provides a measure of how students statewide perform in literacy and mathematics. To enable the examination of performance over time, we have used a common standardizing procedure to track the relative performance of different groups.  We transform these scores into percentiles to aid in interpretation. The statewide average percentile in literacy or mathematics at each grade level is the 50th percentile each year. Note that the percentiles are standardized within year and state, meaning that they are not indicators of how ‘true’ achievement has changed over time or how performance compares to students in other states. Percentiles are used to compare the relative annual performance of Arkansas student groups over time and examine if the gap in achievement has changed (details about the methodology can be found at the end of this blog).

Are there differences in achievement by gender?

Most national studies find that, on average, males outperform females on math tests and females outperform males on reading or English Language Arts (ELA) tests. Recent work by Reardon et al. out of Stanford’s Center for Education Policy Analysis found that the average school district across the county evidenced female-favoring achievement gaps in English Language Arts, but no gender achievement gaps in mathematics. We find similar patterns when we examine the Arkansas assessment data.

Figure 1. Average Literacy Achievement by Gender, 2008-09 through 2017-18

Lit Gender

As can be seen in Figure 1, female students consistently perform above the state average on literacy assessments, and there is a persistent gap of 13-15 percentage points between the average performance of females and males. This difference of 0.33 standard deviations is somewhat larger than the national estimates found by Reardon and represents approximately a year of learning. The difference closed by 50% in 2014-15, the only year that the PARCC assessment was administered. Overall, average literacy achievement by student gender has been consistent over the time examined.

Figure 2. Average Math Achievement by Gender, 2008-09 through 2017-18

Math Gender

In contrast to the differences in literacy achievement, Figure 2 illustrates that Arkansas’ female and male students have performed similarly on the state mathematics assessments since 2009. Female students scored slightly higher than their male peers, and there was no evidence that the gap was different in the year the PARCC assessment was administered.  Overall, average math achievement by student gender has been consistent over the time examined.

Are there differences in achievement by race/ethnicity?

CEPA has also published work on the racial achievement gaps throughout the country.  The researchers find differences in average math and reading achievement by race, and note that one potential explanation for racial achievement gaps is that they are largely due to socioeconomic disparities between white, black, and Hispanic families. The research finds, however, that achievement gaps are still present even in states where the racial socioeconomic disparities are near zero (typically states with small black or Hispanic populations), suggesting that socioeconomic disparities are not the sole cause of racial achievement gaps. Although the CEPA research used data from the National Assessment of Educational Progress (NAEP) as the basis for the research (NAEP), we find similar racial achievement differences using the Arkansas state assessments.

Figure 3. Average Literacy Achievement by Race, 2008-09 through 2017-18

Lit Race

As presented in Figure 3, White students consistently score 7 to 8 percentage points above the state average performance in literacy. The literacy achievement of Hispanic students reflects improvement from an initial percentile of 40, and has been holding steady in the 44-46th %ile range since 2011-12.  Average black student achievement on state literacy assessments is in the low 30th percentiles, beginning at the 31st in 2008-09 and demonstrating a slight increase to 33rd %ile by 2018. The difference between white and Hispanic performance is about a third of a standard deviation, representing approximately a year of learning. The difference between white and black literacy achievement is double that.  Similar to the patterns we have noted for students participating in the FRL program, black student achievement reflects a noticeable relative increase (+7 percentage points) in the 2014-15 school year, which was the year that the PARCC assessment was administered. There is a slight narrowing of the differences in achievement over the past decade, but overall, average literacy achievement by student racial groups has been consistent over the time examined.

Figure 4. Average Math Achievement by Race, 2008-09 through 2017-18

Math Race

Similar to literacy achievement, White students consistently score 7 to 8 percentage points above the state average mathematics performance. Hispanic students’ mathematics achievement reflects improvement from an initial percentile of 43, although there was a slight decline from a high of 48 in the 2013-14 and 2014-15 school years to an average of 46th percentile in 2017-18.  Average black student achievement on state mathematics assessments are in the low 30th percentiles, beginning at the 32nd in 2008-09 and declining to the 31st percentile in 2018. Mathematics differences are similar in magnitude to the literacy gaps. Similar to the pattern we noted for literacy, black students’ math achievement reflects a noticeable relative increase (+8 percentage points)  in the 2014-15 school year, which was the year that the PARCC assessment was administered.


Are there differences in achievement by race/ethnicity and gender combinations?

We were interested to learn if there were differences in average achievement on the state assessment within race by student gender. Based on what we saw in earlier figures, we anticipate that females will perform 15-17 percentage points higher than their male counterparts in literacy, while achievement in mathematics will be essentially the same.   In Figures 5-7, we present literacy achievement by race/ by gender, and in Figures 8-10 we share mathematics results.

Figure 5. Average White Literacy Achievement, by Gender, 2008-09 through 2017-18

White Lit

As expected, Figure 5 illustrates that white females achieve about 15 percentage points above their male peers in literacy, a difference of about a year of learning.  White females score well above the state average in literacy, while their male counterparts achieve just at the 50th percentile, which is average performance for the state average of the 50th percentile.  Interestingly, white female students evidenced the decline in achievement in 2014-15, the year of the PARCC assessment, but white male achievement did not reflect increased achievement that year.

Figure 6. Average Hispanic Literacy Achievement, by Gender, 2008-09 through 2017-18

Hispanic Lit

Similar to white females, Hispanic females demonstrate higher literacy achievement than their male peers in the state by around 13-14 percentage points, representing approximately a year of learning.  Hispanic females performed slightly above about the state average in literacy in 2018, up from 4 percentage points below the average in 2009.  The average literacy achievement of Hispanic males have also increased since 2008-09, leading to a slight closing the difference in achievement with Hispanic females. Unlike their white peers, Hispanic males evidenced increased relative achievement on the PARCC test in 2014-15.

Figure 7. Average Black Literacy Achievement, by Gender, 2008-09 through 2017-18

Black Lit

As is other racial groups, black females demonstrate higher average achievement in literacy than their male peers. Figure 7 illustrates that black females achieve about 15 percentage points above the achievement of black males, which represents approximately a year of learning. With average literacy at the 49th percentile in 2018, however, the average achievement of black females is lower than females in other racial groups. Black males are the lowest relative achievers in literacy – consistently scoring at the 14th-26th percentile since 2008-09 with the exception of a sharp increase of 12 percentage points on the PARCC assessment in 2014-15.  Unlike females in other racial groups, who experience slight decline in 2014-15, black females evidence a score increase of 3 percentage points on the PARCC exam.

Figure 8. Average White Math Achievement, by Gender, 2008-09 through 2017-18

White Math

As anticipated based on the overall similar performance of female and male students in mathematics, Figure 8 represents how white females narrowly out perform their male counterparts on the state mathematics assessments.  Achievement of both white females and white males is above the state average mathematics achievement and has been consistent over time.

Figure 9. Average Hispanic Math Achievement, by Gender, 2008-09 through 2017-18

Hispanic Math

Hispanic math achievement by gender is depicted in Figure 9. Similar to their white peers, Hispanic females narrowly outperform their male counterparts on the state mathematics assessments. Math performance for Hispanic females and males is slightly below the state average, and average mathematics performance has increased slightly  over time for both groups.

Figure 10. Average Black Math Achievement, by Gender, 2008-09 through 2017-18

Black Math

Figure 10 reveals that the difference in average mathematics achievement between males and females is larger for black students than for Hispanic and white students. Black females achieve, on average, 5 percentage points higher than black males. Performance for both females and males, however, is well below the state average. The gap is consistent over time, and both females and males evidenced increased relative performance in 2014-15 with the PARCC assessment.

So, what does this tell us?

Literacy: Arkansas state assessment data reflect differences by gender and race in average literacy achievement that are similar to the national achievement differences. Differences in average literacy achievement between racial groups have narrowed slightly over the past decade.  Differences in literacy achievement between females and males is consistent across racial groups.

Math: Arkansas state assessment data reflect essentially no difference in average mathematics achievement by gender overall, although there are differences in average math achievement by racial groups.  Hispanic students, both females and males, are demonstrating increased math achievement, narrowing the difference between their performance and that of white students. Interestingly, black females and males evidence a greater gap in the average math performance that was present for gender difference among white and Hispanic students.

PARCC: We are really interested in what was different about the PARCC assessment that resulted in higher relative achievement in both math and literacy for Black (and poor) students in Arkansas? Remember that scores from all assessments were standardized, and should be comparable across time.  The changes are interesting, but we have just one year of PARCC results, so we cannot interpret beyond just noting the change.  The performance trends present since 2009 returned when student began being assessed on the ACT Aspire, and the consistency of group achievement (apart from the PARCC year) is a good reminder of how important it is to focus on the long-term trends.

Coming out soon: Achievement trends by grade level!  Find out if the differences in achievement start large and get smaller, start small and grow over time, or stubbornly persist same over the decade.


About the methodology:

Students in grades 3-8 were consistently assessed in both content areas, but there was variation in when high school students were assessed. For our analysis we used all grades assessed in literacy and/ or math in a given year, and limited our analyses to students completing the general assessment.

Scale scores were standardized for each year within grade level and content area, creating a Z-score (mean of 0, standard deviation of 1).  These z-scores were then averaged across student groups and transformed into percentiles to ease interpretation. The statewide average percentile in literacy or mathematics at each grade level is the 50th percentile each year. Percentiles are standardized within year and state, meaning that they are not indicators of how actual achievement has changed over time or how performance compares to students in other states. Percentiles are used to compare the relative annual performance of Arkansas student groups over time and examine if the gap in achievement has changed. 

Leave a Comment

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: