Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Dec 1.
Published in final edited form as: Soc Sci Med. 2023 Oct 15;338:116319. doi: 10.1016/j.socscimed.2023.116319

State-level desegregation in the U.S. South and mid-life cognitive function among Black and White adults

Katrina M Walsemann a,*, Nicole L Hair b, Mateo P Farina c, Pallavi Tyagi a, Heide Jackson a, Jennifer A Ailshire c
PMCID: PMC10872867  NIHMSID: NIHMS1941513  PMID: 37871395

Abstract

Rationale:

Black adults experience worse cognitive function than their White peers. Although educational attainment is an important predictor of cognitive function, other aspects of education, including school desegregation, may also shape this relationship. For Black adults who grew up in the U.S. South in the 1950s–1970s, exposure to school desegregation may have altered life course pathways critical for later cognitive function.

Objective:

We determined if state variation in exposure to school desegregation in the U.S. South was associated with cognitive function at mid-life, if the association varied by race, and if the association remained after adjustment for state-level education quality and respondents’ educational attainment.

Methods:

We linked historical data on state-level school desegregation to the Health and Retirement Study, a nationally representative sample of U.S. adults aged 50 and older. We restricted our sample to Black (n=1,443) and White (n=1,507) adults born between 1948 and 1963 who resided in the U.S. South during primary school. We assessed three cognition outcomes: total cognitive function, episodic memory, and mental status. We estimated race-stratified linear regression models with cluster adjustment and a final model using state fixed effects.

Results:

Greater exposure to desegregated primary schooling was associated with higher cognitive function and episodic memory among Black but not White adults. Among Black adults, the association between school desegregation and cognitive function and episodic memory remained after adjustment for state-level educational quality and educational attainment.

Conclusions:

Our findings suggest that state-level school desegregation efforts played a consequential role in shaping the cognitive function of Black adults who grew up in the U.S. South.

Keywords: education, life course, historical data, school segregation

Despite well-documented Black-White inequities in cognitive function, the reasons for these persistent inequities remain poorly understood. In both national and community-based samples, Black older adults consistently score worse on tests of cognitive function, including episodic memory, executive function, and working memory, than White older adults (Walsemann, Kerr, et al., 2022; Walsemann, Ureña, et al., 2022; Weuve et al., 2018). While educational attainment is often cited as a key explanation for these inequities, less attention has been paid to the ways in which early life educational experiences shape cognitive function across racial groups. For Black adults who grew up during the 1950s-1970s, school desegregation represents a central educational experience that may be relevant for understanding cognitive function, particularly because about 64% of this cohort grew up in the U.S. South (Ruggles et al., 2022).

A growing body of evidence suggests that school desegregation may have positive benefits for cognitive function among Black Americans (Aiken-Morgan et al., 2015; Peterson et al., 2021; Walsemann, Kerr, et al., 2022; Whitfield & Wiggins, 2003). These studies, however, have limited generalizability – most use community samples that may not be representative (Aiken-Morgan et al., 2015; Peterson et al., 2021; Whitfield & Wiggins, 2003). Conversely, studies that use nationally representative samples rely on people’s retrospective reports on whether they ever attended an integrated school (Walsemann & Ailshire, 2020; Walsemann, Kerr, et al., 2022) limiting our ability to examine how duration of exposure to school desegregation during early childhood – an important developmental period for brain development – might matter for cognitive function in mid-adulthood.

Our study considers the role of school desegregation on midlife cognitive function among Black and White Americans who resided in the U.S. South during primary school. To address limitations of prior studies, we link state-level administrative data on school desegregation in the U.S. South for the twenty-year period following the 1954 U.S. Supreme Court decision declaring segregated schooling unconstitutional to individual cognition data from a prospective, population-based sample. We assess if more years of exposure to school desegregation is associated with improved cognitive function at midlife and if the association differs between White and Black Americans.

School Desegregation and Cognitive Function in Historical Context

Explanations for Black-White inequities in cognitive function have mostly focused on individual-level factors, including educational attainment (Chen & Zissimopoulos, 2018; Xiong et al., 2020; Yaffe et al., 2013). Individual-level explanations, however, often miss (or negate) the historical and contextual factors that structure the life course exposures important for later life cognitive function. In response, a growing number of health and aging scholars have articulated the need to consider how structural racism – defined as the interconnected social and institutional systems that construct the norms, opportunities, and policies that define and reinforce racial hierarchies in the U.S. (Bonilla-Silva, 1997; Gee & Hicken, 2021) – shapes the life chances and, in turn, the cognitive function of Black older adults (Barnes et al., 2012; Lamar et al., 2020; Peterson et al., 2021; Pohl et al., 2021; Walsemann, Kerr, et al., 2022; Walsemann, Ureña, et al., 2022).

The public education system is often the site of key political and legal struggles, and as such, is a central actor in the socialization of American children (Lewis, 2003). Schools not only develop the cognitive skills that are linked to later life cognitive function (Crowe et al., 2013; Mantri et al., 2019; Sisco et al., 2015) but they also serve to perpetuate race inequities in U.S. society through policy and practice (Lewis, 2003). Before 1954, all Southern states legally mandated racially segregated schools. Legal justification for school segregation derived from Plessy v Ferguson (1896), which allowed segregation in public areas if facilities were equal (Bell, 2004). Segregated schools were, however, vastly unequal. In the 1920s, for example, Southern states spent 5–8 times more on White schools than Black schools (Tushnet, 1987) and provided school terms that were 50–100% longer (Walsemann, Ureña, et al., 2022). Among cohorts who completed school before 1954, race inequities in schooling duration explained more of the Black-White gap in cognitive function than educational attainment (Walsemann, Ureña, et al., 2022).

In 1954, the U.S. Supreme Court in Brown v Board of Education declared school segregation unconstitutional. In their follow-up decision – Brown II – they ordered states to desegregate schools “with all deliberate speed,” a vague ruling that allowed Southern states to delay school desegregation for years (Bell, 2004; Johnson, 2019). Many Southern states and local school districts implemented policies that ensured school segregation was maintained (e.g., neighborhood schools, zoning, within school assignments, pupil placement statutes, closing of public schools) or only minimally addressed (e.g., desegregating schools one grade at a time often starting with the primary grades). Thus, Black school children’s exposure to school desegregation varied significantly across Southern states in the twenty years after Brown.

Recognizing the slow response of southern states to desegregate schools, the U.S. Congress enacted a series of legislative acts in the 1960s that helped to accelerate school desegregation in the U.S. South. For example, The Civil Rights Act of 1964 prohibited discrimination in public schools receiving federal funding and provided avenues for legal redress (Johnson, 2019; Wilkinson, 1978). The Elementary and Secondary Education Act of 1965 significantly increased federal funding for public schools, making school desegregation more fiscally appealing. In the early 1970s, courts began requiring that school districts provide them with detailed plans for meaningful desegregation (Hudgins, 1973). Thus, by the late 1970’s, the U.S. South had made significant progress in desegregating their schools. This progress would peak in the late 1980s when 45% of Black students were attending majority-white schools but would slowly reverse in the subsequent decades (Orfield & Eaton, 1996).

Within this political and historical context, school desegregation efforts shaped Black children’s educational experiences and subsequent life outcomes in positive, but also sometimes negative ways. On the one hand, school desegregation may have reduced racial inequities in life chances, and in turn, cognitive function, by dismantling the separate and unequal school system that had relegated Black students to under resourced schools. In fact, school desegregation often resulted in Black students attending better resourced and higher quality schools. School resources allow for the implementation of policies and programs that improve student achievement and educational attainment, including smaller classes (Fredriksson et al., 2013; Krueger, 1999; Rivkin et al., 2005), employing and retaining highly qualified teachers (Darling-Hammond, 2000; Goldhaber & Brewer, 2000), and providing more instructional time to students (Johnson, 2019; Reber, 2010). Well-resourced schools and greater educational attainment are each independently associated with better cognitive function (Crowe et al., 2013; Glymour et al., 2008; Moorman et al., 2019; Sisco et al., 2015; Walsemann, Ureña, et al., 2022).

On the other hand, school desegregation often resulted in Black students being taught by White teachers who held lower expectations of them, a form of interpersonal racism (Beady & Hansell, 1981; Bell, 2004). Teacher expectations are important predictors of student achievement (Dee, 2004, 2005). For example, Black students who are never taught by a Black teacher are less likely to graduate from high school than their peers who are taught by at least one Black teacher (Gershenson et al., 2016). Older adults who completed fewer years of schooling score lower on tests of cognitive function than those who complete more years of schooling (Walsemann, Kerr, et al., 2022; Weuve et al., 2018), a relationship that also shows indirect influence via other risk and protective factors, such as social and cognitive engagement, health behaviors, and cardiometabolic conditions (Livingston et al., 2017).

Exposure to desegregated schooling in the U.S. South varied across states and school entry cohort, and it is important to consider both aspects of school desegregation implementation to understand its role in shaping the cognitive function of Black Americans. Some prior work points to the importance of state and cohort variation, though the strength and direction of the effects vary across samples (Aiken-Morgan et al., 2015; Allaire & Whitfield, 2004; Lamar et al., 2020; Peterson et al., 2021; Walsemann, Kerr, et al., 2022). For example, among older Black Baltimore residents, attending a desegregated school was positively associated with cognitive function as measured by processing speed, but the association depended on respondents’ age. Those who were older when Brown was decided were less likely to show cognitive benefits from desegregated schooling perhaps because they were exposed to fewer years of desegregated schooling and/or exposed to desegregated schooling later in their educational careers (Aiken-Morgan et al., 2015). Possible cohort differences were also found in a study using a national sample of midlife and older adults; attending a majority-White school had significant positive associations with cognitive function and episodic memory among Black adults who completed their schooling before or immediately after Brown, as well as with working memory among those who completed schooling after 1964 (Walsemann, Kerr, et al. 2022). Conversely, two community samples showed poorer cognitive function among Black adults who grew up in the U.S. South (Lamar et al., 2020) or were among the first states to desegregate their schools (Allaire & Whitfield, 2004).

Developmental timing of exposure may also be an important aspect linking school desegregation to cognitive function. Early education is thought to be critical for brain development (Burger, 2010; Goldfeld et al., 2016; Shah et al., 2017); thus, exposure to school desegregation earlier in school may have a stronger relationship with cognitive function than exposure later in school. One recent study using the STAR cohort of Black older adults in Northern California found a positive association between desegregated schooling on executive function and semantic memory for those who attended desegregated primary schools, but not for those who attended integrated middle schools or high schools (Peterson et al., 2021).

Finally, across most studies, desegregated schooling is related to better cognitive function for Black older adults. The relationship is less robust for White adults and is often null (Walsemann, Kerr, et al., 2022), perhaps because school desegregation did not appreciably change White students’ access to educational resources. In fact, most southern districts that desegregated their schools did so by reassigning Black students to formerly all-White schools that were better resourced (Bell, 2004; Johnson, 2019). This finding has been replicated in other economic and health-related research (Guryan, 2004; Hanushek et al., 2009; Johnson, 2011; Reber, 2010).

Research Questions and Hypotheses

In this study, we determine if exposure to school desegregation in the two decades following Brown is linked to cognitive function at midlife. Given the importance of early life education for later life cognitive function (Borghans et al., 2015) as well as prior work that suggests the possible benefits of desegregated schooling earlier in childhood (Aiken-Morgan et al., 2015; Peterson et al., 2021), we focus on exposure to desegregated primary schooling. We hypothesize that more years of exposure to desegregated primary schooling will be associated with higher cognitive function at midlife. We also expect that this relationship will be found for Black but not White adults. Finally, associations between exposure to desegregated primary schooling and cognitive function may simply reflect the higher quality schools and greater educational attainment that accompanied school desegregation (Johnson, 2019; Reber, 2010). Thus, we examine the association between exposure to desegregated primary schooling and cognitive function accounting for differences in state-level educational quality and individuals’ educational attainment.

Methods

Data

Individual-level data come from the Health and Retirement Study (HRS), a nationally representative, longitudinal study of U.S. adults over age 50 (Sonnega & Weir, 2014). The HRS is a multi-stage area probability sample of age-eligible households selected from primary sampling units chosen from U.S. Metropolitan Statistical Areas (MSAs) and non-MSA counties, with an oversampling of minorities and the oldest-old. Since 1992, the HRS has conducted core interviews with age-eligible respondents and their spouses approximately every two years. Data collection is ongoing and using a steady-state design, the HRS sample is replenished with younger cohorts about every 6 years.

State-level data on school desegregation come from two historical sources – a 1967 report published by the Southern Education Reporting Service that documented the annual level of school desegregation in each of the 16 southern states from 1954 to 1967 (Campbell et al., 1967) and biennial reports from the U.S. Census Abstracts published in 1968, 1970, 1972, and 1974. State-level data on educational quality come from two historical sources: 1) the 1953/54 to 1957/58 Biennial Surveys of Education (BSE) of the United States and 2) the 1959/60 – 1973/74 Statistics of State School Systems (SoSSS). We linked state-level administrative data to HRS via a single measure that asked respondents the state they lived in most of the time they were in school or around age 10.

Sample

We restricted our sample to Black and White HRS respondents who reported residing in one of 16 states in the U.S. South when they were around age 10, were born between 1948 and 1963, and provided a baseline measure of cognitive function (n=3,281). Next, due to known testing bias on the first cognitive assessment (Cooley et al., 2015; Jendryczko et al., 2019; Rabbitt et al., 2004), we used respondents’ second observation of cognitive function in our analysis. Thus, we excluded respondents who did not provide a valid second observation on cognitive function after cohort entry (when they were approximately 52–63 years old (n=331)). Importantly, rates of mortality or attrition from the first to second (expected) observation were low (~10%) and did not differ by race (see Supplemental Table A1). Our final analytic sample included 2,950 older adults (1,507 White adults, 1,443 Black adults).

Measures

Cognitive Outcomes.

HRS administered the Telephone Instrument for Cognitive Status (TICS) to assess cognitive function either by phone or face-to-face. The cognitive assessment consists of tests that evaluate the respondent’s memory, using 10 word immediate and delayed recall, and attention and processing speed, using a serial 7s subtraction test of working memory and counting backwards. We assessed total cognitive function by summing across all items, resulting in observed scores ranging from 0–27. We assessed episodic memory by summing items across immediate and delayed 10-word recall, resulting in observed scores ranging from 0–20. We assessed mental status by summing items across the serial 7s subtraction test and counting backwards, resulting in observed scores ranging from 0–7. We used the imputed cognition variables provided by HRS so there was no missing data on our dependent variables; however, individual TICs items were only imputed for respondents who participated in the TICs (i.e., non-interviews and proxy respondents do not have imputed data) (Fisher et al., 2017).

Expected Exposure to Desegregated Primary Schooling.

Because local (e.g., county/city) data on school desegregation was not reported, we used state-level administrative data to construct a cumulative measure of respondents’ expected exposure to desegregated primary schooling by summing the proportion of Black students who were attending public schools with White students across a six-year period corresponding to grades 1 through 6 for each school entry cohort and state. A school year where 40% of Black students were attending desegregated schools, for example, contributed 0.4 years to the expected duration of exposure. Our measure represents the expected (average) number of years that members of each school entry cohort would have spent in desegregated primary schools, which theoretically ranged from 0 in a state that did not desegregate schools during the six-year period to 6 in a state where 100% of Black students were attending desegregated schools during the six-year period. Observed values ranged from 0 to 5.7 expected years.

State-level Educational Quality.

State-level measures of educational quality included: pupil-teacher ratio (average ratio of students in daily attendance to primary and secondary teachers in a given year), teacher salary (average annual salary of primary and secondary school teachers each year), per-pupil spending (total spending per pupil each year), per-pupil revenue from local sources (total local revenue per pupil each year), per-pupil revenue from state sources (total state revenue per pupil each year), percent of revenue from local sources, and percent of revenue from state sources. All expenditures were inflation adjusted to 2021 dollars. We calculated z-scores for each indicator by standardizing the six-year forward average of each indicator for all years between 1954 and 1974, ensuring that they were comparably scaled. To summarize state-level educational quality across these seven standardized indicators over time, we used factor analysis with orthogonal rotation. The resulting factors can be thought of as measures of relative state-level educational quality during the 1954 to 1974 period. The scree plot and broken stick test (Jackson, 1993) suggested that two factors were recommended and explained approximately 97.5% of the variation in the state-level educational quality indicators. Generally, the first factor captures well-resourced education systems, and the second factor captures the locality of revenue (i.e., state vs local). We provide the factor loadings and scree plots of the eigenvalues following the factor analysis in Supplemental Table A2 and Supplemental Figure A1. Analyses using individual, non-standardized indicators were similar to those we present.

Education.

Respondents reported the number of years of schooling they completed (range 0 to 17). Covariates. All models included gender (female or male), birth year (range: 1948 to 1963), interview mode (face-to-face vs phone), highest level of education completed by either parent (range: 0 to 17), and respondents’ self-reported childhood health (1=excellent, 5=poor).

Missing Data

Delaware, Oklahoma, and West Virginia were missing information on school desegregation in 2 years. To assess patterns of item non-response, we visually inspected the data for each state by year. Next, using linear regression we used observed data to predict estimates of school desegregation for the missing years by state, and modeled time as either linear or quadratic based on model fit. We followed a similar strategy when handling missing data on per pupil expenditures for North Carolina, which did not report this information in 1950.

We also had biennial data on school desegregation from 1968 to 1974 and state educational quality indictors from 1953/54 – 1973/74; thus, for these measures and time periods, we did not have data in school years that ended in an odd number (e.g., 1968/69). Visual inspection of the data indicated a linear or quadratic pattern for each measure by state. We therefore interpolated data for the odd years using data from the two adjacent years in each respective state.

Complete data were available on individual-level covariates except for parents’ education, childhood health, and respondent’s education. Item non-response ranged from <1% for respondent’s education to 9% for parent’s education. To address item nonresponse, we imputed data using the mi impute command with chained equations in Stata, version 17 (Stata-Corp LP, College Station, TX). Imputation models included all analytical variables as well as variables that were theoretically related to item nonresponse (e.g., childhood moves, childhood financial difficulty, missed > 1 month of school due to health) (Heeringa et al., 2017). Analyses were replicated across 30 generated data sets and combined using mi estimate (Graham et al., 2007).

Analytic Approach

We used linear regression with cluster adjustment for state of childhood residence. All models were race-stratified. Model 1 estimated the relationship between expected years of exposure to desegregated primary schooling and each dependent variable, after adjustment for covariates. Model 2 included the two state-level educational quality factors (well-resourced and revenue source). Model 3 included respondent’s educational attainment. We compare estimates of exposure to desegregated primary schooling from the baseline model (Model 1) to estimates from models with state-level educational quality (Model 2) and individual educational attainment (Model 3) to determine if exposure to desegregated primary school has an independent relationship with cognitive function after inclusion of these other factors. We estimated a final model (Model 4) that included state fixed effects rather than cluster adjustment to account for possible omitted bias due to time invariant differences between states. We applied person-level weights from the interview year when a respondent’s cognitive function was measured to account for complex sampling. If the respondent lived in a nursing home at the time of their interview, we used their nursing home weight to retain them and ensure sample representation.

Results

State-level School Data

Prior to linking state-level school desegregation data to HRS, we examined state and year variation in school desegregation from 1955 to 1974 (Figure 1). All states had de jure school segregation in 1954 and thus, this time point is not shown in Figure 1. Several important patterns emerge. First, the well documented delay in desegregation is evident in Southern states that originally seceded from the United States during the American Civil War – Alabama, Arkansas, Florida, Georgia, the Carolinas, Louisiana, Mississippi, Texas, Tennessee, and Virginia. The remaining states – Delaware, Maryland, Kentucky, Oklahoma, and West Virginia – showed earlier and steadily increasing rates of school desegregation. Second, there is a clear inflection point of accelerating desegregation that occurred after 1964, which has been linked to legislative and judicial actions that were enacted or handed down around this time (Bell, 2004; Johnson, 2019).

Figure 1.

Figure 1.

Open in a new tab

Percentage of Black public-school students who were attending school with White students by state and year, 1955 – 1974.

Data Source: The Southern Education Reporting Service (annual data from 1955 to 1967) and the U.S. Census Abstracts (1968, 1970, 1972, and 1974).

Sample Characteristics

Table 1 presents sample characteristics. At midlife (mean age = 56 years), White adults had higher mean scores on total cognitive function (mean=17.2, SE = 0.12), episodic memory (mean = 11.2, SE = 0.10) and mental status (mean = 5.9, SE = 0.05) than Black adults (mean = 14.4, SE = 0.16; mean = 9.7, SE = 0.12; mean = 4.7, SE = 0.07, respectively). White adults could expect to spend slightly more of their primary schooling in desegregated schools than Black adults (1.5 years vs 1.2 years, p<0.001). Compared to their White peers, Black adults lived in states during primary school with less well-resourced educational systems (mean z-score = −0.3 vs 0.1) that received less of their funding from local sources (mean z-score = −0.1 vs 0.1) and completed fewer years of schooling (mean = 12.8 vs 13.8).

Table 1:

Sample characteristics of Black and White adults who lived in the U.S. South during primary school and were born between 1948–1963, Health and Retirement Study.

White Black
n=1,507 n=1,443
Mean (SE) or % Mean (SE) or %
Cognitive Function
 Total Cognitive Functiona 17.2 (0.12) 14.4 (0 .16)**
 Episodic Memoryb 11.2 (0.10) 9.7 (0.12)**
 Mental Statusc 5.9 (0.05) 4.7 (0.07)**
Expected Years of Exposure to Desegregated Primary School 1.5 (0.06) 1.2 (0.06)**
Demographics
 Age 55.9 (0.07) 56.2 (0.10)*
 Birth year 1955 (0.15) 1954 (0.17)
 Women, % 54.2% 54.0%
 Census Division, %d
  South Atlantic 50.0% 49.4%
  East South Central 25.2% 29.0%
  West South Central 24.8% 21.6%
 Interview Mode Face-to-Face, %e 53.2% 56.5%
Childhood Conditions
 Self-reported Child Health (1=Excellent, 5=Poor) 1.7 (0.03) 1.8 (0. 04)**
 Parent’s Education (years) 12.4 (0.10) 10.3 (0.14)**
Quality of State Education System
 Well-Resourced (z-score) 0.1 (0.03) −0.3 (0 .03)**
 Locality of Revenue Source (z-score) 0.1 (0.03) −0.1 (0.03)**
Respondent’s Education (years) 13.8 (0.08) 12.8 (0.10)**
Open in a new tab

Notes:

a

Values range from 0 to 27, with higher values representing higher total cognitive function.

b

Values range from 0 to 20, with higher values representing higher episodic memory.

c

Values range from 0 to 7, with higher values representing higher mental status.

d

Per our restricted data agreement, we can only present individual-level data no lower than the Census division.

e

Reference group is phone.

*

p<0.05;

**

p<0.01

Regression Models

Table 2 presents estimates from linear regression models predicting total cognitive function, episodic memory, and mental status for White adults. Expected exposure to school desegregation was not significantly associated with any of the cognitive outcomes and the regression coefficients were close to zero (total cognitive function: b=0.02, SE=0.11; episodic memory: b=0.06, SE=0.09; mental status: b=−0.03, SE=0.04) after adjusting for demographics and childhood conditions (Model 1). These associations did not change with inclusion of the two state education factors – well-resourced and locality of revenue (Model 2) – or educational attainment (Model 3). Associations between expected exposure to school desegregation and the three cognitive outcomes were also non-significant in the state fixed effects model (Model 4).

Table 2.

Weighted coefficients and standard errors (SE) from linear regressions predicting cognitive function with cluster adjustment for state of childhood residence (Models 1–3) or state fixed effects (Model 4), White adults who lived in the U.S. South during primary school, Health and Retirement Study, n=1,507.

Model 1 Model 2 Model 3 Model 4
b (SE) b (SE) b (SE) b (SE)
Total Cognitive Function
Desegregated primary schooling (years) 0.02 −0.05 −0.10 −0.17
(0.11) (0.09) (0.10) (0.18)
Well-resourced state (z-score) 0.58* 0.63* 1.45*
(0.16) (0.18) (0.70)
Locality of revenue (z-score) 0.05 0.17 0.40
(0.11) (0.12) (0.60)
Education (years) 0.53* 0.53*
(0.06) (0.06)
Intercept 16.46* 16.54* 16.81* 17.42*
(0.37) (0.32) (0.28) (0.61)
Episodic Memory
Desegregated primary schooling (years) 0.06 −0.01 −0.03 −0.04
(0.09) (0.08) (0.09) (0.15)
Well-resourced state (z-score) 0.47* 0.50* 0.74
(0.15) (0.17) (0.59)
Locality of revenue (z-score) 0.03 0.11 0.22
(0.09) (0.11) (0.57)
Education (years) 0.34* 0.35*
(0.04) (0.05)
Intercept 10.30* 10.36* 10.54* 10.74*
(0.31) (0.27) (0.25) (0.55)
Mental Status
Desegregated primary schooling (years) −0.03 −0.05 −0.06+ −0.13+
(0.04) (0.04) (0.04) (0.07)
Well-resourced state (z-score) 0.11 0.13 0.71*
(0.09) (0.09) (0.28)
Locality of revenue (z-score) 0.02 0.06 0.18
(0.07) (0.07) (0.24)
Education (years) 0.19* 0.19*
(0.02) (0.02)
Intercept 6.16* 6.18* 6.28* 6.68*
(0.13) (0.13) (0.13) (0.22)
Cluster adjusted ✓ ✓ ✓
State fixed effects ✓
Open in a new tab

Notes: All models adjust for gender, birth year, interview mode, childhood health, and parents’ education. Continuous covariates were centered at their means. 95% CI are presented in Supplemental Table A8.

+

p<0.10,

*

p<0.05

Table 3 presents estimates from linear regression models predicting total cognitive function, episodic memory, and mental status for Black adults. After adjusting for demographics and childhood conditions (Model 1), an additional year of expected exposure to desegregated primary schooling was associated with higher scores on total cognitive function (b=0.46, SE=0.12) and episodic memory (b=0.34, 0.09), and marginally higher scores on mental status (b=0.13, SE=0.06, p<0.10). Adjustment for well-resourced educational systems and locality of revenue did not change the size of the associations (Model 2). Inclusion of educational attainment slightly attenuated the association between expected exposure to desegregated primary schooling and total cognitive function (b=0.42, SE=0.13) and episodic memory (b=0.33, SE=0.11), but these associations remained significant (Model 3). Estimates from the state fixed effects model (Model 4) showed significant and positive associations between expected exposure to desegregated primary schooling and total cognitive function (b=0.52, SE=0.21) and episodic memory (b=0.53, SE=0.16). The difference in total cognitive function between a Black adult with an expected exposure to desegregated primary schooling of 5 years compared to 0 years was 2.6 points (Ŷ5yrs = 0.52 × 5 years of desegregated schooling); Table 3, Model 4), which is equivalent to 0.6 standard deviations on the total cognitive function score for Black adults (mean=14.4, SD=4.4). We also confirmed, using post-hoc equality of coefficient tests, that the association between expected exposure to desegregated primary schooling and total cognitive function and episodic memory for Black adults was statistically different from White adults (Supplemental Table A3).

Table 3.

Weighted coefficients and standard errors (SE) from linear regressions predicting cognitive function with cluster adjustment for state of childhood residence (Models 1–3) or state fixed effects (Model 4), Black adults who lived in the U.S. South during primary school, Health and Retirement Study, n=1,443.

Model 1 Model 2 Model 3 Model 4
b (SE) b (SE) b (SE) b (SE)
Total Cognitive Function
Desegregated primary schooling (years) 0.46* 0.45* 0.42* 0.52*
(0.12) (0.16) (0.13) (0.21)
Well-resourced state (z-score) 0.04 0.35 −0.11
(0.26) (0.24) (0.83)
Locality of revenue (z-score) −0.00 −0.03 −0.93
(0.27) (0.25) (0.89)
Education (years) 0.63* 0.62*
(0.07) (0.07)
Intercept 13.67* 13.68* 13.87* 13.03*
(0.39) (0.42) (0.39) (0.80)
Episodic Memory
Desegregated primary schooling (years) 0.34* 0.35* 0.33* 0.53*
(0.09) (0.12) (0.11) (0.16)
Well-resourced state (z-score) 0.04 0.22 −0.74
(0.16) (0.16) (0.62)
Locality of revenue (z-score) −0.07 −0.09 −0.53
(0.19) (0.19) (0.75)
Education (years) 0.35* 0.34*
(0.05) (0.05)
Intercept 9.14* 9.12* 9.23* 8.41*
(0.32) (0.35) (0.32) (0.66)
Mental Status
Desegregated primary schooling (years) 0.13+ 0.10 0.09 −0.01
(0.06) (0.08) (0.06) (0.09)
Well-resourced state (z-score) −0.00 0.14 0.62+
(0.15) (0.15) (0.37)
Locality of revenue (z-score) 0.07 0.06 −0.40
(0.09) (0.08) (0.34)
Education (years) 0.28* 0.28*
(0.03) (0.03)
Intercept 4.53* 4.56* 4.64* 4.62*
(0.16) (0.16) (0.17) (0.32)
Cluster adjusted ✓ ✓ ✓
State fixed effects ✓
Open in a new tab

Notes: All models adjust for gender, birth year, interview mode, childhood health, and parents’ education. Continuous covariates were centered at their means. 95% CI are presented in Supplemental Table A9.

+

p<0.10,

*

p<0.05

Using margins in Stata 17 and estimates from Model 4, we calculated average predicted scores on total cognitive function and episodic memory at 0, 2.5, and 5 years of expected exposure to desegregated primary schooling, separately for Black and White adults. Margins estimates average predicted scores at fixed values of the exposure variable holding all covariates constant at their race-specific means (Figure 2). We conducted independent sample t-tests to determine if the Black-White difference in predicted values at each level of expected exposure to desegregated primary schooling was significantly different from zero and report these in Supplemental Table A4. The Black-White gap in total cognitive function and episodic memory is reduced with increasing exposure to desegregated primary schooling. For adults with 5 years of expected exposure to desegregated primary schooling, the Black-White difference in total cognitive function and episodic memory was essentially zero (Supplemental Table A4), a reduction in the initial race inequity of 3.4 and 2.8 points, respectively.

Figure 2.

Figure 2.

Open in a new tab

Predicted scores on total cognitive function and episodic memory at 0, 2.5, and 5 expected years of exposure to desegregated primary schooling for White and Black adults who grew up in the U.S. South between 1954–1974.

Notes. Predicted scores and 95% confidence intervals were calculated using margins in Stata 17 from race-stratified Model 4 estimates that hold covariates constant at their race-specific means. The 95% confidence interval describe the uncertainty around the predicted value and cannot be used to determine if the predicted value for Black and White adults differ from one another. See Supplemental Table A4 for estimates of race differences in predicted values and their associated 95% confidence intervals.

Sensitivity Analysis

We conducted several additional analyses to determine the robustness of our findings. First, a handful of respondents did not complete primary school (n=37). We re-estimated our models after excluding these respondents and our results were essentially unchanged (Supplemental Table A5). Second, we included additional childhood conditions that could serve as confounders, including residential moves due to financial difficulty (yes/no) or receiving financial help from family or friends (yes/no). These variables were generally unrelated to cognitive function and did not improve model fit; thus, we did not retain them in our models (Supplemental Table A6). Finally, 12.5% and 33% of Black and White respondents, respectively, moved across state lines between birth and age 10. We re-estimated our models after excluding movers from our sample. Our inferences were comparable to those we report (Supplemental Table A7).

Discussion

Before 1954, southern states legally mandated segregated schools, resulting in large and persistent inequities in school resources and educational opportunities by race. The Brown decision declared separate schools unconstitutional and ordered states to desegregate with “all deliberate speed”, but even so, in the two decades after Brown, many southern states actively resisted these efforts, delaying Black students’ access to desegregated schooling and better resourced schools for years. Our study exploited state and temporal variation in state-level school desegregation efforts to determine if expected exposure to desegregated primary schooling was associated with midlife cognitive function among Black and White adults who grew up in the U.S. South in the two decades following Brown, if this association differed by race, and if state-level educational quality and adult educational attainment explained the association.

As hypothesized, we found that, among Black adults, more years of expected exposure to desegregated primary schooling was associated with better cognitive function. These effects were largely driven by higher scores on episodic memory. Several prior studies using community and national samples of older adults also showed a positive relationship between attending a desegregated school and cognitive function in later life among Black, but not White, older adults (Aiken-Morgan et al., 2015; Peterson et al., 2021; Walsemann, Kerr, et al., 2022; Whitfield & Wiggins, 2003). For example, a study of Black Northern Californians found that, compared to Black older adults who never attended integrated schools, Black older adults who only attended integrated schools had higher scores on semantic memory, whereas Black older adults who transitioned to integrated schools during their primary years had higher scores on semantic memory and executive function (Peterson et al., 2021). Similarly, in a national sample, Black older adults who reported ever attending a majority White school scored higher on total cognitive function, episodic memory, and working memory than their counterparts that never attended a majority White school (Walsemann, Kerr, et al., 2022). The relationship between attending a majority non-White school and cognitive outcomes for White older adults, however, was null.

Although our findings align with most studies, two community samples showed poorer cognitive function among Black adults who attended a legally desegregated school in the U.S. South (Allaire & Whitfield, 2004; Lamar et al., 2020). These findings may be due to two factors. First, the studies sampled current residents of Chicago (Lamar et al., 2020) and Baltimore (Allaire & Whitfield, 2004); thus, they may not be representative of Black adults who grew up in the U.S. South in the 1950s–1970s. Second, and perhaps most importantly, given the age of sample respondents, attendance at desegregated schools occurred very soon after Brown (i.e., 1954–1957). As such, respondents who reported attending a desegregated school most likely did so in the handful of states that desegregated early and had relatively small Black populations (U.S. Census Bureau, 1966). In contrast, our study used a large, representative sample of Black and White Americans who grew up in the U.S. South in the two decades following Brown. This time span provides a more complete examination of the role of school desegregation on cognitive function, particularly given that most southern states did not significantly desegregate schools until the mid to late-1960s.

Given prior work documenting a positive relationship between school desegregation and Black students’ access to quality schooling and educational attainment (Johnson, 2019; Reber, 2010), both of which are consistently and independently associated with higher levels of cognitive function at older ages (Crowe et al., 2013; Glymour et al., 2008; Moorman et al., 2019; Sisco et al., 2015; Walsemann, Ureña, et al., 2022), we examined whether the relationship between expected exposure to desegregated primary schooling and midlife cognitive function persisted with the inclusion of these other factors. The relationship was robust to their inclusion. It is important to note, however, that measuring educational quality at the state-level may obscure variation within states and may not adequately capture individual schooling experiences that would have accompanied school desegregation (Johnson, 2019). This level of data, however, is not readily available for older cohorts in the United States. Thus, our measures of educational quality provide an appropriate test of this potential association.

Our state-level measure of school desegregation may be capturing aspects of structural racism within southern states across this period. For example, states that resisted school desegregation may have created and reinforced a broader social and political environment that was actively hostile to Black students and that served to undermine Black students’ learning and educational opportunities. Several studies have shown that racism is related to poorer cognitive function among Black older adults (Coogan et al., 2020; Grasser & Jovanovic, 2022) and that racial discrimination as a chronic stressor can alter the brain’s neural structure and function (Fani et al., 2022; Okeke et al., 2022; Zahodne et al., 2023) in ways associated with poorer cognitive function later in life (Barnes et al., 2012).

Strengths and Limitations

In addition to those already noted, our study includes several innovations and strengths. First, we used twenty years of administrative data on state-level school desegregation to create a measure of an individuals’ expected exposure to school desegregation during primary school. Prior work on school segregation and cognitive function relied on retrospective self-reports which may be prone to recall and/or measurement bias, particularly if people are using different definitions of school segregation. Second, we used a large population-based sample with representation across the U.S. South during a twenty-year period. Thus, our results are generalizable to Black and White individuals who were born between 1948 and 1963 and grew up in the U.S. South. Finally, most studies have not investigated the timing of exposure to school desegregation and cognitive function, even though early exposure to desegregated schools may represent a sensitive period for cognitive development and later cognitive function (cf. Peterson et al, 2021). Our data allowed us to examine exposure to desegregation during primary school, which better aligns with how states often implemented school desegregation plans (i.e., adopting desegregation plans one grade at a time, starting with primary school grades).

We should also mention limitations. First, we linked state data to the HRS using the respondents’ state of residence when they were around age 10. While we were able to determine that state moves between birth and age 10 did not alter our inferences, we cannot account for state moves that happened after age 10; however, since we measured desegregation during primary school (ages 6–11) these moves should not substantially impact our findings. Second, our measure of school desegregation does not capture a respondent’s actual attendance at a desegregated primary school; however, our measure uses high-quality administrative data on how many Black students were attending schools with White students in each state. Although a sub-sample of HRS respondents provided retrospective reports on the racial composition of the schools they attended, using these measures would have reduced our sample size and introduced measurement error, limiting generalizability and power. Third, although the HRS collects information about respondents’ memory and impairment from proxies when respondents were unable to complete the TICs, we cannot incorporate proxy reports into the summated cognitive scores. Only 77 respondents provided proxy interviews; thus, this should not bias our results. Finally, we estimated cross-sectional associations between school desegregation and cognitive function but did not assess associations with cognitive decline. Cognitive decline, however, is uncommon in midlife and mainly occurs at older ages (e.g., ≥ 65 years) (Murman, 2015). Given when our respondents were born, only those born before 1954 provided any observations after age 65. Moreover, our youngest respondents provided at most 2 observations of cognitive function, precluding our ability to model decline since a minimum of 3 observations is required to assess linear change (Singer & Willet, 2003). It will be important, however, to determine if our findings extend to cognitive decline as respondents enter older adulthood and more of them, particularly those in the younger cohorts, provide enough observations to accurately assess the role of school desegregation on cognitive decline.

Conclusion

A large body of work has investigated the role of educational attainment in explaining Black-White inequities in cognitive function (Chen & Zissimopoulous, 2018; Xiong et al., 2020; Yaffe et al., 2013); however, older cohorts of U.S. Southerners were attending school during a time of major historical change. Our findings suggest that one of these changes – the desegregation of public schools – may have had far reaching implications for midlife cognitive function. Our study also demonstrates why historically situating the lived experiences of Black Americans is vital for understanding their later cognitive function and points to the need to conduct withinpopulation analyses that is rooted in history.

Supplementary Material

1

Highlights.

  • Black older adults have worse cognitive function than their White peers.

  • State resistance to school desegregation in U.S. South post-Brown may play a role.

  • Twenty years of school desegregation data linked to Health and Retirement Study.

  • Desegregated primary school related to better cognitive function for Black adults.

  • This association remained after adjustment for education quality and attainment.

Acknowledgments.

This work was supported with funding from an Alzheimer’s Association Grant (AARG-NTF-20-684252), the National Institute on Aging (R01AG067536; K02AG075237, P30AG066589; P30AG043073; K99AG076964), and the Eunice Kennedy Shriver National Center for Child Health and Human Development (P2CHD041041). No funding providers played a role in study design/conduct, analysis/interpretation of data, or manuscript preparation.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Aiken-Morgan AT, Gamaldo AA, Sims RC, Allaire JC, & Whitfield KE (2015). Education desegregation and cognitive change in African American older adults. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 70(3), 348–356. 10.1093/geronb/gbu153 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Allaire J, & Whitfield K (2004). Relationships among education, age, and cognitive functioning in older African Americans: The impact of desegregation. Aging Neuropsychology and Cognition, 11, 443–449. 10.1080/13825580490521511 [DOI] [Google Scholar]
  3. Barnes LL, Lewis TT, Begeny CT, Yu L, Bennett DA, & Wilson RS (2012). Perceived discrimination and cognition in older African Americans. Journal of the International Neuropsychological Society, 18(5), 856–865. 10.1017/S1355617712000628 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Beady CH, & Hansell S (1981). Teacher race and expectations for student achievement. American Educational Research Journal, 18(2), 191–206. 10.2307/1162381 [DOI] [Google Scholar]
  5. Bell D (2004). Silent covenants: Brown v. Board of Education and the unfulfilled hopes for racial reform. Oxford University Press. http://site.ebrary.com/id/10266466 [Google Scholar]
  6. Bonilla-Silva E (1997). Rethinking racism: Toward a structural interpretation. American Sociological Review, 62(3), 465–480. 10.2307/2657316 [DOI] [Google Scholar]
  7. Borghans L, Golsteyn BHH, & Zölitz U (2015). School quality and the development of cognitive skills between age four and six. PLoS ONE, 10(7), e0129700. 10.1371/journal.pone.0129700 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Burger K (2010). How does early childhood care and education affect cognitive development? An international review of the effects of early interventions for children from different social backgrounds. Early Childhood Research Quarterly, 25(2), 140–165. 10.1016/j.ecresq.2009.11.001 [DOI] [Google Scholar]
  9. Campbell RF, Flake T, & Lesson J (1967). A statistical summary, state by state, of school segregation-desegregation in the Southern and Border area from 1954 to the present. https://files.eric.ed.gov/fulltext/ED019382.pdf
  10. Chen C, & Zissimopoulos JM (2018). Racial and ethnic differences in trends in dementia prevalence and risk factors in the United States. Alzheimer’s & Dementia: Translational Research & Clinical Interventions, 4(1), 510–520. 10.1016/j.trci.2018.08.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Coogan P, Schon K, Li S, Cozier Y, Bethea T, & Rosenberg L (2020). Experiences of racism and subjective cognitive function in African American women. Alzheimer’s & Dementia: Diagnosis, Assessment & Disease Monitoring, 12(1), e12067. 10.1002/dad2.12067 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cooley SA, Heaps JM, Bolzenius JD, Salminen LE, Baker LM, Scott SE, & Paul RH (2015). Longitudinal change in performance on the Montreal Cognitive Assessment in older adults. The Clinical Neuropsychologist, 29(6), 824–835. 10.1080/13854046.2015.1087596 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Crowe M, Clay OJ, Martin RC, Howard VJ, Wadley VG, Sawyer P, & Allman RM (2013). Indicators of childhood quality of education in relation to cognitive function in older adulthood. The Journals of Gerontology, Series A: Biomedical Sciences and Medical Sciences, 68(2), 198–204. 10.1093/gerona/gls122 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Darling-Hammond L (2000). Teacher quality and student achievement. Education Policy Analysis Archives, 8, 1. [Google Scholar]
  15. Dee TS (2004). Teachers, race, and student achievement in a randomized experiment. Review of Economics & Statistics, 86(1), 195–210. 10.1162/003465304323023750 [DOI] [Google Scholar]
  16. Dee TS (2005). A Teacher Like Me: Does race, ethnicity, or gender matter? American Economic Review, 95(2), 158–165. 10.1257/000282805774670446 [DOI] [Google Scholar]
  17. Fani N, Harnett NG, Bradley B, Mekawi Y, Powers A, Stevens JS, Ressler KJ, & Carter SE (2022). Racial discrimination and white matter microstructure in trauma-exposed Black women. Biological Psychiatry, 91(3), 254–261. 10.1016/j.biopsych.2021.08.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Fisher PGG, Hassan H, Faul JD, Rodgers WL, & Weir DR (2017). Health and Retirement Study Imputation of Cognitive Functioning Measures: 1992 – 2014 (Final Release Version) Data Description.
  19. Fredriksson P, Öckert B, & Oosterbeek H (2013). Long-term effects of class size. The Quarterly Journal of Economics, 128(1), 249–285. [Google Scholar]
  20. Gee GC, & Hicken MT (2021). Structural racism: The rules and relations of inequity. Ethnicity & Disease, 31(Suppl 1), 293–300. 10.18865/ed.31.S1.293 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Gershenson S, Holt SB, & Papageorge NW (2016). Who believes in me? The effect of student–teacher demographic match on teacher expectations. Economics of Education Review, 52, 209–224. 10.1016/j.econedurev.2016.03.002 [DOI] [Google Scholar]
  22. Glymour MM, Kawachi I, Jencks CS, & Berkman LF (2008). Does childhood schooling affect old age memory or mental status? Using state schooling laws as natural experiments. J Epidemiol Community Health, 62(6), 532–537. 10.1136/jech.2006.059469 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Goldfeld S, O’Connor E, O’Connor M, Sayers M, Moore T, Kvalsvig A, & Brinkman S (2016). The role of preschool in promoting children’s healthy development: Evidence from an Australian population cohort. Early Childhood Research Quarterly, 35, 40–48. 10.1016/j.ecresq.2015.11.001 [DOI] [Google Scholar]
  24. Goldhaber DD, & Brewer DJ (2000). Does teacher certification matter? High school teacher certification status and student achievement. Educational Evaluation and Policy Analysis, 22(2), 129–145. 10.2307/1164392 [DOI] [Google Scholar]
  25. Graham JW, Olchowski AE, & Gilreath TD (2007). How many imputations are really needed? Some practical clarifications of multiple imputation theory. Prevention Science: The Official Journal of the Society for Prevention Research, 8(3), 206–213. 10.1007/s11121-007-0070-9 [DOI] [PubMed] [Google Scholar]
  26. Grasser LR, & Jovanovic T (2022). Neural impacts of stigma, racism, and discrimination. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, 7(12), 1225–1234. 10.1016/j.bpsc.2022.06.012 [DOI] [PubMed] [Google Scholar]
  27. Guryan J (2004). Desegregation and Black dropout rates. American Economic Review, 94(4), 919–943. 10.1257/0002828042002679 [DOI] [Google Scholar]
  28. Hanushek EA, Kain JF, & Rivkin SG (2009). New evidence about Brown v. Board of Education: The complex effects of school racial composition on achievement. Journal of Labor Economics, 27(3), 349–383. 10.1086/600386 [DOI] [Google Scholar]
  29. Heeringa SG, West BT, & Berglund PA (2017). Applied survey data analysis (2nd ed.). Chapman & Hall/CRC Press. [Google Scholar]
  30. Hudgins HC (1973). Public school desegregation: Legal issues and judicial decisions. (No. 24). National Organization on Legal Problems of Education. [Google Scholar]
  31. Jackson DA (1993). Stopping rules in principal components analysis: A comparison of heuristical and statistical approaches. Ecology, 74(8), 2204–2214. 10.2307/1939574 [DOI] [Google Scholar]
  32. Jendryczko D, Scharfen J, & Holling H (2019). The impact of situational test anxiety on retest effects in cognitive ability testing: A structural equation modeling approach. Journal of Intelligence, 7(4), 22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Johnson RC (2019). Children of the dream: Why school integration works. Basic Books. [Google Scholar]
  34. Krueger AB (1999). Experimental estimates of education production functions. The Quarterly Journal of Economics, 114(2), 497–532. [Google Scholar]
  35. Lamar M, Lerner AJ, James BD, Yu L, Glover CM, Wilson RS, & Barnes LL (2020). Relationship of early-life residence and educational experience to level and change in cognitive functioning: Results of the Minority Aging Research Study. The Journals of Gerontology: Series B, 75(7), e81–e92. 10.1093/geronb/gbz031 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Lewis AE (2003). Race in the schoolyard: Negotiating the color line in classrooms and communities. Rutgers University Press. https://muse.jhu.edu/pub/176/monograph/book/6127 [Google Scholar]
  37. Livingston G, Sommerlad A, Orgeta V, Costafreda SG, Huntley J, Ames D, Ballard C, Banerjee S, Burns A, Cohen-Mansfield J, Cooper C, Fox N, Gitlin LN, Howard R, Kales HC, Larson EB, Ritchie K, Rockwood K, Sampson EL, … Mukadam N (2017). Dementia prevention, intervention, and care. The Lancet, 390(10113), 2673–2734. 10.1016/S0140-6736(17)31363-6 [DOI] [PubMed] [Google Scholar]
  38. Mantri S, Nwadiogbu C, Fitts W, & Dahodwala N (2019). Quality of education impacts late-life cognition. International Journal of Geriatric Psychiatry, 34(6), 855–862. 10.1002/gps.5075 [DOI] [PubMed] [Google Scholar]
  39. Moorman SM, Greenfield EA, & Garcia S (2019). School context in adolescence and cognitive functioning 50 years later. Journal of Health and Social Behavior, 60(4), Article 4. 10.1177/0022146519887354 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Murman DL (2015). The impact of age on cognition. Seminars in Hearing, 36(03), 111–121. 10.1055/s-0035-1555115 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Okeke O, Elbasheir A, Carter SE, Powers A, Mekawi Y, Gillespie CF, Schwartz AC, Bradley B, & Fani N (2022). Indirect effects of racial discrimination on health outcomes through prefrontal cortical white matter integrity. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging. 10.1016/j.bpsc.2022.05.004 [DOI] [PubMed] [Google Scholar]
  42. Orfield G, & Eaton SE (1996). Dismantling desegregation. The quiet reversal of Brown v. Board of Education. The New Press. [Google Scholar]
  43. Peterson RL, George KM, Barnes LL, Gilsanz P, Mayeda ER, Glymour MM, Mungas DM, & Whitmer RA (2021). Association of timing of school desegregation in the United States with late-life cognition in the Study of Healthy Aging in African Americans (STAR) Cohort. JAMA Network Open, 4(10), e2129052. 10.1001/jamanetworkopen.2021.29052 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Pohl DJ, Seblova D, Avila JF, Dorsman KA, Kulick ER, Casey JA, & Manly J (2021). Relationship between residential segregation, later-life cognition, and incident dementia across race/ethnicity. International Journal of Environmental Research and Public Health, 18(21), Article 21. 10.3390/ijerph182111233 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Rabbitt PMA, McInnes L, Diggle P, Holland F, Bent N, Abson V, Pendleton N, & Horan M (2004). The University of Manchester Longitudinal Study of Cognition in Normal Healthy Old Age, 1983 through 2003. Aging, Neuropsychology, and Cognition, 11(2–3), 245–279. 10.1080/13825580490511116 [DOI] [Google Scholar]
  46. Reber SJ (2010). School desegregation and educational attainment for Blacks. The Journal of Human Resources, 45(4), 893–914. [Google Scholar]
  47. Rivkin SG, Hanushek EA, & Kain JF (2005). Teachers, schools, and academic achievement. Econometrica, 73(2), 417–458. 10.1111/j.1468-0262.2005.00584.x [DOI] [Google Scholar]
  48. Ruggles S, Flood S, Goeken R, Schouweiler M, & Sobek M (2022). IPUMS USA: Version 12.0 (12.0) [Data set]. Minneapolis, MN: IPUMS. 10.18128/D010.V12.0 [DOI] [Google Scholar]
  49. Shah HK, Domitrovich CE, Morgan NR, Moore JE, Cooper BR, Jacobson L, & Greenberg MT (2017). One or two years of participation: Is dosage of an enhanced publicly funded preschool program associated with the academic and executive function skills of low-income children in early elementary school? Early Childhood Research Quarterly, 40, 123–137. 10.1016/j.ecresq.2017.03.004 [DOI] [Google Scholar]
  50. Singer JD, & Willet JB (2003). Applied longitudinal data analysis: Modeling change and event occurrence. Oxford University Press, Inc. [Google Scholar]
  51. Sisco S, Gross AL, Shih RA, Sachs BC, Glymour MM, Bangen KJ, Benitez A, Skinner J, Schneider BC, & Manly JJ (2015). The role of early-life educational quality and literacy in explaining racial disparities in cognition in late life. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 70(4), 557–567. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Sonnega A, & Weir DR (2014). The Health and Retirement Study: A public data resource for research on aging. Open Health Data, 2(1). 10.5334/ohd.am [DOI] [Google Scholar]
  53. Tushnet MV (1987). The NAACP’s Legal Strategy Against Segregated Education, 1925–1950. University of North Carolina Press. https://search.ebscohost.com/login.aspx?direct=true&db=nlebk&AN=104016&site=ehost-live [Google Scholar]
  54. U.S. Census Bureau. (1966). U.S. Census of Population: 1960. Supplementary Reports, Series PC(S1)-52. Negro Population, by County: 1960 and 1950. U.S. Government Printing Office, Washington, D.C. https://www2.census.gov/library/publications/decennial/1960/pc-s1-supplementary-reports/pc-s1-52.pdf [Google Scholar]
  55. Walsemann KM, & Ailshire JA (2020). Early educational experiences and trajectories of cognitive functioning among us adults in midlife and later. American Journal of Epidemiology, 189(5), 403–411. 10.1093/aje/kwz276 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Walsemann KM, Kerr EM, Ailshire JA, & Herd P (2022). Black-White variation in the relationship between early educational experiences and trajectories of cognitive function among US-born older adults. SSM - Population Health, 19, 101184. 10.1016/j.ssmph.2022.101184 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Walsemann KM, Ureña S, Farina MP, & Ailshire JA (2022). Race inequity in school attendance across the Jim Crow south and its implications for Black–White disparities in trajectories of cognitive function among older adults. The Journals of Gerontology: Series B, 77(8), 1467–1477. 10.1093/geronb/gbac026 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Weuve J, Barnes LL, Mendes de Leon CF, Rajan KB, Beck T, Aggarwal NT, Hebert LE, Bennett DA, Wilson RS, & Evans DA (2018). Cognitive aging in Black and White Americans: Cognition, cognitive decline, and incidence of Alzheimer Disease Dementia. Epidemiology, 29(1), 151. 10.1097/EDE.0000000000000747 [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Whitfield KE, & Wiggins SA (2003). The Impact of desegregation on cognition among older African Americans. Journal of Black Psychology, 29(3), 275–291. 10.1177/0095798403254209 [DOI] [Google Scholar]
  60. Wilkinson JHI (1978). Supreme Court and southern school desegregation, 1955 – 1970: A history and analysis. Virginia Law Review, 64(4), 485–560. [Google Scholar]
  61. Xiong C, Luo J, Coble D, Agboola F, Kukull W, & Morris JC (2020). Complex interactions underlie racial disparity in the risk of developing Alzheimer’s disease dementia. Alzheimer’s & Dementia, 16(4), 589–597. 10.1002/alz.12060 [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Yaffe K, Falvey C, Harris TB, Newman A, Satterfield S, Koster A, Ayonayon H, & Simonsick E (2013). Effect of socioeconomic disparities on incidence of dementia among biracial older adults: Prospective study. BMJ, 347, f7051. 10.1136/bmj.f7051 [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Zahodne LB, Sharifian N, Kraal AZ, Morris EP, Sol K, Zaheed AB, Meister L, Mayeux R, Schupf N, Manly JJ, & Brickman AM (2023). Longitudinal associations between racial discrimination and hippocampal and white matter hyperintensity volumes among older Black adults. Social Science & Medicine, 316, 114789. 10.1016/j.socscimed.2022.114789 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1941513-supplement-1.docx (85.6KB, docx)

RESOURCES