David Brooks: In Praise of Segregation

It's not often that NYTimes conservative columnist, David Brooks, finds any way to use science try to justify the lifestyles of the rich and separate, but someone sent him news of a study that he obviously felt he could manipulate to rationalize the continuing resegregation of America and its schools.

Good fences make good neighbors. When ethnic groups clash, we usually try to encourage peace by integrating them. Let them get to know one another or perform a joint activity. This may be the wrong approach. Alex Rutherford, Dion Harmon and others studied ethnically diverse areas and came to a different conclusion. Peace is not the result of integrated coexistence. It is the result of well-defined geographic and political boundaries. For example, Switzerland is an ethnically diverse place, but mountains and lakes clearly define each group’s spot. Even in the former Yugoslavia, amid widespread ethnic violence, peace prevailed where there were clear boundaries.

Too bad Brooks did not read to the bottom of the abstract, where he would would have found this:

The theory and the data also show that people who are in fully integrated societies will also successfully live in peace. 

What most people know, even those who do not read the results of scientific studies, is that diversity is important for healthy communities, whether business communities concerned with productivity or artistic groups in search of new forms, problem solving groups in search of solutions, or school systems in search of higher achievement for all children.  

A clip from TMoE (pp.  37-41) on Coleman's findings almost a half century ago:

The mandate for the Coleman report came from the Civil Rights Act of 1964, which required a research study be con- ducted within two years of passage to identify where educational resources in public educational institutions were lacking due to “race, color, religion, or national origin”  (Lagemann, 2000, p. 193). Almost overnight, previous education program evaluation criteria  based on resource inputs  shifted  to program outputs as the mandate for tracking  education program effective- ness was written into federal  legislation.

While everyone,  including James Coleman, expected to find large dif- ferences  in achievement based  on  large  differences in resources between the  600,000  children that  his  study  included in  4,000  Black and  White schools,  the  findings  confounded expectations. As Coleman scholar,  Ger- ald Grant  (2009)  points  out Coleman found discrepancies in spending be- tween Black and White schools to be less than  expected, due to infusions of cash by Southern states in hopes  of maintaining the  “separate but  equal” apartheid systems. But, even where  resource differences were large,  Cole- man  found these  disparities  in Black schools  influenced student achieve- ment  differences much  less than  “who you went to school with”:

Simply put,  Coleman found that  the  achievement of both  poor  and  rich children was depressed by attending a school  where  most  children came from low-income families. More important to the goal of achieving equal ed- ucational opportunity, he found that the achievement of poor  children was raised by attending a predominantly middle-class school, while the achieve- ment  of affluent  children in the school was not harmed. This was true even if per-pupil expenditures were the  same at both  schools. No research over the past 40 years has overturned Coleman’s  finding. . . . (p. 159)

Coleman and his team (1966)  found, too, that non-Asian minority  chil- dren are more  affected  by social composition than  are White children, and that  “if a minority  pupil  from  a home  without  much  educational strength is put  with schoolmates with strong  educational backgrounds, his achieve- ment  is likely to increase” (p. 22). Though this finding  is commonly  cited in  analyses and  interpretations of the  Coleman study, the  dynamics  that shape  this  social fact are  most  often  attributed to  the  social capital  that accrues  for various reasons  when  poor  children go to school  with middle class children. Coleman, however, clearly introduces a race element beyond socioeconomic status  that  is related to the  effects  of oppression and  de- moralization that  is rarely cited. Therefore, we include this rather lengthy quote  below, which if attended to by policy makers, would doubtless create an added urgency  to dust off long-neglected integration plans. Notice that Coleman remains loyal to and supportive of the charge  given to him under Section  402 of Civil Rights Act of 1964 to provide  data related to “the  lack of equal  educational opportunity for individuals  by reason  of race,  color, religion, or  national origin . . . ,” even  though his investigations  have  led him to findings that even Coleman could not have predicted:

This analysis has concentrated on the educational opportunities offered  by the schools in terms of the student body composition, facilities, curriculums, and teachers. This emphasis,  while entirely appropriate as a response to the legislation  for the survey, nevertheless neglects important factors in the variability between individual pupils within the same school: this variability is roughly  four times as large as the variability between  schools. For example, a pupil attitude factor, which appears to have a stronger relationship to achievement than do all the “school” factors together, is the extent to which an individual feels that he has some control over his own destiny [italics added]. . . . The responses of pupils, except  for Orientals, have far less conviction  than  Whites that they can affect their  own environments and  futures.  When  they do,  however,  their achievement is higher than  that of Whites who lack that conviction.

Furthermore, while  this  characteristic shows  little  relationship to  most school factors, it is related, for Negroes,  to the proportion of Whites in the schools. Those Negroes  in schools with a higher proportion of Whites have a greater sense of control. This finding  suggests that  the  direction such an attitude takes may be associated with the pupil’s school experience as well as the experience in the larger community. (p. 23)

Coleman found hope, then,  strongly correlated with the presence of a sense of autonomy, which is more  easily demonstrated, measured, and  retained where racial and economic mixing prevails, rather than  in racially and eco- nomically segregated environments—whether that segregation is sustained by antiquated beliefs, legal maneuvering, or by outdated school assessment practices.  And it was this “pupil  attitude factor” of hope  that had a greater effect on achievement than  all other school  effects examined in the  Coleman  study, which remains the  largest  research undertaking of its kind  in U.S. educational history.

The  Coleman findings  on  socioeconomic status  and  school  achievement  echoed the  findings  of another large, longitudinal study a few years earlier,  whose similar results on the  topic  were similarly ignored (for  similar reasons,  we may assume).  The federal  research project in 1960, Project Talent,  involved detailed questionnaires in over 1,300 high  schools  and  a series of tests for 440,000 students that  included achievement, attitudinal, interest, and  aptitude tests, surveys, and  questionnaires. Instruments were administered in 1960 when students were ninth graders  and again in 1963 when they were seniors.  By 1973, it is clear that  Washington’s elite had  digested the implications of these studies, as expressed here  by fiscal and monetary expert, Alice Rivlin (1973):

The most general result of these  statistical studies [the  Coleman Study and the  Project  Talent  study]  has been  the  finding  that  variables reflecting the socioeconomic characteristics of students and their families explain  most of the variation  in test scores, and variables reflecting school characteristics or resource inputs  explain  very little.

These  results should  not  be exaggerated—they do not  prove that  “schools don’t  matter”—but they certainly  provide  a basis for considerable skepti- cism about  using  test  scores  as measures  of the  output of the  education industry  as such.  Test score  changes  may primarily  reflect  changes  in the school  population and  the  way it is mixed,  rather than  the  productivity  of school resources themselves. (p. 424)

Lagemann (2000)  recounts the  drama  surrounding the  release  of the Coleman Report’s  initial findings  in 1966, and  the subsequent “firestorm” set off within the Johnson Administration, which knew that Coleman’s findings could sabotage  the Administration’s strategy of using the federal  purse to buy Southern support to end  apartheid schooling in the  South,  as set forth  by ESEA the  year before.  Johnson knew  that  Republicans, already resistant  to more  federal  spending, would  seize and  exploit  the  counter- intuitive  fact that  spending levels were clearly not  the  prime  factor  in performance discrepancies. Coleman’s  findings,  too, offered  a swipe at a core component of the American  secular faith in education and educational op- portunity as “the chief instrument for redressing the inequalities of American life” (Kantor, 1991, p. 50).

This lofty notion had,  indeed, fed the  Jeffersonian belief,  later  transferred to Horace Mann,  that education may provide  solutions  to social problems that  were thought to be the  result  of the  poor’s  own shortcom- ings. Blaming  the  poor  for their  poverty is as traditional as our  Calvinist forefathers of Puritan New England, who  viewed the  socioeconomically unfit as having earned their  lack of status through their  own moral failings (Rippa,  1996). These shortcomings, in turn,  might signal the column of the celestial tally sheet  to which all souls had  been  added who were not a part of the Elect, or God’s elite. From this early theological base, there eventu- ally grew the Protestant economic catechism of the Gilded Age, with ample doses  thrown  in of Social Darwinism,  which  “held  that  responsibility for poverty lay not with the business cycle or the existence  of a capitalist reserve army of the unemployed, but with the moral failure of the poor  themselves to conduct proper family economy”  (Dawley, 1993, p. 27).

By the 1960s the poor’s personal flaws and the lifestyles they spawned were bundled within a new and encompassing concept known as the “culture of poverty,” which acknowledged structural barriers as well as the traditional blaming  of the victim:7

First, though it acknowledged the structural sources of deprivation, the culture of poverty thesis  tended to focus attention more  on  the  personal characteristics of the poor themselves than  on the economic and social con- ditions  that  shaped their  lives (Aaron, 1978, p. 20). Consequently, and  this is the  second  point,  because  it implied that  people were poor  due  to their own attitudes, behaviors,  and life-styles, it suggested  that changing the poor rather than  redistributing income  or creating jobs was the best way to elimi- nate the problem of poverty. (p. 55)

The  third  characteristic that  Kantor  (1991)  attributes to the  “culture- of-poverty thesis” was its belief that,  since the economic and  psychological conditions left the poor  without the “will and capacity to attack the sources of their  own deprivation” (p. 55), professional intervention was required, which  assured  a powerful  role  for  the  liberal  public  policy  makers  during  the  1960s. Such  interventions, however,  did  not  disrupt the  underlying assumptions of economic order, systems of privilege, or existing power relations, as initiatives  to help  the  poor  focused  more  on  education and training programs. As noted earlier,  these kinds of compensatory solutions could  be provided without  disrupting the  social and  economic structures that would have been  challenged by job creation programs or other alterations to economic and socio-cultural patterns. The preferred compensatory strategies  adopted by liberals could “compensate for capitalism’s inevitable flaws and omissions without  interfering with its internal workings” (Kantor quoting Brinkley, 1991, p. 56).

Coleman’s  findings,  however, were not governed by any of these assumptions. His findings  clearly suggest the  need  for structural alterations to  the  racial  and  socioeconomic organization  of  schools,  while  clearly pointing to the limited  value of simply adding  resources without  structural modifications. The initial findings  of the Coleman Report,  therefore, were appropriately muted by Johnson’s  White House;  the  media,  with no open controversy  to sell copy and  with its accepted narrative wisdom to protect, largely ignored the  complete findings  when  they did  appear late in 1966 (Coleman et al, 1966).  Both  liberal  and  conservative  policy people, then, read  the  Coleman Report  looking  for ideological ammunition, and  they found it. Conservatives centrally  concerned with cutting costs and conserv- ing the  status quo  cherry-picked Coleman’s  findings  (Alexander, 1997)  to argue  that  “throwing money”8  at educational problems couldn’t fix them, while liberals used Coleman’s  findings  related to social capital  and  the importance of racial and economic mixing to argue for mandatory busing policies to achieve racial balance.  Coleman remained disappointed (Coleman,1972) at the reception of the study, and he remained throughout his life an advocate for removing all barriers to socioeconomic integration, even as an interim measure toward achieving equity and equality (Kahlenberg, 2001). Kahlenberg (2001)  cites Coleman from  a rare  interview in 1972, in which his claim for the significance  of social capital is made unequivocal:

Coleman said that research continued to show that “a child’s performance, especially  a working-class child’s  performance, is greatly  benefited by his going to a school with children who come from educationally stronger back- grounds.” Coleman declared flatly: “A child’s learning is a function more  of the characteristics of his classmates than  those of the teacher.” (p. 62)

How different today’s education reform  agenda might  be if Coleman’s core finding had been acknowledged and taken to heart for its central truth: schools alone  can never consistently  close the gaps in achievement that reflect deep  differences in levels of autonomy and  privilege, wide disparities in opportunity, deep  veins of racism, and an ongoing and deepening hope gap. How different our  schools might  be if we were to take seriously what good research already tells us, or if we as a society were to fund other social science research with the potential to matter in the health of our neighbor- hoods  and  our  world. Or, how differently  our  schools and  our  perception of schools might  be if we were to conceive  of educational improvement as one  important component of a comprehensive commitment to social and economic renewal,  in a way that  acknowledges the  wisdom expressed by Jean  Anyon’s (1997)  quip:  “Attempting to  fix inner city schools  without fixing the city in which they are embedded is like trying to clean the air on one side of a screen  door”  (p. 168).