-
.4
ED T17 252
AUTHORTITLEINSTITUTIONREPORT NOPUB DATENOTEAVAILABLE PROM
DOCUMENT RESUME
Coleman, James S.; And Othe sTrends in School Segregati ,Urban
Inst.,,Wastington D C.URI-722-03-01ling' 75
150p.Publications Office, The UrbanStreet, N.W.,
Washington,.D.t.
H. UD 015 653
1968-73.
Institute, 2100 M20037 ($3,50)
DIMS PRICE ME-$0.16 Ht-$6.97 Plus/Postaget
Educational Policy; ElementaryCivil Rights; *Compailtive bata
Analysis;
mentary Gtades; MetropolitanDESCRIPTORS
Areas; *National Surveys; Racial Segregation;, SchoolDistricts;
*School,Igtegration; *School Segregation.;Secondary Grades; So ial
Discrimination'; StatisticalData; *Trend Analys s
Thisipaper repoi*s work in-progress concerningstudent
desegregation among elementary and secon4Fy,schools indistricts
regardless of the source of segregation, and between
school.districts.for the period of 1968-73. The data sources,
the.statistical reports collectedbby DREW, are stated to allbw.for
adetailed statistical analysis of the status and trends in
schoolsegregation by race throughout the U.S.' First; the state of
racialintegration amongschdois within a district in 1968 is
examined,'followed by an examination of the changes that occurred
over theperiod 1968-1973, The differential changes that occurred
over thatperiod of time in different kinds of school settings--in
differentregions of the country, in school districts of different
sizes, andin particular large cities -s -is seen, to be of special
interest. Dataindicate.that, by 1968, desegregation of schools was
a far fromaccomplished task in cities and towns of all sizes in the
South, butthat in the largest cities, it was.equally *high in
Many'places wheredual schOol systems had never existed: yet the
trend for the next
ABSTRACT
V
..
4,,
four years tends toward desegregation. It is concluded (1) that
theemerging problem With regard to school desegregation is the
pioblemof segregation, between central city and, suburbs, and (2)
that currentmeans by which school's are,beinTdesegregated are
intensifying,_rather than reducing the-problem. (Author/AM)
*#************14****,*******************************3;******************-
* ' Ddcuments acquired byERIC:include many informal unpublished
*
* materials not available from other sources. ERIC makes
every,efforA ** to obtain the best copy available. Nevertheless,
items of marginal *!!! reproducibility ate often enconntered and
this affects the quality * t*,pf the microfiche and hardcoly
reproductions ERIC makes available *,* Via the ERIC Document
Reproduction Service pus), BpBs is not ***responsible-far the
quality,of the original document. Reproductions ** supplies by EDRS
are the best that can be made from the original.
**************************************************Ic*******************
-
ri;
Trends in School.Segregation, 1966-73 ,
James S. ColemanSara D. KellyJohn A. Moore
722-03-01.August,1975.
e
b s DEPARTMENT OF HEALTH,EDUCATION A WELFARENATIONAL INSTITUTE
9F
EDUCATIONTHIS DOCUMENT HAS SEEN REPRODUCED EXACTLY AS RECEIVED
FROMTHE PERSON OR ORGANIZATION ORIGINATING IT POINTSOF VIEW DR
OPINIONSSTATED DO NOT NECESSARILY REPRESENT OFFICIAL NATIONAL
INSTITUTE OFEDUCATION POSITION OR POLICY
9
THE URBAN INSTITUTE'WASHINGTON, D.C.
I
.r
I.
Se
4..4
-
BOARD OF TRUSEES
William W. Scranton, ChairmanJean FairfaxEugene G. FubiniWiiliam
GorhamKatharine GrahamRobert V.flansbergerVernon E. Jordan).
Jr.Richard Llewelyn-DaviesBayless A. ManningStanley Marcus
Robert S..McNamara
Arjay MillePJ. Irwin MillerJohn D. Rockefeller, IVWilliam D.
RuckelshausHerbert ScarfCharles L. Sch.ultie, Vice-ChairmanFranklin
A. ThomasCyrus R. VanceJames Vorenberg
INSTITUTE OFFICERS
William Gorham, PresidentRobert Harris, Senior
Vice-PresidentHarold W Guthrie, Vice-Preiident for ResearchHerman
T. Boland, Jr., Vice-President for AdministrationEdward E Wallace,
Controller
The Urban Institute is a nonprofit research corporation
established in 1%8 tostudy problems of the nation's urban
communities. independent and non-pdrtisan, the institute responds
to current needs for disinterested analysesand basic information_
and attempts to facilitate the of this luxisvl-ette.14_ part of
this iOnet ttt;t1;444: ;;lg- Qratrrii*;ltittik Met.
(*kit Write ii:10600ornitliiiiitifiltiathe general public.,-;\
The Institute's research findings and a broad range of interpretive
view-ikAnti are, published as an educational service. Conclusions
expressed inInstitute publications are those of the authors and do
not necessarily reflectthe views of other staff members, officers
or trustees of the Institute, or oforganizations which provide
funds toward support of-Institute studies.
These research findings are made available in three series of
publica-tions: Books and Reports, Papers, and Reprints. A current
publications list isavailable on request. 4
.`,..1.11bIlcations *Office)
The Urban Institute2100 M Street, N.W.Washington, D.C. 20037
4.
7.
-
V 01,
Opinions expressed are those of die.authors and do not
necessarilyreflect the views of The Uiban Institute or its
sponsors.
ISBN 87766 - 147 -02
UI 72'2-0401
PLEASE REFER TO URI-12300 WHEN ORDERING.
Available froi:
Publications OfficeThe Urban Institute2100 M Street, N.
W.Washington, D. C. 20037.
List price: $3.5D
A/75/2000
A,
,
-
,INSERT FOR
TRENDS Ift.A110N., SEGREGATION, 1968-73
F - 0* James $. ColemanSara D. :Kelly
' John A. Moore
Urban Institute Paper 722-Q3-0
ti
4
.
40
-
'8
p
ERRATA AND MODIFIED-CALCULATIONS
After this report was printed, we pliscovered some errorsJand
some places
where procedures used may have introduced a bias. We have
recalculated, results
that might have been affected by. these, correcting errors and
using procedures
which should eliminate the bias. The resulting changes affect
some nupbers in
the text and tables, but with one exception, noted below, do not
affect_the
general conclusions ofthe report. The modifications arise from
sourad as
indicated below, with effects .as noted.
1. In the analysis of segregation at elementary and
secos4iary.164fm1s, it
was originally decided, as explained in footnote 12, page 21, to
measurksegrega-7
tion at elementary levels by,including all schools containing a
sixth Ogde and
to measure segregation at secondary levels_ by including all
schools Containing
a tenth grade. This was done for reisons descri6ed in foOtnote
12, page, 21.
This, as i.45,qrned out, was not a wise decision. A number of
school
systems, pafticularly those undergoing desegregation of some
sorts reorldnizea
their schools to cover different grade spans, or in some cases,
to become un-
graded where 'the school had before.been a graded school. This
reorganization
meant that for elementary/secondary analysis, different
fractions of the
1students in the system and different fractions of the schools
in4the system
were included in different years, in general a smaller fraction
:of the'odhools
in later years because of the conversion from graded to ungraded
schoAls, and.
t'. ,the ieduction in grade spans. Reorganization was primarily
at the el entary
level. For example, in 1968 the OCR -data show that qallas
had.125 shools
containing a sixth grade, 5 graded ejementerry schools ending
below grade six,
q o ,,and 2 ungraded schools. In1970 this was 123 with a sixth
grade; 13 :graded ,
...i,.J'!"6
-
2
ending below grade six 'and 1 ungraded. In 1972, the numbers
were 109, 29,t
and 9. Thus in the later years, a.larger number of elementary
schools was
excluded for being ungraded or failing to contain grade six.
In modifying the analysis, it was evident that because of school
reorgan-..
ization, no gpcedure for including a school in the elementary
school sample
or inthe..secondary school saae would be fully satisfactory. The
procedure
.used for t)le modified calculitions contained here was to
classify any school AS
.
a secondary school if it included 4grade 9, 10, 11, or 12. .X11
Olver schools
were claasified as elementary. Inspection of a number of systems
which,I'
r.,J
undergone come reorganization indicated that this procedure
would give fewest
misclassifications that might result in bias and that the amount
of mfsclassi-
.0fication was very small. .
The effects of the revised procedure are given below:
la: On page 22, the numbers in Table 5 change as indicated in
the
attached revised Tattle 5. The -revisions change no
substantive
conclusions, and lead toidnly minor rewording on page 21
where
the table is described. 1
'lb: On pages 34, 35, and 36, Table 9, Figure 3 and Figure 4
change
as indicated in the attached. The substantive conclusions on
Nc.
Rage 34,about change at the two levfls are modified
slightly.
. r"In the U.S. as a whole, the reduction in segregation was
greater
at the secondary level, rather than equal to that in elementary,
°
as stated there, But as described- At the original text,
there
are opposite differences in diffeient regions, with greater
desegregation .occurring in the secondary schools than in
the,
elementary in thq South, buts grdater resegregation
occurring
in secondary schools io the North.1
-
WS
3
lc: On.page 47, Table 12 changes as shown in attached revised
Table 12.
The conclusions on page 48 are affected somewhat. It is no
longer
the case that in the 3 regions where within-district segregation
:
was most reduced, between-district segregation increased more
at
the elementary level. Thus the suggestion cannot be made
here
that the movement of white families from a district in
response
to desegregation appears to be greater at elementary level.
Id: On pages 77 and 78 are described the results of analyses
at
elementary and secondary levels. This analysis must be
deleted,
eliminating any conclusions about the relative.loss of whites
at
elementary and secondary levels. This is the one most
important
change affecting a general cbnclusion.of the report. That
is,.result
number 10 on page 79inust be deleted. A comparable analysis has
not
yet been carried out for elementary and secondary as redefined,,
sv.
the question' of relative loss raised in that section c annot
yetbe
answered. An attempt was made to get some idea abodthe answer to
.
this question by examining white.loases,in the year of
deiegregstion
in schools with a grade ten and those without in cities listed
on
IRKpages 62 as having undergone a 'drop of more than .1clin
segrega tion
A A .in .any one-year. This gave an average loss of 14% of
whites at
secondary school and 13% at elementqry (with 14% for-all
schools).
There *is greater lobe' in 5 of the 9 instances at secondary
school,
acid `air 4 of the 9. in elementary school. From this analysis
it would.1
app that the losses were about the same. However, this cannot
be
interred because in general, desegregation was greater at
tecondary. 0
aVals. In some cities, desegregation occurred primarily at tone
or
8
-
4
the other level (e-g., in Dallas in 1971, primarily at the
secondary
level Tend in San Francisco in 1971, at the elementary level).
Thus
the question must be left ;open.
2. Seieral modifications have been made affectihg the analysis
of the
size of individual segregating responses to desegregation, the
section qn pages
53-80. In general the effects of these modifications are minor,
but the,earli
calculations did contain.sources of biasor error, so that the
recalculatio s are
,included here, even when they make little difference in overall
interpre ation.
The first modification has to do with eliminating schools with
,teachers,
reported: For 1968-72 OCR obtained information on race of
teacher , and we
included that in basic calculations of segregation though none
.f'those results
are included in the report. When a school gave np report o
teacherEl, the
school was eliminated from the anaiNio for that year. chools
were not
identifiible from year to year, so that meant eliminaty on only
for that year).
Thil- failure to report teachers was very.infrequen except in
the case of a
`few districts, such as Los Angeles and-Greenvill , S.C., which
gave no teacher
reports in 1969, and thus were eliminated front the analysis for
that year.
This procedure would not in and of itself affect the analysis
except that
in 1973 OCR no. longer obtained information on race of teachers
so that all
-
schools were included in our analysis in 1973. Because this
meant we used a
slightly different, procedure for retrieving information in
1968-72 and iii 1971,
it could `lead to possible biases! in the result. Because the
difference was
very slight, the measures of changes in segregation were
essentially unaffected.
But the analysis of white loss upon desegregation might be
affected so those,
' analyses were recalculated. The effects on the results are
indicated below.
However, first, twolother changes should be noted:4
-
5.
and 169, and separate elementary and secondary districts in
subsequent years_.
s learned that the OCR, data on Tucson, one of the next 47
cities ,. ,
the largest 23, showed a combined elementary/secondary district
in 1968
(There are in fact two separate districts, butgoverned under a
single echool
board.) Because of Shift change in reporting,'Tucson was
eliminated in the
'7*modified calculations. (the' "next 46"" designation, however,
is correct for
the, modified calculations beicatise the designation should have
been "next 47"
with Tucson, since Albuquerque from the first 23..replaced
Richmond. See foot -
nOtrir5and 23 onpage 56).
An error was grade in tde lagged equatiOns for 71-73 and 72-13
reported in
Table 15. Although thirdid not affect concltisions drawn from
the table, it
along with the changes discussed above,
affectthe'calculatiOns.
The effects of these changes are given below:
Table 14 is changed as attached. The general effect of-the
changes is
rather small, and will be described in discussion of the text on
pages 60-65,
which elaborates those tables.
The effects shown in the simple equation, Equation 1 in Table
14, described
on pages 60-61, change fittie,,, with a very slight increase in
estimated white
A
loss with a .2 reduction in segregation for the largest 21 (from
5.5% to 5.6%)
and a decrease from 1.8% to I.1% for the next 46. The greater
changes for the
next 46 cities in these modified calculations,than for the
largest 21 is due
to the.exclusion of Tucson and the inclusion of 1969 data for
Greenville, S.C.
(the year befoie desegregation took place):
For Equation 2, discussed on page
For the largest 21., the lass becomes 6
6.0X and 3.9%, only 0.1% in the North).
63, the changes are slightly greater.,
.8%(foi South and 4.0% for North (from
For the next 46, the loss becOmes 4.8%
for South and no reliable estimate possible for the North,.
replacipg2.6% for
0
-
6
0
.. . . ,I.
.s .,
_ ,.. . .
South and 0.2Z for North. the absence of a reliable estimate for
the Noith for
the next 46" result's from the fact that nearly 411 desegregatio
n occurred in thea
8uth, resulting in a correlation of .98 between R and R x South.
(See' skein(
.
footnote 26, on page 65:) , A
The fact that the loss for the next 46 '(in the South) mm
ettimated.by
Equation 2 and Equation 3 is so much greater than as estimated
by Equation.1
results from the inclusion in Equation 2 and Ecination 3 of the
between-district
segregation. This is especially low in some desegregating
districts in. the
South,'thus depressing the white lots in those cities.
On page 64, there, are changes of,O.l; in nmObers in the
,tahulifion, with no
changes in interpretation. On page 64, the revised calculations
from Equation 3
'.
, ,t.,,(see attached) shows somewhat stronger effects of
desegregation on white lois .
.
than before for the smaller cities and.stronger
intentification'Of.the desegreka-:
tion loss Uthinmrease in proportion black and between-distriit
segregrition.
Revised calculations for Table 15 are attached with ho changes
ininter-,
pretations resulting. Similarly on page 71 there are nuaeric
changes which
do not_change interpretation (tee attached).
&i. .
Table 17 on page 72 changes as indicated- in the attached. As
pointed iA .,
out in the footnote to the revised table, the estimated gains,
for Houston and
2, ., .42 .DenVer are very likely spurious, due_tq territory
anneited during the peridd. .- .
),
t.
of desegregation. Table 17 should, however, be taken with"sobe
caution as
.
providing ao.c4trate estimates fqr individual cities tecause the
high degree ofo . , - .
'.
aultic011inearity crAates some instability in estimates as
described in foot$
note 26, page 65.. -
At strong caret :about %e projected long-term effects as
estimated _on
' . 0_ -- . ' "
pages 74-75 should, be added. These estimates' must be regarded
as conservative
eltimates,of the effect of desegregation, because of the
liesSumption,, based on
4.
-
,,e,
,.
. . ,,,weak evidence, that the direct effect, of desegregatioii
on-white loss is a one -,
year..
time ,effect which does not continue beyond, the year
of_desegregation. The. $.
r ' ', iz..exprerience is not sufficient for strong infeiences.
However, examination of, .44
... . . . .
. the losses in particular cities where desegregation occurred
at a clear-cut...
. .Eoinasuggests theb possibility of a continuing effect riot
shown by Table 15.
--,,
. For ,example, in Dallas, where the effect of desegregation in.
1971 was not
particularly strong (an increase in loss from 3% in 1970 to 8%
in 1971) the
subsequent years showed losses comparable to that of 1971. (9%
and 7%, and data
for 1974 not containa in the. OCR data shows a i gs of 9%). In
Denver there was.
a steady growth in loss after the point of greatest reduction in
segregationk +I.. /
(1969) from -2% to 7%. But_in general, there are simply not
enough'data. WhitI '
is important to note, however, is that a 8011, but continuing
irictement inI ( ,
1.1hite loss can have a much greater effect than the one-time
loss. Thus the\-_,......
estimates of long-term impact of desegreation on pages 74 and 75
may be... . $
serious. underestimates. ,, I . .,
. The modified procedures described above lead also to changes
in the. ..'
tabulations fOr the largo central 'city, districts in
Appendix3.with minor\ .., .4
changes.. for all schools and major changes for the separate
elementary and4 . . . .
secondary schools. Revr Appendix 3 may be obtained from The
Urban Inst-'.
tute upon request.
12
i"
"4
(6
-
.
*
.
. -
TErrup. RE IONS
PO1. A.
Page' 21, line 4: ,..
/ ,.,: .
. ,
"As the table showselementaxy'schools are mere segregated than
highschools, in every regionemeept the Sbutheast where
within-district segre-gation at" the two leVelsAa.the same.r '.
'
''' < .' ' ,e ,.. , JPage 21, line 11: "...: only a fifth
(0.20) ...."
_
%. ,..-, .
,,: .'.. ..
Page 34, lines 1-14: i . .I , ...
"The aftwwer at first,appeaFs to be that,the degree of
desegregation wasgreater at the secondary level, for set:able,
shows, the redUption in degreeof-segregation in the country as a
wholeinis!greaer at the secondary'thanthe elementary level: . ..('
. . t .4. 3-'
However, this apparent greater reduction in segregation at the
secondary'' 100. masks differences among regions.' Figures.3 and 4
show the. changes front' 1968472 1n-elementary and segondaXY
4.chools by .region,. In the two regions
where federal and court sett* toward integreation were
strongest.,.. theSoutheast and West South Central` and in the
Border states, the drop insegregation was greater inhighschoold
thanilmelementary ones. ', Hut in:e:Ach of iheoth4r
regions'thedecrease, in segregation Wasgreater in elemewr
_tary schools. 'In fact, in three Ot,the northern regions (New
England, Middle
, , ., ,
Atalntic, East North Central), segregation increased among
secondary schoolsfrom 1968 -70, while no region showe. d an
increase in segregation among elemen-tary .schools." , . ,
. ,
-
4 7
Page 46/1ine,21:, pelete,"., ...but the Southeast-..."
.. .., .7!.,-.%.".
Page 48, lines' 4-18: ... -?
.
. ,.. . .
"When we look at Changes from 1968 to 1972, there is an increase
in everyregion but Border states at both levels. But the increases
vary by region,'and, by level. In all regions,. the increase was
either the same at both ...
'levele pr greater atthe secondary level. . . . .'- . . .What
appears to occur is ihis: -,As suggested_hy thp earlier data,
the
general movement-ofyhites to areas 4th fel' blieks during this
period wasgreater at ,the secondary level, v-ry,likelY due to the
greater age and attic-7,
.ence of families with children of, gh school age. The result of
that'greatermovement was to increase thebetwee district
segregation-more among secondary...school students,than.among
elementary students. Whether the loss of whiteChildren when
desegregation occurred was greater,at secondary than at
elementary
. .
1levels cannot, however, .)e inferred from these reauits:
That.question will bediscuised again i.diubsequent,section."'.
,,,
.1.4
.
-
7
\rage 58, line 14; "large negative values for Ar
*Page 60, line 2 ihiotigh' page 61# line 6:3,. . tor a city with
the average number of students,'with no blacks
. end,no reduction in segregintioia, the expected losskper year
is:a) Largest 21: (gain of) 0.9% of whites present at beginning
of ,yeer (average huipber of students is 169,000)-b) Next 46:
1.2% of whites present at beginning of year
(average number of 'students is 58,0(50)
2. Additional expected loss if the city is 50% black:a) -Largest
21: 6.8% cif whites ptesent at beginning of-year
.,b) Next 46: ..4.5% of whitesepresent at beginning of year3.
Additionarexiected loss if the city experiences a decrease, of,
.2 in the index of segregation in that esr:25.,a) Largest
21:..5.6% of whites at beginning of yearb) Next 46: 14% of whites
at beginning of year
4. Additional expected- lose' ice city were twice its size:a) .
Largest 21: 0% of whites present at beginning .
of yearb),. Next 46: 2'.9% of whites present at beginning of
year
Taking the first three losses together, the expected loss of
whitesfrOm a city system with 50% blacks would-be:For the largest
21:-
with. reduction of .2 in segreation: 0.9% + 6. 8%' + 5:6% =
11.5%. With no change in segregation: (-) 0.9%* 6.8%'= '5.9%For the
next '46:
with'reduction of .2 in segregation: 1.2% + 4.5% + 1.1% =
6.8t'-With no ChangeAn segregation: 1.2k -1- 4.5% =I
Page 63, -line 11 trough 'page 64, line 7:"Estimated increase in
loss of whites in one year as a 'functionof reductibn :2.-in ,index
of segregation:.
South ;.North
taigest 21 6. 8Z 4.0%1*.Next 46' 1.9% *, ...,*No reliable
estimate for the North can bemade since the correlation
between Ar 'and Arx South is,.983 (i.e., nearly all changes in,
segre;gation occured in the. South in these 46 Pities). See
footnote 26 for
, . .. 'further .er discussion. '' ' .-" ....
.4-4 These results show that indeed there has been a greater
los,s of whites when,. desegregation has ,taken place in large
southern cities than when it has taken
place in large northern cities, with the esimate nearly twice
for the south-ern. cities what it is for northern ones. For the
smaller cities, there is a.smaller loss for the Southern cities
though no effect can/be estimated for _the North in these smaller
cities. '
' Fax ttlis analysis with the two. additiontil variables, we can
also ask..A4 3; J 4
, ,-, r
t,
-
what differences in loss of whites are
associatedmithAdifferencelpetween0 and 50% black in the city
schools mild a difference betweer0 between-district segregation and
.4' between - district segregation;:,,
-
Estimated increase iikoss of whites in one, year as
i'lunctioi:of50% black in city school district and between -
district segregationof .4:
Between-district ,50% black egregation of .4
largest 21 2.2%. 6.6%....Next 46 1.7% 4.4%
'. Page 65, tabulation in center of page:
.
Between-district): . Largest 21,segregation proportion black
,25 .50,,.'05.
Next 46proportion black'.25' -.50 .75
0. 2% 10% 17% ; `.3% 6% . 9%.2 9 3.6 24 lit AI 15,.4 15 t 23 30
14 17 20
- . -,,,
t.
Page 66, lines 1. thrbash 3: ' . . - , ,/.
"These estimates arefor.a city in the South. In the North thee
losses... .at the time of reduction in segregation are estimated to
be3,6% less in the. ,
largest 21 Cities with no reliable estimate possible in the next
46.
.. . -,,,.
A
Page 66, line 22: "and. three more equat-ions,' w
'
Page 71, lines 10 through 12: .
"The results ;of the analysis give coefficient's for Ar of .262
(.057) for
the largest, 21 city districts, an0 .098 (.025) for the smaller
cities.29"
!Page 71, footnote 29:
."R
2in these equations are .65 and .60 respectively."
., .4,/
.
.
,
Pages 77 and 78:' Delete ection on-Elementary and tecondary
Sehools, Whichcontinues through sixth line from bott6m on page 78.
-
Page 79: Delete number 10.
y1I
-
- t22
Revised Table 5. BLACK-WHITE CONTACT AND SCHOOL SEGREGATION IN
1968 BY.J. REGION, POR-.ELE14ENTARY SCHOOLS'AND SECONDARY
SCHOOLS
.
. proportion ' SchoolmatesBlack-whitesegregation.
.
. white black
.whites for
average black
blacks foraverage :white
withindistrict 'total
.. . -ElempntarySecondary --
:78
*81"
.16
;15,.1 -
es
.20
.25
:_',
.04,-
.05 ,-
* t
.66
.59.....
,
, ..'75..69
..,
New England-
ElementarySec'ondary
.92
.95
.06
.04.
... .
.42 .
.674,_
' .03.03 -
'.42,
.16
.55
.29
.
Middle, Atlantic
,. ElementarySecondary
- .78
.85
.16'
.12
.26
.42
' -:05'
.06 ..
.
.50 ---",
.30
.67
-.50
,B der . ,ElementarySecondary-
.77Al
.22:19 '
:IV ..31, .
".
.06
..07....4-
-, -
.53
.41
-,,71'
.62
-
.- . .
--Soutliekse:-
-- Elenentary-Secondary,
. _.
. 1.70,...4-8 /
- .28 -. .31 .
.15
.16
.
.
,' .
.06
.07
.
-..
.74
-.74 s,
.78
.76
, West South Central
Elementary'Secondary
t''''.)7
'- 1-8,
. t.16
.15
.-..- ,--'
.1.,
...' : '. F
:03
.9.4 ' .. r's-
%
.73
.63
.80
.71
East North Central
ElementarySecondary
,
.87
..88
...,.
' -.12.12
.
..25.
.35
-:, -
.:04.
0-5,.
,,
...".,....
.50 .60 ..
West North Central
Elementary.
Secondary
.87
.93
.
-.11.06
.22-
.44.
:.
.
`
,
.03
,03.
..-..,
.67:.44**'4..
.75
.153
.,
Mountain
ElementarySecondary
.79
.84
.
,03.02
.29
.53 .
,..
.01
.02 ,'.57
.3-2
.64
.37
PacificElementarySecondary
.
.77
.80
.
.08
.07
.23
.30
' '
.
.02 '
.03
,..59
.50.
. .71'.62
6
,
4
A
-
I
Page 34: Revised Ta. e 9.... ithin-di,stris.t Segregation in
1968? 70,
72 at El nta and Secondary Levels ,for the U.S'. as a Whole
ElementarySedondary
1968 1970 1972 1972-68,
.66 ..48 ..42 -.24
..59 .33 .29 -.30
Page 4 : Revised Table 12. Between-district Segregation in 1968
and.1972 in Each Region for Elementary and. Secondary Schools
aY
- Elementary , Secondary
1968 - 1972 Change 1968 1972 &sage(1) (2) ((3) - (4) . (5) ,
-(6) 1
United ttates .34 ' .37 +.03 .11 .' .34 +,80k..
New England V.20 .34 +.06 -17 . - .23 '-W.06 /-Middle Atlantic
,, .42 .46 +.04 .33 .41 +:.04''
Border 2 . 4.49 .48 . -.01 . .45 -`.47 +.Q2Southeast .21 .23
+.02 .20 .22 . --I-62
West South Central .34 -.-38 +.04 .29 .34 .05
East:North Cehtral .31 .32 +.01` .29 :33 +.04
West 'North Central .39 -.41 +.02 .27 .35 , +.08
\Mountain .19 .20 +.01 .09 ,324/ +.03
Pacific .31 .34.- +,03 .29 3*/-0.05
4
-
la
k
35'tI
t4
, %
1970,
Figurt 3 (Revised),
- Within - district Elementary School Setegatimi,-'by'
Region
-SE
1972
t:19
-
.70
.60
.50
:To
36
OP,
0
-- 1970I
ky Figure 4 (Revised)
- Mtr:NE
.Within-district Secondary School Segregation,- by Region*
20
SE
1972
-
-:-. .
.
,_
Page 59: 'Revised Table 14. Regression Coefficients ,for
Analysds
offthite Student Loss to Central 0kties
Equation 1
,All .
Prop. blackIn NConstant -R2Nu:Ober ,of
Observations
'Largest,21
). --
'Meat 46
.279 (.062)
-.133 (:028).000 (App..013,.29
(105)',
.t.
.056'(.026)-:090 (.014)-.042 (.010.452
..:' .26
l''
; (226)v- 4
Acludingriter-district aegregationil SMSA, and inter-action of
desegregation with South: :
Equation 2
1MR .'
Prop. black.In N
R SMSAAR x t'Constant
.
v-).199 (.156) -,:g6 (.157)
(.039) -;.035 (.016)-.044.066 (,008) --.041 (.010)
-4165 (.050) -.110 (.021)-.143 (.170) .242 (.47).-.059 ,.418
,--
.R2 .36 . .35.
,
.
Including interactions of desegreOtion with proportionblack And
inter-district segr4atinn; and also includingSouth-as a dUinmy
variable: , '
. .
Equation 3
AR ,-.459 (.184)Prop. blank. .051-(.037)In N .003 (.006),
R SMSA -.210 (.044)OR x South '.148 (.198)AA x Pepp...111 ck
1.770 (.307):JRx*RSMSA , -.561 (.494)South .., '!-.00e(.010).'
Constant'', -.039 .-.,
;
1.1
174-349 '(.151)
-.026 (.019)
,-.039'(.009)-I -.102 (.025)
r :244*(.145).511 (.215) ..894 0.314)
-.002 (.L06).-
.414
.40
2i. .
a
-
"4,
Pare : Ravised,Tproportion.
Barge 21'
Tears of d
I '5
le' 15, F4rther Analysis Results (Equationes include ,black, and
inter- district segregation)
AR - ARVSk'\2,
egrep;ation
69-7370-73 I71-73 i72-7.
46 . !
69-7370-7371=7A'72=73 1
f
-ARt
.320.(:060)'.330 (.069), .009 (.OSO).279 (.065) -.035 (.078)
-.022 (.075) .43..603 (.096) :082 (..068) ,-.048 (070) ;71
7
f-
R2
.
.35
.089 (025)
.076 (.026)
.102 (.032)
.130 (.050)'
:034 (.026.024 (.025)
.051 (.033)
timated Added losses of whites due to deSegregatid segregation,
in second year and third,year, ass
Secondyear
in-segregatiOn index.*
'First
year-
Larg 21 7.7%'Nfxt 46 .
*Unwei htederrors ere
averages of above estimates were usednearl alike.'
\ .34
4,, .31
-.024 (.027)..42-.045 (..020N..40'
\, ,
n in first year ofng reduction of
Third.year
v
.0.7%(gain)11, 9,7X(gain)
because standard
7
-
4
v
Page 72: Revised T lXe 17. Eatiiated Additional Loss of ,White
Studentsin Speci ed Cities
i
. ..
(L s during4esegegation in cities which tad a 'Ar in oneear
of-1-.1, beyond general loss of whites in those
cities.Desegregation assumed is Ar = -.2.)
S
4
City, ,
.
....--
. --
, _
!..-
,
Estimated 1pss as a percentof'white students presentat beginning
of year .t
Houstone
t)ailas
MemphisTampaIndianapolisAtlantaDenveiSan -Francisco.
__. ....... ...
Average
(gain)
.
, .....
,,,A
.(gain)
.
,
9.1%7.9%
15.6%2.6%6.7%16.7%CO%5.1%
5u2%
..
.
NOTE: Professor Reynolds Farley (personal communication 10
September,1975) has pointed out to us that Houston, Dallas,
Memphis, and ,Denver annexed substantial.amounts of territpry
during thepericid 1970 -73, so that the losbes for those cities
maybeunderestimated due to an undetermined numbg of white
childrenadded through annexation., Thus the apparent gains for
Houston.and Denver may well be due to annexation.
,
V
23
4
a
-
-CONTENTS
4
' PREFACE..
TABLES... ..
FIGURES..
Introduction
' The'Measures of School integration'
- ,1* iii
vii
.....: . ....
........Integration In 1968
Within-District Segregation 4041600Regional .Variations .
ge6oa000Variation by Size of District Y400006.0 . *00
. .
et Variation in Segregation by District Size ein Eich Region.,
.....Segregation at-Elementary.and Secondary.Racial Contact and
Segregationin'the 22 Largest CentDal
City Districts.... ,
Trends in Within-System Segregation... o40000sChanges in
Segregation at Elementary and Secondary LevelsChangei in 22 Largest
Central-City Districts.
v..- . sA , .
Changes in Total Segregation and 'Segregation Between
Districts.....
Within- and
.4
The Size of
. REFERENCES
APPENDIX f:
APPENDIX
APPENDIX 3:
iit
7
11,
1214'
1618
20
-.
Between-District Segregation in Metropolita&-Axeas. ..-...
49..
indiv1;dual ',SegregatingResponses to Desegregation ... . 53
0 .....Office for Civil Rights. *piing Plan and ReportForm,
1968, ...,....
Aiterritte Use of SI;b" ...........
Basic Data for. 70 Largest Central City School Districts- -All
Schools, EleMentaty and Secondary, 1968-73.. ... .
. 24
-
4.
-
-a
IL
Proportion of Schoolmates from Each'Group
- -ofEich Group
for the Average Chil 12
- 2- Black-White Contact "."and School Segregation'in 1968b'',
Region' 15
*-
3, Black - White Contact. and School Segregatiorvin 1968 by
District. . T7
,
, .
,...;,.--
, -,
Size ...., . . ...,. - , .--.)
`Avarage Within-Disttict Segregation in 1968-in,Each legion- '--
19
According to Dis-triCt -,: .:
'
7,5'Black-White Contact and School Segregation in 1968 by Region
12t2
,
fOr Elementary and . - ,and,. Schools'',-, ,...
,. . ,
Contact, 6,
Black-White ontact and School.-Sdgregation in 196$". for 22
Largaat.
24
'Central City SChool Districts ,"/":-''- ,,
,..',,,:,,,
--.,---- N,
..*,
Black - White. Contact and SChool
Segregatienjp1972shy.Region:
8 Average Within-District Segregation in 1972 in Each Region
Accordinkto District. Size
33
5'
:lathin-District-Segregation in 1968, 1970, 1972 at Elementary s
34 '
and Secondary Levels for the U.S. as a Wfible
. - , , , , e
10 :4
Black-White. Segregation afid Contact of ,Blacks with Whitea'A.n
38
, . 22 Largest Central, SchOol DiatriCts, 196.8 -73A.. .
- e .--1 .7..Z->
11 Changes in Within-Districtand Betweem-District
Segregation,,91.;
1968 and;1972 in _Each 'Region : 4k'
\ . !, 11
.'2, Between-District Segregation, in 1968 and 19'12 in Each
Region 47
for Elementary and Secondary Schools .
. lt Blaqk-White Segregation among Schools within Central Cities
and 50'
Betwaen-Distrits in the.Metropolitan Area" ,
45'
14 Regression Coefficients for Analysis of White StUdent Loss to
59
Central Cities
15 giirther Analysis Results
16 Reduction in Segregation 1968-73,'EXpected and Actual-Loss
of
White Students 1969 -73, 22 Largest Central City School
Districts
17 ',Estimated Additiona1fLoss of}White Students jn Specified
Cities
1,1 '2'vii
68
70
'72'
-
.
M
I4
FIGURES
#
Average Within-Dist ilci 4egiegation,fl968-72 .by
Region--(Alaska 'andcHavaii:omittedl '\
,2 'Avergge Within-District Segregation, 1968-72:by
DistrictSize
Average Within - District, Elementary School. Segregation
by Region,..1168-72
30
31'
35
Average Within-District Secondary School SegregatiOn.by Region.
1968-72 36
5 Average Total Segregation by ,Region, 1968-72.4SAlaska
andliawali omitted)
4
2G
44
1
-
.
I
INTR ODUCTION .
School desegregation has. been a major issue in the United
States, in
the 1980's and 1910's. In 1954, the Supreme Court decision -iii
the Browri
case initiated.a set of activities which has culminated in the
-current.
desegregation efforts in large cities of .the North.
"Desegregation" has-.meant meat things during. the period since
1954.
The term initially referred to elimination
which one set of attendatce zones was used
of dual school 67sCems, in
to assign white children
set of schbals, and a second set of attetidanCe zones' was used
to assign
black children to a different set of schools. The classic and
plaintive
ne.
query of the black mother in the .8duth was why shoUld her 'ch
be busSed
to a school far away, past al nearby school, merely because of
the color of,
his skin...The extent 'of -ohangeAs that tbe sate plaintive
query is now
,.heard; primarily from white- mothers,, primarily in large
cities, where
bussing has begun to be4ised,, uPt to segregate children by
race, but to
integrate the-M.
This change is reflected in a change in meaning of the term
-desegre-z<
gation. From the initial Raaninvapt eliminating a system of dual
assign-
ment, the term desegregati9n has come to mean reduction of any
segregation
within a system, and in the strongest meaning-of the term,
elimination of
any racial ,imbalance.among -schools in the sygtem. Thus
desegregation,.
which initially meant abolition of a ,Jegally-impOsed
segregation, has c,ome,
.ta.mead., in many cases, affirmative integration.4
2. ,
-
I,
However, except for one court case (in Detroit) which
was,later
xs-
reversed in the Suprebe Court, desegregation' has not come to
mean
fion of racial imbalance between school districts. Nor, except
in, a few.1_
-instances; have two or more school systems combined or
cooperated to
reduce segregation due to residence indifferent districts. Thus
social
policy in school desegregation, althOugh changing over time
andndifferent
,..,.
in different districts4has almost wholly been confined to
desegregation
, of sch6ols within a schoolistrict.
..
.,....,,
.
'Given the policies that have been applied, by local.school
systeMs,
by the Department of Health, Education, and Welfare, and by the
courts,
we can ask a series of questions concerning the actual state of
racial
integration in schools, and recent,trends in that state. For
actions
,taken by one branch of government and at one level of
government interact -
with actions taken by individuals and by other branches and
levels of
goriernmene. The actual State a sc io`o a tie az. ron- lila
. interaction. It is different thati it would be in the absence
of theN.
policies designed to bring about integration; but it is more
taan a simple
consequence of the policies. Inteed, there are numerous examples
of
government policy in which. the resul of the interaction between
policy
and ,response is precisely the opposite of the result intended
by those
initiated the policy, It is especially important in the case of
schoo
desegregation to examine this interaction, because many of the
actit6ns
taken by individuals, and some'of those taken by their local go
ernMent,
bodies, have precisely the opposite e4fect on school
desegregation to
that inelended.by federal government policy. The most ob ious
suchindivid-,
who
ual action, of cmurse, is a move of *residence to flee sehool
integration.
ion ie Jr.. _kr.
-
3
To examine the status and trends in school segregStion,
the`primary (and
virtually singular) data source are the statistical
reporti'collected bY the:
'Department of Health, Education,jand Welfare. Beginning in 1968
and continuing
its
to the present, theOffice for Civil Riihts.,(OCRYafTIEW has
obtained from school;
systems throughout the United States statistics showing the
racial compositiOn
of each school in the district, the racial composition of
teaching staffs, and. V.
related information* The data.,for 1968, 694 70, 71, 72, and 73
have been
processed and are available for fmalysis.- These data allow a
detailed.-
statistical,ana ysis of the Status and trends in school
segregation by
r ce throughout, the United States; They Are unique in(this; and
-the
opportunity they ogg= is the opportunity to' examine what has
actually
occurred throughout the period 1968-73 during which there have
been
policies-at local, state, and federal levels, in courts,
legislatures,
executive and administrative. branches of government related- to
'school"
desegregation. }lost of these policies have been aimed at
bringing about
desegregation., though in a few cases, st h as anti-bussing
actions in
4
Congress, they have,, been aimed at preventing certain kinds of
desegregation.
Not all the questiOns surrounding-pchoel desegregation can be
answered
by these data, as will be evident in subsequent pages, but
pome,*n be, in. ,
. . -
a mare complete` way than before. -- .
Of the various policy aims that have been the objects of school
desegt
regation policies, these statistical data can give evidence only
on a subset
-of the aims. And fram this subset, we will examine a still
smallet subset:
'. 4
the aim of eliminating racial, segregation among schools within
a system,40,
whatever its source, and the aim of, eliminiting racial
segregation between
districts.. The data gathered by OCR allow elect for the
examination of
teacher assignment, and thus racial segregation siang Staff
and
29
-
4
etween staff ati_studentt. However, we will not pursue that
ewamination-
.'.
.. ..
' The data do 'not allow, on, the other hand, for a study of
segregation t.. . .
among/classes within a school (often known as '!tracking"),
because there is _
:no good information on pupil assigmmentto classes Within a,
00004 The---.
. .
Office for Civil Rights attempted,, in its 1971
qutsiionnaire,.-to obtain
I
,.
.
....$
...k, ...
.-. .these data from school systems,,.but abandoned the effort
in 1971. A ?
more detailed and ihtensive modepl ...;//
4/
if dat of sufficient quality on
of data collection ,is probably' necessary_
assignments within school are to be obtained.
/to implication is attended by the examination_to be.carried opt
below
that, the policSr aim; of eliminating, segregation among schools
-with a'system,
-whatever.its-source, 1s the "correct" one, and.other policies
whin would_
161th-eV go less far (such as eliminating only that school
segregation not flue
."to residence) ors further (such as eliminating all segregation
amont classes
b P ). .4'.
within a school) are not correct. The question' of what is the
correct.
. -_pogcy depends not only on the
,
implicitly aimedfor social consequences,; .
,
but ,upon the palmrof legitimate authority of the governmental
units .b
a
' applying the policy: .This in turn depends just which
individual rights
citizens, have vested in. their goVernmelit for collective use,
throUgh the
4'Constitution and legislative acts. For example,'tb.aaconplish
the policy
r ,
aim, of-iiiminatingall segrogatip among schools, whatever its
source,c ,,,
the most Offectiive iFplementation would be federally-specified
pupil
-
assignment to'schools to create' precise racial balthice,
disregarding
school district and state] lines. However, such a policy would
be using
collectively certain rightsthat individuals have retained to
themselves
or vested in a More local level of, government. At another
example,
3 0.
YAW
6.
-
citizens have vested certain authority in the court,isuch as
constitutional
p'rotection, but a wider Lange of authority in elected
legidlatures. Thus
certain,policy ams such as elimination of segregation _ among
schools
whatever its source may be appropriate for legislative action if
it achieves
_certain desired consequences, but not appr Rriate for court
action, which...
must be directed not toward achieving desirSble cial'goils, but
insuring
. -.
. ..
tt..
constitutional, protection for; all citizens. It useful alsoto
pdfut."
..
.7:
.
out that data such as these whiCh show the indiiect and
unintended conse-,
.
, . - r.. .
quences of school desegregation actions may be relevant for
certain
desegregation decisions, but not for others. Tbey,Ale-relevant
for an-
.
t.
.;.= .
executive or legislative body which is attempting in its action
to achieve. ,
A....desirable _sol*A-ccmsequenCe. They are not relevant for a
court
decision which is acting to insure eqUal protection. der the
14th.
-Amendment.
, Despite the. fact th
-
-tt only. two _aims, stvd.enticlepegregation arao'rig
stole In. a.district regat le4s*of the source of segregation,
and
desegregation between scho districts, can be studied there 'are
-a number.
ff important questions that an be answered with these data. In
patiicular,'. .
these data show the result of vernment desegregdtion actions
and.individ-,
41.1a1 segregating actions taken together, and fallow some
assessment of the
effects of.each. In' this way, they suggest the limits- of
government
policytot at least the limits of policies' carrlol'out in the
Conflict. ,
mode that has characterized school desegregatiOn'policy.,
or,
We will begin by examining the state of rSCIal'integration among
Schools
within ndistrict in 1968, and then move to an examimation'of
the'chihges
that,occurred over the period 1968-1973. What will beiofspecial
interest
31
41,
-
6
. 1
I ..
is the diff ential changes that occurred over t at period of
time in differ-
ent.kinds:df schdol settings: id different-regions of the
country, in
school,dfstricts of different sizes,.and in particdlar large
cities. For
different, things were happening in different places,during this
time,
giving/riseo very different trends .in different, places./
For much of the analysis in examining trends, two separat. e s
eries must
be used., The even years, 1968, 1970, 1972, constitute a census
of U.S.
. .. .
-school districts, coverin4 90% of the'childr in school; and
excluding only.. ,
. . -
,
r a few very small districts. . The odd years include only a
sample of school
/systems, representing those districts in which most minority
pupils are
4
found.' When examining Vends over time districts tthe odd-
numbered years can be safely included, because each district is
either.4
included .4s.a whole, orPexcluded. Butfor average across
across regions, across states, and even metropolitan areas
:
cannot be included, and the series must end with 1972.-
'See Appendix rfOr OCR sampling plan.
.". V
f.
,
4
0,
the country,
; the odd yegrs
5
-
03,
1
THE MEKSURESOF sumac. INTEGRATION
A principal consequence of school desegregation that is of
major
societal in st is the amount of contact between children of
different
40
racial groups.2
Furthermore, most of the attention has been focussed on
the amount of contact of "minority" children (principally blacks
and
SpanishJAmertcan children.3). with "majority whites." Much
attention both
of courts and legislatures has been directed toward eiiMination
of patterns
f, -
that result in schools which are overwhelmingly or predominantly
minority.
, -..For these reasonsa directly. relevant statistical measure
on a
.. ,
school systeni is_the proportion of white chtldren in e same
school with
the, average black child. This giveg a measure of t e experience
of the4 -
"average black child" in that school district with whites. 'A
similar
2A different consequence may be of legal interest:, the degree
towhich segregation resulting from action of any level of
government(thus failing-to provide equal protection under the 140
Amendment)is eliminated. Still other consequences are of interest
to particular'groups, and these may depend on the particular way
that segregationor integration arises. For example, if either
segregation or integra-tion is achieved through assignment
of-children to schools at somedistance because of their race, theh
the parents affected may feel agreater deprivation of rights than
in the case when such assignmentdoes not occur;'even if the
school's racial composition does not
_ differ. However a study of the kind carried out here cannot
eXamine,these consequences.
t -A
1 CR surveys measure enrollments of the following categories:We
s, American Indian, Oriental, Spanish Surnamed Americans, and
er. White non7minCirity And undesignated minority groups are
.
included in the category "Other."J
',-,, .
tt.
337
-
8
..
measu4 may be calculated for thfliroportio4 of children of each
racial. 4. .
t ,, *
group in the school of the average child from each radial:
group.4
This measure is affected not only by the degree of segregation
between
two groups in different schools in the system, but also by the
overall
proportion of children in each group. If there are few white
children in
the system, for example, then whether or not there is the same
proportion
of whites in each school, the average black child will have a
small propor7,,
tion of white children in his school. Because of this, it is
valuable
also to have a measure of just how far from an even distribution
across?-7:N
the schools the actual distribution is, that is, a measure that
is stare-
4
dardized for the number of whites in the system. Such a measure
can 15Pei,
constructed, having a value of 0 if there is, no segregation
between the4.
4Themeasure of interracial school contact may be constructed as
follows:If we number the schools ikthe system I, ...k, ...n, and
consider the firstschool, there is a given proportion of whites in
this school,. Call this
pu. There are a certain number of blacks in the school. Call
this nib.
Then for this number of blacks, the proportion. of Whites in
their school is
piw. If we average this proportion over all schools, weighting
by the
number of blacks, we obtain the desired' measure, which we may
call sbre
the proportion of white children in the school of the average
black childOpt more generally, labelling the group i and j for
generality):
E
S '2..k nkbPkw 1
(1.) 3
.,
bw ,* E
k nkb 4-." or fpr any groups i and j
1 E
14,nkiPki(2)
sib ck nki
d
34
to*
I
I
-
"^--. two groups in
It is imp
uestion, and a value of 1.0 if segregation is complete.5
rtant to note, however, that although the standardized mea-
sure is a measure of,segregation of children in one, group from
those of
.another, it is the =standardized measure which measures
directly theF.
presence orChildren of a group in schools attended by children
of another
group. Thus the proportion of white schoolmates for the average
black
child may be low, as in Washington, D.Ct, where only 3% of the
children
are white, without' the measure of segregation being especially
high.6
5 The standardized measure of segregation is constructed as
follows. If
the same proportion of children froii63roup j were in each
school, then,
sij (see preceding footnote for notation) would be equal to pj
..If, the
children of group j were all in schools by themselves,
totallyAsolated
from children of group i, sij
would be 0.' Thus a measure of how far sij
is from pj. is (pj s jj)/p.. this we will call rij'which may be
thought
9
of as a measure of the degree of segregation, or the degree to
which
segregation between schools is responsible for the value of sij.
'The
formula is
p - s
rij = 3sij
pi
6For some purposes, ft is preferable not to standardize sijto
create a
, . .
"measure of segregation," rij
, but rather to let sij
be a dependent vari-4,
,
able in an 'analysis, with one of the independent variables the
proportion
of group j in the System. Using this alternative, we a not begin
with a
concept of "begregation," but rather with a concept of
proportion Of the,
average member of group i's'schoolmatet that are of group j. The
degree
to which this is accounted for by the proportion of group j in
the systemis a measure of the integration-of group i with j. In a
regression
equation, if the coefficient on the propOrtion of group j is 1,
there is
no segregation. ',Insofar as it is belmel, there is.
4
-
I
U
INTEGRATION IN 1968
In 1968 in the United States, 15% of the children in public
schools
(grades 1-12) were black, 6% were of another minority, and 79%
were majority
whites. But the average black child in U.S. schools went to a
schoOl which
4,4
had 74% black children in it, and-only 22% white children (and
4% other
minorities). Meanwhile, the average majority white child was in
a school
.which was 93% white and only 4% black.
These numbers show' that the interracial contact in American
schools
in 1968 was quite low. Black children had more contact with
whites than
whites had.with blacks, due to the disparity in overall numbers;
but the
separation was quite marked. Using the standardized measure
described. ,
earlier, rij, the segregation between blacics and whites is
.72.7
Although in the subsequent examination we will focus exclusively
on
black-white segregation, it is useful to note here the
proportion of
schoolmates. from each.of the five racial-ethnic groups fbr the
average,
child from-eacil group. Table 1 shows this for 1968.8
As this .table shows, the average white child it the U.S.-has
far less
contact with any minority children than any of the minorities
have with child-.
ren from other groups. Among the4ainorities,. black
children,have least con-\,
tact with children from.other groups. Construction of
standardized measures
-7 Using'equation (3), this is calculated as r =
.75 .2i- .72 .
bw .79t ,i
8(Ustng equation (3), standardized measures of segregation may
be calculatedfor each pair of groups, from the mbulatiori
presented.
11
.3G
-
Tattle 1.
s-
412 ,
PROPO6ION OF'SCHOOLMATES FROM EACH GROUP .F THE AVERAGE CHILD OF
EACH GROUP
Proportion of schoolmates who are:
For theAverage:
----,
American.
-r Indiane
1
, r
Negro'
704
Oriental'American
0+
SpanishSurname
.
.06
Majority
White
,
.59
Sum
1.00
AmericanIndian .31
Negro- °
I
5.74 0+ .03 .22 .99
,......f..
D'rl, enial
American .11
.
.11.
.12.
.66 1.00
Spanish /Surname /-
,
.
0+ .11
0
.01
.
.43:
.44 .99
MajorityWhite 0+'
-,
.04 0+ , .03 .93 1.00
.
Proportionof each group .0035 .1530
.
.0047
_
.0466 .7923
of segregation frot these figures would show that blacks and
whites are the
most segregated, both from one another and from the other
groups.
6-'
Within-district segregation
The segregation reflected it the value of .72 for'black-white
segregation
is composed of two parts: segregation among schools within the
same school
district; and segregation due to blacks and whites living in
different school
3 IIS
-
\
13
districti. Thd average of the within-district segregation, when-
weighted
according to the number of blacks in the distriCts, is' somewhat
lower, at
..63. However, this figure represents a degree of segregation
nearly as
large as that for total segregation. (If everydistrict in which
blacks
. lived had the same racial composition; then eiven this same
segregation
of .63 within districts, the average black child would have had
29%
white schoolmates, rather than 227.. This would have been
somewhat more
contact, bit still not a high amount.9)
A different picture is evident when we examine/the ,,averagf
within-
district segregation weighted by the -number' of whxites in the
district.
This figu're is .23. The low number compared the high .63 for
the
average weighted by the number of blacks reaects/thb fa at
most
whites live in smaller districts with few blacks, while most
blacks-.in districts with many whites. If segregation in the
smaller districts
was low in 1968 (as we shall see shortly it was), then the
within-district
__sggr.egat4en-fer-the-avera-ge black would be high. Or to ,put
it differently,
the segregation within the districts where most blacks live is
high, while
the segregation within the districts where most whites live is
much lower.
This reflects also the fact that most whites live in districts
different
from those in which Rost blacks live.
We shall .use as a measure of the average within-district
segregation
in the subsequent analysis the average weighted by numbprs of
black ,
students, since desegregation policy within districts has
focussed on the
districts with many blacks, and upon the interracial contacts of
blacks.
9The value of 29% is calculated by use of equation (3) with rbw
= 6.3.79 - sb
pw = . Thus .63 = .79 w , or sbw = .29 .
38 , 1
-
14 SA
Mb
Regional Variations
However, the degree -of bc 0oI uegregation differed considerably
atong
regions of the country in 1968., Table 2 shoWs,.-for the Census
geographicA
-
,regiong,-the-contact of blacks and "whites, and the
segregation, in each of.
the regions.10
-
.=Table 2 shows, 'comparing columns 1 and 3, the disparity
betweenthe
proportion white in each region and the propaitton white among
theaverage
black'ssChooimates. AlthOugh the proportion white ranges from
.69 io :93,
in no region except the Outlying states does the 'average black
have a majority
of white schoolmates. Only in New England does the proportion
approach -this.
Comparing columns 2 and 4 shows a similar disparity for.whites:'
although...-
the yroportion black reaches .29 in one region, in no region
does the 6efrage
white have more tha 7% black/4choolmates.
These disparities a shown in column
J.
the measure of-total segre-
gation. It is greatest in the.two Southern regioni, though not
so much. .
t Agreater than several northern regions as might be expected,
given
Ihistoric
al .., .,
differences between North and South in school policy. With-this
relative. . 1
similarity betOeen total segregation in North and South in
1968,/it'is some-
.
what puzAling that when the goals of desegregation shifted froM
elimination,
of'dual systems to more ambitious ones, attention continued to
be concentrated
on the South until the early 1970's. The ex6lanation'probably
lies in the
16Several regions have been reclasiified, because' the character
of racialSegregation has differed.within the region. Hawaii Alaska
have beenseparated as "outlying" states from.the Pacific regi6.4;
and the SouthAtlantic. and East South Central havegoeen combined
and redivided intoBorder (Delaware, Maryland, West Virginia,
Kentucky) and Southeast (allothers in these two regions). It all
tabulEitions'beyond Table 2, theOutlying states,'Hawaii and Alaska,
are dropped, because as Table 2shoWs, there is no
black-whitesegregation in theirachoOls, and thenumberiof placIs in
those states ,is very small.
I
39.
-
*
15
Table 2. *BLACK-WiiITE, CONTACT AND SCHOOL SEGREGATION IN 1968
BY REGION
PiOportiOn. - . ' Schoolmates
Blackt-white .,,segregatiou
1
white
2
black
k.3 ,'whites foraverage 41ack.
4, blaaks foraverage white
5withindistrict
6
totak,
U.S. .79:7 .15 t.22 - .04 ,.63.. ,
.72
New England.
.93-.-
,.., 4 05....:.,',
-, -...
. .49 : .03\,. -.:..-..;
'". .34 .47 :
Middle -Atlantic .81
. .7
v14 .31
.
.05 .43
r ,.
:62
Border.
.79 .21 .26
'
.07 .48 .67 ,:.*-4
Southeast .69 .29 . .16..
.
,---. .07
,..,
...:75 -.77
-West South ..,
Central .78
r
.16
I
.18 . -.134- .69 .77
East North.,
Central
C .
.
-
.12 .29J
',..t,
',- .04-.
' .58,
.67
' .
West NorthCentral ,
.. ,c,
.09 4. .27
. -ca.'.03
. .,..t,
. ', 4c ,
.61 .10''
Mountain . .03 -.36".'
3.01. .49 .*-.56':.:>
Pacific' .:..r
,.i5 ., .02'; .56 ._. -- .68-, .,
a
Outlying'.
, 413 .83 {'.,
.,,.,,, ..
. 035
(-).04 .
.) .05
?r,
-
:.,, . ..--, -, ,-,. . ..,.. , _ . . .copflitt surrounding
desegregation: The desegregation movement was part of. . ,.,.;, 4 I
-..,-,--4.
..,, .a larger 'Movement of th,e1.960' s of liberating the
southern Negro. Only wits_ . . ._,.. -.,
4:1
the success of 'that movement' did attention turn. to:the
North.
In the 'measure Of within-district segregation..(c .0 the. two.
.
- .,southern regions are more' distinct frpia the northern
regions..., The,
..-..southeast shows the c).asSic at tern of southern
segregatio'n, with nearli,-,..
- ..,..,,!,-,..r. .-, . . . .. ,.all the segregatf
occurringwithin districts, while ',New England, Middle
.
Atlantic, and Birder regions show what lea emerged as .th4,
northern. p4te'rn.-A
of segregationsegregation...due to' blacks and whites attending
schools in;.,,
different distriCts. 'Ili those three, reiiOns, the, fatio. 44f
the within- district
segregation to the, total segregation . ,lowest, to .7*; '',.
.t.:,, , ,.,.
s -Variatfon by ,$).ze of districtV,-.
r ,The policies .of desegilhation; as weltas the individual
r:eqpiinses
,
go ,t, vary gieatly.ty ,iiStrict size. Desegregation in an
urban,district
different pro ese froia. that in *small district inn-a rural ar.
4 - ,:. .
astlt is important., to examine regional variations (bee,ause
of
sf
.
fferenos and because of the different desegregation policies
applied* in
;
earea is a vety- .F.:.,
eat .Thus, just ...', Y ,,
- .NOrth.and.SOuth),;, it is iiPottant tp'.examine ,variations
by:,.distriCt,`-..
, , :SinCe cliitricr` bounciaties..ract'Ot* Often.coincide' with
central ci4!., boundaries,
;--, , '"variationa' in districl. sizk' are _largely cOincident
With itariatian in city_,.
,.. , .. ., ,...:size.,
14hawathe IVO interracial contact and,,se.gregatioil by
diatrici
..
, t'' -'size , T);(e 'cOiruMns have the Same_ meaning as columns
1-5 of 'able,,, 2., "tOoluMn
,.._''':-
6 itt, not inaticied iiefe 'sinde iltotal segregatiOn' has
meaning only for a .-', \ .' .' ::::',..,,, ,:: s4
g'
ao.
g' r... a.,--
P, h i c
't ; ;tr...'
i.%
f i,
t.,
i#
"4S.
U,g,
h -'.,,.. ,M.
,
. SA'.'' \
o'.
r r eli.:.,-, o,n . .,
'r,
g,., t , tolumrts ana t'.-.2,.Is'h
s
ow
the tiy. ,sizes,the smalle*the district, the....
t e r the t. op or t io n white and 8.,th,ane.ra4propOt0on.
Maces, resultt ti 4. ,, .,. 1 .' I
..
' s t,461
:
-
-.*
O
"..
'"!.
"ar!
*-f
.Table
4fac.F:wgt.yp..copt4tT AND SCHOOL sggRgctafl0rlIN,X966.13Y
DZATItit9T,SIZE*
.,
Fv
qf.
7
r
:.:
,:
,'-:
%::
. -..-
.-
tFop
ozgl
..an-
WI'
:7: a
ll tia
liiiia
indistricts af-,9':',-.:c1;rst,rict-of
ii
-"
.-..;
..et Zs- site_
..,,,
;,,:.
..
'-Proportion .qf
n .f
ail liiicks JAI
this
site
-- %
.4.
,
:4*.
Txp
p,?-
rt10
13"
- ;
=',:
:.,
. ',
..;> pchoolmates
.
.
.5Blac4.-whitg
,
within
district
1i:"
2--:::
'whi
t-bl
ack
-blic
ic--
...3iihites for.t'
'''.
average-bl,aujc.
,,.,
-....
.-
Irigacks for
."-
-average white
segregation
f. g
.14
.'
3..0
- -.
4.-
4.00
.9:
5:
,...-
.:-
2,=
.63
.
.4 ''
'l''
'..
D'istrictiliie
'''''..
....i
',
:1
.;*.
-:-
., .
-:,
.e4.;
.,
-...
,,. .. ...
%
..::.:-
.
."
.4.
8 '.
.12.
....
,
*-.L
,-.... '
.
.09
..7
1.
. -
..
,,..
,,,
.,
. ,,...-
.!>
'100
',O
W).
" `
.'..
.-...
).-,
'''"
.,''.
1-...
.-'
2$7,
100,
0g0
.._
,....
.t*-,
-.:::
;,...,
-- - -.
. ?
.
.* -
'',.:,--,...,
r
,%
.a3
,.
- .... ::.
.,
,.
.-.-
'.73
.20
..
:.:1-r; '.
.,
.22
,
.
., 06
. .66
'...
.
-.
i'l.P
:2$?
00 0
:, ,....:-.;
.- .
-4 .1
24,\
,
.:,..
',I4.
'...
....:,
.... 11...
'..
.. .' .3
. 0;
S.6
4.
/
.54
5-10
;63
f 7.
'i.
.1'.8
..
'''-
.11
.9.0
4
,
...-.59
2.5-5,000.
.
.
..
. -..,
1 /1
7.."
.:.
4.
-.09
....
,...
,.*
09
.34
.
.03
.
. 56
<.2
,500
c...
.16
,.06
.90
.06
.35
/.40
244
,
S
* The size classification for districts was carried out only
once, forcomparabikity across years.
Sizes are based on 197-.nrollments.
-
18
of this is shown in columns 3 and 4: the average black child has
as in-
creasing proportion of white schoolmates as district size
decreases, and the
average white has .a decreasing proportion of black schoolMates
as district'
size decreases.
'Column 5r shows thatgiyen the racial distributions in the
districts,
the average segregation is_greatest within the largest diStricts
and declines
somewhat as district size, decreases. Ibus not all the increase
in the pro-
portion of white schOOlmites for the average black child in
smaller districts
is due totthe greater proportion of whites in those districts.
Part is due
to the lesser segregation in'the smaller districts.J
Variations in segregation by district size in each region
The differing patterns of segregation in North and'South suggest
the
uf3efulness of examining segregation in different size districts
in each of
the regions. Table"4 shows the measure of segregation (column 5
in Tables
2 and 3). in each size class in each region. First, lucking at
the largest
districts, the most striking point is that segregation is high
not only ina
! .
the two southern regions; it is equally high in three of the
five northern
. .
regions. Only in Middle Atlantic and Border states is the degree
of segre-
gation in the largest districts lower.
As district s1ze decreases, however, segregation decreases
markedly out-
side the two Southern regions. In the Southeast, segregation
remains almost
constant among all size disVricts, appi in the West South
Central region, it
declines only slightly as distOct size decreases. Thus in 1968,
the dif-
ference between South and North in segregation is not at all 1n
the largeSt
cities, but in the smaller cities and towns. A caution should be
introduced,
however: the measures of segregation do not tellithe levels of
contact be-
-
41
TAW A4.- AVERAGE WITHIN-DISTRICT SEGttEGATION IN 1968 IN
EACHREGION ACCORDING TO DISTRICT SIZE
District
Size (000)
U.S.
New
England
Middle
Atlantic
Border
ISouth-
,east
..
West
South
Central.
East
North
Central
West'
North
Central
Mountain
.
.Pacific
>100
..71 t
---
.53
.
..
.84
.19
.79
.82
__
.84-
25-100
.66
j?,48
.54
sf46
i
.77
.74
.61
.64
..
.56
.45
10-25
.54'
'
.28
''
.30
.23
'.69
..61
'
.33
.22
.44
.35
. 5-10
'.59
,.08
s.15
,.
.24
.74
.64
.18
..
.45
.20
2.5-5
.56
.06
.08
.13
.74
,.52
.09
.19
.
.11
.08
2.5
.44
.01
4.04
.04 -
.70
'
.36
.02
.09
.
.06
.05
p
.
-
,
Ir
4,
420
tween the two racial groupR: as Table 2 showed, the average
white' child in
the Southeast has, despite the higher levelslof segregation, a
higher propor-,.
tion of black schoolmates than in-any other region, except in
the Border states
whete it is equal. These tables suggest, in fact, that it.is the
poten-
tially high proportion of black schoolmates, due to a high
proportion black
in the region or district, that generates high leVelof
segregation. Table
2 shows a rather strong relation between the proportion black in
a region,
and the total or within district segregation..
Segregation at elementary and secondary levels
In the tabulations up to this point, allustudents iii all.
schools were
included, whatever grade they werein: However, inmost
localities, schools
s.
are divided into, different levels,"at least into an
elementary.school and
high school, though sometimes into three levels (a 6-3-3 plan,
with junior
high schools,, or a 4-4-4 plan or a 5-3-4 plan).11
Elementary schools are
characteristically smaller than high schools, having smaller
geogra i
attendance zones, with several elementary schools feeding.into
asingle
high -school. Because of,residentfal segregation by race, we
would expect
. .
the attendance zones of elementary schools to be more racially
homogeneous,, ..-
and thus to be more segregated than high schools.
;
11In recent years, the &fart to achieve. integration without
bUssing hasled. to even finer divisions in some cases,*with a
school building whichonce covered. the four years of high school,
for example, now coveringonly two years, with twice as large an
attendance,zone.
.
45
e
-
r
fe-
#c_.
(
Table 5, which is comparable to the first six columns of Table
2,
show the degree of interracial contact and the degree of
segregation
for t e U.S. as a whole and for each region, in elementary and
seconciary.
21
schools. As the table show, elementary schools are more
segregated than high
schools; in every region. Only in the Southeast is the
segregation at the
--two levels nearly the same. This reflects the remains of 'the
historical de
jure segregation of the South,Jihich.segregates beyond the
segregation induced
by residence, and thus segregates the high schools .as fully, as
elementary.
.
The erence in segregation.at elementary and secondary levels is
rather
subsirtial in most regiohs, and in the U.S. as a whole: the
average black ele-
'mentary school child has only a sixth (0.17) of his schoolmates
white, while the
average black secondary' school child has a quartei.(0.25) of
his schoolmates
white. Whether the greater segregation at the elementary level
includes a
greater tendency, on the part of whites to segregate their
elementary school
children, beyond the due tb small attendance zones, cannot be
inferred from
these data. However, the examination .of trends from 1968-72 -in
a subsequent
section will give some indications of different processes at
elementary and
,
.
12Because of the varying organization of sChoold intolevels, and
because of.the way,data were collected by OCR (the grade levels
covered by each schooland the ;Lumbers of each racial group in each
school, but not by grade level)a fiked rule for classifyink a
schools as "elementary" or "secondary" wasrequired. This was to
count every school with a sixth grade as elementary,and every
school with a tenth grade as secondary. This excluded some
schoolswith less than the first six grades; but to inclUde them
along with thesixth7grade schools for which they were feeders would
have incorrectlyassessedsegregation between, say, a,grade 1-4
school and a 5-8 school,rather than segregation across all sdhople
serving a- given .grade level.Soule schools in email communities,
which cover grades 1-12, are includedboth aselementary and
secondary, uUsing this method, however, we obtainthe segregation
at,two levels in the community, the 6th and 10th grade level.
4.6
-
22'
Table .imAckwHrrE CONTACT AND SCHOOL SEGREGATION IN 1968
BY4;.'REGION., FOR ELEMENTARY SCHOOLS AND SECONDARY SCHOOLS
,
. .
-Proportion
.
. Schoolmates,Black-whitesegregation
1.
white'
2
black3 whites for.
average black
.
4. blacks foraverage white
.
5 withindistrict
,
6total
U.S.
--iltimentary
Secondary.77
.81,'
I. .
.17_
.15
-,--..
.
.
- .17'
.25.
.
.04.
.05
. .70
.61
.78
.70.
New England
SlementarySecondary
.92
.95.
,
.06
-, .04.43.68
- .
.03
.03
.
,,
.40
.14,..53
.29
Middle Atlantic
ElementarySecondary
'
.79
.87
.15
.10 , ty
: .05
.05
.
.50
.27'''A
-.67
.46
.
Border. .
ElementarySecondary
.78
.85
.
. .21
.1\
.
.
. .21
.39
. i
.06
.07
.
.56
.35
.73
.54
Southeast
ElementarySecondary
.
.66
.67
.32
.32
. .12
.15
.
.06
.07
,
4
.79
.75, .81
.78'. .
West South Central
Elementary
. Secondary.76
.81 _
.18
.15
.14-
.23 ,
,
. .03
.04
.
.76
.63
:
1g
East North.Central..-
ElementarySecondary .
.86
.90
.13.
.10.24
.41
. .04
'.04
,
a
.
.
.65
.43
.32
.55
West North Central
ElementarySecondary
,
.89
.93
.10 ''''
.06
.22
..38
.
,03.03
.,
, .66
.47.75
.59
.
Mountain
Element'ary
Secondary
.
.80
.83
.03
..'.02-
-
.28
.53 ,
'.01
.01 ..
. :
.59
..32
.
.66
.36
Pacific
ElementarySeco npry
.7
.
.s
.08
.0 .
,
.21
.36
Y
.02
.02
Y'^
.
,
.62
.44
'%- .'
.73
.57
.
47
t
-
23.et
secondary levels.
Racial contact and segregation in the 22 largest central-city
districts
As Tables 3 and 4 show, segregatiOR is most pronounced in the
largest
11,-,- t.school districts, which tend to be located in the
largest cities. Table 6,
'in the same format as columns 1-6 of Table 2, shows for the
twenty-two laig- .
est central-city districts (1972 enrollment) the proportion of
schoolmates
of the other race in columns 3 and 4,1.and the measure of
segregation in
13column 5, The first seventeen of these are in the 100,000+
size category
in Tables 3 and 4; the last five are in the 25-100,000 class. In
only three-
cities (Columbus, Boston, and San Diego) did the average black
child have more
,than a quarter of his achoolmateswhitynd in only six
cities(Philadelpbia,.
Detroit,.Baltimore New Orleans, New York, and San
Francisco,,pgcluding
Washington, D.C., which is an aberrant case, almost racially
homogeneous) did
the 'average white child have more than 15% of hia schoolmates
black. This
v,law degree of contact is reflected by the segregation
measures, eight of which
are .80 or above, and only three of which are beloil .60. These
figures re- c.
emphasize what Table 4 shows: that segregation in large cities
in 1968
was not concentrated in any region of the country, but appeared
to a similar
degree. in all regions.
Altogether, the picture of racial segregation in U.S. schools in
1968
is one with several components:
13. .
,
-These 22 largest central city. school districts are classified
accordingto 1972 enrollment and So Office of Education metropolitan
status 'Class-ification. They represent 22 of the 23 largest
central city districts;Albuquerque is excluded (the 22nd largest)
becaUse it is not among the .largest 50 cities' in total.
population. , ,
1-.
ti
-
24
Table 6: BLACK-,:IHIT CONTACT MID SCHOOL SEGREGATION IN
1968FO-R- 22 LARGEST CENTRAL CITY SCHOOL DISTRICTS
(Districts ranked by 1972. Enrollment)-
.
. .
Proportion-
Schoolmates. ,,
Segre-gatiotwithindistrict
.
1While 2BlackWhites foraverage black
Blacks foraverage white
1. New York
2. Los Angeles.
3. Chicago
4. Philadelphia1
5. Detroit
6. Houston
T. Baltimore
8. Dallas
9. Cleveland
10. Wash., D.C.
11.« Memphis..
12. Milwaukee
13. San DiegoP
14. Columbus .-
--,
15. Tampa ...'
/6. Se..Louis
17. New Orleans.
18. Indianapolis
19. Boston
20. Atlanta
21. Denver
22. San Francisco
'
..
.44
6. /
.54
.3&
.39
.39
N.53
.35
.61
.42
.06
.46-
v.73
.76
.74
.74
.36.
.31
.66
.68
.38
.66
.41
;
.31'
.23
\ .53
.59
. .
.59
.33
.65
.31
....56
.93.
.54
.24
- ,-,12
.26
.19
.64
.67
.34
.27
-..16i
414
.28'
.
_.
-
'.31
.67.
..05
.14"..
'.I.
.06
.10
.06
.06
.03
,04
.18
.26
.30
..16
.07
.09
.22
..27
06,*
.20
.25
)
.
0
r
.17
.03
.08
., ..2L
.20
Y.04
.19..c
.03
.09 --
.44
.04,
.06
.04
.10
.04
.12
.' .19
.11
.11
.09
.04
.17'
.
...
,
.
e
.
4'
-
.
'
.47
.86-'
.86
.64f
.66
.89
al.91
.85
.53, ..
.92
.75
:66
.60
I
.78
.82 ,'
.72
.67
.60
.85
.69
.38
-
9
49 /
-
4 . 25
1. High segregation in the largest cities of the country, where,
the
proportions of blacks are greatest;
Sharply lower segregatiod in smaller'districts
everywhefe-butqhb.
South (and slightly lower there),
blanks in these smaller districts
but much smaller proportions of
-- except in the South;
3. A large contribution.to total segregation in some northern
regions --
due to blacks and whites living in diff4rent districts, so that
the
'difference in total segregation between North and South is
consider-
ably less than their diff,erence.in segregation within
districts;
4. Greater. segregation, at elementary than at secondary levels,
'due
_ aCleast in part col the smaller, more homogeneous areas
served_by
elementary schools.L . .-
5. A seeming,paradox: the region with the highest degree of
segrega-
-tion, the Southeast, is also the region in which thelaverage
white
child had the highest proportion of black schoolmates (.07).
The
reason, of course, lies in the higher proportion of blacks in.
the
Southeast., .......
..,
.. :;
''. It is clear from .these data that by 1968,-desegregation.of
schools was
a far from accomplished task in cities and/towns of_all sizes in
the South;
but that in the largest cities it
school systems had never existed.
Pthe major thrust of desegregation
was equally high in many places where dual
But this was the picture in 1968, before
id schools had'occurred: The next four
years show strong trends toward desegregation. It is these
trends to which
. We now turn.
... 50 4
-
.,
27
TRENDS EN.WITHiN-SYSTEM SEGREGATION
Between'1968 and 1972, there was a sharp reduction in black
-white
segregation in the United States. In 1972, 16% of public school
children
Were black, and 77% white. The average'black ckild in 102 went
to a school
that was 61% black compared to 74% in 1968) and 34% white. And
the average
majority white child was in a school which was 89% white and 7%
black: The
"comparison below shows the change from 1968 to 1972:
1968
.
1972
Proportion»
, SchoolmatesBlack -white
I segregation
white black
.
whites for .average black ,
blacks foraverage white
.
,
witdis
inrict
.,.,
Total
.79
.77'
".15
;16
.
.
,22
,
.34
.04
.07 -
.63
.37
l
.72.
.
, .56
The change from 1968 to 1972 is substantial. Indeed, the averaie
with-
"in- district segregation in 1972 between blacks and whites may
not be greater
than that between some, pairs of white ethnic groups. But the
change from
168 to 1972 consists of very different changes in,different
locales. or
reference in making comparisons with Table 2, Table 7 shows the
irit rracial
-
Table 7.
BLACK - WHITE CONTACT AND
SCI;
4.,001 SEGREGATION IN 1972.5fRiGION
0.
.,
Prollortio
Schoolmates
Black,-White
segregation
IWhite
2Black
,-;White's for
'average black
Blacks for
'average wh1te
Within
district
s'Total
U.S.
New England
.,
,
Middle Atlantic
Border,
South
t
West"South Central
.
East. North Central
West North Central
Motilatain
Paciff.c
4,1
.77
.92
.78
:79
.68
.76
.86
.89
.80
.75
.16
.06.
.16
.21--
.30
.17
.13
.09
.03'
.08
-%
,
.34
.47.
.28
.28
.28
.
.28
.30
.
.52
.29
.
`
.07
.03
.b6
.07
-.19.
.06
.03
.02
.03
,
, 2. .
\,
e37
.
.33
.43
'.44
.19
.48
.57
.56
-....
-.-
.25
...42
,..6.5
.'.;-35
.56
.49
.64
.35
.63
.67
,
.-6
.61
I.
-
29
contact and the segregation in each of the regions. But the
varying changes
"
can best be seen via a graph. figure 1 shows the trends i \...n
average segre-. ,
. , .
gation within school districts(comparable to column 5 of Table 2
and Table,- , ,
7) in each region over the three points in time, 1968, 1970, and
1972. There
is a radical drop in the Southeast; froth highest at .75 in 1968
to lowest