The Widening lncame If we do not find ways to reduce thegrowinginequalityin '1 education outcomes?betweentherich andthep'oor?e schools will no longer be the great equalizer we want them to be. Sean F. Real-don as the academic achievement gap between students from high-income and low?income families changed in the last few decades? And if so, why? Historically, low?income students as ?3 a group have performed less well than high~income students on most mea- sures of academic success?including (3 standardized test scores, grades, high I school completion rates, and college enrollment and completion rates. Countless studies have documented these disparities and investigated the many underlying reasons for them. But no research had systematically inves- tigated whether these income-related achievement gaps have narrowed or widened over time. To answer this question, I con? ducted a comprehensive study of the relationship between academic achievement and famiiy income in the United States over the last 50 years. I used data from 12 nationally represen- tative studies that included information on family income and student perfor- mance on a standardized test in math or reading. Because each of the tests measured reading and math skills on a different scale, I standardized all the test scores and expressed the income achievement gap in standard deviation units (Reardon, 2011ievmet Gap Striking Findings Findng 1: The income achievement gap has grown signi?cantly in the last three decades. Among children born in the 1950s 1960s, and early 19705, the reading achievement gap between those from high-income families (at the 90th percentile of the income distribution) and those from low-income families (at the 10th percentile) was about 0.9 of a standard deviation. As illustrated in Figure 1 this gap began to widen beginning with the cohorts born in the mid~19705. Among those born 20?25 years later, the gap in stan? dardized test scores was roughly 1.25 standard deviations?~40 percent larger than the gap several decades earlier] Although the trend in the income achievement gap is striking in its own right, it is even more. striking when compared with the concurrent trend in the black?white achievement gap (see fig. 1). The black-white achievement gap considerably larger than the income achievement gap among cohorts born in the 19505 and 1960s but now it is considerably smaller than the income achievement gap. This change is the result of both the substantial progress made in reducing racial inequality in the 19605 hot ll rennin hath. economic equality now exceeds racial equality in education outcomes. and 19705 and the sharp increase in economic inequality in education out- comes in more recent decades. indeed, Figure 1 encapsulates two important trends in US. history over the last 50 years. In the 19505 and 19605, racial inequality was high in virtually every domain of life?edu? cation, health, earnings, residential seg? regation?whereas economic inequality was lower than it had ever been in the last century (Piketty Sir Saez, 2003). By the early part of the 2lst century, racial inequality was much lower (although far from eliminated) in terms of wages, health disparities, and residential segre- gation. Meanwhile, economic inequality reached historic highs (Saez, 2012). Although both remain high, economic inequality now exceeds racial inequality in education outcomes. Finding 2: Income gaps in other measures of education success have grown as well. Academic achievement, as measured by standardized test scores, is not the only education outcome for which disparities between high~income and low-income students have been growing. The collegewcompletion rate among children from high-income families has grown sharply in the last few decades, whereas the completion rate for students from low?income families has barely moved (Bailey (Sr Dynarski, 2011). Moreover, high?income students make up an increasing share of the enrollment at the most selective colleges and universities (Reardon, Baker, &r Klasik, when compared with low-income stu? dents with similar test scores and aca~ demic records (Bailey Dynarski, 2011; Belley Lochner, 2007; Karen, 2002). A related trend during the last 20 years is the growing social-class gap in other important measures of ado- lescents? ?soft skills? and behaviors related to civic engagement, such as participating in extracurricular 12 EDUCATIONAL LEADERSHIP MAY 2013 1 Achievnn'mnt Gap and Auhinwrm?nt (3:03 in [landing for 19-13 2001 flirtll Cohorts 1.50 1.25 .00 .75 in standard aawarron umls .25 .00 1940 1950 Average Di?arenae in Standardized Scores" 190100 Incmre Gap or Black-White Gap) 1960 Income Gap Blaek~Whna Gap i 1 1970 1980 1990 2000 Cohan Birth Year I ., Source: Adapted 'rcm The Widening Samoamnomc Status Aehmvament Lian New Ewdenca ?y?lp?and Exptanahons? 1p. QBI by F. Heard-an, in Fl J. Murnane (3. Duncan iEds.]. 2? Whither Omar-runny? Rising Inequality, Schools, and Children '5 Life Chances. 2t}! 1, New York; . Russell Sage Foundation activities, sports, and academic clubs; volunteering and participating in com- munity life; and self-reports of social trust (Putnam, Frederick, Snellman, 2012). Finding 3: The income achievement gap is already large when children enter kindergarten, and it does not grow signi?cantly as they progress through school. One possible explanation for the wid- ening income achievement gap is that K-12 schools have grown more unequal in quality over the last few decades. If this were true, then the gap should grow larger the longer students are in school. But when I examined the data, I found little evidence that this occurs. In one study, the Early Childhood Longitudinal Study-Kindergarten Cohort (ECLS-K), roughly 25,000 students were tested in math and lit? eracy skills in kindergarten in 1998 and then were reassessed as many as six more times between 1998 and 2007', when the students were in 8th grade (Tourangeau, Nord, Le, Pollack, Br Atkins-Burnett, 2006). 1 used this study?s data to examine how the income achievement gap changed as this cohort of students progressed through ele- mentary and middle school. As Figure 2 15) shows, the gap in reading grew very little during this periodwit was 1.15 standard deviations when the children entered kindergarten and 125 standard deviations in 8th grade. Other longitudinal studies that assessed stu? dents multiple times dun?ngmiddle and high school show the same pattern: The achievement gap changes little during the years. The fact that the income achievement Income inequality has risen dramatically XII lla',? "irfr .il H: II il! {tin Lita.- w: -: l'l" ill' 4 .. i'f Ilii?Hi'Iiu? 111th la tin-tilt I gap is large when children enter kindergarten?and does not grow substantially during the school yearswsuggests that the primary cause of the gap is not unequal school quality, In fact, the data in Figure 2 show that schools may actually narrow academic achievement gaps, rather than widen them. The data show the gap narrowing between the fall and spring of the kindergarten and 1st grade years?periods when stu? dents were in school?and widening in the summer between kindergarten and grade?~when they were not in school. Although we can?t assume that the same pattern holds in later grades, the data do suggest that schools may reduce ?12? MARK WEBSTOCK inequality rather than widen it. This finding is consistent with other research on the ?summer setback? that has been con? ducted in smaller, more localized samples (for example, see Alexander, Entwisle, Olson, 2007'). Why Has the Income Achievement Gap Grown? To understand the reasons for the growing income achievement gap, it is necessary to look at the social history of the past 50 years in the United States. A few key trends are worth considering, First, income inequality has risen dramatically in the last 30?40 years, making the gap in income between high-income and low?income families much greater. In 1970, a family with schoolwage children at the 90th percentile of the family income distribution earned 5 times as much as a family at the 10th percentile; today, the high?income family earns 11 times more than the low?incorne family? This rapid growth in income inequality means that high-income families now have far more resources, relative to low~income families, to invest in their children?s development and schooling. Second, upward social mobility has become far more dif- ficult and far less certain than it was 50 years ago, partly because of rising income inequality and partly because of deciining economic growth. While the economy was growing rapidly in the 1950s and 19605, the vast majority of children in the United States (particularly white children) grew up in families in which they were much more economically secure than their parents (most of whom had grown up during the ASCD 13 STEPHEN MCSWEENWSHUWERSTOCN Great Depression and World War II) had been. But beginning in the 19705, economic growth slowed dramatically, and upward social mobility became far less certain. Third, the economy has become increasingly bifurcated into a low-skill, low?wage sector (for example, service jobs and routine production jobs) and a high?skill, high~wage information sector (for example, engineering and financial analysis). Largely gone are the manu~ factoring jobs that provided a middle- class wage without a college degree. As a result, education success has become increasingly essential to economic success (Autor, Katz, 5: Kearney, 2008; Murnane, Willett, s: Levy, 1995). Fourth, popular notions of what con? stitutes education success have changed. In the last few decades, test scores have become increasingly central to our idea of what schools are supposed to produce. As test scores have played a more dominant role in education policy over the last decade (and have become more important in college admissions), they have become increasingly salient to parents concerned with their children?s education success. Fifth, American families have changed in several important ways in the last four decades. Children in higheincome families are increasingly likely to be 14 Mar 20$} High?income families have far more FESOUFCQS, ivl il l: l-ll?llill? to invest in their children's development in til at ll]. :9 FM lAH/l VOLC-IJI HERE: I ULT- raised by two parents, both with college degrees, whereas low-income children are more likely than ever to be raised by a single mother with a low level of education (McLanahan, 2004; Schwartz (Sr Mare, 2005). This means that family income has become increasingly corre? lated to other family characteristics and resources that are important for chil- dren?s deveIOpment. The combination of these broad social trends has had important conse? quences for children?s academic success. increased uncertainty about children?s likelihood of upward social mobility, coupled with the increased importance of education for career security, has made parents increasingly anxious about their children?s education. This has led to greater competition among families for their children?s academic success. In summary, the growth in income inequality and in the correlation of income with other family resources means that family resources have become increasingly unequalat the same time that families are increasingly focused on their children?s education, a constellation of trends that has led to a rapidly growing disparity in the extent to which families invest their time and money in their children?s education. Indeed, high-income families now spend nearly 7' times as much on their children?s development as low~income families, up from a ratio of 4 times as much in 1972 (,Komrich $1 Furst? enberg, 2013). What Role Can Schools Play? US. schools have historically been thought of as the great equalizer?the social institution best suited to ensure that all children have an equal oppor~ [unity to learn, develop, and thrive. it is unrealistic, however, to think that school-based strategies alone will eliminate today?s stark disparities in Educational success has become increasingly essential t? i n: i! if"; Lift" i; academic success. Economic policies that reduce inequality; family support policies that ensure children grow up in stable, secure homes and neighbor- hoods; and early-childhood education policies that promote cognitive and social development should all be part of a comprehensive strategy to close the economic achievement gap. Nonetheless, schools do have a key role to play in the efforts to reduce this gap. Among the school-based strategies that might be most effective, I suggest three specific areas. First, states and school districts could devote a greater share of their resources and efforts to the earliest grades, including kindergarten and preschool. Because achievement gaps are self perpetuating, the earlier we intervene to reduce them, the more effective we will be at eliminating them in the long run. Second, growing evidence suggests that more time in school (for example, extending the school day or year or pro- viding after-school or summer?school programs) may help to narrow academic achievement gaps?if the added time is used effectively (Dobbie or Fryer, 2011; National Center on Time and Learning, 2012). Although the evidence is far from conclusive at this point, it appears to be a strategy worth pursuing. Third, states and school districts can FIGURE 2 Kil'ltliritinrlen 81h {Halli} 1.50 a is? . 33 .33 125 spring 63 59129 GE a: ?9 E, FallG1/ 5;:an G5 .3 1.00 1 n. 3 Springlt Spring student Snore. Adapted from ?The Widening SocIoeconomic Status Acmeveman: Gap. New and Possible Explanazions? Io. 1001b?, 5. Heavdon, Ft. Murnane E: .1. Duncan {?th 1. Whither Opportunm-?P Rismg Inequality: Schools. and L'fe Chances. 2m 1. New York..- Russe't Sage Foundation. do more to ensure that all students have equal access to high-quality teachers, stimulating curriculum and instruction, and adequate school resources (com? puters, libraries, and the like). The United States has grown more residen- tially segregated by income over the East four decades (Reardon (SI Bischoff, 2011), meaning that schools have, in many places, become increasingly segregated by income as well. School districts can work against this growing segregation by developing student assignment systems that promote socio- economic diversity within schools. Important Consequences The widening income achievement gap is a of a confluence of trends that have accompanied and exacerbated widening income inequality in the United States over the last four decades. But it is a with real and important consequences. If we do not find ways to reduce the growing inequality in education out- comes, we are in danger of bequeathing our children a society in which the American Dream?the promise that one can rise, through education and hard work, to any position in society-His no longer a reality. Our schools cannot be expected to solve this problem on their own, but they must be part of the solution. 1Analysis of data from the National Assessment of Educational Progress (Rampey, Dion, (Sr Donahue, 2009) and from the 12 studies suggests that the income achievement gap is not widening ASCD 15 because 01 declines in low-income students? performance In fact, average test scores of both low-income and middle-income students have risen substantially in math and very modestly in reading. But they have been outpaced by high?income students, whose scores have risen even faster. 2Author?s calculations, based on Current Population Survey data (King et al., 2010). References Alexander, K. L., Entwisle, D. R, (St Olson, L. S. (2007, April). Lasting consequences of the summer learning gap. American Sociological Review, 72(4), 167?180. Autor, D. H, Katz, L. H., (St Kearney, M. S. (2008). Trends in US. wage inequality: Revising the revisionists. Review of Eco? nomics and Statistics, 90(2), 300?323. Bailey, M. 1, (SI Dynarski, S. M. (2011). Gains and gaps: A historical perspective on inequality in college entry and com- pletion. In G. Duncan 8: R. Mumane (Eds), Whither opportunity? Rising inequality, schools, and children?s life chances (pp. 117?132). New York: Russell Sage Foundation. Belley, P., (St Lochner. L. (2007). The changing role offarnily income and ability in determining educational achievement (Working Paper 13527). Washington, DC: National Bureau of Economic Research. Dobbie, W., 8: Fryer, R. G. (2011). Getting beneath the veil of effective schools: Evidence from New Yorh City (Working paper 17632). Washington, DC: National Bureau of Economic Research. Karen, D. (2002). Changes in access to higher education in the United States: 1980?1992. Sociology of Education, 75(3), 191?2 10. King, M, Ruggles, S., Alexander, 1., Flood, 5., Genadek, K, Schroeder, M, et a1. (2010). Integrated public use microdata series, Current Population Survey: Version 3.0. [Machine-readable database]. Minneapolis: University of Minnesota. Kornrich, S., (St Furstenberg, F. (2013). Investing in children: Changes in parental spending on children, 1972 to 2007. Demography, 50(1), 1?23. McLanahan, S. (2004). Diverging destinies: How children are faring under the second demographic transition. Demography, 41(4), 607?627. Murnane,R.j.,Willett,j.8,81: Levy. F. (1995). The growing importance of cognitive skills in wage determination. 16 Mar 2013 Upward social mobility has become far more dif?cult .nn'l llir cirr?tair'i ll. itp'). Review of Economics and Statistics, 78(2), 2 51?266. National Center on Time and Learning. (2012). Time well spent: Eight powerful practices of expanded~tirne schools. Boston: Author. Piketty, 1., 8r Saez, E. (2003). income inequality in the United States, 1913-? 1998. Quarterly Journal of Economics, 118(1), 1?39. Putnam, R. 13., Frederick, C. B., (St Snellman, K. (2012). Growing class gaps in social connectedness among American youth. Cambridge, MA: Harvard Kennedy School of Government. Retrieved from _2orzoaoa.pdf Rampey, B. D, Dion, G. 5., 8r Donahue, P. L. (2009). NAEP 2008 trends in I . TRENDS of the TIMES What is the poverty I I thruhold in the Unittd I Status for a family of four? . $13 one I a 523,000 I . . 533,000 Source {for current response}: U3. Camus I Bureau. Hattimd from censusgov I nihemmwartwdatahhreshid?ndax academic progress. Washington, DC: National Center for Education Statistics, US. Department of Education. Reardon, S. F. (2011). The widening aca? demic achievement gap between rich and poor: New evidence and possible expianations. In G. J. Duncan R. j. Murnane (Eds), Whither opportunity? Rising inequality, schools, and children?s life chances (pp. 91?115). New York: Russell Sage Foundation. Reardon, S. E, Baker, R., 69: Klasik, D. (2012). Race, income, and enrollment pat- terns in highly selective colleges, 1982?2004. Stanford, CA: Center for Education Policy Analysis, Stanford University. Reardon, S. F, 81? Bischoff, K. (2011). Growth in the residential segregation of families by income, 1970?2009 (Project US 2010 Census Brief). Washington, DC: National Bureau of Economic Research. Retrieved from Saez, E. (2012). Striking it richer: The evo? lution of top incomes in the United States (updated with 2009 and 2010 estimates). Berkeley: Department of Economics, Uni? versity of California, Berkeley. Retrieved from Schwartz, C. 11,6: Mare, R. D. (2005). Trends in educational assortative marriage from 1940?2003. Demography, 42(4), 621?646. Tourangeau, K, Nord, C, Le, T., Poliack, j. M, (SI Atkins?Burnett, S. (2006). Early childhood longitudinal study, hinde rgarten class of 1998?99 combined user?s manual for the ECLS?K?_fth?grade data ?les and electronic codeboohs (NCES 2006?032). Washington, DC: National Center for Education Statistics, US. Department of Education. Sean F. Reardon (sean.reardon@ stanfordedu) is a professor in the Stanford University Graduate School of Education, California.