 

AGE-GRADE RETARDATION AMONG
ﬁEXlCANe AMERRCAN YOHYHS
IN HERMAN

“mat: {or “N Dogma 05 M, A.
MICHIGAN STATE UNIVERSE”

William Patrick Boss
1972

 

’I’HESI:

University -

 

ABSTRACT
AGE-GRADE RETARDATION AMONG MEXICAN-AMERICAN
YOUTHS IN MICHIGAN

By

William.Patrick Boss

This thesis deals with the measurement and analysis of the concept
age-grade retardation among a sample of Mexican—American youths in.the
state of Michigan.

The data is a secondary analysis of data collected in November and
December of 1967 and January 1968 for a study sponsored by the U.S.
IEpartment of Labor.

The results indicate that the MexicanPAmerican children are relatively
disadvantaged in terms of age-grade progress when compared to the population
of Michigan from the 1960 census. The Mexican-American children are
comparable to the Michigan non-white population of 1960 in terms of their
retardation.

The remainder of the analysis attempts to explain the variation of
age~grade retardation among the Mexican-American.population. The
independent variables which appear to explain the most variation in the
dependent variable, age-grade retardation, are, (1) the head of household's
educational attainment, (2) the head of household's income, (3) language
spoken at home, (A) the migrant stream experience of the head of household,

and, (5) how long the child has been in the Michigan school environment.

AGE-GRADE RETARDATION AMONG MEXICAN—AMERICAN
YOUTHS IN MICHIGAN

By

William.Patrick Boss

A THESIS

Submitted to
Michigan State University
in.partia1 fulfillment of the requirements
for the degree of

MASTER OF ARTS

Ibpartment of Sociology

1972

* ‘.
,’ ,
&'
in.
_J"\‘ ,9
\

:5

ACKNOWLEDGMENTS

I wish to express my appreciation to Professor Grafton D. Trout Jr.
and Professor Harvey M. Chouldin, for providing me use of their data
for this thesis and for their suggestions and encouragement while
pursuing this research.

My thanks are also extended to Professors J. Allen Beegle and

Wilbur Brookover for their helpful comments.

ii

Chapter

TABLE OF CONTENTS

INTRODUCTION . . . . . .

THE CONCEPT "AGE-GRADE RETARDATION"

DATA COLLECTION

SOURCES OF BIAS IN OUR CONCEPT "AGE-GRADE RETARDATION

COMPARISON OF RETARDATION RATES

Age-Grade Retardation as a "Risk" Variable .
Comparison of Michigan Total, Non-White and Sample

Mexican-American Retardation .

SEX AS A FACTOR IN “AGE-GRADE RETARDATION" .

VARIABLES ASSOCIATED WITH "AGE-GRADE RETARDATION" AMONG

MEXICAN-AMERICANS . . . . . . . . . . .

Family Characteristics

Education of the Head of the Household .
Family Income .

Occupation and Socio- Economic Status
Family Size . .

Educational Aspirations

Cultural Factors

Language Spoken At Home.
Assimilation Preference

Migration Variables

SUMMARY

Number of Moves .

Birth Place and Place of Early Residence
of Head . . .

Recency of Arrival . . .

Number of Years in the Migrant Stream

CONCLUSION .

iii

Page

.P'

«a «a -q \n

10

Table

10

ll

l2

13

1h

LIST OF TABLES

Percent Age-Grade Retarded by Age . . . . .
Percent Age-Grade Retarded by Age and Ethnicity .
Percent Age-Grade Retarded by Age and Sex

Percent Age-Grade Retarded by Age by Educational
Level of Head of Household . . . . . . . . . . .

Retardation by Age and Family Income
Percent Age—Grade Retarded by Age by SEI . . . .

Percent Age-Grade Retarded by Age and Language
Spoken.Among Adults . . . . . . . . . . . . . .

Percent Age-Grade Retarded by Age and Language
Spoken With Children . . . . . . - . . . . . .
Percent Age-Grade Retarded by Age by Assimilation
Preference . . . . . . . . . . . . . . .
Percent Age-Grade Retarded by Age by Assimilation

Preference Controlled for Language Spoken by Head
with Other Adults . . . . . . . . . . . . . . . .

Percent Age-Grade Retarded by Age by Year of
Arrival in.Nﬁchigan of Head of Household

Percent Age-Grade Retarded by Age by'NUmber of
Years in the Migrant Stream . . . . .

Percent Age- Grade Retarded by Age by Years in
the Migrant Stream . . . . . . . . . . . .

Summary of Characteristics Typical of Age-Grade
Retarded and Non-Retarded Children

iv

Page

12

11+

17
18

22

23

26

28

31

33

35

38

LIST OF FIGURES

Figure Page
1 Graphic Representation of the Meaning of
Age-Grade Retardation . . . . . . . . 3
2 Percent Age—Grade Retardation by Age for Nﬂchigan
. . . . 11

Total, Non-White, and Mexican American

INTRODUCTION

Various studies have shown that Mexican-Americans have lower levels
of educational attainment than other segments of the U.S. population.
Julian Samora has suggested that the educational system in the United
States is one of the major vehicles for social mobility, but that
Nbxican-Americans have not taken full advantage of this opportunity.1
Our concern at present is to attempt to discover what factors cause some
Nbxican-Americans to progress through the educational system and what

factors keep others at a relatively low or retarded level.

THE CONCEPT "AGE-GRADE RETARDATION"

The basic idea for the concept "age-grade retardation" is adopted

from Bernert's America's Children:

 

The school experience of children and youth in the United
States is basically one in'Which the pupil enters a school
system in a certain minimum age level and from then on is
expected to proceed within the system from grade to grade

at a regular pace of achievement. It is generally accepted
that the pupil is to spend one year in each grade, that in
eight years he shall have progressed through eight grade
levels, and by the end of a twelve-year period he shall have
completed his elementary and high school training. But this
is not always the result. Although the general pattern of
achievement and experience may be one of yearly promotion to

 

1. Julian Samora and Richard A. Lamanna, "NbxicaneAmerican in a
Midwest Metropolis: A Study of East Chicago", Mexican-American Study
Project, Advance Report, 1967, University of California, Ios Angeles,

p. 51.

2

the next grade, there are pupils who move fast and "skip"

grades, and there are pupils who are held back and have

to repeat grades.

. Retardation is defined as a slowness of progress
through school as a result of nonpromotion--a lagging behind

from the expected pattern of progress through the school

system.2

Samora's study of Mexican—Americans in East Chicago reports a high
rate of age-grade retardation. He suggests that Mexican—American
children are older than the modal age for each grade level, and that
they, as a group, quickly fall behind their age peers in school.3
They are said to have difficulty in the system.because of language
problems, discrimination, and a vast number of other social and cultural
problems. Not only is their final educational level lower, but also at
each grade level they tend to fall further behind.

Normal progress is defined as progression from grade to grade for
each year a child progresses in chronological age. If a child enters
grade 1 at age 6, he is expected to enter grade 6 at age 11, grade 10
at age 15, etc. If age and year are plotted on a graph, our ideal is
represented by a straight line. If a child falls below the line, then
he is age-grade retarded; if he is above the line, he is educationally
ahead of his age peers, i.e., accelerated. Thus, we may view age-grade
retardation as performance which is not in line with the expected age
and grade level.

The question then is what is the expected age and grade level?
Bernert's census monograph suggests that, "pupils in specific age groups

are not to be found in one grade, but usually spread out over two

~ A
adjacent grades." Therfore an 8 year old is expected to be enrolled in

 

2. Eleanor H. Bernert, America's Children, New York: Wiley, 1958,
p. 1M8.

 

3. Samora and Lamanna, op. cit., p. 65.

h. Bernert, op. cit., p. lh9.

F‘F’F’
uasrv1

Accelerated

F’P‘
rein

----Idea1
// .Age-Grade
Progress

   

—--Age Minus 6 Allowance
///_

 

Education

Age-Grade Retarded

H
Ol—‘fUUJ-P'U'IGQCDOO

 

 

5 6 7 8 9 1o 11 12 13 1h 15 16 17 18 19 20

Figure 1.

Graphic Representation of the Meaning of Age-Grade Retardation.

R

either the second or third grades; a 9 year old is expected to be in
grades 3 and A, etc. The formula then is age minus 6 years. If that
number is greater than his grade level, he is retarded; and if that
number is less or equal, then is is not retarded. For example, a 15
year old must be enrolled at least in the 9th grade (age (15) - 6 = 9)
to be non-retarded. If, however, he is only enrolled in the 8th grade,

then he is retarded.

DATA COLLECTION

The present paper is a secondary analysis of data collected during
November and December, 1967, and January, 1968, for "Mexican-Americans
in Transition, Migration and Employment in Michigan Cities" by Harvey M.
Choldin and Grafton D. Trout, Jr.5

This research project used a sample of Mexican-American families
in eight counties outside the Detroit SMSA. After the eight counties
were selected, the Nbxican—American households within them.were identified
and listed by contacting informed persons in various organizations and
through the use of city directories and telephone books from which persons
having Spanish surnames were listed. A pilot study indicated that the
two main errors in the lists were the non-locatable addresses and names
of non-MexicanpAmericans. Names were then selected from these lists
for interviews. The interviewing was performed by bilingual and
biliterate MexicaneAmerican residents of the counties under study after

receiving interview training.

 

5. Sponsored by the U.S. Department of Labor, Office Manpower Research,
Contract No. 8l2h6632.

SOURCES OF BIAS IN OUR CONCEPT "AGE-GRADE RETARDATION"

We wish to compare retardation rates of our Mexican-American sample
with those of other populations. Are these really comparable? The
problem of what date a sample or census is taken arises. Census data
are collected on April 1, while the sample data on MexicanaAmericans
used here was collected over a three-month period from November 1967
through January 1968, thus the census population was given from three
to five months more time for its children to have birthdays and become
one year older. This extra year of age could in many cases lead to
classifying a student retarded. The problem is that our sample of
Mexican-Americans was not allowed this three to five months to become
older and in some instances to be therefore classified as age-grade
retarded. It seems that approximately one-third to one-fourth more of
those students in the lowest expected grade would be classified as
age-grade retarded if the sample had been taken as of April 1. Thus,
in an effort to make our survey data more comparable with U.S. census
data, 25 percent of those students in the lowest accepted non-retarded
grade were classified age—grade retarded. In other words, we are
assuming that 25 percent of those students who are just one year above
non-retarded status will have a birthday during this three to five month
period and thereby become classifiable as retarded.

The measurement of age-grade retardation with this survey data on
Nbxican-Americans poses another problem. The interview included age
and year completed in school for every child in every household; and not
year enrolled in school. In order to make year completed equal to year
enrolled, the definition must be changed to age minus 7 for our survey

data.

6

A second source of bias is the comparison of‘populations at
different time periods, i.e., 1960 population compared to a 1967 popu-
lation. From 1950 to 1960 there was a decline in the rates of age-
grade retardation.6 We probably can assume that the population of 1960
will be less retarded as of late 1967 or that a 1967 population would
have been more retarded as of 1960. This form of bias is non-correctable.
It also appears that our entire sample has an urban bias and it has
been shown that urban retardation rates are lower than that for rural
areas.

To sum.up, it seems that we are under-representing Mexican-American
age-grade retardation rates by some factor. This must be kept in mind
When we compare other populations with our Mexican-American sample.
However, we have attempted to alleviate the major flaw in our measure

and feel that the measurement, although not perfect, is very useful.

 

6. James D. Cowling, Age Grade School Progress of Farm and Non-Farm
Youth: 1960, Economic Research Service, U.S. Department of Agriculture,
Agriculture Economic Report No. MO.

 

 

7. Ibid.

 

COMPARISON OF RETARDATION RATES

Age-Grade Retardation as a "Risk" Variable.

 

Table 1 indicates that age is one of the most important factors in
age-grade retardation. .At younger ages, we find smaller proportions
retarded and as age increases, especially in the teens, retardation
rates start to expand at a rapid rate-~doubling from group 11-13 to
lh-l6, and more than doubling from group lh-l6 to 17—19. Retardation
is clearly a "risk" conditioned variable since each additional year of
school presents one more possibility of failure of promotion to the
next grade. Therefore our analysis of associated variables will be
done separately for each of the low age groups or "risk cohorts." We
cannot say at this point if MexicaneAmericans have high retardation

rates unless we compare these rates to other populations.

Comparison of Michigan Total, Non-White and Sample Nbxican—American

 

Retardation.

 

Table 2 includes all people enrolled in school plus all those not
enrolled between the ages of 1h-19. The survey data for MexicaneAmericans
does not ask only about those enrolled, but asks about all people in the
household, thus dropouts must be added to those enrolled in order to
make the Michigan population comparable to our MexicaneAmerican sample.
By including dropouts, the rate of retardation is increased from 8.7
percent to 10.3 percent in the lh-16 age group, and from 15.5 percent
to 2h.5 percent in the 17-19 age group for the Nﬁchigan.population. For
the nonewhite population, retardation rates go up from l9.h percent to
20.9 percent in the lh-l6 age group, and 35.0 percent to h2.2 percent
in the 17-19 age group. What this combination of data suggests is that

retardation may be one of the prime causes of dropping out.

7

Percent Age-Grade Retarded by Age.

 

 

 

 

Table l.
Retardati n
Age % Adjusted % Base N
8-10 8 10 (369)
11-13 8 10 (382)
1h-16 16 20 (329)
17-19 37 M1 (267)
Total Base N (13h7)
8. Both sets of retardation.percents are given because both are

significant. For purposes of comparisons we need a more refined measure,
thus the adjusted figure. However, in cross-classification analysis we
use the non-adjusted figures.

Table 2. Percent Age-Grade Retarded by Age and Ethnicity.

 

 

Michigan Non-White Mexican-

Age Population Michigan Population Americans
% Base N % Base N % Base N
8—10 h (h8u,oh2) 6 (50,990) 10 (369)
11-13 7 (A59,7oh) 13 (h2,676) 10 (382)
1h—16 10 (361,h88) 21 (32,832) 20 (329)
17-19 2h (320,990) A2 (26.773) hl (367)
Total Base N (1,626,22h) (153,271) (1,3h7)

 

Source: United States Census of Population, 1960, Michigan, Detailed

Characteristics.

 

It is a combination of Table 101 - Year of

School in Which Enrolled and, Table 102 - Years of School
Completed by Persons 1A to 2A Years Old and Not Enrolled in

School.

10
Mexican-Americans are relatively disadvantaged in terms of their
age-grade progress--clear1y more retarded than the Nﬁchigan.popu1ation.
There exists some similarity between Mexican—Americans and the non-white
population, but Nbxican—Americans are more retarded in the earliest
years--lower in the 11-13 year old age group--and very similar to the
non—white population in the older age groupings (see Figure 2 for a

graphic illustration of this phenomena).
SEX AS A FACTOR IN "AGE-GRADE RETARDATION"

Is age-grade retardation a male phenomena? Bernert states that,
”backwardness in age-grade school progress occurs to a considerably
greater extent among males than among females."9 For all children
8-18 years old in the U.S. in 1960, the sex ratio is 103 males for
every 100 females. The sex ratio for retardation is 1A9 males per
100 females.10

Table 3 shows that Nbxican—American age-grade retardation is
consistent with Bernert's finding regarding the total population. At
each age group males are more retarded than females. The sex ratio
for our sample is 99 males per 100 females. The retardation sex ratio

is 250, 115, 135, and 118 for the respective age groupings; and 133 for

the entire sample.

 

9. Bernert, op. cit., p. 71.

10. Ibid., p. 71.

ll

%
ho

3o
20
10 w
o IIIEEE

8-10 11-13 1h-16 17-19

 

 

 

 

Age Groups

 

 

 

Michigan

I Non-White

Mexican American

 

 

 

 

 

Figure 2. Percent Age-Grade Retarded by Age for Michigan Total,
Non-White and Mexican American.

12

Table 3. Percent Age-Grade Retarded by Age and Sex.

 

 

 

 

Sex

Age Male Female

% Base N 47% Base N
8-10 11 (187) h (189)
11-13 8 ( 191 ) 7 (196 )
11+-16 19 (162) 1h (170)
17-19 38 (138) 3A (131)
Total Base N11 (678) (686)

 

 

11. The total sample size will vary on account of missing data for
each independent variable used.

 

 

VARIABLES ASSOCIATED WITH "AGE-GRADE RETARDATION".AMONG

MEXICAN-AMERICANS

Age-grade retardation will be the dependent variable in the following
analysis. The independent variables will be attributes of the heads of
the household which may affect school performance. Since age is so
strongly related to age-grade retardation, it will be held constant,
as a control variable in all our cross-tabulations. Also, the explanation
of age-grade retardation may be different for the different age groups,
i.e., one variable may explain retardation for the older groups, but
not the younger or vice versa. Thus the control for age is considered
very important.

The results will be recorded according to the grouping of our
independent variables into three classes of variables: (1) Family

Characteristics, (2) Cultural Factors, and (3) Migration Variables.

(1) Family Characteristics.

 

To what extent does a favorable family setting aid a child's
ability to ”stay up" in school?

(a) Education of the Head of the Household. Does the education

 

of the parents--in this case the head of the household—~influence the
school performance of their offspring? Do better educated heads have
children who are less likely to be age-grade retarded?12

In general, the data indicated (Table A) that better educated heads

have children who are less retarded. Those heads with a high school

education or greater--12+--have significantly less retardation among

 

12. Patricia Cayo Sexton, Education and Income, The Viking Press
Inc., 1961.

 

13

1h

Table A. Percent Age-Grade Retarded by Age by Educational Level of Head
of Household.

 

Educational Level of Head
Age 0 l-h 5-8 9-11 12+
%7 Base N %' Base N %’ Base N %’ Base N %’ Base N

 

 

 

 

 

 

8-10 3 (37) 12 (106) 5 (137) 10 (50) 5 (37)
11-13 2 (N5) 10 (112) 11 (1AA) h (N8) 0 (32)
1h-16 25 (52) 2A (102) 12 (125) 7 (29) o (21)
17-19 35 (A9) ho (86) 37 (86) 38 (29) 2h (17)

Total Base N (183) (hO6) (A92) (156) (107)

 

15
their children than does any other group. A close inspection of Table A
indicates that the relationship is not perfectly linear. The children
of heads having no education appear to be doing quite well, especially
at the younger ages--only 3 percent and 2 percent retarded at the 8—10
and 11-13 age groups. Joseph Kahl13 suggests that among working-class
people there are those who feel they are "getting by" and put little
pressure on their children to do well, while there are those who View
themselves as partial failures because of a lack of education.
"Consequently they encourage their sons to take school seriously and to

"1“ Kahl's sample of wokingmen's sons and our sample

aim for college.
of MexicaneAmericans may not be comparable, but the notion of‘pressure
to succeed and educational failure of’parents may be significant for
both groups. Perhaps those heads with no education may apply more
pressure for their children to do well than do the heads with low
educational achievements (l-h or 5-8). Similarly higher-educated heads
will apply more pressure to do well.

When mother's education is examined, no significant relationship
with age—grade retardation is found. However, those mothers with no
education have children with lower retardation rates, especially in the
younger age groups. The 8-10 year old group has only 3 percent
retardation and the 11-13 group, A percent. This finding seems to give
added confirmation to our notion that parents with no education are

influencing their children, perhaps by negative example, to do well in

school. This appears true especially among young children.

 

13. Joseph A. Kahl, "Educational and Occupational Aspirations of
Common Man Boys", in Harvard Educational Review, Vol. 23, 1953, pp. 186-
203.

 

1h. Ibid., 13. 19h.

16

Kahl’s work suggests that the mother is a crucial motivating force.
The Mexican-American culture is predominantly paternal in orientation
and only after a period of acculturation can we expect the mothers
influence to play a more pervasive role.

(b) Family Income. Is family income related to educational

 

achievement? Do MexicaneAmerican children from.higher income families
have a lower rate of age-grade retardation? Our data answers both these
questions affirmatively. Higher family income is strongly and inversely
related to retardation.

Family income and head's education among our sample of Mexican-
Americans is not highly inter-related. Both of these variables possess
independent explanatory strength, and it is our contention that together
the two variables have a cumulative or additive effect. Those families
with high income and high education among their Children.wi11 have less
retardation than either high income or high education alone.

(0) Occupation and Socio-Economic Status. What effect does

 

occupational prestige and/or socio-economic status have on a child's
ability to "keep up"? Occupation and socio-economic status were used
as two measures. Occupation is a very difficult variable to interpret,
because the vast majority of MexicanaAmerican heads of households fall
into the blue collar categories. Thus an SEI index to measure socio-
economic status was used with a range from 1-99. This was considered a
better indicator, and the retardation rates were more easily inter-
pretable.

Table 6 suggests that heads having higher SEI's have significantly
less age-grade retardation among their children. It must be noted,
liowever, that the SEI measure explains little variance in the retardation

:rates at the earlier years.

17

Table 5. Retardation.by Age and Family Income.

 

Family Income

 

 

H

 

Age less than 37,000 $7,000 plus

% Base N %7’ Base N
8.10 12 (165) h (200)
11-13 12 (16h) 5 (213)
1h-16 2A (156) 9 (169)
17-19 #5 (132) 30 (127)

Total Base N

(617)

(709)

 

18

Table 6. Percent Age-Grade Retarded by Age by SEI.

 

 

SEI
Lower Low Middle Low Upper Low Low Middle

Age (1-10) (ll-20) (21-30) to Upper (31+)
% Base N % Base N % Base N % Base N

8-10 11 (57) 8 (96) 7 (128) 7 (59)

11-13 9 (A5) 8 (111) 9 (139) A (57)

lh-l6 22 (A6) 21 (85) 11 (105) 10 (52)

17-19 A7 (38) A3 (61) 35 (9h) 22 (37)

Total Base N (186) (353) (N66) (205)

 

19
When we controlled for family income, head of household's education,
recency of arrival in Michigan, and number of years in the migrant
stream; SEI’S ability to predict retardation remained. Thus, we have
some confidence that the socio-economic status, or more simply the
occupation, of the head influences their children's ability to "keep
up" in school.

(d) Family Size. Do larger families have greater retardation

 

rates? We wished to explore the relationship between family size and
retardation in order to give some empirical confirmation to the
relationship between family size and age-grade retardation.

Our data does not give us any clear answers. There exists neither
a linear relationship between family size and retardation, nor any
easily interpretable relationship at all. The only indication from our
data is that children from families of 3 and h seem to be slightly less
retarded than other family size groups (0-2, 5—6, and 7+). We view
family size as a weak.predictor variable.

(e) Educational Aspirations. Do children.whose heads aspire to

 

a college education have less retardation than those children whose

heads aspire less than a college education? First of all, it is important
to point out that among our heads of households almost 90 percent

indicate a desire for their sons to go to college. Mexican-Americans,
like other Americans, say they desire their children to go a long way

in the educational system. This may simply be an expression of an
American society norm, i.e., universal higher education. Retardation
rates of children among college-aspirant heads and non-college aspirants,
show little significant differences. Aspirations as measured in this

survey appear to be meaningless as a predictor of retardation.

20
Although a child's retardation level does not appear to be related
to the educational aspiration of his parents, this finding does appear
significant. It is significant in that there exists a real gap between
reality and aspirations. The educational system is failing to educate
Mexican—American children when their parents desire this education.
The failure of the sdhool and of their children thus becomes a major

area of concern for Mexican-American.parents.

(2) Cultural Factors.

 

To what extent is the retention of Mexican or Mexican-American
subcultural traits by the parents detrimental to a child's age-grade
progress? Is the desire for assimilation toward the "American Style"
helpful in a child's ability to keep up to his age-grade peers?

(a) Language Spoken At Home. If a different language is spoken

 

in the home from that which is used in school, it seems that this might
be a detrimental factor in a child's ability to perform in English in
school. Thus, English-speaking in the home should yield lower rates of
retardation, while Spanish-speaking may handicap a child's ability to
progress in the educational system. What about bi-lingualism in the
home? Are children of parents who speak both Spanish and English
intermediate in rates of age-grade retardation?

In our survey, three questions concerning language at home were
asked of the head: (1) What language did you speak with your parents
when you were a child? (2) What language do you usually speak with
other adult members of your family here at home? (3) And what language
do you usually speak with your children.here at home? In an attempt
to discover what language is spoken in the home at present and how this

will affect retardation, we excluded the question concerning language

21
spoken by the head with his parents when he was a child. If we used
only the question concerning language used with children, we were
afraid that we might be masking some forms of bi-lingual behavior.
In some cases subjects might well speak English to their children, but
might speak Spanish or a combination of the two to other adults. An
analysis of both of these questions will tend to give us more meaningful
answers.

The data indicates (Tables 7 and 8) that those heads who speak
only Spanish have children Who have higher rates of retardation. Also,
English or a combination of both languages are superior to only Spanish.
Finally, the differences between English and a combination of the two
are neither great nor consistent. The first two conclusions are
expected, i.e., speaking mostly Spanish by the head leads to higher
retardation by his children, while speaking mostly English or a
combination leads to lower retardation. The third, however, is intriguing;
and worth looking at for a moment.

When we view language used with other adults, we see that speaking
both is equal to if not better than English for the age group 8-16, and
only in the 17-19 age group does English show a significant margin over
use of both. When we look at language spoken with children, the same
pattern emerges. "It is virtually impossible to avoid the conclusion
that children should be started off in their formal education in their
mother tongue . . . the evidence is overWhelming that the home-language
should be the springboard for the proper development of the second
"15

language. Young Mexican-American children raised in a Spanish-speaking

 

I

15. George I. Sanchez, "History, Culture, and Education,’ in
Julian Samora (ed.), La Raza: Forgotten Americans, University of
Notre Dame Press, 1966, pp. 19—20.

 

22

Table 7. Percent Age-Grade Retarded by Age and Language Spoken.Among

 

 

Adults.
Spanish Both English
Age % Base N % Base N % Base N
8-10 12 (156) 6 (12h) 3 (89)
11-13 11 (158) h (130) 8 (89)
lh—l6 2h (lu2) 10 (125) 11 (62)
17-19 M5 (137) 33 (8h) 20 (A6)

Total Base N (593) (M63) (286)

 

23

Table 8. Percent Age-Grade Retarded by Age and Language Spoken With

 

 

Children.
Spanish Both English
Age % Base N % Base N % Base N

8—10 6 (80) 10 (121+) 6 (16h )
11-13 9 (67) 7 (151) 9 (16h)
lh-l6 28 (75) 12 (131) 15 (123)
17-19 38 ( 85) M1 (97) 25 (81)
Total Base N (307) (503) (532)

 

2h

home need time before they can read adequately and gain the verbal skills
required to use the newly required language.

When language spoken with other adults was controlled by other
relevant variables, it retained its explanatory strength. However,
some interesting points emerge. It appears to be very important for
the most recent arrivals (1960's) to speak English or their children
will be at a great disadvantage. It also seems that head of household's
education and language have a cumulative effect, but the cell sizes are
small and this conclusion is tentative.

In an attempt to probe deeper in the question of language, we
explored whether it was more advantageous for one's children if the
head had learned EngliSh before coming to Michigan, or after his arrival.
Our data indicates that it makes no difference at all where the head
learned his English.

(b) Assimilation Preference. We constructed a variable to indicate

 

attitudinal preferences for group homogeneity among our Mexican-American
sample. The variable, assimilation preference, is an inverse measure
of cosmopolitanism, i.e., those who are low on assimilation preference
are high on cosmopolitanism, while those high on assimilation.preference
are more ethnocentric in orientation. In an attempt to differentiate
between a preference for in-group and out-group interaction, respondents
were asked to rate their preferences regarding the following ten items:

1) Have mostly Mexicanquerican friends.

2) Have your children play mainly with other Mexican-American

children.
3) Have a Mexican-American as a next door neighbor.
h) Have a majority of Mexican-Americans in your (neighborhood)

(farm area).

25

5) Have a majority of Mexican—American children in your children's

schools.

6) Have a church where most of the members are MexicansAmerican.

7) Have a majority of Mexican—Americans in any organization of

which you are a member.

8) Have a separate political organization for NbxicaneAmericans.

9) Have your children only date Nbxican—Americans.

10) Have your children marry only Mexican-Americans.

On each of the items, the response was a preference score from 1 to
7. A score of 7 indicates the strongest preference, while a score of 1
indicates the strongest opposition. The respondents were assigned a
score on this variable by an addition of his scores for all ten items.
Thus the range of scores is from 10 (all l's) to 70 (all 7's).

The distribution of actual scores indicates a high proportion in
the middle range (ho-N9). This seems to be the typical response set.
Thus, for analysis we divided assimilation preference into three
dimensions: high (a score of 50 or greater), medium (a score of h0-h9),
and low (a score of less than ho).

Are those who are more favorable to separate Mexican social inter-
action more likley to experience age-grade retardation among their
offspring?

Our data suggests they are. By reading across Table 9, we see that
the relationship is linear. Those heads low in preference for Nbxican
exclusivity have children with retardation rates which are lowest for
each age group. Those who are high have the highest retardation rates

for each age group, with the medium group in between as expected.

26

Table 9. Percent Age-Grade Retarded by Age by Assimilation Preference.

 

Assimilation Preference

 

 

 

 

Age High Medium Low

% Base N % Base N % Base N
8.10 7 (122) 7 (172) N (57)
11-13 9 (133) 8 (173) 5 (61)
lN-l6 21 (133) 16 (13N) 9 (N7)
17-19 N2 (95) 36 (122) 2N (37)

Total Base N (N83) (601) (202)

 

27

Is this a spurious relationship? When we control for language
spoken with other adults, some interesting results emerge (Table 10).

When Spanish is the language spoken with other adults, we notice
that assimilation.preference still shows some relationship to age-grade
retardation. However, when English becomes the spoken language, in
combination or alone, we see that the relationship reverses. If one
views the bi-lingual group (probably the best group to view because
the cell sizes are larger and more meaningful) it is seemed that from
the medium to the high assimilation.preference groups there is a drop
in retardation.percents at every age group instead of an increase
which existed in the cross-tabulation of assimilation preference by
retardation.

When family income is used as a control, we see again the same
general pattern. For low family income, assimilation preference seems
to hold up, but in the high family income group, the relationship
vanishes and seems to reverse itself.

We do not view assimilation preference as an independent explanatory
variable. It acts as a conditional or intervening variable, while
family income and language are the more powerful explanatory variables.

As a next step we wished to see if composition of the community
would have an effect on our dependent variable. Ethnic density, i.e.,
whether the neighborhood is NbxicaneAmerican, Anglo, or Negro, was one
variable; and number of MexicaneAmericans on the block was another
variable. The data indicates that there exists no relationship between

these variables and age-grade retardation.

 

28

Amen image Aoav Aoeav Amman Aron Amoav Ammmv Amen a seam

 

 

 

 

ﬁance
Amav mm Amav em Amav HH AHmV mm Amev om Ami mm Amov ms Ammv mm AOHV em mH-AH
Amﬁv m Ammv m Aamv ma Acmv 0H Amov ma AJHV o Amsv mm Ammv Hm ANHV o mH-:H
Ammv ma Aemv m Ammv m Awsv m Assv : AAHV o Aosv m Awmv ma Amav s mH-HH
Aamv m Aeev o Ammv m Asav o Asmv a Aemv 0 Away ma Aasv Ha ANHV o oa-w
z e z s e s z a z a a a z a z a a a
boom ommm omom ommm ommm ommm ommm ommm comm om<
smug assoc: 30H amen assoc: 30H swam assoc: sou
mononoaohm coﬂpmﬁﬂsﬂmm< oocosomohm qoﬁpoaﬂsﬂmm< oosohomohm noepoaﬁsﬁmm<
anaamem epom amaeoam
owoomsmq

 

.mpasod 90390 apes doom 59 Qoxomm
ommomsmq How ooHHOAPQOQ mosohomopmAQOﬂpmHﬂsﬁmm< an owa an oooaopom cooau|0w< psoopom .OH manna

29

(3) Migration Variables.

 

(a) Number of Moves. What happens to children who have parents

 

who move a great deal? Do these children have higher retardation rates
or lower? We have three variables to see if movers have more retardation
among their children than do non-movers. It is assumed that those
children who have parents who are moving a great deal will be readjusting
to different school systems and may well be put back for a number of
reasons upon reaching a new school system. The three variables we used
are: (1) Number of moves in lifetime, (2) NUmber of moves in Texas,
and (3) Number of moves in Michigan. It appears that these variables
have no effect on the retardation rates of the children. Retardation

does not seem to be determined by how much a child's parents move.

(b) Birth Place and Place of Early Residence of Head. Does birth

 

place of the head or where the head lived longest until he was 16 years
of age effect a child's retardation rates? The distinction seems more
useful than the use of one indicator that, among a population which
moves a great deal, many times birth place and residence until a
certain age may have a low correlation.

We suspected that the heads having a Michigan birth place or being
raised in Michigan would be more advantageous in terms of age—grade
progress than would either Texas or Mexico. Whether Texas or Mexico as
a birth place, or a place lived longest until 16, was more advantageous
was left open.

The data indicates that those heads born in Texas or who lived in
Texas until 16 years of age have a considerably higher retardation rate

among their children than do heads born and/or raised in either Michigan

30
or Mexico. Nﬁchigan and Mexican-born and raised heads seem to have
similar retardation rates among their children. The only firm
conclusion that we draw is that for a child's head of household to be
born or raised in Texas increases his probability of being age-grade
retarded.

(c) Recency of Arrival. Does the time exposed to modern, industrial

 

living experienced in Michigan reduce the retardation rates of children?
If'people have just arrived in Michigan, will they carry values and
beliefs which will be detrimental to their children's school performance?

Table 11 indicates that the number of years a head is exposed to
the Michigan environment affects the age-grade performance of his
offspring. Those with the lowest amount of exposure (1960-1967 group)
have clearly the highest retardation rates. It should be noted that
recency of arrival explains a great deal of the variation for the
younger age groups. That is, recency of arrival is very important in
determining how a young child will keep pace with his peers.

When we control for other relevant factors (family income, head's
education, and number of years in the migrant stream) recency remains
an important factor. Again, when income and number of years in the
migrant stream are combined with recency, we get a cumulative effect.
Late arrival and low income result in higher retardation together than
either one suggests alone, as does experience in the migrant stream and

late arrival.

(d) Number of Years in the Migrant Stream. Approximately 50 per-

 

cent of our sample of Mexican-American.heads of household did not
participate in migrant labor activities. Do Mexican-Americans who were
part of the migrant stream have greater age-grade retardation among

their children than do those who were non-migrant workers? The impression

31

 

 

Table 11. Percent Age-Grade Retarded by Age by Year of Arrival In
Michigan of Head of Household.
Year of Arrival of Head of Household in Michigan
Age 1900-1939 l9Nl-l960 1961-1967
0 Base N % Base N 0/0 Base N
8-10 0 (21) 5 (233) 23 (57)
11-13 N (26) 7 (263) 15 (N7)
lN-l6 18 (28) 12 (222) 39 (NN)
17-19 10 (N2) 30 (165) 55 (38)
Total Base N (117) (883) (186)

 

32
is that children of migrant workers are sporadic attenders of school
and would tend to be behind upon settling out of the migrant stream.
Now that these former migrant workers are more permanent in their
residence and employment, how are their children doing in school?

Concerning migrant agricultural families, Paul Blackwood states,
"The tradition of going to school is not strong among them, nor is the
expectation strong in many communities that these children should go
to school."16 It appears clear that attendance and retardation should
be highly related. Those children who have poor attendance records are
those most likely to be held back. Any experience which develops a
tradition on non-school attendance seems most detrimental to a child's
ability to progress through school at the expected rate. The number
of years in the migrant stream will be associated with the tradition
of non-school attendance. The higher the number of years as a migrant
worker, the greater will be this non-school attendance tradition. It
should be noted that, regardless of the attendance question, we would
hypothesize that the experience of heads in migrant work situations is
not conducive to their children's progress in school.

Table 12 indicates that the lack of migratory experience by the
head of household is clearly advantageous for their child's school
progress. It also appears that the longer the head is in the migrant
stream, the higher becomes his children's retardation rate, to a point
of saturation around ten years. For purposes of clarity, let us view

the differences between no migratory labor experience and some.

 

16. Paul E. Blackwood, "Migrants in Our Schools," in Bernard N.
Meltzer, Harry R. Doby, and Philip M. Smith, Education in Society:
Readings, Thomas Y. Crowell Co., New York, 1958, p. N92.

 

33

Table 12. Percent Age-Grade Retarded by Age by Number of Years in the

Migrant Stream.

 

Years in Stream

 

Age 0 l-N 5-9 10+

% Base N % Base N % Base N % Base N
8-10 N (189) 6 (80) 12 (N2) 17 (58)
11-13 5 (196) 9 (91) 8 (39) 16 (56)
1N-16 8 (172) 20 (71) 39 (33) 27 (53)
17-19 30 (130) 37 (52) 52 (27) N5 (58)
Total Base N (687) (29N) (1N1) (225)

 

3N

The differences of retardation rates of children Who have heads
with no migratory experience is most significant (Table 13). The
differences are large at the younger ages and are greater at the older
ages. Our data does not allow us to check attendance rates, but it
would be of interest to see if attendance rates of children are correlated
with the head's migratory experiences. Again note that the number of
years in the migrant stream is a good predictor at the younger ages.

We view experience in the migrant stream as a predictor of
retardation. After controls are made, migrant stream experience retains

its strength as a predictor variable.

35

 

 

 

 

Table 13. Percent Age-Grade Retarded by Age by Years in the Migrant
Stream.
Years In Stream
Age None Some

% Base N % Base N

8.10 N (189) 11 (180)
11-13 5 (196) 11 (186)
lN-16 8 (172) 26 (157)
17-19 30 (130) N3 (137)

Total Base N (687) (660)

 

SUMMARY

NbxicanrAmerican youth are relatively disadvantageous in terms of
age-grade retardation. In this respect, they are somewhat similar
to non-white Nﬁchigan residents.

In attempting to explain why some Mexican-American children are
more or less retarded, we conclude that there are a number of causes
or determinants of retardation and that this phenomena is only explained
by a combination of factors and that some of these factors are more

important at different age groups.

FAMILY CHARACTERISTICS

 

It appears that income, SEI and head of household's education are
important variables. It is true that these three are interrelated, but
we believe that each is an independent predictor to some degree. Our
data shows that age-grade retardation is disproportionately a male
phenomena as reported in other populations. Mother's education, head
of household's educational aspirations, and family size appear to be
poor predictors. However, none of the family characteristic variables

explain retardation at the younger ages very convincingly.

CULTURAL VARIABLES

 

The number of Mexican-American families on the block, and ethnic
density proved of little importance. Language spoken by the head with

other adults was a powerful explanatory variable. Assimilation preference

36

37
appears to be an intervening factor which helps explain retardation
when family income is low and the language spoken with other adults
is Spanish. We do not find it to have unconditional explanatory

strength.

MIGRATION VARIABLES

 

Number of moves in lifetime, number of moves in Michigan and in
Texas were not associated with age-grade retardation. Birth place of
head and where one lived before age 16 (where one was socialized)
indicated that Texas is the environment most conducive to age-grade
retardation. Recency of arrival in Michigan and number of years in
migrant stream were powerful variables, and of all the independent
variables these two were the most powerful predictors of retardation
at the younger age groupings.

From this list of explanatory variables we can suggest two types
of Mexican-American children; namely the age-grade retard and the

non-retard child.

38

Table 1N. Summary of Characteristics Typical of Age-Grade Retarded and
Non-Retarded Children.

 

Retarded
Mexican-American Child

Non-Retarded
Mexican-American Child

 

Sex

Male

Female

 

Age

17-19

8-10

 

Family Income

low (($7,000)

High (> $7,000)

 

 

Socio-Economic Low (1-20) High (21+)

Index

language Spoken

by Head with Spanish Bi-lingual or English

Other Adults

 

Recency of Arrival

Late arrival (1960's)

Early arrival (Before 19N0)

 

Head of Household's
Education

l-h years (not 0)

High school graduate

 

Number of Years in
the Migrant Stream

5 or more years

No mirgrant stream
experience

 

CONCLUSION

A multi-variate analysis was performed in order to determine which
of our variables produced the greatest amount of explanatory power. The
results from this analysis confirmed the results obtained from cross-
tabulation and the use of partials or controls. Age, income, language,
stream year experience, recency, and head of household's education were
factors which emerged from this analysis. This form of electronic
analysis gave confirmation to our extensive manual searching approach
in the analysis of age-grade retardation.

The one issue we feel most uncomfortable with at this time is the
comparability of our data. We have manipulated data for other purposes
to fit our needs. Some of the techniques used were most interesting,
but for purposes of explanation we would hope to gather data directly
related to the problem of age-grade retardation; and not the use of
secondary data.

The next step in exploring this area might be (I) gather data
directly related to age-grade retardation, (2) attempt to discover
the relationship of age-grade retardation to other phenomena e.g.
dropping out, and (3) examine over time what happens to those children
who have fallen behind in school. What kind of jobs do they get and
what happens to the education of their Children.

Although this paper has been aimed at the explanation of age-grade

retardation among a sample of Mexican-Americans, we feel that our real

39

 

N0
task has been to raise a number of hypotheses which need further
empirical confirmation. Let us hope that other populations are compared
to the Mexican-American sample used in this paper, in order to arrive
at a more convincing level of explanation of the phenomena of age-grade

retardation.

 

f.’);Cir-UG.AN STATE UNIVERSITY LlBRAFHES I
((1: at ma: :1 I,
H I (I) “.H‘NH I I
31293 03196 2008

 

ﬂ‘. ‘- l