We investigate associations between community water fluoridation (CWF), adolescent IQ, and cognition across the life course using representative data from the US state of Wisconsin. Exposure is inferred from historical records on community water fluoridation; adolescent IQ is ascertained from state testing records; and cognition in later life is assessed as part of the Wisconsin Longitudinal Study [P. Herd et al., Int. J. Epidemiol. 43, 34-41 (2014).]. In contrast to studies cited in recent decisions to end CWF in Utah, Florida, and elsewhere, we find no evidence that CWF is negatively associated with adolescent IQ or adult cognitive functioning.

OBJECTIVE:National surveys of health and aging increasingly collect data via multiple modes, including via face-to-face, phone, and the web. However, mode effects-differences in measurement due to how the data are collected, rather than to underlying differences in cognition-undermine comparability in performance for different segments of samples, impeding analysis of cognitive function in the full sample. This article reports on the construction and evaluation of a measure of general cognitive functioning comparable across diverse phone and web cognitive tasks for the 2021 follow-up of the High School and Beyond Study of 1980, a nationally representative survey of individuals who were U.S. high school sophomores and seniors in 1980. METHOD:We used a hierarchical explanatory item response theory model to equate general cognition scores across phone and web samples based on the relationship between midlife cognition and covariates from midlife and high school. RESULTS:The full model replicated task scores from unidimensional models, and the explanatory component of the model did not bias general cognition scores, indicating that the model was a faithful representation of patterns of cognitive task performance. We found no evidence of a violation of our assumption of the independence of regression coefficients and survey mode. CONCLUSIONS:This article provides an empirical example of how to construct a psychometrically sound cross-mode cognitive score from data collected through multiple modes. Given the increasing demand for cost-effective large-scale survey design, our method may find greater application in other nationally representative aging studies. (PsycInfo Database Record (c) 2026 APA, all rights reserved).

I quantify the degree to which people’s self-assessed overall health (SAOH) predicts their subsequent mortality. Prior research has justified using this measure in substantive research by claiming it has high predictive validity; many authors subsequently suggest that the measure has clinical utility because of its validity in predicting mortality. I computed risk of death within 1, 3, 5, and 10 years as a function of responses at baseline to a question about whether people’s health is generally excellent, very good, good, fair, or poor. Data come from the 1986 to 2018 U.S. National Health Interview Surveys and the 2014/2015 High School and Beyond study—each linked to mortality records. People offering worse evaluations of their health were substantially more likely to die within 1, 3, 5, and 10 years. Nevertheless, SAOH has low sensitivity and a high false positive rate (regardless of how it is dichotomized). Results are consistent across samples and sociodemographic groups. SAOH is an important measure of well-being and should continue to be used in epidemiological and other research. However, on its own it is not predictive of individual-level mortality; claims about its clinical utility for targeting interventions at people at highest risk of death are not supported.

This study examines the relationship between early-life bilingualism and midlife cognitive health. Using data from the 1980, 1982, and 2021 waves of High School and Beyond: 1980 (N ≈ 13,980), we find that bilingual adults show higher midlife cognition scores, but the association weakens after adjusting for early-life academic achievement and educational expectations. Becoming bilingual before school entry is not associated with stronger midlife cognition than monolingualism or becoming bilingual after school entry. English-first bilingual speakers do not exhibit better midlife cognition than either English monolingual speakers or non-English-first bilingual speakers. By situating the debate about the cognitive benefits of bilingualism within broader educational and stratification contexts, our results highlight early-life educational inequalities as a key mechanism driving the apparent relationship between early-life bilingualism and midlife cognition.

We document the role of premarket human capital in explaining racial and ethnic income disparities in labor market outcomes for individuals from their late 20s to mid-50s. Using comprehensive measures of the human capital students possess at the end of high school—including scores on achievement tests in reading comprehension, vocabulary, and math, GPA—and detailed school context measures, we study Black-White and Hispanic-White gaps across various economic outcomes. We find that human capital at high school completion largely determines racial and ethnic differences in mid-life labor market outcomes, explaining up to 87% of gaps in some outcomes. Our analysis examines men and women separately to provide a comprehensive picture of how human capital relates to racial and ethnic disparities across genders. We find that human capital explains a larger proportion of Black-White income gaps for women (95%) than men (69%), with women’s racial gaps primarily explained by test scores while men’s gaps show greater contributions from educational attainment and school characteristics. Finally, we show that the Black-White income gap for this cohort widened from early career to mid-life, with the explanatory power of human capital increasing as people grew older. Our research extends previous literature by considering longer-term outcomes, employing comprehensive and refined human capital measures, and broadening the scope of racial and ethnic comparisons. These findings emphasize the critical and persistent role of human capital development in shaping long-term economic outcomes and racial and ethnic disparities, reinforcing the need for targeted educational interventions to address persistent inequalities across the life course.

Educational attainment has consistently been identified as an important determinant of later-life cognitive functioning and recent studies, moreover, suggest that additional aspects of the educational process, particularly school quality, may be related to later-life cognition. Many studies find a strong association between childhood socioeconomic status and later-life cognition and that part of the association is mediated by educational attainment. What is not known, however, is whether the associations between the different aspects of the educational process and later-life cognitive functioning vary by childhood SES: Is education more beneficial for the cognitive functioning of individuals from lower or higher socioeconomic backgrounds? The “cumulative advantage” or “resource multiplication” perspectives would hypothesize that individuals from higher socioeconomic backgrounds are likely to benefit more from higher school quality and a higher educational attainment. The “compensatory leveling” or “resource substitution” perspectives, in contrast, would hypothesize stronger effects for individuals from lower socioeconomic backgrounds. Using long-spanning nationally representative data from the U.S. High School & Beyond (HSB) cohort study, we examine whether the associations between school quality, educational attainment, and midlife cognitive functioning differ by childhood SES. HSB has followed a nationally representative sample of ∼25,500 people from high school in 1980 through midlife in 2021-22. HSB data from the 1980s (when sample members were adolescents) includes rich information about educational contexts, opportunities, and outcomes; early life socioeconomic and family circumstances; spatial location; and demographic group memberships. HSB data from 2021-22 (when most were approaching age 60) include high quality measures of cognitive functioning. The findings will have implications for our understanding of the life course processes underlying disparities in later-life cognitive functioning. None yet None yet

Are the cognitive and physical health returns to education uniform across racial and ethnic groups in the United States? Past research has been inconclusive. A few studies posit that the benefits to higher levels of education are greater for Black people than they are for White people (Barnes et al. 2011, Sherman-Wilkins and Thierry 2019) while others point to lower returns to higher education for Black (Assari 2018a, Assari 2018b, Holmes and Zajacova 2014, Kimbro et al. 2008) and Latinx (Turra and Goldman 2007) people. Still other studies report few differences in the returns to education across racial/ethnic groups (Eng et al. 2021, Vable et al. 2018). We evaluate the returns to education by race and ethnicity across several outcomes of relevance to ADRD. We hypothesize that variation in the health returns to education are modest and are largely accounted for by variation in the quality of the high schools’ students attend; the curriculum to which students are exposed; the standing of the colleges where they earn their degrees; the fields of their degrees; and their levels of academic achievement in high school. Past studies have mostly been unable to observe variation in either academic achievement or the quality of educational opportunities, treating all credentials as identical. If opportunities or achievements are unequally distributed across racial/ethnic groups, however, these differences may lead to the erroneous conclusion that returns to credentials are unequal. We use data from the U.S. High School and Beyond (HSB) cohort study, which has followed a nationally representative sample of ∼25,500 people from high school in 1980 through midlife in 2021-22. HSB data from the 1980s includes rich information about educational contexts, opportunities, and outcomes; early life socioeconomic and family circumstances; spatial location; and demographic group memberships. HSB data from 2021-22 include high quality measures of a variety of cognitive and health outcomes from extensive surveys, cognitive assessments, and biomarkers. Our outcomes measures will include a global measure of cognitive functioning at age ∼60; blood pressure / hypertension; A1C levels and diabetes; history of stroke; and self-assessed overall health. None yet. None yet.

Educational gradients in health conditions are profound, but our understanding of how to address them is limited by the data available to us. In large-scale population-based research, we most frequently evaluate the relationship between educational attainment (or years of schooling) and either (a) self-reported incidence of adverse health events or diagnoses of health conditions or (b) biomarkers of those diseases. Self-reports are limited, however, by inequities and access to care; people with limited access to care may not know about health problems or conditions (at least until they reach acute levels). This means that inequities in (for example) self-reported hypertension conflate actual inequities in disease with inequities in awareness of that disease. Relying solely on biomarkers presents a different challenge for understanding health inequities ─ mainly because biomarkers often reflect both disease incidence and disease treatment. For example, a person with normal A1C levels may have never had diabetes or they may have been effectively treated for diabetes. Until now, we have not had ─ for large, population-representative cohort studies ─ all three of (a) people's self-reports of disease; (b) biomarkers of disease; and (c) pharmaceutical records that indicate disease treatment. How do inferences about educational inequities in disease ─ specifically hypertension and diabetes ─ change when we have full information about all three? To address this question, we use new data from the U.S. High School & Beyond (HS&B:80) cohort study, which has followed a nationally representative sample of ∼25,500 people from high school in 1980 through midlife in 2021-22. HS&B:80 data from the 1980s includes rich information about educational contexts, opportunities, and outcomes; early life socioeconomic and family circumstances; spatial location; and demographic group memberships. Data from 2021-22 include all three of (a) self-reports of hypertension and diabetes; (b) blood pressure and A1C levels; and (c) pharmacy records on treatments for both conditions. The results of this study will inform our understanding of health equity, and either reinforce our confidence in the reliability and validity of self-reported health conditions or qualify how we interpret them. Nothing yet. Nothing yet.

Does intergenerational occupational mobility impact health and well-being net of occupational origins and destinations? Are the net effects of mobility homogeneous across demographic subgroups? Using data from High School and Beyond cohort (HS&B:80) and two mobility effect models (Luo's Mobility Contrast Model and Sobel's Diagonal Reference Model), we find no evidence of mobility effects on physical (i.e., hypertension, diabetes), mental (i.e., depression, loneliness), self-rated global, or cognitive health in mid-life at the population level. However, we find heterogeneity in mobility effects across gender and racial subgroups. The effects of origin and destination statuses vary across health outcomes, and the (dis)advantages associated with origin status are sometimes fully offset by those of destination status for intergenerationally mobile individuals.

Importance:The incidence and prevalence of clinical Alzheimer disease (AD) are higher among Black and Latinx older adults than among White older adults. Past studies that compared plasma AD biomarker concentrations among groups minoritized by their race and ethnicity yielded inconsistent findings; however, these efforts did not include population representative samples or statistical procedures to ensure population representation. Objective:To examine race and ethnicity differences in plasma AD biomarker concentrations and in the association between biomarkers and medical conditions in a US population-representative cohort of middle-aged adults (approximately 58 years of age). Design, Setting, and Participants:Data for this cohort study came from the High School and Beyond sample, a nationally representative cohort of high school sophomores and seniors who were recruited in 1980. In 2021, a subset of participants provided blood samples that were assayed for amyloid-β (Aβ42/Aβ40 ratio), phosphorylated tau-181 (pTau-181), neurofilament light chain (NfL), and glial fibrillary acidic protein (GFAP). The analyses for the present study were conducted between July 2, 2024, and August 26, 2025, using data collected during the 2021 follow-up study. Exposures:Race and ethnicity groups and common medical conditions. Main Outcomes and Measures:General linear models with Wald tests were used to compare biomarker concentrations between race and ethnicity groups and to test interactions with common medical conditions using unadjusted biomarker values and models adjusted to ensure population representation with inverse probability weighting and multiple imputation. Results:The sample included 4340 adults (mean [SD; range] age, 58.1 [1.1; 56-63] years; 2400 [55.3%] women); 630 (14.4%) were Black, 900 (20.7%) were Latinx, and 2610 (60.1%) were White. Black participants had a lower Aβ42/Aβ40 ratio (d = -0.002; 95% CI, -0.004 to -0.000; P = .04) and lower NfL concentrations (d = -1.16; 95% CI, -2.15 to -0.16; P = .02) than White participants, but these differences were attenuated when models were adjusted for population representation (d = 0.000; 95% CI, -0.002 to 0.002; P = .85 for Aβ ratio; d = -0.88; 95% CI, -1.78 to 0.02; P = .05 for NfL). Latinx participants had lower GFAP concentrations than White participants (d = -3.87; 95% CI, -7.30 to -0.45; P = .03), but these differences were also attenuated when models were adjusted for population representation (d = 3.36; 95% CI, -3.13 to 9.86; P = .31). In general, estimated biomarker means were similar between race and ethnicity groups. History of type 2 diabetes was associated with increased NfL concentration (d = 0.19; 95% CI, 0.07 to 0.30; P = .04), and high body mass index was associated with lower Aβ42/Aβ40 ratio (d = -0.13; 95% CI, -0.21 to -0.06; P = .02); whereas high cholesterol was associated with lower pTau-181 concentration (d = -0.18; 95% CI, -0.25 to -0.10; P = .01) and high BMI was associated with lower GFAP concentration (d = -0.30; 95% CI -0.44 to -0.16; P = .01). No differences in associations between morbidities and AD biomarker concentrations were detected across race and ethnicity groups. Conclusions and Relevance:In this cohort of middle-aged adults, the use of appropriate statistical estimation to ensure population representation indicated that blood-based AD biomarker concentrations were not distinguishable among race and ethnicity groups. Common medical conditions were associated with plasma biomarker concentrations similarly across race and ethnicity groups. These results highlight the importance of considering population representation and comorbid conditions in AD research to ensure accurate characterization of disease pathophysiology and improve precision of diagnostic and treatment strategies for populations that experience AD disparities.

In recent years, American political figures, media pundits, and others have argued that undocumented residents in the United States should not be included in census data used for congressional apportionment. They argue that including them unfairly benefits some states-especially politically Democratic-leaning (or "blue") states, according to most arguments-at the expense of others. Some people have speculated that many more House seats and Electoral College votes would have been apportioned to politically Republican-leaning (or "red") states and that many fewer would have been apportioned to politically "blue" states if undocumented residents were excluded from apportionment data. The only systematic empirical examinations of this question occurred prior to the 2020 census and projected-mostly inaccurately-how apportionment would differ after the 2020 census under this hypothetical. Does including undocumented residents in US census data used for congressional apportionment alter (i) political party representation in the House of Representatives or (ii) presidential Electoral College votes? We use annual state-level estimates of the undocumented resident population at each census date to examine the consequences of including them in official apportionment data since 1980. If undocumented residents had been entirely excluded from census data used for apportionment, no more than two House seats and three Electoral College votes would have shifted between political parties in any year since 1980; this would have had no bearing on party control of the House or the outcome of presidential elections.

A substantial body of research points to the persistence of educational credential gradients in cognitive function across the life course, from early adulthood through old age. Fewer studies have considered the roles of educational opportunities and achievements early in the life course in shaping (1) educational credential gradients or (2) variation in cognitive function among those with the same credentials. As a result, researchers underestimate the contribution of education writ large to variation in cognitive performance later in life. We analyze data from the nationally representative High School and Beyond (HSB) cohort, which has followed ∼25,500 people from high school in 1980 through age ∼60 in 2021-22. The 2021-22 round of HSB gathered high quality measures of cognition across multiple domains; the 1980-1992 rounds gathered extensive information about educational opportunities, achievements, and attainments. We demonstrate that high school educational opportunities, and especially academic achievements, account for most of the credential gradient in cognition on selected cognition measures among adults who attended high school in the U.S. in 1980 as they approach 60 years of age. High school educational opportunities and academic achievements also help account for some of the variation in cognition among those with the same educational credential. Educational credentials commonly employed to understand the contribution of education to cognitive aging therefore risk both mischaracterizing and understating the contributions of education to cognitive function and change over the life course.

Development of blood-based biomarkers makes measurement of neurodegeneration possible in populations that have not traditionally been represented in ADRD research. Investigation of the relationship of blood-based AD biomarkers to cognitive functioning in longitudinal population representative samples is critical. This study examined the relationship between plasma AD biomarkers and cognitive functioning in middle-aged adults, and determined whether educational experiences collected in 1980 while they were in high school (including school contexts, high school academic performance, and educational attainment) moderate this relationship. We also investigated whether these associations differed by early life SES, sex/gender, race and ethnicity, and US region. High School and Beyond (HS&B:80) assessed a nationally representative cohort of people who were in high school in 1980, and 41 years later, collected telephone- and web-based measures of word list learning and memory, paired associates learning, fluency, and number span at age ∼60 years. A measure of global cognitive functioning was calculated using IRT. Amyloid beta 42:40 ratio, pTau-181, NfL, and GFAP levels were assayed from plasma. Regression models examined linear and nonlinear associations between cognitive measures and plasma biomarkers. Participants ( n  = 4,340) ranged in age from 56 to 63 (mean = 58), were 55% women, 60% were White, 21% were Latinx, and 14% were Black. Better global cognition was associated with lower plasma p -tau-181 (β = -.003, 95% CI -0.110 to -0.015) and NfL (β = -.004, 95% CI -0.006 to -0.002), and with higher AB 40/42 ratio (β = 1.388, 95% 0.207 to 2.568). Higher GFAP was associated with better cognition (β = .001, 95% CI 0.000 to 0.001). Among the individual measures, paired associate learning was most consistently associated with plasma biomarkers. There was no effect modification by educational experience, early life SES, race and ethnicity, sex/gender, or region of the US. At age 60 in a population representative sample, blood based biomarkers account for almost none of the observed variation in cognitive function, and these relationships do not vary by early life SES, race and ethnicity, sex/gender, US region, or educational experience.

INTRODUCTION:Educational attainment is associated with midlife cognitive functioning. However, degree attainment is the culmination of complex and unequal processes involving students' backgrounds, the opportunities that schools provide them, and their performance within those schools─all of which may also shape midlife cognition. What do educational gradients in midlife cognition look like using a richer conceptualization and measures of "education?" METHODS:We use data from High School and Beyond (HS&B:80)─a large, nationally representative sample of Americans followed from high school through age ∼60─to assess the role of education in stratifying midlife cognition. RESULTS:High schools' academic and socioeconomic environments predict midlife cognition primarily through their associations with their students' academic performance. Student academic performance strongly predicts midlife cognition, partially through its association with degree attainment. DISCUSSION:Inequalities in educational opportunities and in students' performance in schools shape midlife cognition─even among students with the same attained degrees. HIGHLIGHTS:Degree attainment predicts midlife cognitive functioning, but a large portion of that association is accounted for by students' high school academic performance as measured by test scores, grades, and course completion. High school contexts and learning opportunities predict midlife cognition mainly because they play a role in shaping students' academic performance. Understanding the potential benefits of education for later-life cognitive functioning requires attention to broader schooling processes and to students' academic performance beyond degree attainment.

How are children’s fluoride exposures associated with cognitive test performance in adolescence and midlife? Whereas most prior research has estimated effects of exposure to extremely high levels of fluoride, we consider exposure to levels of fluoride within the range typical in most places and of greatest relevance to policy debates about government water fluoridation. We use data from the nationally representative (United States) High School and Beyond cohort, characterize fluoride exposure from drinking water across adolescence, adjust for confounders, and observe cognitive test performance in both secondary school and at age ~60. We find that children exposed to recommended levels of fluoride in drinking water exhibit modestly better cognition in secondary school, an advantage that is smaller and no longer statistically significant at age ~60.

BACKGROUND:Educational gradients in health conditions are profound, but our understanding of how to address them is limited by the data available to us. In large-scale population-based research, we most frequently evaluate the relationship between educational attainment (or years of schooling) and either (a) self-reported incidence of adverse health events or diagnoses of health conditions or (b) biomarkers of those diseases. Self-reports are limited, however, by inequities and access to care; people with limited access to care may not know about health problems or conditions (at least until they reach acute levels). This means that inequities in (for example) self-reported hypertension conflate actual inequities in disease with inequities in awareness of that disease. Relying solely on biomarkers presents a different challenge for understanding health inequities ─ mainly because biomarkers often reflect both disease incidence and disease treatment. For example, a person with normal A1C levels may have never had diabetes or they may have been effectively treated for diabetes. Until now, we have not had ─ for large, population-representative cohort studies ─ all three of (a) people's self-reports of disease; (b) biomarkers of disease; and (c) pharmaceutical records that indicate disease treatment. How do inferences about educational inequities in disease ─ specifically hypertension and diabetes ─ change when we have full information about all three? METHOD:To address this question, we use new data from the U.S. High School & Beyond (HS&B:80) cohort study, which has followed a nationally representative sample of ∼25,500 people from high school in 1980 through midlife in 2021-22. HS&B:80 data from the 1980s includes rich information about educational contexts, opportunities, and outcomes; early life socioeconomic and family circumstances; spatial location; and demographic group memberships. Data from 2021-22 include all three of (a) self-reports of hypertension and diabetes; (b) blood pressure and A1C levels; and (c) pharmacy records on treatments for both conditions. The results of this study will inform our understanding of health equity, and either reinforce our confidence in the reliability and validity of self-reported health conditions or qualify how we interpret them. RESULT:Nothing yet. CONCLUSION:Nothing yet.

Structural racism shapes educational quality by perpetuating unequal access to school resources, creating disproportionate exposure to disciplinary actions, and limiting access to advanced courses among racially minoritized children. School quality is linked to later life ADRD risk, but the benefit may vary across groups. We asked 1) whether the same high school social contexts and academic resources predict midlife cognitive functioning across race, ethnicity, and gender, and 2) how much could disparities in midlife cognitive function be narrowed if everyone had equal access to high quality high schools? Data was from the nationally representative High School and Beyond cohort, which prospectively followed 12,530 Americans from high school through age ∼60 and administered telephone- and web-based measures of memory, language, and attention in 2021-22. Extensive information about family, high school social and academic context, and student achievements was gathered directly from students and their schools in 1980-1982. OLS regression models estimated associations between specific school variables and an IRT-derived cognitive composite, before and after adjustment for confounders in race, ethnicity, and sex stratified models. Higher average SES of students in each high school predicted better midlife cognitive test scores among White people and Latina women, but not among Latino men or Black participants. This association weakened after adding school academic context variables. A composite measure capturing school-level academic engagement and rigor (academic press) predicted better cognition among White and Latinx participants but not among Black participants. If everyone had access to high schools of the same quality as the group with the highest test scores (White women), and adjusting for family confounders, disparities between White women and participants from racially minoritized groups would be substantially reduced (by 9% among Black women, 5% among Black men, and 6% among Latino men and Latina women). Investment in high schools that serve Black and Latinx children could meaningfully narrow disparities in cognition four decades later, particularly among Black women. Later life cognitive benefit of specific aspects of school context, such as high SES peers and academic rigor, had differential benefit depending on race, ethnicity, and sex.

OBJECTIVES:The 1940 Census is a valuable resource for understanding various aspects of historical populations in the United States. Recently, the National Social Life, Health and Aging Project integrated 1940 Census data into its extensive data set, providing researchers with an opportunity to explore new avenues of life course investigation. We leverage the newly introduced measures of childhood residential environment and evaluate their potential predictive utility in older adult cognitive functioning net of childhood and adulthood characteristics known to be key risk factors for poor cognition. METHODS:We analyzed 777 respondents who were children in 1940 (age <17) that have been linked to the 1940 U.S. Census. We used childhood geographic location, homeownership status, household composition, and parental nativity as predictors. Cognitive function was measured using the Montreal Cognitive Assessment. RESULTS:Regression analysis showed that growing up in an urban area was associated with better cognitive function, while being born in the South was linked to poorer cognitive function, even after controlling for childhood health, parental education, educational attainment, stroke, and smoking status. Additionally, childhood multigenerational household was associated with better cognitive function, and childhood family size was associated with poorer cognitive function. However, these associations became statistically insignificant with the inclusion of educational attainment. We did not find homeownership and parental nativity to be associated with cognitive function. DISCUSSION:The findings may shed light on the potential long-term effects of childhood circumstances on cognitive aging processes. Implications for current literature and directions for future research are discussed.

Education is associated with cognitive functioning and risk of Alzheimer’s Disease and related dementias (ADRD). However, researchers rarely consider anything about education other than highest degree completion. Do schools’ social and academic contexts and students’ other (non-attainment) schooling outcomes independently predict cognitive outcomes? If so, this opens the possibility of manipulating educational policies and practices to improve long term cognitive well-being. This study included a nationally representative sample of 12,530 Americans followed prospectively from high school through age ∼60 as part of the High School and Beyond cohort study. OLS regression models estimated associations between various aspects of education ─ including schools’ social and academic contexts; students’ grades, test scores, and course taking; and degree attainment ─ and multiple dimensions of cognition, before and after adjustment for confounders. Educational attainment predicts cognitive functioning at age ∼60. However, the social and academic contexts of schools and other measures of students’ educational performance predict cognition at least as well—even after adjusting for attainment. Education is more than just how far in school people go, and degree attainment is not the only ─ or perhaps even the primary ─ thing about education that has long run consequences for cognitive functioning and ADRD risk. Advancing our understanding of how and why education shapes cognitive functioning and ADRD risk requires analyses of data with nuanced information about schools, schooling, and student performance.

The IPUMS Multigenerational Longitudinal Panel (MLP) is a longitudinal population panel that links American censuses, surveys, administrative sources, and vital records spanning the period from 1850 to the present. This article explains the rationale for IPUMS MLP, outlines the design of the infrastructure, and describes the linking methods used to construct the panel. We then detail our plans for expansion and improvement of MLP over the next five years, including the incorporation of additional data sources, the development of a "linkage hub" to connect MLP with other major record linkage efforts, and the refinement of our technology and dissemination efforts. We conclude by describing a few early examples of MLP-based research.