Skip to content
ascd logo

April 1, 2019
Vol. 76
No. 7

A "Color-Aware" Approach to Data

premium resources logo

Premium Resource

April 2019 Roegman Header image
When Martin Luther King Jr. called us all to judge children by the contents of their character, not the color of their skin, he did not mean that we should ignore their race. This is true especially in our schools, where educators need to acknowledge that students' racial identities matter—and that they profoundly affect students' experiences in school. Race, along with social class and zip code, continues to be an important predictor of students' educational experiences, from pre-kindergarten through high school (Grissom & Redding, 2016; Skiba, 2015). 
Race matters in terms of access, discipline, and outcomes (Grissom & Redding, 2016; Skiba et al., 2011), and to address this, the federal Every Student Succeeds Act requires that states and districts publish student test scores disaggregated by major racial and ethnic groups, among other characteristics. Nevertheless, a recent study we conducted suggests that even though their districts follow these federal reporting requirements, K–12 school principals may, intentionally or unintentionally, avoid considering race when they collect, interpret, analyze, and make decisions about data (Roegman et al., 2018). 
In our study, we interviewed 18 elementary and secondary principals from three Midwestern school districts about their data use. One district is small and rural, with about 2,000 students, 90 percent of whom are white and 5 percent are Latinx; about 25 percent qualify for free or reduced-price lunch. The second district, also rural, has almost 7,000 students, 75 percent of whom are white and 15 percent are Latinx; about 50 percent qualify for free or reduced-price lunch. The third district is in a suburban area, serving 11,000 students; 80 percent are black and 10 percent are white; more than 99 percent qualify for free or reduced-price lunch. 
We found that overall, principals did not think it was useful to disaggregate data by students' race. They believed in taking a color-neutral approach, saying that they did not see students' race nor regard race as a factor to consider when analyzing data. Several principals shared that they had not even thought about disaggregating assessment data collected in their own school, such as benchmark assessments or semester grades. During our interviews, one responded, "Makes me think, because you asked me that question," while another pondered, "I've never done that one. I wonder if I should. That's interesting." 
Our results suggest that educators have more work to do in unpacking the reasons behind and highlighting the importance of disaggregating data by race and ethnicity.

Color-Neutral Attitudes

The principals who exhibited a color-neutral attitude focused on each student as an individual, or they focused on groups of students in terms of whether or not they achieved proficient scores on state assessments. In other words, principals with color-neutral attitudes disaggregated data by performance outcomes alone rather than by race. 
A color-neutral mindset allows educators to ignore the "presence of different social realities due to identities that are different from a white social identity" (Fergus, 2017, p. 32). For example, one of the principals told us, "I don't pay any attention to [disaggregated state data]. I don't know if that's right or wrong, but I don't. I tend to take the view that we're going to treat all kids the same." 
This is a problem because the lens through which principals look at data can disguise or hide potential disparities in educational outcomes for students of color. If they do not disaggregate, patterns may remain invisible—however, even if they do disaggregate, or review state-level disaggregated data, principals may wrongly interpret disparities in test scores and not consider the systems of oppression that affect students of color (Bonilla-Silva, 2017). For example, they may blame students of color for low test scores, instead of considering the students' lack of access to qualified teachers and rigorous curricula. 
Seemingly neutral processes often have profoundly disparate and inequitable impacts. Grissom and Redding (2016), for example, investigated reasons that students of color were less likely to be included in gifted programs. They found, even when controlling for standardized test scores, socioeconomic status, school characteristics, and other variables, that when it came to teacher discretion, black students taught by non-black teachers were significantly less likely to be referred as potentially gifted. The authors suggested that teachers may have "biases in their judgements or expectations" (p. 14), and they also considered that students may behave differently in the presence of own-race or different-race teachers. 
When principals examine their school data to look for patterns such as those found by Grissom and Redding, they may discover that seemingly neutral processes such as how students are identified for placement in a gifted and talented program are in fact inequitably impacting students of color. Unless principals disaggregate data by race, these types of patterns remain hidden and unaddressed.

A Lack of Understanding and Technical Limitations

Several of the principals in our study were not aware of the benefits of disaggregating or the myriad ways disaggregated data could be used to gain insight into student performance. For example, the purpose of disaggregated data is not limited to identifying disparities between groups. Duke (2017) argues that "within-race comparisons are likely to reveal important causal factors that may go unnoticed when between-race comparisons are made" (p. 100). Disaggregated data analyses can reveal where and to what extent students are doing well, countering anecdotal experiences or bias-based ideas such as a belief that all black students are low-performing. Disaggregated data can also be used to identify a group of students' strengths—or a strength of the school more generally, providing evidence to sustain current practices. A principal might assume that aggregated low test scores are the result of the performance of students of color; disaggregating the data by race may show that students of color overall are performing well. More specifically, it might show that Latinx students in a bilingual program are performing significantly higher than Latinx students in a pull-out program. 
In addition to not understanding the different ways it could be used, some principals were not sure that disaggregating data was even possible given their student demographics. Principals in majority-minority schools (where the majority of students are not white), as well as those in schools where most of the student body was white, tended to express the view that data disaggregation would not yield useful information for instruction. For example, one principal noted: 
The predominance of my school is low socioeconomic status and mostly black, so to split off, I'd be looking at 10 kids compared to 100 kids because it's about 10 percent that are white or other … and then the rest are African American. … It would be like comparing a watermelon to a lime in size of groups. So, I don't do that. 
None of the principals we spoke with shared any methods that they knew of to quantitatively compare different sized groups of students (though such methods exist, such as composition analysis and relative risk analysis, as discussed in Fergus's 2017 book). This, combined with a somewhat narrow view of disaggregating data, reinforced these school leaders' decisions to not disaggregate data by race. 
Another obstacle for principals was the limitation of technology and technical support. One principal said that disaggregating his school's data would be useful, but that the data-management software did not easily allow for it. He wanted a program where he could "hit this button" and the analyses would come up, but explained with frustration in his voice that "it's not that easy." Another principal noted that: 
I had to go to the [Assessment A] website to pull [those] scores. Had to go to [Assessment B] website to pull that. Had to go to all these different areas to pull [all the data] and then I had to enter them one at a time for 500 kids. It became very, very time-consuming. 
Investing in technology and supports for disaggregated data analysis with a variety of student characteristics is a key step that can support principals in bringing considerations of race into their site-based decision making. 
Here are some other ways that leaders can begin to bring considerations of race into their data use practices.

Create a Sense of Urgency

When everyone involved understands the intersections of race and education, it can productively move conversations forward within a school or district. First, school leaders can use data to create a sense of urgency for change. This is not always easy, as conversations about disparities in assessment scores, for example, can lead teachers and administrators to blame students for not achieving well and then reinforce deficit beliefs about students from different groups. Brown and coauthors (2011) found that conversations about achievement gaps led educators to attribute low performance to the characteristics of students and families, instead of reflecting on how race and educational experiences might intersect to perpetuate disparities. 
However, disaggregating data could be framed from an equity perspective to highlight patterns of high performance and to identify potential disparities in students' outcomes or access (for example, participation in a gifted program), without using the language of the achievement gap. Principals in our study were right to be concerned about how focusing on race might be problematic, but they did not consider how a color-neutral approach is also problematic. They did not bring up any historical, structural, and cultural oppressions within K–12 educational systems that impacted students' experiences and performance. Awareness of how racism and other inequities impact students of color provides a foundation for racial equity (Banks-Wallace, 2000). Having high expectations or standards for all students does not mean ignoring systemic racism.

Increase Districtwide Expectations to Identify and Address Racial Inequities

None of the interviewed principals in our study reported any central office or state mandate that required them to examine their school data with an attention to potential racial disparities. Yet when central office and site-based administrators lead the way in identifying patterns, they support their entire systems in developing plans to address inequities, instead of blaming students or families. 
This support needs to include both expectations to analyze data through an equity lens and professional development on how to do so. When principals look at data across a school, they can identify areas of concern that may be masked in an individual teacher's classroom. For example, if each classroom has only one or two students learning English, it may be easier for individual teachers to excuse each student's low performance as something unique to that student. However, if the principal looks at the performance of English learners across the school, patterns of inequity may emerge. Superintendents can set the stage for this type of analysis by requiring principals to disaggregate data and look for disproportionalities in student access and performance.

Involve Teachers in Data Analysis

Teachers can also play a critical role in identifying and addressing inequities within their own classrooms by using disaggregated data—and by data, we do not just mean state assessments, but also student writing samples, formative work such as exit slips, and even participation patterns and students' oral responses in classrooms. This can help teachers recognize areas of strength and concern and get beyond initial assumptions that involve blaming students or families. Lachat and Smith (2005), for example, described how a deep dive into disaggregated achievement data helped high school teachers in high-poverty urban districts understand that student achievement patterns in their classrooms were not a result of poor attendance—their initial assumption—and instead were related to the quality of instruction that they received. 

The Promise of Brown

Ladson-Billings (2004) contended that the Brown v. Board of Education decision should be viewed as a first step, one that dismantled de jure segregation but must be followed by continued work to dismantle systems of oppression in K–12 schools that harm students of color. "The nation has never fully and honestly dealt with its 'race' problem" (p. 10), she argues. 
Moving from color-neutral to color-aware approaches to data use supports principals in examining race and racism in students' educational experiences so that their individual schools and districts, and the United States as a whole, can continue the journey to fulfill the promise of Brown

Guiding Questions

➛ Does your school include racial breakdowns when assessing data? Why or why not?

➛ Are there places in your teaching or leading where you could be more aware of possible racial equity issues?

➛ Did you find the authors' recommendations helpful? What would you add as a step that leaders can take to bring considerations of race into their data-use practices?


Banks-Wallace, J. (2000). Womanist ways of knowing: Theoretical considerations for research with African American women. Advances in Nursing Science22(3), 33–45.

Bonilla-Silva, E. (2017). Racism without racists: Color-blind racism and the persistence of racial inequality in America. Lanham, MD: Rowman & Littlefield.

Brown, K. M., Benkovitz, J., Muttillo, A. J., & Urban, T. (2011). Leading schools of excellence and equity: Documenting effective strategies in closing achievement gaps. Teachers College Record113(1), 57–96.

Duke, D. L. (2017). Can within-race achievement comparisons help narrow between-race achievement gaps? Journal of Education for Students Placed at Risk22(2), 100–115.

Fergus, E. (2017). Solving disproportionality and achieving equity: A leader's guide to using data to change hearts and minds. Thousand Oaks, CA: Corwin.

Grissom, J. A., & Redding, C. (2016). Discretion and disproportionality: Explaining the underrepresentation of high-achieving students of color in gifted programs. AERA Open2(1).

Lachat, M. A., & Smith, S. (2005). Practices that support data use in urban high schools. Journal of Education for Students Placed at Risk10(3), 333–349.

Ladson-Billings, G. (2004). Landing on the wrong note: The price we paid for Brown. Educational Researcher33(7), 3–13.

Roegman, R., Samarapungavan, A., Maeda, Y., & Johns, G. (2018). Color-neutral disaggregation? Principals' practices around disaggregating data from three school districts. Educational Administration Quarterly54(4), 559–588.

Skiba, R. J. (2015). Interventions to address racial/ethnic disparities in school discipline: Can systems reform be race-neutral? In R. Bangs & L. Davis (Eds.), Race and social problems (pp. 107–124). New York: Springer.

Skiba, R. J., Horner, R. H., Chung, C. G., Rausch, M. K., May, S. L., & Tobin, T. (2011). Race is not neutral: A national investigation of African American and Latino disproportionality in school discipline. School Psychology Review40(1), 85–107.

Learn More

 Ala Samarapungavan is a professor of educational psychology in the Department of Educational Studies at Purdue University, College of Education.

 Yukiko Maeda is an associate professor of educational studies at Purdue University.

 Gary Johns is a doctoral student in the Department of Curriculum and Instruction at Purdue University, College of Education.

ASCD is a community dedicated to educators' professional growth and well-being.

Let us help you put your vision into action.
Related Articles
View all
Navigating Shifts in College Admissions Is an Equity Matter
Tiffany Jones
4 weeks ago

Related Articles

From our issue
Product cover image 119041b.jpg
Separate and Still Unequal: Race in America's Schools
Go To Publication