THFSlS

 

W“"‘“:i\‘illlll mm-
l\\\\\\\\\\l\\\9l\l\\\\\\ iim , “W“
Michigan State _
l Umvcrsuy

This is to certify that the

dissertation entitled

Maternal Child—Rearing Attitudes and Practices
and Child Behavior in Mexican-American/Chicano Families

presented by

Estella Martinez

has been accepted towards fulfillment
of the requirements for

Ph.D. Family Ecology

degree in

 

 

M ajor professor

 

DateOctober 12, 1984

 

MSU is an Affirmative Action/Equal Opportunity Institution O~ 12771

 

 

IVIESI_J RETURNING MATERIALS:
Place in book drop to
LlBRARJES remove this checkout from
w your record. FINES_ will
‘ be charged if book is
returned after the date

stamped below.

 

 

 

 

JUN 0 2.52004

 

~mL O 8 2003
.\_ fi . A a
:lXOi' 12% 9000

 

 

 

 

 

MATERNAL CHELD—REARING ATTITUDES
‘AND‘PRACTICES AND CHILD BEHAVIOR

IN MEXICAN-AMERIC AN/CHIC ANO FAMILIES

BY

Esxflla.A.lManihez

A DISSERTATION

Submittedto
Mhflfigantazme‘Unhnxsty'
in.pmdialthflfflbment<jfthe:namfireumnus
‘fimrthe<iqpmxacf

DOCTOR OF PHILOSOPHY

Ikgxutmentcf’Fanfihrand Chfld.Ecdkxnr

1984

ABSTRACT
MATERNAL CHILD-REARING HITITUDES

AND PRACTICES AND CHILD BEHAVIOR
IN MEGAN-AMERICAN/CHICANO FAMILIES

BY
Estella A. Martinez

The problan of this study was to describe naternal child-rearing
attitudes and practices, the relationship of such practices to child
behavior, and the relationship of family derographics to the maternal
child-rearing attitudes and practices of Mexican-American (Chicano)
families residing in an urban area of the Midwest. The research took
the form of a ocmbined ex post facto (field) and attitudinal survey
design. It was based upon structured observations of forty-seven
mother-child dyads in their homes and interviews of the mothers.

The majority of Mexican-Marian anthers expressed authoritarian
child-rearing attitudes, however, their practices were divided alnnst
equally between authoritative (49 percent) and authoritarian (47
percent) child-rearing patterns. There was a significant relationship
between expressed child-rearing attitudes and observed child-rearing
practices, x2 (7df) a 15.48, p = .03. Attitude variables, Acceleration
of Autonany, Lack of Control, Casual Use of Time, and Equalitarianism
were found to best discriminate between authoritative and authoritarian
practices. Univariate analysis of authoritative and authoritarian
patterns of maternal child-rearing practices and each of the child
behavior variables were significant only for: Imitation of Mather.

The effect of paternal arployment status on maternal child-rearing

attitudes was highly significant. In families with euployed fathers,

Estella A. Martinez

maternal child-rearing attitudes here 20 percent permissive, 40 percent
authoritative, and 40 percent authoritarian. In families with
unatployed fathers, mthers' patterns of child-rearing attitudes were 7
percent authoritative and 93 percent authoritarian.

I dedicate this disertatim to my .E_‘_a_1_n_fl1_'_a—
especially Gmndma, Mom, Dad,
Kathy and Lily,

Gerry and Baby Joel,
and, i m, Melinda

ii

ACKNOWLEDG MENTS

Many individuals have mearfingfully conudhlted to the educational grocea
ofmydoctoralgrogramaswellastlfisdisertatficn. Dr.Eil.eenEarhart
gradanlysenredadiairpersonofmyguidanceoomnurteeanddissertation
director. Throughout my program she has modeled an enthusiastic, positive, and
optimistic approachtogoblem solving. Aftersheleft Michigan State Univasity
macceptaposifiminanotherstate,Dr.EarhartcontinuedtobehelpfuL Her
generosiryinretnmfingtodiairthea‘alexaminadmindefenseithe
disertafimandamtinuingtoglfidemethmughitsoompledmarevery much
appreciaed. Dr.LilJianPhenicediJigentlyconU:thbedhertimeandaaistance
while Dr. Earhart was out of the country for one term. Cmsequently, Dr.
Phenice'sefifortswererecognizedbynaminghadisertmjmoo-directa'. Iam
especiallygratefultoDr.Phem'.ceforencmraging myprofesn’onalgrowthby
invdfingmemgivingaworkstnpwithleramiforemfinganoppcrmtfityto
co-autlxrapublicafimwithher.

Drs. Damis Keefe, Enen Strommen, and Raywin .Huang also gave
generously ti theirtime and expertise as members of my committee. Dr. Keefe's
shared research interm in Mexican-American families and his theaetical
applicadmcfthefamflyecmystem mdelwereveryhelrfulthroughoutmy
program. Dr. Strommen's outstanding oommenGmtheg'oposalanddraftcftkfis
(fissertadmwereveryeducadumImpemleamtogiveflueldndcffeedback
to athersthat Dr. Strommen has given me. Dr. Huang's coracientious guidance

during the statistical analyses of this research was highly motivating. I am

iii

thankfulto Dr.Huangforhe1pingmetogainabetterunderstarxiingcf
statistical procedures and computer applications.

I also acknowledge Drs. Linda Nelson and Verna Hildebrand for being
great sources cf encouragement and advice. Dr. Hildebrand has thoughtfully kept
meinformedcipotendalptofsmonalopporturfiifies. Inaddifim,Man'a Moon
cartdbutedleracperdsemtheSpmfishu'ansladoncftheresearchmsuumerm
fa‘thisstudy.

Tremomersanddfildrenwhoparfidpatedasaibjectsareadmowledged
formaldnguiissuxiypcsa'ble. Iespeciallythankthemothersandfathersfor
appedadng my research interests in their families. Among several Hispanic
mmmmfityleaderswmidendfiedpctaifialmbfecm,8isterMafiadelcsArgeles
of Cristo Rey Catholic Church and Cmnie Man’n of Cristo Rey Community
Cmterwereparticrflarlyhelpful.

Ihavealsohadinvaluablesipport‘andencozragementfromveryspecial
fderidsaridconeagueamtheirsincsrecmisdansfidLJmnananucmBen
madeitpcaihlefcrtlusdisertaticntobegrocesedmamicrocomputer. Our
Midfiganf_a£li_a,EfrainandFrancisca Marinezaxxiantheirchildrenhave
staredtheirwamthandbveparticulaflyinuienmmrancecfmydaughter,
Mindy. mianarxiDianaareespecianyapg'eciatedfcrtreafingmeatheir
cclleagueandsiSLer-evenifIamfromtiew Meidco! Ivy(Lungie)chukaand
Amy Hunt have been loyal friends. Together we have prayed, studied, nearly
cried,andlaughed. Iapg'eciatetheirnurturancechindyaswell.

Finally,Ihavebeenrichlyblsedbyawmderfunymppcrfiveexteifled
famflydqfitethemilesthathaveseparatedta. Whatsmore,mydearchildren,
eachintheirown way,havebeenasourceofinspirati.m. Toallwhohave
sharedwithmeinthisPh.D.experience,manythanks.

iv

TABLE OF CONTENTS

Listchables

ListiFigures

CHAPTER

I. INTRODUCTION
Statement cf the Problem
Nad for the Study
Purpose cf the Study
Reearch Questions
Theoretical Framework
Conceptual and Operational Definitions
Research Hypotheses
Asumgn'cns

Overview Statement

II. IBERATURE REVIEW
Child-Rearing
Awicfirarian/‘I‘radifional Findings
Pa'miave Child-Rearing Finch’ngs
A Range d Child—Rearing Findings

Literauire Surveys

Page

PH

11
15
16
17

18
18
19
23
28

31

CHAPTER
11'. LITERATURE REVIEW (continued)
Demographic Variables
Maternal Age
Maternal Educational Level
Maternal Generations Removed from Mexico .
Child's Gender

Summary

III. METHODOLOGY
Design
Design Over Time
Population
Sample Strategy and Characterisdrs
The Raearch Instrument-s
Interview Instruments
Background Informatim
The PARS and Ccrnell Attitude Survey
omervaticn Muments
The Matemal Teaching Observation Technique
The Child Behavior During MTOT
Observer Reliam'lity
Validity
Design Over Variables
Statistical Models of Analysis
Z Tet cfSignificance for a Single Sample
Pearson's Chi-Square Test

vi

Page

32
32
34
35
36
38

40
40
40
41
42
43
44
44
45
49
49
S3
53

55
56
56

57

CHAPTER
III. Madndcbgy (continued)
Statistical Models of Analysis
Student's t
Discriminant Analysis
Multivariate Analysis of Variance
Pearson Product-moment Correlation Coefficient
Analysis of Variance
Mahcds Used in the Analyses of Data
Research Hypotheses
Procedures
Ethical Cmsideraticns

IV. RESULTS OF DATA ANALYSES

Hypotheses Tests
Hypothesis 1 Child-Rearing Aditudes
Hypothds 2 Child-Rearing Practices
Hypotha's 3 Aujmdes & Practices
Hypothesis 4 Practice a Child Behavia'
Hypothesis 5 Atfiimdes & Demograptfirs
Hypothas 6 Practices & Demographics

Statistical Resins '

Summary Data

vii

Page

' 57
S7
58
59
59
6O
60
62
64

66

66

70
74
78
80
80
82

CHAPTER
V. SUMMARY, CONCLUSIONS, DISCUSSION, IMPLICATIONS

Simmery d the Study

Cmclmims

Discusa’cn
Raearch Question 1
Research Question 2
Research Question 3
Research Question 4
Research Quesion 5

Implications
anxkalApdmxmkms

APPENDICES

Appendix A Conant Fania
In English
In Spanish

Ammdix B Badcground lhfcrmdzim

Appendix c Child-Rearing Atdmds Marvin:
In English
In Spanish

Appendix D Otnuvatlm when:
Maternal Teaching Obervaiian' Technique (MTOT)
Child Beiavior During MTOT
Protocol

REFERENCES

viii

Page
84
84
89
92
92
97
98
99

101
102
106

108
109
110

111
115

1.19

123

125

127

LIST OF TABLES

Table I Page
3.1—Matemal Child-Rearing Attitudes Scares on Interview 47
3.2—Expected Scores and Ranks on Child-Rearing Attitude Measures 48
3.3-Maternal Behavior Measures Ranked According to

Child—Rearing Patterns 52
3.4—Methods Used for Analyses of Data 61
4.1—Child-Rearing Aujmdes Based on PARS and Ccrnell Interview 68
4.2—Chin-Rearing Practices Bead on MTOT Observation 70

4.3-Cancnica1 Discriminant Function Bazween Child-Rearing Attitudes

a Measlred bythe PARS and Cornell Attitude SurveyInterview and

Child—Rearing Practices Measured by the MTOT 73
4.4-Pearscn Ccuelation Coefficients ch omerved Maternal and Child

Behavior Variables 75
4.5—Muliiwriate and Univariate Analyses cf Child Behavior with

Authaitadve and AW Maternal Child-Rearing Practices 77
4.6-Inferentia1 Tests cf Significance cf Demographic Variables by

Mothers' Ctn'ld-Readng Austria and Practices 7 79
4.7—Statn'stical Results cf Hypotheses Tests - 81
4.8—Cromtabulaticn of Mexican-American Maternal Patterns of

Child-Rearing Attitudes and Practices 83

ix

LIST OF FIGURES

Figure Page
ll-Jrransactive Cycle cf Child-Rearing Attitudes, Child-Rearing

Practices, and Child Behavior 7

M I
was

Statelent of the Problan

The principal task of parents is to rear children from dependent
infants to young adults capable of functioning independently in society.
Muchof humandevelognent is accnsequenceof themanner bywhich this
is accalplished. As a vital and cmplex task, child-rearing has genera-
ted a substantial body of research, notably studies by Baumrind (1967,
1971, 1975, 1978); Sears, Maccoby, and Levin (1957); and Sears, Rau, and
Alpert (1965). Other studies in child developnent suggest evidence of
cultural and ethnic diversity along various dimensions of child-rearing
(Bamurind, 1971, 1975; Hess, 1970, 1972; Lindholm & Padilla, 1981;
Missen, Cmger, & Kagan, 1974; Stramen, McKinney, & Fitzgerald, 1983).
that of the research on the topic, waver, has been conducted using
middle-class White or lower-class Black populations with relatively
little attention given to other cultural or ethnic goups.

Additional infornation is needed about child-rearing in other eth-
nic minority families, especially Maxican-Pnerican families living in
the 0.8., whose mmbers represent the largest group of Hispanics, a
young , diverse, and changing population that is experiencing rapid
growth (Bureau of the Census, 1983);( The problan of this study, then, is

to describe Mexican-Anerican maternal child-rearing attitudes and

2
practices, the relationship of such practices to Child behavior, and the
relationship of family demographics to naternal attitudes and practices
of child-rearing .

Need for the Study

The few studies on Child-rearing in Mexican-American families that
can be considered enpirical or verifiable fail to provide consistent
cnnclusicns about those child-rearing attitudes and practices. For
exanple, Durrett, O'Bryant, and Pennebaker (1975) and LeVine and Bartz,
1979) conclude that Mexican-Anerican parents have permissive attitudes
toward rearing their children. The findings of other studies, however,
suggest that traditional values and authoritarian structures are more
characteristic of Mexican-Anerican child-rearing (Kagan & finder, 1975;
Kearns, 1970) .

Although lacking sufficient enpirical research, the literature is
replete with anecdotal and nonetpirical descriptions of interaction in
Mexican-Pnerican families and socialization of their children. Descrip-
tions, sane of which are written by mxican-Amrims (Alvirez, Bean, &
William, 1982; Delgado, 1980; Miranda, 1980; Miranda & Ehriquez, 1979;
Statcn, 1972; Vigil, 1980), characterize child-rearing in tandem-
Anerican families as warm, nurturing, and affectionate within a patri—
archal, authoritarian family structure with its traditional respect for
males and the elderly.

Ehpirical research on Mexican-Anerican child-rearing, then, is
needed to resolve the inconsistencies in prior research and ethnographic

descriptions. Additional research evidence may provide a better

3
understanding of how children are reared in Mexican-American families.
Such knowledge has the potential for practical application in many
settings where it is important to understand the manner by which parents
rear their children. For example, parent education programs could be
designed specifically for Mexican-Merican families with content based
on empirical evidence of their mild-rearing attitudes and practices. In
day care settings , knowledge of the care Mexican-American children
receive at tune can be applied to staff interactions with them and their
parents. Similarly, in educational institutions, etpirical information
about Marleen-American child-rearing attitudes and practices can have
practical application to teacher-child and teacher-parent interactions.
This knowledge can also be applied to interactions with Mexican-Marian
children and their families in health care, social services, and

recreational settings .

Purpose of tbs Study

The purpose of the study is to examine Micah-American maternal
Child—rearing attitudes and practices, their relationship to one an-
other, and their relationship to child behavior. Another plrpcse was to
investigate the relationship between demographic variables in the
Medan-American family environment and maternal child-rearing attitudes
and practices. Attitudes were measured by maternal responses in m oral
interview conducted in the language of the mother's preference i.e.,
English, Spanish, or a canbination of the two. Child-rearing practices
and child behavior were measured through observations of mother-child
dyads during a structured teaching task. The demographic variables were

4
mother's age, years of schooling, generations removed from Mexico, and
religious peference; father's years of schooling and employment status;
the child's gender; and number of male and female children in the

family .

Research Qnesticns

This expirical study of Mexican-American maternal child-rearing
attitudes and practices, child behavior related to such practices, and
the influence of family demographics was designed to determine answers
to the five research questions listed below. These questions were
formulated to give specific focus to the research activities and ensure
that the major goals of the study were met.

1. What patterns of Child-rearing attitudes do Mexican-American
mothers express about rearing their Children?

2. that patterns of child-rearing practices are Mexican-American
mothers observed to use during interactions with their children?

3. Are the child-rearing practices of Mexican-American mothers
related to their child-rearing attitudes?

4. Is there a relationship between bendcan-Pmerican maternal child-
rearing practices and child behavior during observed interactions be-
tween mother and child?

5. Are there differences in maternal child-rearing attitudes and
practices related to derographic variables in [Vatican-American family
ecosystems, such as mother's age, generations removed from Mexico, and
religious preference; mother's and father's years of schooling; father's

5
anployment status; the Child's gender, or the number of male and female
children in the family?
Theoretical Framework

The conceptual framework for this study is the ecological approach
to study of the family (Andrews, Bubolz & Paolucci, 1980) contained with
Baumrind's (1968, 1971, 1978) theoretical model of parental disciplinary
patterns. At the family level, the mother and child are parts of a
whole family system in which they perform different integrative func-
tions so that they contribute differently to the family system. The
mother and child also interact in such a way that they constitute an
environment for one another through their reciprocal interactions . The
child-rearing attitudes and practices of the mother are influenced by
the family resources in ways that affect the growth and development of
the child as well as the whole family system.

The mother and child also interact with other family mariners and
environments in an exchange of physical energy, messages, signs, and
symbols referred to as matter-energy and information. Family members
transform matter-energy and information by storing some of the informa-
tion in the form of attitudes toward rearing children and in using kin-
etic energy which may be manifested in mother-child interaction. The
flow of energy and information between the mother and child serves as an
organizing function relating them independently to one another, to the
family system, and to external environments. Additional energy is re-
quired for transactions with other systems outside the family, e.g.,
educational, religious, health care, economic, and recreational. Much of

the human energy required for those transactions is transformed into

6

attitudes and practices of child-rearing as well as into the behavior
and development of the child. The transactive cycle of child behavior
and child-rearing attitudes and practices is illustrated in Figure l.

Baumrind's (1966, 1971, 1978) theoretical model of parental discip-
linary patterns is integrated into the family ecosystem approach for
further analysis of attitudes and practices of child-rearing.
Permissive, authoritative, and authoritarian are the concepts applied to
the disciplinary patterns of the model. The concepts refer to a process
which influences parental discipline techniques. The concepts also imply
an input by the parent and an output (outcome) in the child. Definitions

of concepts follow in the next section.

' concapmai and Operational Definitions

The following concepts are cperationalized in this study:

Moment refers to the Medan-American culture, the family, the
home, and the roon(s) in which the research interaction takes place.

Emily is a unit of interacting and interdependent individuals who
ravecomongoalsandresources, andacommitment tooneanother over
time.

Interaction refers to behavior in which mothers and their children
exchange meanings and have a reciprocal effect upon each other's
behavior, expectations , and thoughts .

Interface refers to "a meeting ground of two systems or more sys-
tems or subsystems..." (Kantor & Lehr, 1975, pp. 23-24). For example,

employment is an interface between a family member and an enployer.

     
  
    
   
 
    
 

Child-Rearing
Attitudes

Child-Rearing
Practices

Child Behavior

  

External
Environments

Figure l.1--Transactive cycle of child-rearing attitudes, child-rearing
practices, and child behavior.

System refers to the mother-child dyad. System is defined as a set
of interrelated and interacting units or parts in which the action of
one unit affects other parts. Systems can be concrete (e.g., a child) or
they can be abstract (e.g., attitudes).

Fmsystem (ecological system) consists of the mother-child dyad,
their environment(s) , their interaction, and transactions.

Family system refers to the family members of the mother-child
dyad.

Boundary is defined in this study as the mother-child relationship
during the structured teaching task of the study.

Child-rearing refers to the oare and guidance of the child by a
parent. For purposes of this study, child-rearing is synonymous with
parenting, child guidance, and parental disciplinary patterns.

Child-rearing attitudes refer to what a mother thinks she ought to
doandwhatshevalues intherearingofherchildrenasuellashowshe
thinks children ought to act (Baumrind, 1971; Ferguson, 1970). This var-
iable is measured by the interview questionnaire contained in Appendix
C. Different values of the variable are categorized permissive, authori-
tative, or authoritarian according to Baumrind's (1966, 1971, 1978)
definitions of the three child-rearing patterns .

(mild-rearing practices refer to maternal behavior-what the mother
actually does during interactions with her child (Ferguson, 1970). This
variable is measured by observation with the use of the Maternal Teach-
ing Observation Technique developed by Laosa (1980c) . Different values

9
of the variable are also categorized permissive, authoritative, or
authoritarian according to the definitions below.

Patterns of child-rearing are permissive, authoritative, and auth-
oritarian. They are defined according to Baumrind‘s (1966, 1971, 1978)
theoretical model. In the literature, Bannmnrind has referred to them as
patterns of parental control (1966), socialization practices (1967),
parental authority (1971), and parental disciplinary patterns (1978).
The prototypic characteristics of each child-rearing pattern are defined
according to Baumrind (1978) as follows:

Permissive in which the parent attenpts
. . . to behave in an affirmative, acceptant, and benign manner
towards the child's impulses and actions. The permissive parent
seeshimn—orherselfasaresourceforthe childtouseashe
wishes, but, not as an active agent responsible for shaping and
altering the child's ongoing and future behavior. The immediate aim
of the ideologically aware permissive parent is to free the child
from restraint as much as is consistent with survival. Sone
permissive parents are very protective and loving, while others are
self-involved and offer freedom as a way of evading responsibility
for the child's developnent (p. 244).
Pnthoritative in which the parent attenpts

. . . to direct the child's activities in a rational, issue-

oriented manner. He or she encourages verbal give and take, shares

with the child the reasoning behind parental policy, and solicits
the child's objections when the child refuses to conform. Both
' autononous self-will and disciplined conformity are valued by the

10
authoritative parent. Therefore, this parent exerts firm control
when the young child disobeys, but does not hen the child in with
restrictions. The authoritative parent enforces the adult
parapective, but recognizes the child's individual interests and
special ways. Such a parent affirms the child's present qualities,
but also sets standards for future conduct, using reason as hell as
power and shaping by regimen and reinforcement to achieve parental
objectives. But this parent does not base his or her decisions on
group consensus or the individual child's desires (p. 245).
Mthoritarian in which the parent

. . . values obedience as a virtue and favors punitive, forceful

measures to curb self-will at points where the child's actions or

beliefs conflict with what the parent thinks is right conduct. The

authoritarian parent believes in keeping the child in a

anbordinate role and in restricting his autonomy, and does not

encourage verbal give and take, believing that the child should
accept a parent's word for what is right. Authoritarian parents may

be very concerned and protective and they may be neglecting (p.

244) .

The authoritarian parent also ”believes in inoilcating such in-
strumental values as respect for authority, respect for work and respect
for the preservation of order and traditional structure" (1971, p. 22).

Child behavior refers to what a child actually did while inter-
acting with his or her mother during the observed teaching task. Eight
categories of behaviors are identified by the instrument, Child Behavior

During naternal Teaching Observation Technique (CBDMI‘OI‘) . The behaviors

11
(variables). are Positive Verbal Response, Negative Verbal Response, In-
quiry of Mather, Physical Response, Imitation of Mother, Task
Involvenent, Nonconpliance, and No Response.

Moan-Marian and Chicano or Chicane are used interchangeably in
this study. The terms refer to persons of Mexican descent who are United
States citizens. These persons may be descendants of Mexican parents or
more renote ancestors who immigrated to the 0.8. from Mendco, or they
may be descendants of Hispanic or Undo-Hispanic forebearers who resided
within Spanish or Mexican territory annexed by the 0.8. in 1848. Chicano
referstoamalememberof thegroupandis alsousedinreferenceto
the ethnic group. Chicana refers to a fenale person of the culture
91'0“?-

Hispanic is conprehensive term which includes U.S. citizens of
Mexican, Puerto Rican, Cuban, and other Central and South Merican
descent.

Anglo is also a comprehensive term which refers to all White,
English-speaking United States citizens.

Research Hypotheses
From the general research questions for this study and the theor-
etical framework, the following hypotheses are formulated for testing:
801 There will not be a permissive pattern of child-rearing
attitudes expressed by Mexican-American mothers.
all There will be a permissive pattern of child-rearing

attitudes expressed by tandem-American mothers.

12

This hypothesis is based on the research findings of LeVine and

Bartz (1979) that Chicano parents have permissive attitudes toward rear-

ing their children.

H02

312

There will not be an authoritarian pattern of child-rearing
practices observed in Mexican-American mothers.
There will be an authoritarian pattern of child-rearing

practices observed in Mexican—American mothers.

The research findings of Kagan and Ender (1975) suggest that

Mexican-American mothers use authoritarian child-rearing practices.

These findings are the basis for the hypothesis.

Elo3

There is no relationship between madam-American maternal
child-rearing attitudes expressed dnring an interview and
their observed child—rearing practices airing a teaching task
with their young child.

There is a relationship between Moan-Marian maternal
child-rearing attitudes expressed during an interview and
their observed child-rearing practices dnring a teaching task
with their young child.

A basis for this hypothesis is Baumdnd's (1.971) research evidence

that parental behavior is predictably related to expressed parental at-

titudes. The shared variance between these parental attitudes and paren-

tal practices observed in her study, however, was small. In the studyat

hand, a correlation of the expressed attitudes of Mexican-Pnerican moth-

ers and actual child-rearing practices provided evidence for answering

the third general research question.

13
04 There is no relationship between the observed child-rearing
practices of Mexican-American mothers during a teaching task,
and the W behavior of their children.

314 There is a relationship between the observed child-rearing
practices of Mexican-American mothers during a teaching task,
and the observed behavior of their children.

The rationale for this hypothesis is based on the research findings
of Baumrind (1971, 1973, 1975). Her extensive studies “provide evidence
for differences in child behavior related to patterns of parental child-
rearing practices . Enamination of the interface between child behavior
and maternal child-rearing practices provided evidence for answering the
fourth general research question.

HOS There is no relationship between Mexican-American maternal
child-rearing attitudes and the demographic family variables,
(a) mother's age,

(b) mother's years of schooling,

.(c) mother's generations removed from Mendco,
(d) mother's religious preference,

(e) father's years of schooling,

(f) father's employment status,

(9) subject child's gender,

(h) number of fenale children in the family,
(i) number of male children in the family.

15

14

There is a relationship between Mexican-American maternal

child-rearing attitudes and the demographic famnily variables,

(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)

mother's age,

mothers years of schooling,

mother's generations removed from Mexico,
mother's religious preference,

father's years of schooling,

father' 3 employment status ,

subject child's gender,

number of fenale children in the family,
number of male children in the famnily.

The rationale for this and the following hypothesis is derived from

the family ecological model (Andrews, Bubolz, 8: Paolucci, 1980; Bubolz,

Eicher, 8 Sontag, 1979), in which the interaction between family and

environment is one of reciprocal influence. Families affect their envi-

ronments, which in turn affect families.

HOG

There is no relationship betwesn Mexican-American maternal

child-rearing practices and the famnily demographic variables,

(a)
(b)
(c)
(d)
(e)
(f)

. (g)

(h)

mother's age,

mother's years of schooling,

mother's generations removed from Mexico,
mother's religious preference,

father's years of schooling,

father ' s enployment status ,

subject child's gender,

number of female children in the family,

15
(i) number of male children in the family.
16 There is a relationship between Mexican-American maternal
hild-rearing practices and the family demographic variables,
(a) mother's age,
(b) mother's years of schooling,
(c) mother's generations reroved from Mexico,
((1) father's religious preference,
(e) father's years of schooling,
(f) father's enployment status,
(9) subject child's gender,
(h) number of female children in the family,
(i) number of male children in the famnily.

Assumptions

The assumptions made in this study are as follows:

1. Through continuous interaction over time, children and their
family systems grow, change, and develop within an ecosystem.

2. bother-child interaction is best studied in natural settings.

3. Teaching and learning occur in the family environment.

4. baterml attitudes can be measured by using maternal responses
to statements posed.

5. Mather-child interactions observed during a teaching task are
representative of their interactions during similar teaching-learning
interactions.

6. Children at the age of five years are generally able to success-

fully manipulate Tinkertoys .

16
Limitations

The potential limitations of the study concern the sample, mea-
sures used, and generalizability of the findings. The sample size of 47
may be too small to allow findings to be generalized, with confidence,
to a population of Midwest urban-dwelling, working-class
Mexican-American families that the sample is intended to represent. The
sample may not be representative of the population due to the pnrposive
and snowball sampling deemed necessary to meet research objectives. The
nature of the structured observation limits the findings to
child-rearing practices involving similar teaching tasks . General
statements about the overall effects of child-rearing patterns cannot be
derived from this sttdy.

The fact that there was only one observer posed certain questions
of observer reliability as well as the validity of the data collected.
Despite efforts made to control for observer reliability, one observer
of two subjects engaged in simultaneous interaction poses threats to the
internal validity of the observation data.

Only maternal child-rearing has investigated. Findings apply,
therefore, only to mothers and not to parenting in Moan-American fam-
ilies in general. Similarly, only Mexican-American mothers residing in
an urban setting in the Midwest were included as subjects. Results of
the study, therefore, are only generalizable to similar populations.

17
Overview Statenent

This research is reported in the following four chapters. In
Chapter II is a review of the literature on child-rearing in
Moan-American/Chicano families and on demographic family variables
pertinent to this study. The design of the study and procedures used are
explained in Chapter III. Chapter IV is a presentation of the findings
from the data analyses of the study. In Chapter V, the study is
summarized, conclusions are drawn, findings are dismissed, and
implications for future research and practical application are

presented .

CHAPTEII

mm

The literature pertinent to this study is reviewed under two head-
ings, Child-Rearing and Demographics. Enpirical or verifiable studies on
Medan-American child-rearing are reviewed according to their findings
of authoritarian/traditional, permissive, or a range of child-rearing
findings . Literature surveys on child-rearing in Mexican- American fami-
lies follow. Next, the literature on family demographic variables perti-
nent to this study is reviewed; and a summary concludes the chapter.

Child-Rearing

Over the years, researchers have learned not to generalize their
findings for some Meadow-Americans as characteristic of all
Mendcan-Americans. In the studies reviewed, distinctions were made be-
tween Mexican-Americans of different socioeconomic levels and those in
rural or urban areas of residence. Mexican-American subjects of these
studies, however, have primarily besn urban-dwelling and of low socio-
economic status (SES) . Research findings as to the child-rearing atti-
tudes and practices of these subjects vary widely, so they are divided
for review into studies in which researchers reported authoritarian or
traditional child-rearing, those in which permissiveness was reported,
and the broad range of findngs in one study.

18

l9
Authoritarian/Traditional ChildrRearing Findings

Kearns (1970) studied 50 Anglo, 50 MExican-American, and 50 Papago
Indian Inmflhers (if first-grade: childrenl in iflme umteun areal of 'Tucson,
Arizona. Due to the perceived need for developing rapport, personal in-
terviews were conducted in the homes of the subjects in their preferred
language by interviewers of Hatched ethnicity. Interviews ranged from
twenty-five minutes t1) on: and one-half hours. Interviewers recorded
subjects' responses in shorthand and transcribed them after the inter-
view for rating. The research instrument was a self—report measure of
maternal childrrearing based on a structured oral interview, an adapta-
tion of the child-rearing rating scales for nother interviews developed
by Sears, Maccoby, and.LeMin for their classic study, Patterns of Child
Rear'gg (1957).

Significant differences were found.among the groups on the dimen-
sio1s of strictness, restrictions on children, and permissiveness for
aggression toward other children. Since socioeconomic status (SES) was
controlled by including only lower SES subjects, Kearns concluded that
significant differences anong the Papago, madcan-Anerican, and Anglo
child-rearing practices were cultural. Mexican-American and Papago pat-
terns of childrrearing appeared to be governed to a large extent by tra-
ditional values and practices. Bonds with tradition appeared to be weak-
er among the Anglo mothers. Although Kearns generalized her findings to
patterns of childvrearing practices, it appears that the results of the
study' also apply' to» childrrearing attitudes, since her self-report
interview schedule measured maternal childrrearing attitudes as well as

practices (Sears et al., 1957).

20

Kagan and Ender (1975) investigated the preferred reinforcement
patterns of 16 urban Anglo-American, 16 urban Mexican-American, and 16
rural Mexican mothers. American mothers were selected at random at a
boating and picnicking park in Riverside, California, and ranged from
upper middle to lower class economically, as determined by family in-
come. Rural mxican mothers were also chosen at random from the town of
San Vicente, Mexico, where most families live slightly above subsistence
level. In both rural Mexico and the urban U.S. city, subject participa-
tion was obtained by requesting that a mother-child pair spend time in-
teracting in a gamelike situation at a table, while the experimenter
tallied mother responses. Prior to the experiment, family demographics
were collected on where the family lived, the number and ages of family
members, the types of exployment in which family menbers were engaged,
and the amount of family income. During the experiment, measures were
taken of a mother's responses to the success or failure of her child.
The experiment consisted of each mother giving her child chips, taking
chips away, or doing nothing when her child succeeded or failed to reach
the child's predetermined goal in pressing a tally counter a specific
number of times during timed intervals.

Analysis of the findings suggested a tendency for higher-incore
mothers to discriminate between success and failure more than lower-
incote mothers. Although neither group of American mothers discriminated
between boys and girls in giving chips, they did give significantly more
chips on success than on failure trials. Rural Mexican mothers gave al-
most as many chips following failures as they did following successes.

Mexican-American mothers, however, took more chips away from their

21

children on failure trials. According to Kagan and Ender, the overall
differences between mothers tended to be related to both economic and
ethnic variables. The behavior of both American groups was similar to
one another, except that Mexican—American mothers were generally more
punitive than Anglo-American mothers. Regression analysis of the
results, however, indicated the more frequent use of pmishnent by
Medan-American mothers was a function of their lower economic status
and not of their culture. The researchers found that economic level, but
not ollture, was significantly related to the use of punishment. They
concluded , however , that parental reinforcement patterns may function
to maintain onltural and economic class differences. 'lhus, Kagan and
Ender's findings on punishment slggsst that Mexican-American mothers of
lower SE5 have a tendency toward authoritarian child—rearing practices.

Griswold's findings (1976) also suggest that SE3 cits across eth-
nicity in determining child-rearing attitudes. Griswold studied Anglo,
Black, and Modem-American mothers to gain information concerning simi-
larities and differernces in child-rearing as a function of ethnicity and
SES. Her 114 subjects were from an Arizona metropolis; 52 were Anglo, 32
were Black, and 31 were Vatican-American. SE of the Anglo mothers was
24 upper, 18 middle, and 10 lower class. Among Blacks, there were 11
upper class, 10 middle, and 10 lower class mothers. Of the 31 Medcan-
American mothers, 11 were upper class, 10 were middle class, and 10 were
lower class. Mothers were identified through their second-grade children
of dual-parent families, but the children were not included in the study

as subjects. Data were collected with the use of two questionnaires, The

22
Maternal Information Needs and Attitudes Assessment (MENAA) and the Par-
ent as a Teacher Inventory (PAAT).

Descriptive statistical comparisons of MINAA responses indicated
similarities and differences which existed as a function of ethnic
background and socioeconomic status of the subjects. MINAA data also
provided information on maternal information reeds, frustration,
satisfaction, information sources, and characteristics valued at hone.
Differences. between the groups showed that lexican-American and Anglo
mothers expressed greater need for child-rearing information .
Mexican-Americans also expressed greater concern about their children's
school experiences than Anglos and Blacks. Micah-Americans and Blacks
expressed greater satisfaction with their children's successful
adjustment in self-concept, relationships with siblings and peers, and
relationships within the environment. lexican-Mericans and Blacks also
expressed more pride in their children's increasing independence,
assumption of responsibility, and obedience. Mexican-Americans were the
least likely to consult agencies, services, institutions, and printed
matter for help in child-rearing concerns. Medan-Americans and Blacks
reported that they turned to their mothers and other family members for
help in child-rearing more often than Anglos. Mandcan-Americans and
Anglos regarded participation of both parents in child-rearing,
assurance and support from both parents, religious and moral faith, and
a routine life style as sotewhat more important for proper children's
adjustment than Black mothers.

Conclusions were made based on the statistical analysis of the

PAAT, which measured creativity, frustration, control, play, and the

23

teaching-learning process. Ethnic background had a significant effect on
every variable except frustration; however, the major source of variance
was between Anglo and Black mothers. There were no significant differ-
ences between Arnglo and Mexican-American or between Black and lexican-
American mothers. In addition, sns had a significant effect on all the
variables measured. The major source of variance was between upper and
lower class; however, variance on control, play, and teaching-learning
was also attributed to differences between middle and lower class SES.
Significant variance on control was attributed to differences between
upper and middle class. Due to the higher proportion of Anglo upper- and
middle-class mothers, Griswold was cautious about generalizing her find-
ings. Nevertheless, the results of her study suggest to this researcher
that the Mexican-American mothers tended to have traditional child-
rearing attitudes.
Permissive Child-Rearing Findigg

Some researchers (Durrett, O'Bryant, & Pennebaker, 1975; LeVine &
Bartz, 1979) have concluded that Mexican-American parents have permis-
sive attitudes toward rearing their children. Durrett et a1. , interview-
ed 29 White, 30 Black, and 31 Mexican-American pairs of mothers and fa-
thers about their child-rearing orientations and techniques. The sub-
jects were all low-income families with five-year-old children enrolled
in Head Start. A 91-item Q sort was individually administered to each
parent in the home, in their preferred language of either Spanish or
English. The Q sort itene were arranged by each subject on a continuum
of importance according to goals, methods, and perceived reactions to
his or her child.

24

Measures of child-rearing orientations were achievement,
authority, protectiveness, stressing of individual responsibility,
control of elotions, and the positivity of the parent-child
relationship. Measures of child-rearing techniques were consistency or
lack of it, use of positive reinforcenent, use of guilt, use of
aversive control, and physical punishment. The following differences
were found among the groups on both the orientation and technique
categories:

In the Durrett et a1. (1975) study, White and Black "parents re-
ported being more authoritative (e.g., stressing respect for aiults)
than did bendcan-American parents." (p. 871) The example that stressing
respect for adults suggests authoritativeness is in disagreenent with
Balm'rind's definition that the authoritarian, not the authoritative,
mother "believes in inculcating such instrumental values as respect for
authority." (1971, p. 22) -

Durrett and her associates (1975) also found that fathers of Black
and White children were more achievement and success oriented than their
Mendcan—American counterparts. Mexican-American parents scored lowest on
the category emphasizing individual responsibility and were significant-
ly more protective. taxman-American parents stressed somewhat greater
control of enotions than the other parents and were most consistent in
their methods of reward and punishment. Use of reward when the child
displays. good behavior was a favored method of Black parents, but Black
fathers also reported more use of strict arbitrary rules. Mexican-

American and White mothers reported using significantly more guilt in

25
child-rearing than did Black mothers. The groups did not differ signifi-
cantly on the positivity of their parent-child relationships.

Durrett and her associates (1975) concluded that data on self-
reported techniques of socializing children were similar along the
groups, but the parents' desired ends differed. The researchers offered
no explanation of parental desired ends. The conclusion suggests that
low-income Mexican-American, White, and Black parents expressed differ-
ent child-rearing ”orientations" but they have similar child-rearing
"techniques". If the statistically significant findings are categorized
according to Baumrind's (1971) definitions of permissive, authoritative,
and authoritarian patterns of parental authority, the data suggest that
the American-American parents' child-rearing practices tend to be author-
itarian while their child-rearing attitudes are permissive.

Bartz and IeVine (1978) and IeVine and Bartz (1979) also colpared
the child-rearing attitudes of 152 Chicano mothers and fathers with
those of 143 Anglo and 169 Black parents to identify unique ethnic
child-rearing patterns. Using data from the same study, they focus on
Black parents in their 1978 article, while their 1979 article centers on
Chicano parents. Their study was a component of the multidisciplinary
project of the Family Study Center of the University of Missouri to in-
vestigate various dimensions of ethnic family life. The data were col-
lected in 1973-74 according to a comunication with one of the project
directors, Ronald E. Cromwell (Jilly 1984).

The research subjects, who resided in a low socioeconomic urban
community in Kansas City, Missouri, were randomly selected from public

and parochial elementary school rosters. A survey questionnaire was

26

administered in their homes in English or Spanish by interviewers of
matched gender and ethnicity. In instances when both parents in a family
participated, they were interviewed simultaneously but in different
rooms of their hone. The questionnaire items consisted of 25 statelents
selected from adaptations of the Parent AttitudeResearch Scale (PARS)
and Cornell Parent Behavior Inventory (Cornell) , which IeVine and Bartz
report that Cromwell adapted for use with lower-class parents. The 17
statements selected from the PARS measured 5 factors labeled as Acceler-
ation of Autonomy, Casual Use of Time, Equalitarianism, Value Strict-
ness, and Devalue Penmissiveness. The 8 Cornell itene selected from the
Cornell Inventory were labeled Lack of Support and Lack of Control and
were restated to measure parental perceptions of child-rearing. IeVine
and Bartz's resulting research instrument of 25 items was the interview
schedule used by this researcher in the present study and is discussed
in Chapter III.

Conparisons by ethnicity on the seven factors indicated parents
did not differ in their child-rearing, but in the degree of their alpha-
ses on particular attitudes or desired behaviors (Bartz a IeVine, 1978).
IeVine and Bartz (1979) found that the child-rearing attitudes of
Chicano parents differed from those of Anglos and/or Blacks on six of
the seven factors measured: (1) Chicanos and Blacks, more than Anglos,
pressed for early assumption of responsibility. This finding is in
contrast to the Durret et al., (1975) findings on the same measure. (2)
Blacks, more than Chicanos and Anglos, pressed for wise time use. (3)
Chicanos were least likely of the groups to stress equalitarianism. (4)

Chicanos were more likely to value permissiveness than Anglos. (5)

27

Chicanos and Anglos reported that they offer less support than Blacks.
(6)Chicanc and Arglo parents also reported that they are less
controlling than Black parents. (7) No significant differences were
found between the three groups on the strictness factor. The
researchers indicated that this lack of significance suggest all three
groups value strictness in child-rearing (IeVine & Bartz, 1979, p. 171),
but conclnfle that ”Hispanic child-rearing is quite permissive" (1979, p.
175) in the low socioeconomic class.

To further delineate the relationship of ethnicity to parental
responses on each factor, analysis of covariance was performed with edu-
cational level as the covariant. Education was nonsignificant as a con-
trol variable for all factors except one. “The higher the level of edu-
cation, especially in Chicano families, the greater the belief in equal-
itarianism'i (Bartz a IeVine, 1978, p. 714). No explanation was made of
what the levels of education were for the sample. Nevertheless , IeVine
and Bartz (1979) concluded that Chicanos, as a group, were more permis-
sive and less equalitarian than Arglo and Black parents in their child-
rearing attitudes as well as practices. mile a self-report instrument,
such as the interview instrument erployed by IeVine and Bartz, is
appropriate as a measure of attitudes, it is not considered an
appropriate measure of actual child-rearing practices (Baumrind, 1971) .
Cross-ethnic differences in attitudes, however, were more attributable
to Chicano fathers than to mothers. F ratios on the seven factors
measured eiggested to the researchers that Chicano fathers and mothers
stressed acceleration of autonomy more than Anglo mothers . Chicano
fathers, and not mothers, tended to devalue equalitarianism, support,

28
and control. The conclusions, devalue equalitarianism and devalue
control appear contradictory. ' If equalitarianism is devalued, the
implication is a lack of equality between parent and child, in which
case control on the part of the parent would be enqnected.

The researchers further concluded that since Chicanos were less
equalitarian than Arglos and Blacks, their findings were consistent with
descriptions of the relatively authoritarian Hispanic family. In sum-
mary, the conclusions by IeVine and Bartz on Chicano child-rearing pro-
vide conficting information, some of which may be attributed to the la-
bels they assigned to the categorical variables of their research in-
strument, which was employed in the present study. The instrument and
conclusions of the IeVine and Bartz study are discussed further in
Chapter V.
5.3.4.1299 of mild-MW

In their study of Mexican-American perceptions of parent and
teacher socialization roles, Parra and Henderson (1982) interviewed 95
Chicano parents of lower and middle socioeconomic levels. The 70 sub—
jects classified as middle-class were attending a community college and
were enployed in skilled or higher occupations. The 25 lower-class sub-
jects were receiving welfare and/or pnblic housing benefits. The inter-
view instrument consisting of 22 items calls for open-ended responses
and was administered in each subject's preferred language of either
Spanish or English. It was developed by the researchers to elicit infor-
mation on parents' child management practices, sex-role perceptions re-

lating to children, age-related behavioral expectations , activities and

29
qualities valued in children, learning outcones and focus of responsi-
bility, and aspirations and expectations for their children.

Parra and Henderson found sore differences between the reports of
middle-and lower-class Mexican-American parents. For example, there was
a significant difference in the language preference of the two groups.
While 83 percent of the interviews with middle-class parents were con-
ducted in Erglish, 80 percent were conducted in Spanish with lower-
status parents . Middle-class parents also shifted from one language to
the other (code switching) more frequently than lower-class subjects.

Findings on child management indicated that most parents in both
groups had a number of different means of controlling and guiding the
behavior of their children including induction, positive reinforcenent,
andpunistment. 'I'heuseofexplanationasameansofmnagingdnildbe—
havior was suggested by 68 percent of the parents, and there was no dif-
ference by SES of subjects. The majority of parents in both groups re-
ported the use of positive reinforcement, and about one-half of each
group also reported practices which were classified as negative rein-
forcenent. There was a difference, however, between the two groups in
reports of verbal punishment with middle-class parents reporting signi—
ficantly less use of verbal punistment. Overall, fewer parents reported
the use of physical punishment as a control procedure than any other
category. When compared with practices using positive reinforement,
lower-status parents reported using both verbal and physical punishment
to a significantly greater degree than the middle-class parents. Study-
ing these self-report research findings from the perspective of
Bammrind's (1975, 1978) patterns of parental authority, the practices of

30
lower- and middle-class Mexican-American parents in this study can be
considered as ranging fren authoritative to authoritarian patterns.
Parra's and Henderson's suggested need for both observational and
self -report research to determine the range of Mexican-Anerican child
uanagelent practices is addressed in the study at hand.

Findings on the other areas of investigation by Parra and
Henderson (1982) revealed no statistically significant differences
between lower- and middle-class subjects with the exception of learning
outccmes and focus of responsibility. Despite their awareness of changes
in society toward less differentiation in sex roles, a najority of the
parents reported they held different expectations for boys and girls.
Parents also‘reported that age of a child has a determining influence on
how parents should respond. There was a balance between social and rec-
reational skills and academic ability in the activities and qualities
valued in children. A significant difference was found, however, between
the two social classes in their responses relating to learning outcomes
and focus of responsibility for socialization activities of the home and
school. Most of the parents, 98 percent, perceived the min socializa-
tion task of the hone as fostering socioenutional developnent. The
school was perceived as having responsibility for fostering intellectual
development and acadenic achievenent . Parental expectations for the edu-
cation and future occupation of children were not as high as the aspira-
tions held for then. Of the 74 percent lower-class and 83 percent
middle-class parents who had aspirations for their children to attend
college, 67 percent of lower- and 44 percent of middle-status parents
expected that their children would actually couplete college. Parra and

31
Henderson found further evidence that these Mexican-American parents
perceived that teachers fail to take children ' s cultural background into

consideration .

Literature Survey

From his descriptive literature survey, Staton (1972) determined
that Medcan-Anerican parents erphasize that their children learn
enbmnission and obedience to the authority of their father. The mother,
consequently, becomes the family affectional focus for the children.
While in early childhood, the father would play with the children,
attend to their needs, and demnstrate affection for then, he mintained
his authority, and denanded respect from his children. As they approach-
ed puberty, the father increased his distance frann then in order to
maintain his authority.

Staton determined that Mean-American anthers and daughters
tended to be close as a result of their male-daninated world and the
daughters' training for their adult fetale role. The mother-sen rela-
tionshipwasalsodescribedasclosewiththemotherservingasa
permissive and affectional figure for the 'son.

Mare recent descriptions characterize child-rearing in Mexican-
American families as warm, nurturing, and affectionate with enphasis en
traditional respect for nales and the elderly (Mirande s. Ehriquez, 1979;
Delgado, 1980; Mirande, 1980; Vigil, 1980; Alvirez, Bean & Williams,
1982). Delgado apports Staton's findings about the en'phasis on children
learning submission and obedience to the will of the father. He

describes maternal responsibility for child-rearing, but paternal

32

responsibility for major discipline of the children. Under no circumn-
stances , however , should children talk back to either parent according
to Delgado. Mirande and Enriquez further descibe the Chicano family as a
supportive unit as exemplified by older siblings working so that younger
ones can complete their schooling. The Chicano family is also a basic
source of emotional support for children since close bonds are developed
with grandparents, aunts and uncles, cousins, family friends, as well as
immediate family menbers (Mirande, 1980). In addition, girls participate
early in the child care of younger siblings (Miranda & Ehriquez, 1979) .

We Variables

Maternal Ag

A mother's age has long been considered an influential variable. in
determining behavior and attitudes of mothers. In interviews and direct
observations of 379 mothers, Sears, mccoby, and Levin (1957) found that
younger mothers tend to be somewhat more punitive in their treatment of
young children. The younger mothers, aged 24 to 30 years, were quick to
punish and more likely to express feelings of hostility toward their
children than older mothers, who were 31 to 38 years of age. Younger
mothers' greater punitiveness was attributed to age, which had a greater
bearing an the mothers' use of physical punishment and ridicule than her
social or educational background. .

Mare recent studies have generated inconsistent data on maternal
age and behavior. The results of two studies (Field, Widmayer, stringer ,
& Ignatoff, 1980; Osofsky & Osofsky, 1970) suggest that younger mothers
tend to have less positive maternal attitudes or behavior. Other

33

researchers (Conger, McCarty, Yang, Lahey, & Burgess, 1984) have found
that with increasing age, mothers may develop priorities over child-
rearing. In the Conger et al., study of three independent samples (n a
33, 38, 36) , the data indicated that chronological age was inversely
correlated with positive behaviors of mothers. The erotional tone of
maternal interactions became less positive and more negative as age in
years increased. These findings were significant for the sample of White
mothers, but narginal for the samples with 72 percent Black mothers. The
researchers, therefore, concluded that ethnic differences in the differ-
ent effects of chronological maternal age may have been based on greater
opportunities outside the family for White mothers in the study than for
the Black mothers.

In addition to the effects of chronological maternal age, the in-
fluence of a mother's age at the birth of her first child was also enam-
ined by Conger, NbCarty, Yang, Lahey, and Burgess (1984). Their assess-
ment of maternal age in terms of first birth, revealed that the older a
mother at the time of her first child's birth, the more positive (i.e.,
praise and physical affection) and the less negative (i.e., criticism
and physical punisment) her maternal behaviors. Age at first birth was
also negatively related to level of stressful life circumstances, which
led to the conclusion that an early first birth may have long-term nega-
tive consequences for maternal social, psychological, and material
well-being.

34
mternal mncational Level

Maternal behavior has been found to vary according to socio-
economnic and educational levels (Laosa, 1977, 1978, 1980a, 1980b, 1980c,
1981)). In his studies of the influence of Anglo- and Hispanic enlture
and education on maternal behavior, Laosa found that the manner of.
teaching her child varied as a function of a mother's culture and formal
educational level. The differences were attributed to the average level
of formal education attained by the mothers in the two cnltural popula-
tions. Differences between the anltural groups disappeared, however,
when the mothers' or fathers ' schooling levels were controlled . The
higher the level of formal education, the more inquiry and praise were
used as a maternal teaching strategy. The lower the mother's level of
education, the more a mother used modeling as her teaching strategy.

The educational level attained by the Chicanas observed ranged
fromoneyearofschoolingtotwoyears of college. Twopatternsof
maternal teaching strategies were identified in the Cnicanas. bothers
wlno had completed at least an eleventh-grade education taught with a
conbination of praise, visual ones, and inquiry and less frequently with
directives and modeling. Chicanas with less than an eleventh-grade
education, on the other hand, taught by modeling, using visual ones, and
issuing directives (1981). Laosa also found an inverse relationship
between Chicano mothers' educational levels and their teaching
strategies involving negative physical control with sons (1978) .

Formal education conpleted by Anglo mothers ranged from eleventh
grade to_ eight years post-high school. Laosa (1980a, 1981) found that

Anglo mothers used inquiry and praise as teaching strategies more

35

frequently than Chicano mothers, while Chicanas taught by modeling,
using visual ones, issuing directives, and using negative physical
control more frequently than Anglos . The most frequently observed
teaching strategies among C'hicanas were nnodeling and visual ones,
followed in order of decreasing occurrence by directive, praise,
inquiry, and negative verbal feedback. Among Anglo nothers, the most
frequently observed teaching strategy was praise followed by visual
ones, inquiry, directives, modeling, and negative verbal feedback in
order of their decreasing frequency. Laosa found these onltural-gronp
differences were related to the average level of educational attainment
since the differences in teaching strategies disappeared when maternal
educational level was controlled .

Tleinstrumentdevelopedbylaosaforuseinhis studieswas also
enployed in the present study. A copy of the instrument, Maternal Teach-
ing (bservation Technique (MTOI‘), is contained in Appendix D. The in-
strument and its measures are further disonssed in Chapters III and V.

Maternal Generations Revved from basics
Tiers is common agreement in the literature that the number of

generations renoved from Mexico is a salient variable in the process of
aconlturation. Laosa's findings (1977, 1978, 1930 a, 1980b, 1980c, 1981)
enggest that the generation renoved from Mexico as well as the level of
formal education of Chicanas may influence the teaching strategies which
they enploy with their children. Similarly, the findings of an enpirical
study of the role of onltural awareness and ethnic loyalty in

acculturation, suggest that generational level is an important predictor

36
variable in determining the extent of acculturation among
Mexican-Americans (Padilla, 1980). The definition of olltural awareness
in the study included respondents' olltural heritage as well as that of
spouse and parents , language preference annd use, olltural
identification and preference, and social behavior orientation. Ethnic
loyalty consisted of olltural pride and affiliation, perceived
discrimination, and social behavior orientation. Padilla (1980)
reported that the 584-item survey questionnaire of 68 men and 313 women
indicated that there were individuals who had nnot conpletely
acculturated end who possessed a marked degree of cultural awareness
and/or loyalty to the Mexican ollture even to the fourth generation. As
might be expected, less acculturated respondents had lower educational
levels while higher educated respondents were more acollturated . Incone
and ethnic density of neighborhood were also founnd positively correlated

with level of acculturation.

Child's Gender

Differences in the rearing of boys annd girls are well documented
by Sears, Maccoby, & Ievin (1957), who conpared ratings of 202 mothers
of kindergarten-aged boys and 177 mothers of kindergarten-aged girls on
scales of child training in relation to gender of child. Fer: dimensions
of differential treatment were founnd statistically significant in the
areas of feeding and warmth, toilet training, sex training, aggression,
tasks and chores, school achievenent, and techniques of training.
Findings indicated that weening took longer and mothers were warmer with

infant girls. There were no significant differences in the way boys and

37

girls were toilet trained, but there was a tendency for more severe
training of boys and for it to be connpleted later than with girls. Boys
and girls were treated similarly with respect to sex training at kinder-
. garten age. The greatest gender distinctions were made by parents re-
garding aggression. Boys were allowed to be more aggressive with neigh-
borhood children but not with siblings. Boys were also more frequently
encouraged to fight back than were girls. Similarly, aggression toward
parents was not allowed of girls, but boys were given high freedom of
expression. There were no differences in the amount of tasks annd chores
assigned to boys and girls, but there was evidence of sex-typing in the
nature of chores assigned. In the area of school achievement, mothers
expectedtheirsonstogotocollegeandtheirdaughterstoconplete
high school. Differential treatment was evident in techniqnes of train-
ing and discipline. Among relatively non-aggressive children, there were
more girls who practically never received physical punishment while
annong aggressive children, girls were spanked as often as boys. Girls
received more praise end were more often elbjected to love withdrawal
than boys. Girls were prinnarily disciplined by their mothers, and boys
received more discipline from their fathers when both parents were in
the bone. Sears and his associates (1957) state that there were few dim-
ensions on which boys and girls were differently treated; however, this
study took place nearly 30 years ago. Their conclusions about the find-
ings of their study may conceivably be that there were considerable
differences in child—rearing according to a child's gender.

38
8!!er

In summary, research on Mexican-American child-rearing is limited
and the findings are inconsistent. Results annd descriptions have been
supported by some researchers but contradicted by others. There is evi-
dence suggesting that Meadow-American child-rearing attitudes range
from permissive to authoritarian. There is no documentation, however, of
actual observation data on Mexican-American child-rearing practices
since parents were not observed in interaction with their children in
the studies reviewed. Nor is there evidence of the effects of varying
nnaternal attitudes and practices on child behavior in previous studies .
A predominant criterion for selecting subjects was that parents have a
child between the ages of five and seven years; consequently, child-
rearing attitudes were studied vis-a-vis a young child.

Tiereaderwill note inthenextchapterthatstrengthscfpre-
vious studies are incorporated in its methodology . Child-rearing is ob-
served in the home environment where subjects were interviewed in their
preferred language by a researcher of matched ethnicity. This is salient
to minimizing olltural annd language barriers since the meaning of rele-
vant behaviors and innplicit group norms is likely to be readily under-
stood (Iaosa, 1981; Bennal, North, Rosen, Delfini, & Schultz, 1979).
Slggestions in the literature that demographic variables are determin-
ants of differences in child-rearing are tested. Child-rearing is stud-
ied relative to kindergarten-aged children thereby providing continuity
with earlier studies. An attenpt was also made to improve upon earlier
research by investigating both maternal child-rearing attitudes and

practices separately and relative to one another. Children were included

39
as subjects in the study and the effects of child-rearing practices on

their behavior were examined.

CHAPTERIII

This onapter has two primary sections, design and procedures. In
the design section, the research methods used to investigate the
research problem are described. In the procedures section, the
activities used to implement the design and conduct the study are
discussed.

Design
Design Over Time

The purpose of this descriptive, cross-sectional study was to
describe Mexican-American maternal child-rearing attituda and
practices, the relationships, if any, between them and any
relationships between child behavior annd child-rearing practices. In
addition, the relationships, if any, between attitudes and practices of
child-rearing and family derographic variables were investigated. The
subjects were 47 nnothers paired with their kindergarten-aged children.
The study took the form of a conbined ex post facto (field) and
attitudinal survey design (Kerlinger, 1974). Tie ex post facto portion
was based upon structured observations of mother-child dyads in their
natural home environment . Maternal child-rearing attitudes were

surveyed in interviews with the mothers following the structured

4O

41
observations. The ex post facto outcore measures were child-rearing
practices and child behaviors based on simultaneous observation of the
mother-child dyad engaged in a structured teaching task. A survey

technnique was used to determine the mothers' child-rearing attitudes.

Foundation

The population studied was Mexican-American mothers and their
kindergarten-aged children living in Lansing, Michigan. The Two Factor
Index of Social Position (Hollingshead, 1957) was used to determine
socioeconomic status (SE) of the sample, which was predominantly lower
SE. A lower SE Chicano population was selected because it is most
representative of this ethnic group in the Lansing area as well as in
the 0.8. as a whole (0.8. Census, 1980). The study was limited to
mothers and their kindergarten-aged children for oznparability with
earlier studies in which the child-rearing of Moan—American parents,
primarily mothers, of children in this age group was examined.
Researchers have also successfully used this age gronp to eamine
parent and child teaching-learning interaction (Brophy, 1970; Hess &
Shipman, 1972; Laosa, 1980a, b, c). Furthermore, the instrument, The
Maternal Teaching Observation Technique (MI‘OI'), developed in both
English and Spanish, was previously found to be successful in eliciting
. and maintaining the attention of five-year-old Chicano children and

their mothers (Laosa, 1978; 1980a, b, c).

42
£13 Stratgy and Characteristics

Tie sample consisted of 47 bexican-American mothers who
volunteered to participate in the stndy with their kindergarten-aged
children. A nnonprobability, purposive sample was selected because of
the impossibility of random sampling. Snowball sampling was also used.
Names, telephone nnumbers, and addresses of potential subjects were
obtained from organizations with Chicano members, as well as from
various Mexican-American community leaders.

Potential subjects were contacted by telephone or in person to
request their participation in the study. At that time, the mothers
were informed of the nature of the study and that the investigator was
seeking Chicanas to participate with their kindergarten-aged children.
Tle researcher comunicated with the mother in either English, Spanish,
or both. Of 66 potential subjects who were oxntacted, 50 volunteered to
participate and 47 actually participated. Of the 47 mother-child dyads,
4 (8.5 percent) spoke only Spanish, 7 (14.9 percent) spoke only
English, and 36 (76.6 percent) spoke both English and Spanish to each
otheranndtotheresearcl‘ner.Elevennpercentofthemotherswerebornin
Mexico (first generation), 26 percent were second generation, 40
percent were third generation, and 23 percent were four or more
generations renoved from Mexico. Tie famnilies of 57 percent of the
olbjects lived in predominantly white neighborhoods, 27.6 percent in
mixed, 10.6 percent in Mexican-American, and 2 percent in Black
neighborhoods. Seventy percent of the families were Catholic; the rest
were Protestant. The average number of children per family was 3.4 with

ages ranging from infancy to 26 years. The age range of the mothers was

43
22 to 47 years with a mean age of 30.3, median age of 30 years, and
mode of 23 years. The mean age of the 47 children (28 girls and 19
boys) in the study was 5 years, four months. Eighty-one percent of the
mothers were married, 9 percent were divorced, 2 percent were
separated, and 8 percent were never married of whom 4 percent were
cohabitating. Most (62 percent) of the mothers were not employed
outside their homes; 1 (2 percent) was a proprietor of a fast food
business; 4 (8.5 percent) were clerical workers; 4 (8.5 percent) were
semi-skilled workers, and 7 (14.9 percent) were unskilled workers; the
remaining 4 (8.5 percent) were students. The mean educational level was
tenth grade for mothers and eighth grade for fathers. Their years of
formal education ranged from O to 14 years for mothers and 0 to 16
years for fathers. The majority of the fathers were semi-skilled (l7),
and skilled (7) laborers; 2 were unskilled laborers, l was a clergyman,
and 2 were students. Of the 40 fathers, ll (27.5 percent) were

unenployed .

The Research Instrumts

The data measures obtained for this study were (1) family
demographics, (2) maternal child-rearing attitudes, (3) maternal
child-rearing behaviors, and (4) child behaviors. Family denographics
and pertinent information were recorded during an interview with the
use of the Background Information Instrument developed by the
researcher for use in this study and included as Appendix B. The
instrument used to obtain measures of maternal attitudee expressed

during an interview was selected for its successful use with

44
Mexican-American parents of similar social class. The instrument
consists of adaptations of the Parental Attitude Research Scale
(Cromwell, 1969) and the Cornell Parental Behavior Inventory (Weigert,
1968). It is included as Appendix C with a translated version in
Spanish. The Maternal Teaching Observation Technique or MPOI‘ (Iaosa,
1980 a & b) was used to measure maternal behaviors. The Child Behavior
During the MI‘OI‘ (CBDMI'OI') was developed by the researcher and was used
to measure child behaviors during administration of the Maternal
Teaching Observation Technique (MI‘OI'). The war and CBDMI‘OT are
included as observation instruments in Appendix D along with the

protocol for recording observations.

Interview Instrnnents

W Information. Demographics and other information
pertinent to the study were recorded during the initial
rapport-building interview of the mother in her home. Denographics
include mother's age and marital status; mother's and father's
ocenpation, educational level, and place of birth; child's date of
birthandgender;agesandgendercf allchildreninthefamily; and
place of birth of maternal grandparents and great-grandparents. Other
pertinent information include ethnic predominance in the family ' 5 area
of residence, mother's religious preference and whether or not she
attends church regularly. The presence of absence of Tinkertoys in the
- home was recorded for comparison with the Laosa studies (1978, 1980a,
1980b).

45

'I‘heParentalAttitudeResearchScaleandtheComellParental
Behavior Inventory (PARS and Cornell). Twenty-five statenents
constitute the PARS and Cornell attitudes survey . The statements were
selected by IeVine and Bartz (1978, 1979) to analyze child-rearing
attitudes and behaviors of Black, Anglo, and Chicano mothers and
fathers, and they were also used in this study for measuring
Mexican-American maternal child-rearing attitudes (see Appendix C).
Cromwell (1969) adapted items from both the Parental Attitude Research
Instrument by Schaefer and Bell (1958) and the Cornell Parent Behavior
Inventory (Weigert, 1968) to use with a lower-class population. IeVine
and Bartz selected 17 items from the PARS which were most class
sensitive in that they discriminated at least 40 percentage points
between lower and middle-income samples in Cromwell's item analysis.
They conducted a factor analysis and identified five factors, which
they labeled Acceleration of Autonomy, Casual Use of Time,
Eqnalitarianism, Value Strictness, and Devalue Permissiveness. The
remaining eight items were selected from the Cornell, which had been
restated for parental rather than child responses. Fair of the items
were labeled Lack of Support and four, Lack of Control. Because of the
high reliability of the item on the Cornell in previous research, no
additional factor analysis was undertaken by IeVine and Bartz. Scores
for each interview statenent were generated by assigning numerical
values to Likert scales ranging from "strongly agree” to ”strongly
disagree“ on the PARS items and "never" to ”very often” on the Cornell
itena (see Appendix C). IeVine provided a copy of the instrument to the

researcher .

46

In this study, the PARS and Cornell instrument was erployed to
obtain measures of maternal attitude. Acceleration of Autonomy, Casual
Use of Time, Bqualitarianism, Value Strictness, Devalue Permissiveness,
Lack of Support, and Lack of Control were the variables used for
categorizing mothers ' attitudes expressed during an interview into
permissive, authoritative, or authoritarian child-rearing patterns.
Bammrind's theoretical model (1971, 1975, 1978) was applied by this
researcher to determine how mothers with permissive, authoritative, or
authoritarian child-rearing attitudes would be expected to score on
each of the 25 measures cotprising the PARS and Cornell. For enample,
on Acceleration of Autonomy a mother with permissive child-rearing
attitudes would be expected to have a low score according to Bannind's
(1978) theoretical definitions; a mother with authoritative
child-rearing attitudes would be expected to have a mdium—low score;
and a mother with authoritarian child-rearing attitudes would be
expected to have a high score (Baumrind, 1978, p. 266). Dr. Eileen
Earhart, an expert in the areas of child developent and parent
education, concurred with the researcher's determination of expected
scores, thereby, establishing content validity. The expected scores for
each measure and child-rearing pattern were then ranked as shown in
Table 3.1.

The following procedures were followed in order to determine the
pattern of maternal child-rearing attitudes for each subject:

1. Mean scores of the seven a priori classificatory variables of

the items on the PARS and Cornell Instrument were corputed for each

subject.

47

2. Descriptive statistics for the seven variable categories were
compnted for the sample (see Table 3.1).

3 . The seven categorical mean scores were rank-ordered for each
subject manually.

4. Chi-square values for each child-rearing pattern were
calculated using the observed ranks and the expected a priori
determined ranks for the variables according to the application of
Baumrind's (1966, 1971, 1978) theoretical model.

5. The smallest of the three chi-square values indicated the
child-rearing pattern for each subject.

Table 3 . l—Maternal Child-Rearing Attitudes Scores on Interview

 

 

Variable Mean Variance
Acceleration of Autonomy 2.574* .185
Casual Use of Time 2.415** .330
Equalitarianism 2.915* .330
Value Strictness 2.702* .181
Devalue Permissiveness 2.539“ .167
Lack of Support 1.622*** .244
Lack of Control 2.021*** .252

 

*Scores above 2.5 indicate general agreement.
“Scores above 2.5 indicate general disagreement.
***Scores below 2.0 indicate behavior is often shown.
N s 47

48
Table 3.2—Expected Scores and Ranks on Child-Rearing Attitude

 

 

 

 

Measures
Child-Rearing Attitudes
Permissive Authoritative Authoritarian
masure Score Rank* Score Rank* Score Rank*

 

Acceleration of Autonomy Low 6 Medium 3 . 5 High 2
Casual Use of Time High 2 Medium 3.5 Low 6
Equalitarianismn High 2 ngh 1 low 6
Value Strictness Low 6 Medium 3.5 High 2
Devalue Permissiveness row 6 mdium 3.5 ~ High 2
Lack of Support radium 4 Low 6.5 Medium 4
Lack of Control High 2 Low 6.5 low 6

 

*Mid—ranks were used in cases of ties.

49
daservation Instruments

The Maternal Teaching daservation Technique (morn.

The MI'OI‘ was developed in English and Spanish versions by Laosa
(1980a, 1980c) for use with American subjects from both Chicano annd
Annglo cultural—linguistic populations. Laosa (1980b) determined
parallel form consistency between the two forms of the instrument by
conputing Spearman rank-order correlations , corrected for ties , for
Form A and Form B raw frequencies for each ethnic sample. Estimates of
consistency were significant with a median of .68 for Anglo-Americans
and .70 for Chicanos.

For this study, slight modifications were made to the MIUI‘ by
expanding the definitions of four of the nine categories measured to
include additional verbal and visual behaviors. This was done in the
interest of measuring as many observed behaviors as possible. Laosa
(1980c) originally included sone of these behavioral measures, but
deleted them to enhance interobserver reliability annd parallel-form
consistency. Because there was only one observer in this study,
establishing interobserver reliability could not be done. Birthermore,
t-tests revealed no statistically significant differences between Fonts
A and B, indicating that there were no significant differences between
the behaviors observed during the admninistration of each form of the
MIDI. Forms A and B were, therefore, combined for further analyses. The
Ml‘UI' is contained in Appendix D with the modifications delineated.

The MPOI‘ was utilized in this study as a means by which to measure
observed maternal behaviors . The nine behavioral measures on the

instrument are (1) Inquiry, (2) Directive, (3) Praise, (4) Negative

50

Verbal Feedback, (5) bbdelinng, (6) Visual 0.16, (7) Physical Affection,
(8) Positive Physical Control, and (9) Negative Physical Control. The
frequency of observed behaviors was recorded on a protocol on which the
measures are abbreviated (l) I, (2) D, (3) P, (4) —VF, (5) M, (6) W,
(7, PA, (8) +PC, annd (9) -PC, respectively (see Appendix D). For the
purposes of this study, Baumrind's (1966, 1971, 1978) theoretical model
was then enployed to rank the measures according to their expected
frequency of occurrence in permissive, authoritative, and authoritarian
patterns of child-rearing practices. For example, the expected order
of the behavior measures for a mother with a pattern of permissive
child-rearing practices would be: (1) Praise, (2) Inquiry, (3)
Physical Affection, (4) Visual One, (5) Modeling, (6) Positive Physical
Control, (7) Negative Verbal Feedback, (8) Directive, (9) Negative
Physical Control. Table 3.2 indicates the rank order of the measures
that was expected for each pattern of child-rearing practices. This
rank ordering was applied to the data collected for em individual,
and a determination was made as to which order most nnearly matched.

This rank ordering was applied to the data collected for each
mother in order to determine her pattern of child-rearing practices.
The following procedures were followed to make each determination:

l. Thenine rawfrequenciesonFormAandFormBof theMrOI'were
totaled and verified. .

2. Proportions for each variable were corputed by dividing the raw
frequency of each variable by the aim of the respective nine

frequencies on the forms .

51

3. Thetotalsof FornrsAandBwere combined since there were no
significant differences between the paired variables on the two formns.

4. A grand total frequency of the combined fornns of the MIOI‘ was
used to determine the proportions for each of the new nine conbined
variables.

5. The final nine frequency proportions for each subject were
rank-ordered.

6. The chi-square goodness of fit was conputed between the
observed variable ranks on the MDOI’ and the a priori determined ranks
of the variables according to the criterion of Baumrind's theoretical
model as indicated on Table 3.3.

7. The smallest chi-square for each subject indicated the best fit
for each pattern of child-rearing, i.e., permissive, authoritative, and

authoritarian .

52
Table 3 . 3- Maternal Behavior measures Ranked According to

Child-Rearing Patterns .

 

Child-Rear ing Pattern

 

MI‘OI‘ Measures Permissive Authoritative Authoritarian
Ingliry 2 1 9
Directive 8 2 1
Praise 1 3 8
Negative Verbal Feedback 7 8 5
bbdeling 5 7 2
Visual Cue 4 5 3
Physical Affection 3 4 7
Positive Physical Control 6 6 4

Negative Physical Control 9 9 6

 

53

Child Behavior During MIUI‘ (CBDEOI').

The CBDMI'OI' was developed by the researcher for measuring the
child's behavior during administration of the MCI. Its objective was
to measure child behaviors concurrently with maternal behaviors during
the MI'OT strucured teaching tasks. When the MI'OI' was pilot tested with
five mother-child dyads to measure child behaviors during the teaching
tasks , eight behavioral categories were found . The eight behavioral
measures on the instrument are: (1) Positive Verbal Response, (2)
Negative Verbal Raponse, (3) Inquiry of Mother, (4) Physical Response,
(5) Imitation of nether (6) Task Involvement, n7) Nonconpliance, and
(8) No Response. Frequency counts for each measure were recorded on the
protocol used to record maternal behavioral masures (see Appendix D).
Abbreviations on the protocol for the eight measures of child behavior
are (l) +VR, (2) -VR, (3) Q, (4) PR, (5) IM, (6) T, (7) INC, and (8) NR,
respectively. The measures were, then, used to further evaluate

patterns of maternal child-rearing practices.

W Reliability

To insure efficiency and consistency in data collection, the
researcher trained herself to observe mother annd child behaviors
simultaneously by viewing video tapes of similar parent-child
interactions on video tapes obtained from the Michigan State University
Institute for Family and Child Study. During this process, the
researcher experimented with various forms annd methods of coding and
recording the observations 'until the protocol used in the study was

developed (see Amendix D). A .98 construct validity of measurenent on

54
the simultaneous use of the MI‘OI' and CBDMI'OI‘ was established with the
expert assistance of Dr. Lillian A. Phenice, faculty member of the
Department of Family and Child Ecology, Michigan State University, who
observed annd recorded several video taped interactions with the

researcher prior to data collection.

Validity
Although the Maternal Teaching Observation Technique was

originally developed to measure maternal teaching strategies, in this
research it was also considered a valid measure of child-rearing
practices during a teaching task. An assumption underlying Iaosa's
(1980c) development of the MI‘OT is that a motherYs behavior in an
observed structured situation is representative of her teaching
behavior during everyday situations in which she might teach her child
similar tasks. Research evidence for this type of validity was found by
Stallings annd Porter (1981) as a result of their observations during a
large-scale , government-supported study of day care homes . The
structured observation was , therefore, considered a valid technnique for
measuring child-rearing practices during a maternal teaching task,
particularly since the observation took place in the familiar home
environment.

External validity was established by administering the structured
observation prior to conducting the interview survey of child-rearing
attitudes. Thus, the potential effects of an interview about
child-rearing on maternal behaviors during the structured observation

were controlled .

55

Concurrent validity was established between the CBDMI‘OI' and MI'OI'.
The set of variables developed by the researcher for measuring child
behaviors during administration of the MIDI was found to be related
significantly (.80) to the set of maternal behavior variables on the
MIDI. The observations of the mother-child dyads could, therefore, be
measured concurrently; and scores on one instrument can be expected to
predict scores on the other instrument. The concurrent validity of the
two observation instruments could not be answered since the researcher

was the sole observer for both mother and child.

Design Over Variables

In this section, the inter-relationships of the research design
variables are discussed. The major interdependent variables in the
study were (1) maternal child-rearing attitudes, (2) maternal
child-rearing practices, (3) child behavior relative to maternal
child-rearing practices, and (4) denographics relative to maternal
child-rearing attitudes and practices. The denographics include
mother's age, mother's and father's years of schooling, mother's
generation renoved from Medco, mother's religions preference, father's
onployment status, child's gender, and number of girls and boys in the
family. Three analyses of the interdependencies between data sets were
made. In the first, the interdependent variables are child-rearing
practices with child-rearing attitudes, which are subdivided into
permissive, authoritative, and authoritarian categories. In the second,
the interdependent variables are child-rearing practices and child-
behaviors . The variables are also subdivided into behavioral

56
categories. In the third analysis, the interdependent variables are
child-rearing attitudes and practices with family demographics.

Statistical bbdel of Analysis

Inferential statistical models were used to test the hypotheses in
this study. The inferential analyses included the 2 test of
significance for a single sample, - chi-square, Student's t—test,
discriminant analysis, multivariate annd univariate analysis, Pearson
correlation, and analysis of variance. All hypotheses were tested at
the alpha = .05 level of significance. Descriptive statistics were
calculated where applicable, annd they were used to describe the sample.

The ZTestof Sig’ficance foraSingleSmpLe; This testwas
appropriate for the first two hypotheses since the data were interval
in nature. The z test assumes, conventionally, that a sample size is
equal to or greater than 30, that the population is nnormally
distributed, and that the mean and standard deviation are known
(Ieonard, 1976). To conpute the 2 statistic for Hypothesis 1 annd 2, the

following formula was used:

p-a

where: p = proportion (i.e., n/N)
a = a fixed criteria defining the majority,

0.6 (60 percent) in this case
q= (l -p)
P‘-'-0

l: pfO

cm :11
o

57

Pearson's Chdrsggare Tast. This test is a measure of statistical
sigificance, is a representative statistic for nominal variables, and
helps to determine whether or not a systenatic relationship exists
between variables. In the chi-square test of statistical significance,
which is based 'on the null hypothesis, the chi-square measure assumes
there is no relationship between the variables (Babbie, 1979) . For
purposes of this study, the chi-square tests for goodness of fit and
independence were need. These tests assume:the sample has been randomly
and independently selected; however, they are sufficiently robust to
allow violation of the assumption, as was the case in this study. The
chi-square goodness of fit test statisic was appropriate where the data
consisted of frequency counts . It was appropriate for testing the
degree of association between nominal scales since it is sensitive to
any systenatic departure from independence or total nonnpredictability
(Freeman, 1965). The test of independence was computed for categorical
variables, i.e., permissive, authoritative, and authoritarian
childrrearing patterns.

The Student's Thtest. T—tests for paired observations and
independent samples were computed.for differences between sample means.
The purpose of pairing was to reduce the effect of subject-to-subject
variability. Independent samples refer to subsamples within the single
sample of this study. The 1; statistic assumes a normally distributed
newton variable where the mean is known and the population variance is
estimated from a sample (Nie, et al., 1975).

Discriminant Analysis. This statistical technique is used to

distinguish between two or more groups of cases, which have a

58

collection of discriminating variables that measure characteristics on
which the groups are expected to differ. The effects of a collection of
interval-level independent variables on a nominal dependent variable
are mlculated. The two research objectives of this technique are
analysis and classification. Although the latter was not used in this
study, the analysis aspect provided media for interpreting data. For
enample, there are statistical tests for measuring how successfully the
discriminating variables actually discriminate when conbined into
discriminant functions by forming linear conbinations of the
discriminating independent variables. Weighting coefficients identify
the variables which discriminate most. Further evidence of group
differences can be derived from group centroids, which are the mean
discriminant scores for each group. The centroids summarize the group
locations in the space defined by the discriminant function. There was
one discriminant function, canonical, for this analysis since there
were only two groups. The canonical correlation, which is a measure of
association between the single discriminant function and the set of
variables that define the groups, was used as an aid in judging the
importance of the discriminant function in this study (Nie, et al.,
1975) .

mltivariate Analysis of Variance (HANNA) . MANOVA tests of
significance included univariate F-tests for masuring the association
between several variables (i.e., dependent child behavior variables and
independent maternal child-rearing practices variables). The P value in
the univariate F-test is a function of the ratio (SSH)/ (SSE), where SSH

isthesumofsquaresduetothehypothesisandSSEisthesumof

59

squares due to error. The MANOVA procedure of the Statistical Package
for the Social Sciences (SPSS) program conputes four statistics used
for significance tests. They are Roy's largest root, Wilks' lambda,
Hotelling's trace, annd Pillai's criterion. F statistics are given for
each, with the exception of Roy's largest root. Univariate analysis of
variance results are given for the effect of the independent variable
on the dependent variables (Hull & Nie, 1981).

The Pearson Product-mt Correlation Coefficient (r). The
Pearson r is a measure of association that measures the strength of
relationship between two continuons variables (i.e., observed maternal
and child behaviors). The coefficient has a specfic numerical value for
anygivensetofpairedvariables equal tothemeanof theproductsof
the z-score for the pairs. The strength of relationship refers to the
linear correlation between two variables . Assumptions involved in
testing for significance are random sampling from the population
concerned, bivariate normnal distribution, linearity, annd interval-level
data (Spence et al., 1976).

Analysis of Variance (ANNA). ANOVA is a statistical technique
used to measure group differences. It assesses the effects of
categorical independent variables (i.e. , child-rearing attitudes or
practices) on an interval-level dependent variable (i.e., age). Three
assumptions unnderlie ANOVA: (1) that the populations from which the
groups in the study were drawn are normally distributed, (2) that the
variance values for the populations are equal, and (3) that the
subjects of the study have been randomly and independently drawn from

their populations. According to Spence et al. (1976), practical

60
usefulness of the ANOVA procedure may be nearly as great when one or

two of these assumptions are met as when all are satisfied.

bothods Used in the Analyses of Data

The purpose of each analysis, along with the appropriate data and
statistics, are summarized in Table 3.4. The conputer procedures
applicable to the study are described in the Statistical Package for
the Social Sciences (Nie, et al., 1.975; Hull & Nie, 1981). The Control
Data Corporation Cyber 750 conputer of the Michigan State University
Conputer Laboratory was used to perform the statistical analyses.

Research W
The testable hypothcees for this study are stated below in the
null form. The hypotheses are stated in more detail in Chapter IV.
Hypothesis 1. There will not be a permissive pattern of
child-rearing attitudes expressed by Moan-American mothers.
Hypothesis 2. There will not be an authoritarian pattern of
child-rearing practices observed in bexican-American mothers.
We 3. There is no relationship between the child-rearing
attitudes annd child-rearing practices of Mexican-American mothers.
Wis 4. There is no relationship between the child-rearing
practices of Mexican-American mothers and their children's behavior.
Wis 5. There is no relationship between Mexican—American
maternal child-rearing attitudes and family demographic variables .
Hypothesis 6. There is no relationship between Mexican-American

maternal child-rearing practices and family derographic variables.

61

Table 3.4—Methods Used for Analyses of Data.

 

Purpose of Analysis

TypeofDataUsed

Statistics

 

Test Hypothesis 1

Test Hypothesis 2

Test Hypothesis 3

Test Hypothesis 4

Tat Hypothesis 5

Test Hypothesis 6 .

Interviews

Observations

Interviews &

Observations

Demographics

Demographics

x2 goodness of fit test, 2

test for a single sample

Paired t-test, x2 goodness
of fit test, Z test for
a single sample

X2 test of independence,

D i s c riminant analysis ,

t-test for

independent samples

Paired t-test, Pearson r,

M

X2 test of independence,
ANOVA

X2 test of independence,
ANOVA

 

62
Procedures

A roster of potential subjects was compiled fron various sources,
such as the Cristo Rey Catholic Church, Cristo Rey Conmnity Center,
annd several Cnicano community leaders . Telephone or face-to—face
contacts were made with the potential subjects, informing them about
the study and that the researcher was seeking Mexican~American mothers
and their kindergarten-aged children as participants. The researcher
explained in English, Spanish, or both languages what participation
would entail. All but three of the mothers who were contacted, readily
volunteered to participate. An appointment time was scheduled, and the
mothers were requested to telephone the researcher if the child did not
wish to participate. All those mothers who initially volunteered took
part in the study with their children.

Data were collected in the hones of the subjects during the winter
of 1934. Most home visits took place on weekday mornings or afternoons,
but some were scheduled on weekends or early evennings for employed
mothers. The visits were approximately one and one-half hours in
length. The objective of making tune-visits was to take the data
measures in the hone environment where Internal-child interaction
occurs most naturally.

Rapport was established with the dyad upon the researcher's
arrival. The nnother read and signed the research participation consent
form (see Appendix A) which was written in the language for which she
had indicated a preference. Sane mothers asked the researcher to read
the consent form to them, stating they could not read Spanish or that

they could not read well . Rapport-building conversation continued

63
primarily with the mother as family demographics were recorded on the
Background Information sheet (see Appendix B).

The mther was then asked to sit at a table with her child where
she would teach the child how to make two replicas of Tinkertoy models
(administration of mar Forms A & B). Sore mothers cleared the living
room coffee table. Others led the way to the kitchen or dining room
table. A few preferred the floor; one famnily did not have a table. The
mother was given a Tinkertoy robot model (Form A, MI'OI‘), which was
glued together to prevent it being taken apart. She was asked "to
teach" her child "how to make" a toy like the model and to ignore the
researcher's presence during the process. In Spanish the phrases used
to instuct the mother were, "haga que _ aprenda" or "hagalo(a) que
aprenda." The researcher positioned herself apart from the dyad, but
retained close enough that she could readily observe the faces and
hands of both mother annd child'. If other people were present, they were
asked not to participate and to be as quiet as possible. Siblings were
distracted by the researcher during the visit with other toys included
in the see-through, plastic shopping bag which contained supplies for
the study. The researcher explained to the dyad that she would time her
observation of then for five minutes with a timer, but emphasized that
they were not being timed and could continue the task after the timer
went off if they wished. met dyads were so engrossed that they did
continue the task if the toy was not conpleted within the five-minute
observation . The researcher manually recorded on a protocol the
maternal and child behaviors demonstrated during the observation,

according to the categories on the MOT and CBDMI‘OI‘ (see Appendix D).

64
Because most children were eager to begin the next model, the interval
between administration of the two forms of the MIDI was generally only
longenoughtoputaway FormA. Themotherwasthengivenanother
Tinkertoy model (Form B, MI‘OI'), annd the instructions and observation
were repeated.

Shortly after the teaching tasks , the adapted PARS and Cornell
attitude survey interview was administered to the mother (see Appendix
B). Any children present were given a large cylinder of Tinkertoys, or
other toys more appropriate for very young children, to entertain them
during the interview. Sitting next to the mother, the researcher read
tonerinthelanguageofherpreference,eachof the258tatelentson
the PARS and Cornell instrument and circled the mother's responses on
the Likert scale. In sone instances, the mother read along with the
researcher and circled her responses herself . Explanations of some
statements as well as clarifying enamples were given by the researcher
when necessary. Sane mothers were quite reflective during the
interview, recalling incidents with their own children and/or conparing
their own child-rearing with the ways they were reared. Following
conpletion of the interview, mothers frequently expressed their
enjoyment of the researcher's visit and offered names of other
potential subjects for the study.

Ethical Considerations
The required documentation was submitted for review by the

University Committee on Research Involving Human Subjects (UCRIHS).

65
Approval was granted by UCRIHS prior to the selection of subjects for
the study.

Various ethical concerns were taken into consideration during the
implenentation of this study. Of forenost importance was that the adult
subjects (mothers) understood their option, as well as that of their
children, not to participate in the study. A consent fornm (Appendix A)
was provided to each mother to ensure that she unnderstood the
explanation of the study, that there were no inherent risks to either '
her or her child, and that her participation was entirely voluntary. It
was also made clear that the mother had the option of discontinuing her
participation, and that of her child, at any time. As previously
stated, the procedures of the researcher's home visit, which would last
approximme one to one annd one-half hours, were discussed in the
initial contact. The mother was informed after the observation
technique that the researcher was interested in what Mexican-American
mothers think is important in the rearing of their children and,
thereby, encouraged her to participate in the maternal attitude survey.

The mother was informed during the home visit that any information
she provided and observations made by the researcher would be treated
asoonfidential. Stewasshownthathernamewouldnotappearonanyof
the forms on which the researcher recorded information annd assured that
names would‘not be connected to any data used for analysis, the
dissertation, or any piblication of the information. The subjects were
also told that the results of the study would be made available to them
upon request.

CHAPTER IV

RESULTS OF DATA ANALYSES

wacfthedataanalyseearereportedinthischapterunderthe

sections: Hypotheses Tet-s, Statistical Resnlts, and Summary Data.
Hypo:heseTesls

Inthissection,the hypothesesaredn’scussedindnesameoderathey
arefistedinChaptemIoflmahzhougheadnmmmpothesisisreazatedina
moeoedseform.Areportofwhetheronote:afisdcaldgnmficancewasfound
foflowseachhypothesis.

Hypozhdsl

H01 Ttemajxitydnedm-Amodcanmtheswmnam

poundvechfld-reaong atfihxiesamedbythacomflned

WitheParental ReearchScalemARS) andConen

Parent Behavionnventoymonem oalinterview.

Inodertotestthehypothesis,meanscoreswerefirstcomp.ntedforthe
sevenapioficategodcalvafiables(Acce]erafioncfAutonomy,CanalUsecf
Time, Equalitarianism, Value Strictness, Devalue Permissiveness, Lack of
Support, Lackof ConooDmneasuredbythe 25 itemsonthe combined PARSand
Conell audtude survey. Deceptive statisties were then computed for the seven
categon'calvariablesofscoresonthe PARS and Cornell instrument. Next, the
meanscorescfthesevencategoicalvariables were rank-orderedforeach
subject-

66

67

,Chi.-squareva1naforeachchfld-rearingpattern werecalculatedusing
theobservednadcsandtheenpectedranlscfthesevenoategoocalvadames
discussedinChapterIIL ThesmaDestdthethreedni—squarevahesindicated
thednild-readngpatiernforeachsnbject. ltwasfoundthat7motherswere
permissive, 13 were authoriradve, and 27 were authoritarian in their exposed

Ttepercentagedisbdbutfionofthethreepatteneofdnfld—readngatfi-
undeswascomputedmeTableLDandyieldedacombinedooporfionfor
authoritative and authoritarian cold-rearing attitudes cf 0.85, indicating that
mostcfthe Mexican-American mothers did not expires permisive dnild-reao'ng
attitudes. Finally,theszorasing]esamp]e wasusedtoevalnatethe sig-
nifn’canceofthepnoporo'oncfthose mothers whoenanesedauthon'tativeand
authoritarian dim-rearing altitudes. The result, Z =4.8, was significantatthe
.Olhmmndicafingszafisdcalsignfificanceatonearamajnfitybvelof 0.60
(60 percent). The nullhypothesis, therefore, failedto berejected. The majority
cf Mexican-A merican mothers in the sample exposed authodtative and authori-
tafianomld-reafingaudmdaasmeaanredbythePARSandConenimherfiew
instrument. Thefindingsithishypcthesisfafledtosippottheresearchfinnd-
ingsof IeVineandBartz (1979)upon whichthehypctheeis was based.

68
Table 4.1—Child—Rearing Attitudes Based on PARS and Cornell Interview.

 

Child-Rearing Attitude Percentages

 

 

Permisive Authoritative Authoritarian
.11 .2. 9. 3. 2 l
7 0.15 13 0.28 27 0.57
N = 47
Hypothds 2

 

H02 The majzitychedcan—Ameicanmotheswflnnctbeob—

served to use authoitarian child-rearing pacific:- dining

admniniflratimdthe Maternal Teaching Observuion Technique

(MTOT). A

As with the previous hypothesis, several proosanres were performed in
ordertotestthehypothesis. First,flneninerawfrequenciesonFcrmAand
FomBitheMTOTweretotaJedandverified,andooportioaforeachvanda-
blewereoomputedbydividingtherawfrequencyofeachvariablebytheenm
offlnenninefrequencies for each formcfthe MTOT.

TtettestforpairedsampIeswasflnenneedtoevalnatethedifferencss
between the effects of Forms Aand B across their respective nine variables.
The meanns dtwopairedvarn'ablswerefoundtobedgnificantlydifferent. Inn-
quiryoccun'edwirhgreaterfrequencyduringTask A,whir:hwasadministered
first<t=2.35,df=46,p = .023). Negativeverbalfeedback was more frequent
during Task B (t =I--=-2.21, df = 46, p= .032). Since the remainning 78 percent
variabledifferencabetween unepairedobservafions werenotsignificant,

69
Fonms A and B were combined. A grand total frequency for the combined forms
was used to determine the proportions for each of" the new nine combined
variables, and the final nine frequency proportion for each subject were
rank-ordered.

Chi-square goodness of fit was computed between the omerved variable
ranks on the MTOT and the a priori ranks determined by employing Baumrind's
(1967, 1971, 1978) theoretical model to rank the measure according to their
expected frequency d occurrence in permissive, authoritadve, and authoritarian
patternscfchild—rearingoacdces. Foenample,theexpectedo:derofthe
behaviormeaenresforamotherwithapeomissivepaozemofchfld—reaning
practice would be: (1) Praise, (2) Inqairy, (3) Physical Affection, (4) Visual Cue,
(5) Modeling, (6) Pceirive Physical Control. (7) Negative Verbal Feedback, (8)
Directive, (9) Negative Physical Control. Table 3.2 indicates the rank order if
the measnresthat wasexpectedforeachpaoerndonild-readngpractices. The
smanestchi-squarefor each particularsnb'pctthusindicated thebestfitfora
particularchild—rearing pattern.

Apercentagedistdbutioncfthetmeepatternsofofld-reaohgoacdces
was then computed (Table 4.2). The gropertion for authodtanian practices was
fomdtobe.468,indicafingthatneaflyone—talf<fttemothaindneample
wereobsenvedtouse authoritariandfld—rearingoacficesmdnemcmred
teachingtask. Tre percentagedistobufionresiltedin acombinedooportion cf
0.53 for permimfve and authoritative child-rearing practice, indicating um:
slightlyoverone—halfcfthe Mexican-American anthersdidnotuse authoritarb
ianchild-reao‘ngoactices. Finany,uneZtee.cfsignificanceforasingle
sample wasusedtoevalnnatethea’gnificancecithe53proportionof mothers
whousedpermixiveandauthoritatived'fild-reafingnractices. Theresilt, z =

70
0.96 was not significant at the .05 level, indicating the .53 proportion was not
statistically e.gnificant below the majority criterion of 0.60. The null.
hypothesis, therefore, was rejected. This finding that Mexican-American
maternal child-rearing practices were almost equally chivided into authoritative
andaufinofitafianpattemsleohtentafive supporttotheresearch findingscf
Kagan and Ender (1975) upon which the hypothesis was based.

Tabb 4.2—Chi1d-Rearing Practices Based on MTOT Oteervafion by Percentage.

 

Child-Rearing Practices

 

 

Pemisive Authoritative Authofitarian
£1 3. £1 3. 2 i
2 0.04 23 0.49 22 0.47
N a 47
Hypddnesis 3

 

803 There is no relationship between exomed Mexican-Andean
maternaldnfld-reafingatfiomasmeaanredbyfinePARSand
Conefloalinneview,andtheiroteevedchfld-¢earingpacfices

duringadminisoafionitheMTOT.
.Ttepaliemscfdfild—reafingafijtudesandpracficesdefivedfromproce—
duresfor testing the oevious hypotheses were the nominal variables for the

chi-square test cfindependence employedto test this hypothesis. The resulting

7l

sin-square contingency table revealed important descriptive statistics. about the
anfitudsendpracficescfthemdzheseudiedwrficharedisomedandoe-
sentedapartcftheSummaryDatasecfioncfthischapter. Duetotherela-
fivelysmanoemptycenfrequendesithepermisiveoacficescolumnonthe
dn'i-sqnare table (Table 4.8) which included the total 47 mothersin the sample,
the permissive column was eliminated from the contingencytable in oder to
arfiveatamoreaconratechi-squareresmlt. Becausethetwomotherswith
pe'nmisdyepnacficerepresentedaooporfimdthetotalsampleunatwasmt
ignificanhflnepermussiveofild-reafingoacficacategorywaseimmatedfiom
thisandsubseqnentanalyses.

The on-squaretestreniltforthishypothesisdxzm df) a3.93,p=.14,
wasnote’.gnificant,hdicajng_maternaldnfld-mafingatfinflesondoacfices
wereindependentinthissondy. Thne,themnllhypothesisfailedtoberefected.
This result was incongruent with Baumrind's (1971) findings that parental
Wesweeoedictablyrelatedtoobsevedparentalbenavor,althonghune
sharedvafiancewassmall.

Furtheranalysisdthematernalpatie'nedctfild-reafingafidmdesand
oacficebydiscfiminantfuncfion, however, revealedreanltsenppo‘tingthis
hypothesis. Theeoontrastingontcomesareattributedtodngthescoesfrom
fineatfitudesmeasnneratherflnantheclasificafioncfthesooesinto
permisiye, authoritative, annd authoritarian categories. The chi-square test was
appliedtomminalscalescfmeaanrementwhichrepresetthemostbaiclevel
ofdata. Posthocoocedureswereneedtofurtherexaminefinedataatits
oiginallfigherintervallevelduetoposiblebssofinformafimbyredudng
'mesurement levels. Another remon for further data analy'e was that the

oni-sqnaretestisatestoffignificanceandnotameasuredaociafion;

72
canequtly,itcanbeisedmdetennmetheacceptakflityofanhypcthesishm
cannctbeused aloneforaesessingthe degreecf relationshipbetween variables
(Leonard, 1976).

Disaiminantanalysis was applied,withthecmtinucussca1es¢child-
reafingatfimdescoresemployedasdiscfiirdnafingvafiablesbetweenme
categcdescfauthon'tadveandauthofitafian dfld—rearingg'acticesmee'rahle
4.3). The canonical correlafim (.569) indicated that the canonical function's
ahi'litytodiscciminate between thetwc patterrecf ctfild—zeafinggracticeswas
significant, x2 (7 an a 15.48, p = .03. Based on function coefficients, the
variables meaning attitudes, Acceleration of Autonomy, Lack of Control,
Canal. Use of Time, and Equalitan'am’sm, were found to best discriminate
baween authoritative and authoritarian practices.

Amuhvafiateanalysiscfvafiancetestcfmeatfinldevafiableswith
clfiJd-readngpracticee revealedadgrfificantdifferencefor Accelerafinnof
Autonomy, F (1,43) = 13.24, p = .001, and for Canal Use of Time, F (1,43) =
4.57, p = .038. These findings suggested that informatim may have been lost by
reducing the mewn‘ement level of the atfitude dm:a. Based m these findings,
twodthesevenvafiablesmtheattitudeinstmmemzcolfldbeiisedto
discdminatebetweenauthofitafiveandautmfitafianpauzermcfdfldieadng
evidencetorefectthenull hypothesisthatthere wasno reladonship between
child-rearing attitudes and practice. Conversely, findings from the Cid-square
testcfthe mminallevddatag'cvidedevidencetnretainthenunhypcthesis.
Baedmfindingsmthelfigherleveldatatedzedbymorembustazadsdcal
tests cf significance, the nun hypothesis was rejected.

73
Table 4.3—Canonical Discriminant Function Between Child-Rearing Attitudes as
Measured by the PARS and Cornell Attitude Survey Interview

and Child-Rearing Practices Measured by the MTOT.

 

Canonical Correlation = .569
x2 (7 df) = 15.43, p = .03*

 

 

Coeffidenm Variables

-.84070 Acceleration of Autonomy
.36436 Casual Use cf Time
.35507 Equalitafianism
.13306 Value Sm
.30648 Devalue permissive-less
.27976 Lack & Support
.54141 Lack cf Cam].

snap—cam 9m

.66230 Aer

--.69241 Authan'tarian

 

N845
*p< .05

74
Hypothesis 4

There '3 no relatimship between the observed child-rearing

pactices d Mexican—Andean mothers, determined by the MTOT,

and ma: children's cmaved behavior manned by the Mument,

Child Behavior Dlning the MTOT (canu'ro'r).

The following mcedures were followed to test the hypothesis. First, the
eight raw frequenciesancrmsAandB cit-he CBDMTOT weretctaled and
verified. Proportiiore for each variable were computed by dividing each raw fre-
quency byitsrespectivegrandtctalof the eightfrequenciescn eachcbserva-
tion form.

A matchedpairst-ted: wasthenusedtoevaluatethedifferencesbaween
FamsAanddeieinstrumentacrcstheen‘ghtmatchedvafiablesmthe
forms. One variable, Negative Va'bal Repcnse, was found to be azmzistically
significant, t (46) = -2.01, p = .05. Significance was attributed to a greater
firequency of Negative Va'bal Feedback by children during the second task.
Because d the similarity between them, Farms A and B were combined. Each
newlycreatedvafiablewaaratiodthetwopairedfrequenciestothetctal
frequencies on the forms. Pearson Eroduct-moment correlaticre were calculated
for pairs cf variables 61 the MTOT and CBDMTOT (see Table 4.4). Statistically
a’gnificantpmifivecon'elaficnswerefoundbaween Praiseandlhquirycf
Mdzher, Praise and Talc Involvement, Modeling and No Rapcme, Pcnitive Phys-
ical cm and Pmitive Verbal Response, Pom'tive Physical Cartrol and Talc
Involvement, and Negative Physical Cmtrol and Negative Verbal Response. Sig-
nificant inverse relationships were found between Directive and Talc Involve-
ment, Praise and No Resporee, Visual Cueand No Response, Pmitive Physical
Control and Physical Response, Positive Physical Control and No Response.

'75

a 33m— mocooCEEm 5.
w a ”—96— mccouCESm «

 

 

 

. _ccucou
mmc. "a Mac. "a mom. “a mam. "a mam. "a .c.. "a «~mo. .a com. "a .ou.m»;a
name.-ut oo_o.ut acme.-»t _m__.nt ammo. “t e.a_.-ut mmaN. at ammo.-ut m>_soamz
_ocu=cu
.NNo. "a ~o~. ma ..Mco. "a .MN. ua .ceo. «a cc”. «a ~ao. «a .mwc. "a .ou_m»;a
amam.-nt o-~.-ut mmNM. n. mao_.-nt oun~.-.. o~eo.-.. «has.-.r ".mw. .t m>_u.moa
me_. "a mom. "a owe. "a was. .a "we. as .5”. .a can. ua cm". "a =o_numcc<
amen. at some. "t eoo~. .t -m~.-.t ~m-.-ut ease.-.t aces. at «ems. at .mu_m»;a
.nuo. "a “so. "a “NN. "a was. .a ~o_. .a mac. .a men. .a man. “a
meow.-ur m-~.-ut om~_.-.t ~_oc. .t amm_. .t QQMN. .t ghee. .t snag.-.t one _.=m_>
.._oo. "a one. "a «an. "a c_m. .a ace. «a mc_. "a cos. «a man. «a
Nrmm. “t mae.. at mwao.-ut mesa. "t ~mmo. at ceac.-ut ae_o.-ut mama.-nr a:..oeaz
.uaaummc
amo. "a mNN. "a mac. "a owe. .a _m.. .a ce_. "a «mo. na amm. "a .marm>
cm_~.-ut m~__. at ~¢a_.-ut saws. .t «was.-u. ooe_. at mocN. u. ~mmc. at u>_suamz
...-o. "a -_. «a .omo. "a mom. "a .~_. ua ..~oo. .a mm_. .a “mm. «a
mama.-nt Nm~_.-ut anew. at mama.-.t «can.-ut once. .t can“.-.t _omo.-nt ¢m_cta
mom. "a Nam. "a «.Noc. "a o~_. as am". "a can. as c~m. ua ecu. «a .
mama.-ut mao_. at «on..-ut _~e_. .5 mg... «a ~meo.-ut case. .r came. at m>_cumt_a
mam. na mac. ua Mao. ua ~a.. .a .cm. .a _~o. ua Nam. .a aoN. .a
case. at m_oc. at coa_. .t ma~_.-.r Kano. u. m~_~.-ut _mma.-.t mwac.-ut Reggae.
uncoam «macaw
oncoammz coco—passage: u:a&m>.o>c_ cacao: co omcoamom cannot be _aaLoN¢ possum:
oz swab copaou.§_ .ou_m>;a »c_:cc_ c>puommz u>—u_mca mm—nopgo>
_a:tmua:

m6.¢....> u_.=u

 

 

.3325; .3323 9:5 23 35.38. 32.38 .8.— mu:o_u.toou 5322.39 coflafiivémZoh

76

Muhvafiate tests for sgriificant differences between authoritative and
authoritarian child-rearing practices on the variables measiring child behavior
were then applied (Table 4.5). The multivariate tests of the efifects of child-
reafingg’acficevafiabhsmcrfildbehaviorvadablesasagrouprevealedm
significance (p = .21). Univariate analysis cf variance results for each d the
dfldbehaviorvafiablesrevealedclfild—reafinggacficeeefifecmm cnlycne
variable, Imitation of Mother, F (1,43) = 5.08, p =- .03. 11: was, therefore,
concludedthatme hypothesisfailedtobereficted. Significanceincne cutcf
eightvariablescn the CBDMTOT wasnctcmsideredsufficientto conclude that
there wasarelatiomlfipbetweencmerved maternalandchildbehaviors. The
finding was incongruent with Baumrind's (1971, 1973, 1975) finding that
differencesmdfildbehaworarerelatedtopmiermdparemaldfld-readng
practices.

77
Table 4.5—Multivan'ate and Univan'ate Analyses cf Child Behavior with

Authoritative and Authoritatian Maternal CIfild-Reafing Practices.

 

Child-Rearing Practices

 

Child Behaviors Authoritative vs. Authoritarian
2'. .P.

Behaviom as a Group 1.46 .21
Pdfive Verbal Reponse .77 ‘ .39
Negative Verbal Reqaonse .19 .66
Inquiry of Mother .32 .58
Physical Response .59 .45
Imitatim of Mother 5.08 .03*
Task Involvement 1.56 .22

‘ chompliance 1.08 .30
No Response

 

*Significance level: p < .05

78
Hypotheses
Thereisnorelatimshipbetween Meadow-American maternalchild-
rearingtfimdaandthefamflydemcgraptficvafiables,
805a mother’sage,
HOS!) mother‘syearscfschcoling,

H motha'sgenamzicnremovedfrom Mexico,

05c

3056 mctha's religious weft-truce,

305e father's years of sanding.

i105f father's employment stanza,

3059 chfld's gender,

30511 numba' of girls in the family,

8051 number cf boys in the family.

Tests of statistical significance were applied to the depetflent demo-
graphicvariablestotesttheireffectscn child-rearing attitudes. The one-way
analysiscf variancegocedure wasxnedtotefithekmervalvafiabhsfle,
mother's age, mother's years of schooling, father‘s years cf schooling, number of
male children inthe family, and mimbercffemaled'xildrenin the family). The
chi-squareted: wasappliedto the mminalvadablm (i.e., mb'pctchild'sgender,
father's employment stable, mother's generation removed from Mexico, and
mother'srelimg'eference. Onlycnedependentvan‘able wasfound satiatio-
ally significant. The effect of father's employment status on mother's child-
rearing adrimdes was highly significant, x2 (2, 39) a 10.487, p =- .005. Sub-
hypothesis Hosf of Hypothesis 5 was, metefote, rejected. The other 31b-
hypothesesfailedtoberejected. Descfipl'ivesatistia forthetestieach
variable by child-rearing attitudes are shown in Table 4.6.

79

Ho. v at.

 

 

 

 

OH. v 9:
ohm. me H mHm.um vvm. he N mmo.ns >HHEcw :H coucHch b .nmbsdz
vmm. mv H Nhh.um com. he N Hvo.nm >HHEcw cH cmuoHHcc m umossz
mow. me H vmm.nmx Ohm. he N mHo.~u~x Hoocmu m.UchU nomnbdm

«bmo. mm H mo.mumx «hmoo. an N nmv.oHuNx nounum ucmaonmEm m.umcucm
Ham. me H mHo.um mHm. be m msH.Hnm maeHoocom no what» m.hocnmm
mom. me H mHh.umx mmH. be m OHv.mnmx coccuoumum mSOHmHHmm m.umcuoz
Hes. me m moo.~nmx vmm. be c mo~.mu~x coexmzehxrmmcoeumhmcmc m.umcuoz
mom. me H ooo.um mow. be m Nmo.Hnm scHHoocom mo mummy m.uccuoz
oHN. mv H mmo.Hum mmH. be m vmh.Hus cod m.uc:uoz
anoHc z .m.c oeumeutum ccoHe z .m.o oeumeutum

mmeanc>

chHuocum mmcnuHuu¢

 

 

mcehcomucHeno .mnccuoz so

.mmoHuonum can monouHuud

mcHQMHumS 0H£mcumosco mo cOGMOHMHcmHm mo mumon. HcHuccummcHino.v pomn.

80
HypothesSG

Thaeisncrelafiomhipbetween Mexican-Americanmatemalclfld—
madngpacdcesandthefamflydemognflficm
306a mother'sage,
3061: mdher‘syemofschcoling,
mctha‘sgeneraticnremovedfrom Mexico,
306a mother'srdisianirefaence.
306e fathc'syeuscfschooling,
so“ father'semploymentm.
3069 dfld'sgendx,
mmbercfgidsinthefamily,
mmbadboysinthefamfly.

Testscfiatisticalsignificancewereappliedtothedependentdemo—
graphicvariablestotesttheireffecmmdfild-rearingp'acdces. Aswithtests
cfthesib-hypothesesforflypothesis S,theme-wayanalysisofvatiancepmce-
dumwastnedmtestthemtervalvafiablesme,mouier'sage,mcther'syears
cf schooling, father's years of schooling, number of male children in the family,
andmmberifemaledfildreninmefamfly).Thedfi-sqiaretes:wasappliedtc
the nominal. variables (i.e., abject child's gender, father's employment stats,
mother's generation removed from Mexico, and mother's religious greference).
Sincenaieduietestrewltswerestafimcanysgnificarmansm-hypotheses
cnypothes's6failedtcberejected(seeTable4.6).

Studstical Rants
Stafisficalresilts of the five hypothesestested in this study are summar-
ized in Table 4.7.

81

Table 4.7—Statistical Results of Hypotheses Tests.

 

 

Focus of Hypothesis Hypothesis Statistic Decision
Maternal nui‘mdes 1 z a 2.88* Retain
Maternal Practices 2 z = 6.43* Reject
Attitudes x Practices 3 x2 =- 3.98 Reject
Child Bdiavior x Practices 4 F =- 1.46 Retain
Mother's Age X Atfimdes 5a F = 1.76 Raain
Md'her's Age x practices 6a F a 1.62 Retain
Mothers Years of Schooling x Attitudes so P a 1.65 Retain
Mother'sYearsoESchooiingxpractices 6b 19:0 Retain
Mdzher's Generation x Attinides 5c x2 . 8.203 Retain
Mdzher's Geieration x Pmctices 6c x2 . 2.693 Retain
Mother's Reunions Preference x Attitudes 5d 1:2 . 3.4 Retain
Mother's Religious Prefaence x Practices 68 x2 . .71 Retain
Father's Years of Schooling X Attitude 5e F s 1.78 Retain
Father's Years of Schooling X Practices 6e 1? a .019 Retain
Father's Employment Status x Attitudes 5f x2 a 10.487 Reject
Father's Employment Stacie x practices 6f x2 - 3.63 Retain
Child's Gender x Auitudes 59 x2 =- 2.62 Retain
Child's Gender x Practices 69 x2 = .53 Retain
Numba‘ d Girls in Family X Attitudes 5h F = .041 Razain
Number cf Girls in Family X Practices 6h F = .772 Retain
Number d Boys in Family X Attitude Si F = .058 Razain
NumberchysinFamflnyractices 6i F=.32 Retain

 

the. Hypotheses rejected or retained at the significan'ce levelcf p < .05

82
SummaryData

Ancverallpictureofthefindingsfromthedataanalysesofpattemscf
dfld-reafingattinflesandg'acficesisshowanaNe4.8.Tietameisthevar-
iablematrix(Figure 3.]JfrcmtheDesign OverVariablessectionofChapterllI
withthe appropaate simmary statistics entered. Majorinterdependentvariahies
inthestudy were maternalchild—rearingattimdeand matemalchild—rearing
practice. In table 4.8 a chi-square table is [resentedin which the permissive,
authoritative, and authoritarian child-readng aua'mdes are correlated with the
pattermcfcifild-rean‘ngpracticestfthen mothersinthesample. lhthefirst
mw,theemptyceflindicatesthatcffl1e7motherswhoe:qresedpemfissve
diild-reafingaufitndes,ncne were cbsenredtcusepermissived'n'ld-readng
p'actica. Five cf three mothers had authoritative g‘actices, and the remaining
2werecbservedtohaveauthoritafian child-rearinggractices. Thesecondrcw
dthe table,autha:itative attitudes, showsthatofthe13 motherswho
emceedamhofitafiveatfitudes,8weremnsstentmmeircbservedp:acfioes,
wtfile4hadauthoritafian,and1hadpernfimiveclfild—readngg'acfices. The
thirdrow, authoritarian atfitudes,indicatesthatdthe 27 motherswith
amzhao'tariandiild-reafingaujtifia, 16werealsoobservedtobecons'stently
auundtadaninthdrgracfices,10wereatmhcrfl:afive,mdlwaspermissivem
herchild—reafingpracfices. Rowaxxicolumnpercentagesreflectgoporficnsof
the row and column total. For example, row percentage, 71.4, indicates that
71.4percentcfthe7mdzherswhcexpresedpermiafveau1tudeshad
authcfitative tractice. Under column percentages, 72.7 percent (i the 22
mothers who were observed to have authoritarian practices also expresed
authcdtarianchild—readngauzitudes.

83

 

 

 

 

 

o.ooH m.me m.me m.e “pouch
Ne mm mm N "caspcu
e.em o.em m.HN H.~ a peach
- ~.N~ m.me o.om a esopoo
m.mm o.~m ~.m u so:
ceetouvtoeuz<
N.NN m.m o.mH H.~ a pouch
N.mH m.¢m o.om a esopoo
m.om m.Ho N.“ a :01
mH e m H acaou
m>_pou—toeu=<
m.eH m.e o.oH o a pouch
H.m ~.H~ o a esopoo
m.m~ e.H~ o a gem
N N m 0 “£300
m>_mm_5tma
mme=u_uu<
Hmwm» coeeeuetocu3< m>_ueupto;u=< c>pmmFEtma
amorousta

 

 

.mmceuustc to“ mmezueuu< meetemmtepeeu so mctmuuea postman: couetce<ucsupxmz we :o_aoH=oeummotuttm.e epoch

WV

may, counsmzs , nxswssxw, moulds

In this concluding chapter, the purpose, methodology, and findings
are sunmarized; conclusions are draw from the results; and interpreta-
tions of the findings are discussed. A final section, Implications,

includes recaunendaticns for future research and practical application.

Statuary of the Study

Mich of hunen development is a consequence of the manner by which
parents rear their children. There is evidence of cultural and ethnic
differences on various dimensions of child-rearing. Hence, the rationale
for studying the child-rearing attitudes and practices of Mexican-
Anericans, the largest group of Hispanics who are the second largest and
rapidly growing ethnic minority population in the United States (Bureau
of the Camus, 1980). while the contributions of the paternal role to
button development are recognized, only the maternal role was examined
since the mother continues to be the primary agent in carrying out the
expressive role with children in nany families (Bigner, 1979).

The purpose of this study was threefold. The interface between
Mexican-Pnerican naternal child-rearing attitudes and practices was exam-
ined. The interface between these practices and child behavior was stud-

ied. Finally, madcan-Anerican maternal child-rearing attitudes and

84

85

practices were investigated relative to demographic variables in the
family environment.

The subjects were 47 Mexican-American mother-child dyads. The
children, who were enrolled in kindergarten had a mean age of 5 years, 4
months. The age range of the mothers was 22 to 47 years with both a mean
andmsdian ageof 30years, andambdeof 23years. Themajorityof the
mothers were married (81 percent). The 38 percent who were arployed out-
side their tunes, had occupations ranging from proprietor of a fast-food
business to unskilled service workers. Occupations of fathers ranged
from clergyman to unskilled laborer . The unaxployment rate for the 39
fathers has 30 percent. Years of schooling ranged from 0 to 14 years for
anthers and 0 to 16 years for fathers. The mean educational level was
tenth grade for anthers and eighth grade for fathers. The average mmber
of children per family has 3.4 with ages ranging from infancy to 26
years. The majority (66 percent) of the families were Catholic; others
were Protestant. The families lived in predominantly white (57 percent)
and ethnically mixed (28 percent) neighborhoods.

Five questions were considered in the study. The first addressed
the attitudes Meadcan-Anerican mothers express toward rearing their
children. The second question focused on child-rearing practices
Mexican-AneriCan mothers enploy when they interact with their children.
The third question concerned how the attitudes expressed cmpared with '
practices observed and the fourth concerned the relationship between
maternal child-rearing practices and child behavior. The fifth question
focused on the attitudes and practices of child-rearing relative to fam-

ily datographics such as mother's age, generations removed from mice,

86

religious preference; father's years of schooling and employment status;
the child's gender; and the number of male and female children in the
family.

The study took the form of a carbined ex post facto (field) and
attitudinal design in which the ex post facto portion was based upon
structured observations of 47 mother-child dyads in their homes. Matern-
al behavior was measured with the Maternal Teaching diservation Tech-
nique (MIUI‘), which was developed by Laosa (1980c) for use with two
sociocultural and language groups—Anglo and Chicano. During
administration of the MI'OI', child behavior was measured with an
accompanying instrument developed by the researcher , Child Behavior
During the mar (CBDMI'OT). Mother's attitudes were surveyed in
interviews following the structured observations using an instrument
designed for use with lower-class populations that was comprised of 25
statements adapted from the Parental Attitude Research Instrument (PARI)
by Schaefer and Bell (1958) and the Cornell Parental Behavior Inventory
(Cornell) by Weigert (1968).

Inferential statistical models were used to test six hypotheses
drawn from the research questions . Inferential statistical procedures
included the Z test of significance for a single sample, chi-square,
Student's t, discriminant analysis, multivariate and univariate anal-
yses, Pearson correlation, and analysis of variance. Decisions on hy-
potheses vere made at the alpha = .05 level of significance. Descrip-
tive data here calculated where applicable and were used to describe the

sample of Mendcane-American mothers and their children.

87

The Z test of significance for a single sample was used to test
the first two hypotheses since the data were interval level and the
sample size has greater than 30. Results revealed that the majority of
Mexican-American mothers expressed authoritarian child-rearing atti-
tudes, but that the majority of mothers were not observed to use author-
itarian child-rearing practices. The childrearing practices of Meirican-
American mothers in the sample were split between authoritative (49 per-
cent) and authoritarian (47 percent).

Tho statistical models of analysis were used to test the relation-
ship between attitudes and practices of child—rearing. The chi-square
test of independence was performed an the permissive, authoritative, and
authoritarian nominal-level variables for attitudes and an authoritative
and authoritarian variables of child-rearing practices. Because the two
mothers with permissive practices represented a proportion of the total
sample that was not statistically significant, the permissive child-
rearing practices category was eliminated from this and further anal-
yses. The chi-square test results were not significant, suggesting that
there was no relationship between Mendcan-American maternal child-
rearing attitudes and practices. Post hoc procedures were used to fur-
ther examine the data at its original higher interval level due to pos-
sible loss of information with the chi-square technique' A more robust
test, discriminant analysis, was applied to test the effects of the col-
lection of interval-level variables that measured characteristics on
which the sample of mothers were expected to differ. Discriminant analy-
sis results provided evidence that there was a significant relationship

between expressed child-rearing attitudes and observed child—rearing

88

practices, x2 (7 df) = 15.48, p = .03. Based on function coefficients,
the attitude variables of Acceleration of Autonomy, Lack of Control,
Casual Use of Time, and Equalitarianism were found to best discriminate
between authoritative and authoritarian practices. An accuracy check,
the multivariate analysis of variance (MANOVA) test of the attitude var-
iables, revealed significant differences for Acceleration of Autonomy, F
(1,43) =- 13.24, p = .001, and for Casual Use of Time, F (1,48) =3 4.57, p
= .038. The significant discriminant analysis and MANOVA test results
suggested information was lost by classifying the attitudes data into
the permissive, authoritative, and authoritarian categories of the
Banmrind theoretical model. Findings on the higher interval level
attitude variables provided evidence that there was a relationship
between child-rearing attitudes and the theoretically classified
practices of Deidcan-Anerican mothers; therefore, the null hypothesis
was rejected.

Miltivariate tests of significance were used to measure the
association between independent maternal practices variables and depen-
dent child behavior variables. Findings were not significant, suggesting
that child bdnavior variables, as a group, were not associated differ-
ently with authoritative and authoritarian maternal child-rearing prac-
tices as measured by the Maternal Teaching Observation Technique and the
Child Behavior During the mar. Test results of univariate analysis of
variance between the two patterns of child-rearing practices and each of
the child behavior variables were significantly different on the varia-
ble, Imitation of Mather at the .05 level. The hypothesis that there was
no relationship between child-rearing practices and child behavior

89

failed to be rejected, however, since there was lack of significance in
the other seven child behavior variables.

The statistical technique, analysis of variance (ANOVA), was used
to assess the effects of the interval-level derographic family variables
on attitudes and practices of maternal child-rearing. The results were
not significant, suggaeting that mother's age, mother's and father's
years of schooling, and number of male and female children in the family
did not have a significant effect on the child-rearing attitudes and
practin of the Mexican-American mothers in the sample.

Pearson's chi—square test of independence was applied to the nomi-
nal variables of mother's generation revved from Mexico and her
religious preference, father's exployment status, and the subject
child's gender. One test result, the effect of father's aployment
status on maternal child—rearing attitudes, was highly significant at
the .01 level, mggesting effects from paternal unemployment. In
families with exployed fathers, expressed maternal child-rearing
attitudes were 20 percent permissive, 40 percent authoritative, and 40
percent authoritarian. In families with unerployed fathers, mothers'
childrearing attitudes were 7 percent authoritative and 93 percent

authoritarian .

Omclusims
The following conclusions were deduced from the statistical
findingS:

90

l. Mexican-Mexican mothers expressed a broad range of child-
rearing attitudes-permissive, authoritative, and authoritarian
patterns.

2. Mthoritarian child-rearing attitudes were predominant along
this sample of Meadow-American mothers residing in an urban area of the
Midwest.

3. Mexican-American mothers were also observed to use child-
rearing practices ranging from permissive to authoritarian patterns.

4. Mexican-American maternal child-rearing practices were nearly
equally divided into authoritative and authoritarian patterns.

5. Sale variables of maternal child-rearing attitudes were posi-
tively related to observed maternal behavior.

6. The authoritative and authoritarian patterns of child-rearing
practices were best discriminated by scores on the attitude variable,
Acceleration of Autonomy. A

7. Another attitude variable, Casual Use of Time, was also used to
discriminate between authoritative and authoritarian maternal child-
rearing practices.

8. Only one child behavior variable, Imitation of Mother, was
significant as a discriminator between the authoritative and author-
itarian patterns of maternal child-rearing practices .

9. Based upon Pearson product-moment correlations for pairs of
mother-child behaviors during the teaching task, there were significant
linear relationships between the following pairs of variables:

a. The more directives a mother issued, the less a child
initiated or became absorbed in the task.

91

b. The more a mother praised her child, the more the child
responded to the mother.

c. The more a mother praised her child, the more the child
questioned the mother.

d. The more a mother praised her child, the more the child
initiated or became absorbed in the task.

e. The more visual mes a mother gave her child, the more the
child respmded' to the mother.

f. As the mother's positive physical control increased, the
child's physical responses decreased.

9. As the mother's positive physical control increased, the
child's failure to respond decreased.

h. As the mother's positive physical control increased, the
child's positive verbal responses increased.

1. As the mother's positive physical control increased, the
child's frequency of initiating or being absorbed in the task
increased. I

j. As the mother's negative physical cmtrol increased, the
child's negative verbal responses increased.

10. The significant influence of father's exployment status on
maternal child-rearing attitudes was attributed to the paternal
unetployment rate of 36 percent .

In summary, it is concluded that the majority of taxman-American
mothers in the sample expressed authoritarian child-rearing attitudes
but were divided in their observed child-rearing practices between auth-

oritative and authoritarian patterns. While permissive, authoritative,

92

and authoritarian patterns of child-rearing attitudoe and practices were
not significantly related, some variables comprising the three patterns
of attitudes were significantly related to the predominant authoritative
and authoritarian patterns of practica . The variables, Acceleration of
Mtonamy and Casual Use of Time showed discrimination between
authoritative and authoritarian patterns of maternal child-rearing
practices. Furthermore, significant linear relationships were found
between variables comprising the patterns of practices and variables
which measure child behavior, indicating a relationship between
mother-child behavior which was not significant between patterns of
child-rearing practices and child behavior . Finally, the
Mexican-American family ecosystem had only one detographic variable,
father ' s elployment status , which significantly influenced maternal
child-rearing attitudes or practices . It is concluded that patterns of
maternal child-rearing attitudes were influenced by paternal

unemployment .

Discassicn

This section is organized around the five general research ques-
tions stated in Chapter I. Findings of the study are integrated with the
theoretical framework described in the first chapter. Cmclusions from
the findings are also further developed and interpreted.

Research gestion 1
that patterns of child-rearing attitudes do Moan-mica
mothers express about rearing their children?

93

The results of the data analyses provided evidence that Mexican-
American mothers expressed permissive, authoritative, and authoritarian
child-rearing attitudes . These results were inconsistent with the find-
ings of IeVine and Bartz (1979) that Chicano parents have permissive
attitudes and practices of child-rearing. The following discussion of
the conclusions based on findings for each of the seven inclusive
attitude variables used to classify maternal attitudes of child-rearing
explains the inconsistency between the findings of IeVine and Bartz and
those of this study. The statements comprising each of the seven
categorical variables are listed following the appropriate eibheading.

Acceleration of Autonomy

l. The sooner a child learns to walk the better off he will be.

2. Most kids should be toilet trained by 15 months of age.

3. A child should be take) away from the bottle or breast as soon
as possible. -

4. Themaingoalofaparent istoseethatthekids stayoutof
trouble. '

5. Kids should be nicer to their mothers since their mothers
suffer so much for them.

IeVine and Bartz interpreted their findings a) the variable,
Acceleration of Autonomy, as "Chicano parents . . . pressing for an
"early assumption of responsibility by the child for his or her bodily
functions and personal feelings” (p. 173). The description of Accel-
eration of Autonomy to which IeVine and Bartz assigned the first five
statmnts m the instrument (see Appendix C) primarily assessed

knowledge of child development rather than social expectations of

94
autonomy or assumption of responsibility. For example, a response to a
statement that a child should be toilet trained by the age of 15 months
requires knowledge of biological maturation. If a mother expects that a
child should be toilet trained by the age of 15 months, she may be
lacking in child development information rather than trying to
accelerate the child's autonomy. Interestingly, mean scores on the
variable are equivalent in both studies , but interpretations of the
variable in this study differ from those in the IeVine and Bartz study
(1979) because the statements included in the variable do not indicate

children assuming responsibility.

Casual Use of Time

6. The sooner a child learns that a wasted minute is lost forever,
the better off he will be.

7. There is no excuse for a child sitting around doing nothing,
becausetherearesomanythingsheneedstolearnabontlife.

For the variable, Casual Use of Time, IeVine 8 Bartz (1979) found
that Chicanos deetphasize wise time use. Disagreement with the state-
ment, ”the sooner a child learns that a wasted minute is lost forever,
the better off he will be”, was assumed to imply a casual Hispanic
orientation toward time. To this researcher, towever, the implication
was not all that clear. Agreenent with the statement may be an implica-
tion that one is pressuring a child to think like an adult (Elkind,
1981). The findings in this study suggest that Chicanas were divided
between agreement and disagreement with the variable, Casual Use of

95

Time. Scores above 2.5 indicate disagreerent, implying a casual time
orientation about their children's time use. The mother's in this study
had mean scores of 2.415, which reflect the ambivalence many expressed
about responding to the statements . Sore contented that the qnmtions
were inappropriate for five-year-old children.
finalitarianism

8. A child has the right to his own ideas and should be encour-
aged to tell others about them.

9. When kids think family rules are wrong, they should not tell
parents about it. '

10. When a child is in trouble, he should not talk about it with
his folks.

11. Kids should be able to talk with parents if they think their
own ideas are better. i

12. Parents must earn the respect of their kids by being fair with
them at all times.

IeVine and Bartz suggested that disagreement with the Equali-
tarianism variable was consistent with descriptions of "the relatively
authoritarian Hispanic family” (p. 173). The application of theory to
analyze this variable is in agreement with the IeVine and Bartz inter-
pretation of this variable. According to Baumrind's (1971, 1978) theor-
etical definition of an authoritarian parent, disagreenent with these
statements suggests authoritarian parental beliefs or attitudes .

Value Strictness

13. Kids that have firm rules to obey, grow up to be the best

adults.

96

14. A clnild will thank you later on for strict training.

This label accurately describes one of the two statements
cotprising the variable. In the second statement, however, "firm" is
evidently equated with strict. The Lack of Control variable is a more
accurate measure of strictness than this variable.

Devalue Permissiveness

15. Parents should give in to the kids some, rather than expecting
the kids to always obey the parents .

16. Nbst kids need less discipline and punishment.

17. Kids are not happy under strict training.

Devalue Permissiveness is also an inaccurate descriptor because
the words, ”discipline, punishment, and strict training" are used in
statements measuring permissiveness . Disagreeent with ench statelents
can hardly be equated with a permissive atmnsphere as Levine and Bartz
did. Two of the mothers interviewed in this study cemented on the diff-
erences between discipline and punishment. Several other mothers hai—
tated, scowled, or asked for clarification of Item 16.

Lack of m

18. Ifmychildhas anykindof problem, s/hecancounton meto
help her/him out.

19. I say nice things about my child.

20. I make him (or her) feel I am there if he (or she) needs me.

21. I teach him (or her) things he (or she) wants to learn.

The range of responses, never to very often, to the four state-
ments cotprising the Lack of Support variable clearly indicates parental
support or a lack of it. For example, any of the five responses to Item

97

20 is indicative of a degree of support. There was a contradiction, how—
ever, in the Levine and Bartz finding that there was a lack of support
from fathers-but not mothers— and their conclusion that Hispanic chil-
dren had minimal support from both of their parents.
lack of Control

22. If my child doesn't do what is expected of him/her, I am very
strict about it.

23. I keep pushing him (or her) to do his (or her) best in
whatever he does.

24. I expect him (or her) to keep his (or her) things in good
order.

25. Ikeepafterhim (orher) todowellinschool.

In each of the four statements comprising this variable, strict-
ness is either stated or implied, i.e., "I an very strict about it", "I
keep pnshing", "I expect", "I keep after him or her”. Responses ranging
from never to very often to such statements are eibject to interpreta-
tion and are, therefore, provisional measures of lack of control. IeVine
and Bartz (1979) found that, ”anicano fathers devalue control” (p.
173) and ”Chicano parents offer less support and control than do Black
parents” (p. 173). Such findings contradict Levine and Bartz' conclusion

that Chicanos value strictness in their child-rearing.

Researchmestionz
mat patterns of child-rearing practices do Mean-mica
mothersdelonstratewdniletheyinteractwiththeirdnildren?

98

Authoritative and authoritarian patterns were found to be the ob-
served child-rearing practice of Mexican-American mothers in this
study, although permissive practices were observed in 2 (4 percent) of
the mothers. The process of maternal interactions with children was
transformed into a range of permissive, authoritative, and authoritarian
child-rearing practices during a structured teaching task. Tlnese find-
ings lend support to the Parra and Henderson (1982) findings which sug-
gested that Mexican-Americans had an extensive range of child managenent
procedures, which included explanation, positive reinforcement, and
punishment. 1

msearch on 3

Are the child-rearing practices of Moan-American mothers

related to their child-rearing attitudes?

Systematic application of Baumrind's (1971) theoretical model to
measures of attitudes and practices (i.e., behavior) resulted in classi-
fying continuous scores into permissive, authoritative, and authori-
tarian global measures. Results of the chi-square test of independence
between the categories indicated that the relationship between at-
titudes and practices was independent or not statistically significant.
The theoretical framework for this study prompted further inquiry into
the lack of association between maternal practices and attitudes. The
transactive cycle of attitudes and practices of child-rearing and child
behavior presented in Chapter I graphically pictures a cyclical process
in which child-rearing attitudes, practices, and child behavior are
related to one another and to external environments surrounding them.
Thus, the discriminant analysis statistical procedure was used to

99
further interpret the data. Attitude measures at their original
multivariate, continuous scales were used to discriminate between the
categorical authoritative and authoritarian patterns of child-rearing
practices. The canonical discriminant function for this analysis
revealed a significant correlation between observed maternal practices
and particular attitude variables of the interview survey. The
discriminant function, therefore, indicates that discrimination between
authoritative and authoritarian patterns of child-rearing practices is
possible from scores on the attitude inverview variables , Acceleration
of Autonomy, Lack of Control, Casual Use of Time, and Fqualitarianism.
These four particular variables discriminated best between authoritative
and authoritarian child-rearing practices. Based on this finding, a
prediction formula of patterns of child-rearing practices could be
developed for testing in future research.
‘ Research on 4

Is there a relationship between Man-mica maternal

child-rearing practices and child behavior daring observed

interactions bewesn mother and. child?

Measurement of the combined clnild behavior variables failed to
reveal a significant relationship between observed child behavior and
the categorized patterns of maternal child-rearing practices. than the
particular child behavior variables were analyzed separately, there was
a significant relationship between Imitates Mother and child-rearing
practices. This finding is inconsistent with evidence for differences in
child behavior related to parental disciplinary patterns provided by

Baumrind's extensive research (1971, 1973, 1975). Since there was a

100

significant correlation between the instruments measuring maternal and
child behaviors , it may be that reducing the interval level data of
maternal behavior to nominal level patterns of behaviors caused a loss
of information which resulted in a lack of statistical 'significance. an
the other hand, if measures'of behavior sequences between mothers and
children were available for analysis, the relationship could potentially
be more adequately explained.

The findings were also inconsistent with the ecological view that
the flow of energy and information between the mother and child serve as
an organizing function relating then to one another. From an ecological
point of view, perhaps the lack of a significant relationship between
maternal practices and child behavior was that the maternal and child
observation instruments did not allow for measuring tramactions with
other systems during the structured observation which may have
influenced the reciprocal flow between mother and child. For instance,
ethnographic notes, which were taken by the researcher following each
hone visit, provided evidence of other family menbers permeating the
established boundary around the mother-child dyad engaged in the teach-
ing task. Although lasting only briefly, such boundary permeations may
have been manifested in transactions with other system in the home
environment which were not measured by the observation instruments. It
wasobvious to theresearcherthatwhen theboundarywas permeated, the
system was influenced in such a manner that mother and child behaviors
were transformed, if only temporarily. Examples of boundary permeations
were two father's atterpts to verbally or visually coach either the
mother or child in how to perform the task; older siblings in two

101

famnilies inadvertently connunicated with the mother about their own
needs; in one hone, younger siblings conpeted for the mother's attention
by climbing on her lap, grabbing a part for the Tinkertoy model, crying,
whining, or otherwise demonstrating a desire to be included in the task;
friends or relatives visited unexpectedly in two tunes; and telephone
calls, such as the two which the mother later told the researcher con-
cernedan inquiry aboutorareportofthe grandmotherwhohadbeen in
surgery that day. Undoubtedly, the human energy required for such trans-
actions was transformed into the attitudes and practices of child-
rearing and may have also affected child behavior. Such normal everyday
famnily situations suggest that the mother's and child's behaviors during
the observed structured situation are representative of mother-child
interaction (i .e. , teaching-learning behavior) during similar tasks in
the home environment. This is discussed further in the following section
on implications for further research.
Research Mon 5

Are there differences in maternal child-rearing attitudes

and practims related to daographic variables in

tarian-Marian family ecosystu, such as mother's age,

generation from Medan, and religions preference, mother's

and father's years of schooling, father's aploymnt status,

thednild'sgender,orthemmflnercfmaloandfenale

children in the fanilY?

The influence of the father's employment status on maternal child-
rearing attitudes was found to be the only statistically significant

denographic variable affecting the outcome of this study relative to the

102

family ecosystem. In families with employed fathers, expressed maternal
child-rearing attitudes were 40 percent authoritarian, 40 percent au—
thoritatiVe, and 20 percent permnissive. In families with unemployed
fathers , mothers ' attitudes were 7 percent authoritative and 93 percent
authoritarian. The significant influence of father's employment status
was attributed to the paternal unenployment rate of 36 percent. The
systenatic relationship was reflected by the dramatic differences in
the proportions of patterns of maternal chi ld-rearing attitudes .

The results of the chi-square test of significance showed that a
systematic relationship existed between maternal child-rearing attitudes
and father's employment status. Craner's V, a measure of strength of
relationship, indicated that a .52 degree of association existed between
attitudes and employment status . These findings lend support to the
family ecological perspective of the necessity for viewing the
interrelatedness and interdependence of family memnbers and those
environments which impinge upon the family (Andrews, Bubolz, & Paolucci ,
1980) .

Imlications

The implications of the results of this study for further
research offer several contributions to the field. The methodology of
this study differed frcmn methods used in previous studies in a number
of critical ways. Contributions in the area of Medan-American child-
rearing include the investigation of maternal attitudes and practices
and their relationship to one another as well as the relationship be-
tween child behavior and maternal child-rearing practices. The study of
the interface of attitudes and practices with demographic variables in

103

the family ecosystem are also a contribution to the body of research on
child-rearing in Micah-American families. Further refinement and
replication of the design of the study will help researchers to improve
upon future research in this area.

While this study makes unique contributions to the sparse research
on child-rearing in taxman-American families, the shortcomings in the
study might be avoided by future researchers. The conmpt of boundary
helped the researcher gain insight into the difficulties in making
structured observations in a natural setting. Procedures for erploying
the Maternal Teaching (Inservation Technique (MI'OI') in this study re-
quired establishing a boundary separating the mother-child relationship
from other famnily members or outside individuals during a structured
teaching task for observation. The necessity of such a boundary for ob-
serving maternal teaching strategies, however, may interfere with
observing child-rearing practices as they oconr naturally in the home
setting. For enample, the average number of children in Hispanic famil-
ies was 2.3 compared to 1.9 in nonHispanic families in 1980 (Bureau of
the Census, 1983). In the sample for this study, the average number of
children per family was 3.4. In many families with several children,
siblings and other family members may be as involved in child-rearing as
the mother. Consequently, a somewhat artifical child-rearing situation
is created by structuring a boundary which separates the mother-child
relationship from other relationships in the family. According to the
researcher's ethnographic notes of each home visit, in 70 percent of the
families, the established mother-child boundary was permeated at least

once by a younger sibling, an older sibling, the father, a relative, or

 

 

104

a friend of either the child or mother. The structured observation did
not allow for measuring the behaviors of individuals outside the boun-
dary nor their effect on mother or child. Hence, some information rela-
tive to child-rearing may have been overlooked. If an objective of the
wholistic ecological approach to study of the family is to have a real-
istic ecological view of child-rearing practices, then perhaps the
boundary should not be limited to mother and child in future studies.

an the other hand, the structured teaching task allowed direct
observation of typical mother-child interaction within a very short
time span with minimal intrusion by the researcher. Baumrind and Black
(1967) report that generally a minimum of two unstructured observation
visits in the home are needed for mother-child behaviors to stabilize
and approach typical patterns of interaction. In this study, the struc-
tured teaching task was observed to quickly involve mothers and children
so that their natural modes of interacting were restored almost immed-
iately. The subjects quickly relaxed and apparently forgot or ignored
the researcher's presence. Laosa (1980c, 1981)) contents on evidence
that a mother's behavior during the observed, structured situation is
representative of her teaching behavior during daily situations in
which similar tasks might be taught her child. This was observed to also
apply to the child's behavior during the structured observation in this
study.

Another potential shortcoming of the study, which was mentioned
previously in Chapter III, was having only one observer for both mother
and child. Qnestions of concurrent validity of measurements enploying

the two observation instruments used in the study cannot be answered

105

since the researcher was the sole observer. Two observers, one for the
mother and the other for the child, would correct this shortcoming. On
the other hand, it would be more difficult for two observers to be un-
obtrusive in the home setting, and observer. interactions with the mother
during the interviews would also be affected.

Suggestions for future research are that the study be replicated
to meet the need for a series of studies on Mendcan-American child-
rearing with samples of varied socioeconomic status in different regions
of the United States . Further research into congruence among Mexican-
American parents on child—rearing attitudes and practices is also
needed. Inclusion of fathers in a similar study would extend the know
ledge base beyond maternal and child behaviors into a dual-parent family
measure of attitudes and practices of Mean—American child-rearing. A
study which would focus on direct observations of Mean-American men
interacting with their children would be an important and unique contri-
bution to the increasing number of research studies on fathering be-
haviors in recent years . Judging from the enthusiasm of fathers who were
present during this study, it would not be difficult to sewre a sample
of Medan-American men. It is suggested that interviewer or observer
gender be matched in such a study so that subjects might feel as
comfortable as possible. It is also recommended on the basis of this
study and other ethnic minority ‘ family research that observer ethnicity
be matched (Cromwell, Vaughan, & Mindel, 1975; Iaosa, 1977; Bernal,
North, Rosen, Delfini, & Schultz, 1979).

106

The practical applications of this study are in the areas of par-
ent education, day care, education at all levels, and other human
services. The findings suggest that parent education programs that in-
clude components related to the ethnicity of Micah-American parents
may be appropriate. The heterogeneity of Nexican-Americans, however,
should be considered in the planning of such programs. Also germane, are
the findings of this and other studies that Mexican—American child-
rearing attitudes and practices may vary from permissive, to authorita-
tive, to authoritarian patterns. Special conponents in such programs
should be designed for families who are Lexican-American or are related
to Chicanos through family bonds. A nonformal style of instruction is
recomended for two reasons. First, the literature shows that participa—
tion of Mexican-American parents in general parent education classes is
minimal. Second, research (Rose'nblatt & Wiggins, 1967; mrtinez Byrd,
1978) has shown that parent education programs are least successful in
retaining Mexican-Americans as group menbers, reflected by their attri-
tion rate relative to that of other group members. For these reasons,
nonformal Mexican-American parent education programs should foons on
informal learning experiences rather than on a more formal, classroom
structured progrene. The fundenental goal of such programs would be to
encourage families to think annd talk about their parenting relation-
ships. Family members of all ages should, therefore, be encouraged to
participate. Child care should be provided to encourage participation.

Practical use of the information gained from this study about how

sone Mexican-Americans rear their children is also viewed as germane to

107
individuals working with Mexican-Americans in day care as well as educa-
tional, and health are, and social services settings. Knowledge of
maternal child-rearing attitudes and practices and child behavior in
Mexican-American families, including the languages spoken in the bone,
can be applied by professionals serving Mexican-American children and
their family members to enhance their interactions with and improve upon

intervention in these families.

APPENDIX A

CONSENT F0 R MS

108

Appendix A

CONSENT FORM IN ENGLISH

Child ’5 name:

 

As the legal parent/guardian of the above named child, I
hereby give my permission for her/his participation in a
study conducted by Estella A. Martinez, a Ph.D. candidate,
under the supervision of Dr. Eileen Earhart and Lillian
Phenice of the Department of Family and Child Ecology,
Michigan State University.

b’other/Gnardian ’3 Name:

 

In addition, I have freely consented to take part in this
study.

I urnderstand that I am free to discontinue my participation
and my child ’s participation in the study at any time.

I understand that our participation wil 1 require about one
and one-half hours and that I will be teaching my child to
build a toy model, and I will be interviewed.

I understand that my participation in the study does not
guarantee any beneficial results to me or to my child.

I understand that at my request, I can receive additional
explanation of the study after my participation is
completed.

I understand that the results of the study will be treated
in strict confidence and that I will remain anonymous.
Within these restrictions, results of the study will be made
available to me at my request.

Signed:
Date:

 

 

109

CONSENT FORM IN SPANISH
Fonma de Consentimiento

Nonbre de nii'no/nina:

 

Cono madre o guardian del nifio/nif’na ya mnencionado, doy
permiso que participe en el estudio conducido por Estella A.
Martinez, candidata del Ph.D., bajo la supervision de las
Profesoras Eileen Earhart y Lillian Phenice del Department
of Famnily and Child Ecology, Michigan State Unniversity.

Nombre de madre/guardian:

 

Ademés, he consentido libremente a participar en el estudio.

Entiendo que soy libre de discontinuar mi participacion y la
de mi niho/nifia en este estudio en cualquier tierpo.

Tengo entendido que nuestra participacion requiere una hora
y media y de que voy a ensenarle a1 nifio/nifia a hacer un
juguete y despues participar en una entrevista.

Entiendo que esta participacion no garantiza resultados
beneficiosos tanto para 0 mi 0 mi nifio/nifia.

Ehntiendo que despues de haber participado en el estudio
puedo recibir mas‘explicacion sobre el estudio.

Ehtiendo que los resultados del estudio sera’n estrictamente
confidenciales y que permanecere’ anénima. Los resultados
del estudio estara’n a mi disposicion si yo los pido.

Firma:

 

Fecha:

 

APPENDIX B

BACKGROUND INFORMATION

110

Appendix B

BACKGROUND INFORMATION

 

Age: Mother Child’s D.O.B.* Male __ Female __
Marital Status: single

married

divorced

widowed

separated
Oconpaticn Employed _ Education yrs.
Spouse ’3 occupation Employed _ Education yrs.

 

(what does he do? or job title)

Area of residence: predominantly Moan-American _; white _;
Black _; Mined .

thher’s P.O.B.“ . Spouse ’3 P.O.B.

 

Maternal grandmother’s P.O.B. Grandfather’s P.O.B.

 

Maternal great-grandmother’s P.O.B.
Paternal great-grandfather’s P.O.B.

 

 

Children: Age Boy Girl

Religion: Catholic Attend church regularly?

Protestant Yes
__ Other No

Do you have a Tinkertoy set in your home?

*D.O.B. - date of birth MrP.O.B. - place of birth

APPENDIX C

CHILD-REARING ATTITUDES INTERVIEW

lll

owumnmwa anmsowum
e

mmuonmea >a©d0uum
v

moumnnfio waocowum
H

mouoomwo xaocOHHm
A

mononmwo xaocouum
H

omumnmwo aeoccuum
A

ooumnnfia aumsouum
a

.02 .Q.H

mononnen
m

mononnfia
m

mononmen
m

oouoonea
m

oouoomfin
m

anemone:
N

ooumnnao
m

owned

omum<

mowed

mowed

mmum<

mmuo<

mauve

mowed

mauve

mowed

mmuo<

mowed

oouo<

mount

adosouum
A

aaocOHum
H

auocOHum
e

>emcouum
v

headcoum
v

xumcouum
v

aumsouum
v

.ouaa scone canoe Cu noon: on
nmcesu >an On one when» monsoon
ocqsuo: omeoo ossoun mcwuuem
Ogden s new museum c: we ounce

.oh “so: on mac uneven 0:»
.uo>ou0u once we ounces oounoz n
bass unused edema n smooch one

sees ud on: Hosnou

.Eonu uOu nose Om ummmsn
nwonuoa Henna mosen nuosuos
Hanna 0» none: on oasoan no“:

.oaboouu

we use anus hoax on» unsu mom
Cu we assume o no doom same was

.oehemnoo
on soon an unseen no season one

Eouu hose some» on ousosn oaeso c

.mon mo nousoe me an

cocoon» ooeeou on choose does moo:

.on use: on «mo wouuon on»

xdnz c» nausea cease n unseen one

>Eoscus< no sofiunuouoooc

3mH>mmB2H mQDBHafifl 92Hm<mmIQAH=U

U xeocoood

.HH

112

ooumonea >Hocouum
H

oouoomwo >Hocouum
H

ooumonfia >Hocouum
H

ooumoowa >Hmcouum
H

oouoomHo >Hosouum
v

oouoana >Hmcouum
v

oouoonHa >Hm:0uum
H

ooumomwo
N

oouooneo
m

ooumooHo
N

oouoomwo
.N .

ooumoowo
m

ooumonHa
m

ooumooHo
N

oouoc

oouoc

ooume

oouo¢

ooumc

ooumd

ooumc

oouot

oouod

oouo<

ooumt

ooumd

ooum<

ooumd

>Hmcouum
‘ .

aHocoyum
v

hHocouum
v

>Hmsouum
e

>Hmsouum
H

>Hosowum
H

>Hosouum
v

.ocucHouu unauun new
so wounH 50> scone HHes oHHso <

.ouHsoo anon on» on on on scum
aono ob moHsu shew o>nn use» mon

noocuoHuum ocHo>

.noEHu HHo on Eon»
toe: seem meson so ones names so
uooonou on» sumo uosE nucoune

.uouuoh ouo noooe
:30 uHosu xsenu eon» we mucouoo
:uH3 xHou 0» loo on oHsoso non

.oxeoo as: one:
on ozone seem so: oseoem o:
oeneomo on we oeeeo o doe:

.uH nacho oucouoo HHou
ac: oHsono >onu .mcous own
moses season geese dose can:

.Eozu nacho nuosuo HHou 0»
oooousooco on oHsoso oso noooH

:30 oH: Cu unoHu on» on: oHeso d

.vH

.MH

.>H

.NH

.HH

.OH

.m

EanoHunuHHooom ..HHH

3oH>uoucH

 

113

couuo >uo>
H

couwo >uo>
H

couwc >uo>
H

coumo >uo>
H

oouonowo >Hmcouum

v

ooumoowo >Hocouum

v

oouoooHo >Hoccuum

v

coooo
N .
condo

cooeo

coueo

moeeuoeom
m

noEHuoEom
m

moEHuoECm
m

ooEfiuoEom
m

ooumooHa
m

ooumonfio
m

oouoooea
n

uo>oz onost

uo>oz aoOEHe

uo>oz uooEH<

uo>oz uoOEH<

oouod

oouo<

oouo<

H0>mz

H0>Oz

H0>Oz

HQ>GZ

oouo< aHo:0uum
H

oouoe >Hocouum
H

oouoe >Hocouum
H

.cuooH Ou ouso3
.ozo do. on eaten» am; soooo H

.oe nooo: .osn woe o: uH ouozu
En H Hoou «no: so. EH: oxnfi H
.eeeno as ozone modes» one: and H

.uso no: oHo: Ou oE
so assoc coo one .EoHQOue
we use: see no: oeeeo as we

uuoaosm mo xooq
.ochHowu uonun
woos: zoom: so: ouo neex

.ucoeansso oco

ocHHQHooHo onoH oooc hoax uooz
.mucouom one wono o>o3Ho ca non
on» osHuoooxo coma Hogans .oEom
ocex one o» cH o>Hm oHsono mucouoo

mmmflfl>dmmfishmm 05d0>00

.HN

.cN

.mH

.mH

.H>

.hH

.oH

.mH

.>

30M>HQUCH

114

.muomnnsm nqu 6mm: mEHOH :oHuomHHoo mama use :0 Hammad yo: 6H6 mxcmu HmoHHmssz

.czonm :muwo mH HoH>mzmn umzu mumoHocH c.~ onon mmuooa

:muuo >Hw>
H

:muwc >Hm>
H

:wuwo >Hm>
H

cmuwo Hum>
H

.wuoz

.H .N .m .o .m cmxcmu mmHmom Mom

.ucmsmmummmHfi Hmumcmm oumochH m.~ m>ona umuoom .v .m .N .H cmxcmu monom Hon

.ucmsmmumm Hnumcmo

cmumo

:muHo

cmuuo

:muHo

mmEHumEom
m

mmEHumEom

m

moEHuweom
m

mmEHquom
m

Hm>mz umoEH<
v

Hm>mz umOEH<
e

Hm>mz umoEHd
v

Hm>mz amoeH<
v

H0>mz

H0>02

H®>02

H0>wz

mumoHccH m.~ o>onm umuoom .H .N .m .v omxcmu mmHmom mom

.Hoonom :H HH03 on
0» Ho: Ho EH: Hmuum Qmmx H .mm

.Hmcuo @000 :H mmcHnu Ho: Ho mH:
mmmx ou Ho: Ho EH: uommxm H .vm

.mmoc man no m:
Hm>0um53 :H yawn Hm: Ho mH: 0c
Cu Ho: Ho EH: mcHnmam Qmmx H .mm
.HH usonm uoHHum
>Hm> an H .umn\EHz Ho wouommxm
mH gas: on u.:mmoc cHHso He HH .HH
HoHucou no xomq .HH>

30H>Hmqu

llS

wucmsmuum=m
ovumaommmc cm

mucmfimuumsw
oouwzommmc cm

mucmemuumsm
ocumsommc cm

mucwEouHosu
ocumsommc cm

mucoamuumsm
ocumsommmv cm

mucoemunSH
ovumsocmmc :m

mucmswuumsu
ocuwnommfi am

I‘ll

ooumsom
>0umm

ocumsom
>oumm

ocumsom
>0umm

ooumsom
>0umm

ocumsom
Houmm

ocumzom
Heumo

ocuosom
acumm

.02 .Q.H

06
oz

06
Oz

96
Oz

00
Oz

06
02

m6

06
Oz

.ocumsom

ccuosom

oouosom

oouozom

OQHQDOM

ocumsom

ocumzom

mo

00

mo

00

00

00

concave av
mucwamuuosm

ooumrou ov_

musoamuuosm

ocumsoo cc
mucmsmuuosm

ooumsoo mo
mucoamuumsm

ovuvson mu
oucmEmuHmsm

ovuwaou mo
macmsmuumam

ocumsow on
mucmemuumsm

.wvH> 0H munom Hmc

ncmuam muHmoom: wso mmmoo mmucmu

Ha: oswuoa new: Hmong :Hm moan

OBH: :5 0:0 06 mmsomm an 02

.OHQEmHm

mung ovHuqu mama HocmuumuHmE.
ocuHoHouoauov ouacHE a: 0:0 oucon
Hmvcoumu OSH: Ho mung uoflms mumm

oaemHu Hmc Hmnmmo ow:

.uoHHm Hon Oucmu :muuam

moufims man 030 a» mmucms mam coo

mocwsn awe umm :mnmo monH: no;

.AmmcmuHsoHuHcv mocmucm :0 :mums mm
o: momH: moH mam Hm> mm wucmm
:3 we HmaHocHHQ o>Humnno Hm

.mHnHmoa

Oucoum was OH och a: a 050mm Hm
o mHHouon mH mHmmeuHsc mama

.mmmme mH moH mung ovumsomm
H0 new: a movucmuuco Hmm
:mnmo mocH: uoH on QHHOHME >mq

.HMHHo. Hm damn momma

mama Cancun

mas Hausa a mNmHQEm\ocH: c: Hm

MfifiOflOHfifl 06.:OfiUflHmHOUfl

mocHz moq HMHHU H< macauHuoc no; munom mumH>muucm

.HH

ll6

mucmsmuuosm
ocumzommmc cm

mucmsmuHmSH
ocumaommmc cm

wvcmEQUHmfim
OCHmDOmeU Gm

mucmsmuumzm
ocumsommmo :m

mummEmuHmsw
ovumzommmn :m

mucmemuumsm
ocumsommmc :m

mucmsmuumsm
ooumsommmc :m

ovumaom «c
Heumw oz
ocuosom mo
>Oumm oz
ocumsom mo
>0umm oz
ocumsom we
acumm oz
ocumsommo
Houmm oz
ovumsom ma
Heumm oz
ocumzom we
>oumm oz

ovumson

ocumson

ocumsom

ocumnom

ocumsom

ocuwsom

ocuosom

m0

00

0D

we

m0

00

ocuoson mo
mucosauuwah

Cenozon 06
ouzmamuuosb

ocumaon ow
mucwamuumsm

ocumsoa m6
mucmsmuuoah

ocumsom mu
mucmemuumsm

ooumzom mo
oucmfimuumsm

ocumsom m6
mesmemuumam

.muoHuumo ncHHmHomHv as: Hon
«mammmc mumomcmumm «H oan :m

.mouHacm mmuonms moH umm m
caunvaH umomcmno mung mmEHHH
.uuHmou :mcmHu mam monH: won

WOUUfiHUmm WGHOHM>

.moHHm :00 moumsn mumEmHm
oocmHu monH: moH mu cummmmu
Hm omumcmm :mnmo mmuoma mo;

.mouonofi Goa mmHmoum mmmoH mam
msv :mmcmHm Hm mmuomm mzmcoo
umHnmn Haven :mnmc momH: won

.mmucmHHmm
mam :oo oHHHusoch mama o:
mmcmuHsonHc mamHu omH: c: occmso

.mmucmm mam m Oumm
oHHHUmc cmnmc o: mmuooHHOOCH
cow nHHHEmw mH mo mMHmmH mmH
03w :nmanm.m0wH: moH occmsu

.moHuo a mmemHWucoo m oomucon
Hmm mama H mmwcH mmHQOHQ
mam m ogomumc Hm mcoHu OCH: :3

OEmHHnuHHmsmH

.VH

.MH

.>H

.NH

.HH

.oH

.HHH

mumH>0Hucm

ll7

oanmmm as:
OUHJmmm >9:

ocHsmmm >52

ocHammm >52

mucmEmuHmSH
onumsommmc cm

mucmsmuumsu
ovumnommmc cm

mucmsmuumzw
ocumsommmc cm

ocHammm

ooHdmmm

ocHnmmm

ooHsmmm

ocumsommc
Houmm oz

ocumsom mu
>oumm oz

ocumsom mo
. >oumm oz

m000>

mmcsmH<

mm00>

mmcstm

m00m>

mm::UH<

mmom>
mmcaoH< Hmmo

ocumzom mo

ocumaom mo

ovumsom ma

floflflfl
Homo

moss:
Hmmu

moss:
Hmmo

MUCfiZ

mocsz
mocsz

mocsz

mocsz

ocumsom m6
mucmemuumsm

ocumsom m6
mucmEmuumsm

ocumaom mo
mucmemuumsm

.umocmumm mHmHsa msv
ammoo .mHHm o. Hm m ommmcm mg .Hm

.muHmmom: mE Hm HHHm >oumm
msv mucmHm AmHHm. Hm msv omm: .om

.Amv 00H: HE mo
moumom mmcmzn ammoo och o» .mH

.mHHmc=>m

mama omHE:00.umu:oo momma

Hm o mHHm .mEmHnon mo mmmHo
HmHavazo mcmHu HovmnH: HE Hm .mH

oaomm mu muHmm .H>
.muoHHumm mcHHQHoch ma: onmn
mmoHHmu mmE cow 0: momH: moq .hH
.ovHummo z
mcHHmHoch momma :muHmmom:
mohH: moH mu muuma Hoamfi ma .wH
.mmummm moH m :moumcmno mHQEmHm
moan moH mzv Hmumamm mm HmmsH
:m moflH: moH m A£Oumu Hmc.
omHm Hmvmo :mnmc mmucmm mod .mH

mflmcHuumnHH Hm HmHon>mn .>

mumH>mHucm

118

oanmmm

OUHzmmm

OUHsmmm

ocHsamm

>52

>52

>92

>52

ocHsmmm

ocHsoom

ocHsvmm

ocHsmmm

mmmm>
mmcauH¢

mmom>
mmaame

mmom>
mmcsuH<

mmom>
mmcsmH<

60:3:
Hmmo

60:3:
Hmmo

60:5:
dm6U

60:3:
dm6U

60:32

mocsz

moczz

60:52

.mHmzomm
mH :m :mHn mmmHm .mmmn. mzw
mumm mHHm o Hm mo mmuu Goad

.cmcuo cmsn :m mmmoo mam
mmcmucme mHHm o Hw mac cummmm

.mmm:
msw mmoo HmHamHmso :m momma
msv Momma OH mmm:.mHHm o Hm

mac mumm ovcmnaaem omHm Q—

.omm mm moumom muoHHumm
>55 How .mHHmv Hm mm oummmm
msv 0H mom: 0: .mvonH: HE Hm

Houucoo mu muHmm

.mm

.vN

.mm

.NN

.HH>

mumH>mHucm

APPENDIX D

OBSERVATION INSTRU MENTS

119
Appendix D

THE MATERNAL TEECHIBE
OBSERVATION TECHNIQUE (MIUI‘)

No parallel forms of‘the technique were developed by Laosa
(1980b) for use with mothers and their four- to seven-year-old
children. Laosa’s description of the M'IOT is quoted as follows:

The technique may be administered in the subject’s hcme

or in the laboratory. After establishing rapport, the

mother is asked to sit. at a table next to her child. The

mother is given an assembled Tinkertoy model and all the
disassembled parts necessary for making an identical model.

(The parts of the assembled model are glued together to'

prevent their being taken apart, and the disassembled parts

are ‘worked ’ in and out prior to using them with the subjects
until none is unusually difficult for a child fit into any
other part.) The mother is asked ‘to teach' her child

‘how to make' a model like the one already assembled and to

do this as she would if the observer were not presmt.

If others are present, they are asked not to participate and to

be very quiet. The observer manually records on a protocol

the frequency of occurrence of the following maternal beha-
vior categories :

Inquiry (I). The mother asks the child a question or
otherwise directs a verbal inquiry to the child.

120

Directive (D). The mother verbally commands the child
to pursue a given course of action.

Praise 1;). The mother praises, or otherwise verbally
expresses approval of, the child or the child’s activity or
product. .

wave Verbal Feedback gr; Disapproval _(_-_V_§‘_L. * The mother
verbally indicates to the child that a given course of acticn
taken by the child is incorrect or that she is displeased with
the child or the child ’s activity or product.

Modeling (M). The mother works on the model and the child

 

observes. A behavior unit is considered complete (and a
frequency point is recorded), every time the mother fastens or
unfastens two parts.

muggingflg). The mother atterpts to attract the child’s
attention toward a givei aspect of the task by providing a visual
cue. This category is limited to attempts to attract the child ’s
attention by sliding, pushing, or lifting a part or portion of
the model being assembled (but short of fastening or unfastening
any parts). The behavior unit is considered complete (and a
frequency point is recorded) when the mother releases the part or
portion of the model or otherwise moves her hand avay from it.
(More subtle visual cues, such as pointing and touching a part,
were included as additional categories but were deleted from
subsequeit analyses becausetheir interobserver reliabilities and
parallel-form consistency were, in general, only moderate in

magnitude) .

121

Phy§ica1 Affection (PA). The mother makes physical contact
with the child as an expression of a favorable feeling toward the
child.

Positive Physical Control (+PC). The mother manually con-
trols the child’s:motor behavior to facilitate the child’s
solution of the task, e.g., turning the child’s body toward the
task or restraining the child as the child tries to leave the
task area. _

Egghtive Phy§ica1 Control (-PC). This category includes two
classes of ncnverbal behavior, both displaying the mother's
disapproval of the child’s activity on the task or product: (1)
an action that generally would be interpreted as physical
punishment (e.g., slapping the child’s hand); or (2) manually
restraining or controlling the child’s motor activity as the
childworkson the taskinordertokeephimorrerfranpurming
what the mother apparently perceives as action not conducive to
learning or solving the task, or not appropriate for that
particular time (e.g., she takes or pushes the child’s hand away
from the task material or she holds the child ’3 am as the child
begins to reach for a Tinkertoy part).

For each parallel form, the observation is discontinued
5 minutes after the mother is signaled to begin teaching or when
the task is completed, whichever occurs first. The observation
time in seconds is recorded. The coordinated use of a timer and
a stop watch can allow the observer to monitor tame in a manner

that skould be free of distraction to both observer and subjects.

122
The two parallel forms of the technique differ from one

another only in the Tinkertoy models employed. Both models

appear on page 5 of the Tinkertoy Instruction and Idea Book

(Questor Education Products Company, 1972) . Form A employs the

‘Robot, ’ and Form B the ‘Jet Airplane. ' They cmsist of 27 and

28 parts, respectively. Both are of approximately equal

difficulty. (Laosa, 1980c, pp. 357-358)

The following modifications were made to the last four behavior
categories of the M'IUI‘:

M913. Subtle cues suchas pointingand touchingapart of
the model were included.

Phfiical Affection. In addition to the described physical
contact, affectionate behaviors involving no contact were included
such as facial expressions showing approval, mulling, positive eye
contact, and words of endeanment.

Positive Physical Control. Additicnal means of oarorol were
looked for such as the mother placing her hand on the child’s hand to
facilitate solution of the task.

gave Physical Control. The category was expanded to include
facial expressions slowing disapproval, frowning, ani negative eye
contact.

123

Child Behavior During
The Maternal Teaching Observation Technique

Child behaviors during administration of the Maternal
Teaching Observation Techniquee(vflrflfl are observed and recorded
calcurrently with maternal behaviors . The observer manually records
on a protocol (see page 10) for both maternal and child
behaviors, the sequence and frequency of occurrence of the
following dimensions of child behavior: (coding abbreviations
are in parenthesis.)

Positive verbal Response (+VR). The child makes a positive
verbal response to the mother such as answering an inquiry; making an
inquiry, or asking for the mother’s help.

Egghtivepy§§h§;_Re§pgnse‘L:y§L. The child.mekes a negative
verbal response to the mother indicating to the mother that the child
is displeased with her. - "

Thghggyyggwhghh§£;(gu, The child initiates verbal communication
by asking the mother a question or otherwise directing a verbal
inquiry to the mother.

Phygical Regpgnse‘ighl, The child physically responds to the
mother by attending to the task in an effort to comply with her. This
includes the child doing what the mother has directed or suggested
verbally or nonverbally.

ImitatiOn gf_ Mother (IN). The child rewonds to the mother’s
behavior by imitating or attempting to imitate her.

Task Involvement (T). The child initiates the task at hand, or

124
is absorbed in the task to the extent of being oblivious to what the
mother is doing.
Noncompliance (M2) . The child makes no attempt to comply with
the mother's verbal or nonverbal suggestions or directives.
_N2 m _(I‘_1R_L. The child makes no response to the mother.
The child does nothing to indicate a response to any of the maternal

behavioral categories on the MI‘OT.

125

PROTOCOL FOR MTOT AND CBDMTOT

The gotoool for recording maternal and child behaviaa airing
simultaneanadnfinisumdmofdleMatemalTeaclfingomervameedlm
andCtfldBehavichlmngMTOTisstownmmenemPage. Behavim
firequeldsarerecadedintheoflnmnmde'meappogiatecahegaywitha
eutslahmark. Eachroqumeaemsabehavioralmamm
noteswlfichslbfictl‘mfiatesabehaworalsequelce.

Ondotopdgtt—handoamercfthegotocdhthembfict‘s
idemfifimonmmber. Onuiemelcline,the1e:e."A"a:'B'hinlrtodmt'he
blanklineindicatingtowtmchfiormdmeMTOTunmme
gotccol’ndividedintulfmolatroosmdmdthemare
mademdmleft-hdndmarginddopagmmteaaremadedflnchfldmme
fight-handmargin. Onthedghtmargin,d\erearea]soblanbfaxuingdn
mmbercfbehaviormfitsiniflatedbyachmbjactawanndnmlmmbe:
dbehavia'aqmceaoboerveddlringmedmedmlemmabdzondgm-hand
anaerofthegxotocolfiarTakB,ano:eismadedwhethe:a:nd:the
motha-dflddyadcommunicatminangfish,8pamel,crinbomhnguagudldng
eithercbotholoervaa‘oro.

126

I.D. No. ____
Task

NOTES
on
Mother

 

tiacion

 

127

Alvirez, D.; Bean, F. D.; & Williams, D. (1982) The
Mexican American Family. In C. H. Mindel and R. W. Habenstein
(3:13.), Ethnic Families In America, (pp. 269-292). New York:
Elsevier Science Publishing. Company, Inc.

Andrews, M. P.; Bubolz, M. M.; & Paolucci, B. (1980) An
Ecological Approach to Study of the Family.
Marriage and Family Review.
A, 29-49.

Babbie, E. R. (1979). The practice of social research.
Belmont: Wadsworth Publishing Co.

Bartz, K. w. & IeVine, E. S. (1978, November)
Childrearing by Black parents: A description and
comparison to Anglo and Chicano parents. Journal of
Marriage and Family. 709-719.

Baumrind, D. (1966) Effects of autl‘oritative parental control
on child behavior. Child Development. 3_7, 887-907.

Baumrind, D. (1971) Current patterns of parental
authority. Developmental Psychological Monographs, 3(11) ,
Part 2' 1‘103.

Bammrind, D. (1975) Early socialization and the disci-
pline controversy. Morristown, NJ: General Learning Press.

Batmtrind, D. (1978) Parental disciplinary patterns and
social carpetence in children. Youth & Society. 9(3),
239-276.

Baumrind, D. & Black, A. E. (1967) Socialization practices
associated. with dimensions of competence in prescrool boys
and girls. _Ch, _3_8_: 291-327.

Bernal, M. 3.; North, J. A.; Rosen, P. M.; Delfini,
L. F.; & Schultz, L. (1979) Observer ethnicity
effects on Chicano mothers and sons. Hispanic Journal
of Behavioral Sciences. 1, 151-164.

128

Bigner, J. J. (1979) Parent-child relations. New York:
memillian.

Brophy, J. E. (1970) Mothers as teachers of their own
preschool children: The influence of socioeconomic status
and task structure on teaching specificity. Child
Development. fl: 71-94.‘

Bubolz, M.; Richer, J.; & Sontag, M. S. (1979, Spring) The
human ecosystem: A model. Journal of Home Economics.
28-31.

Conger, R. D.; mCarty, J. A.; Yang, R. K.; Lahey, B. B.;
Burgess, R. L. ~ (1984, May) Mother’s age as a predictor of
observed maternal behavior in three independent samples of
families. Journal of marriage and Family. _4_6_(2):

411-424.

Cook, T. D. & Campbell, D. T. (1979) Wimtation:
Design and analysis issues for field settings. Houghton
Mufflin Co. .

Cromwell, R. E. (1969) Development of a parental
attitude research scale for use with the lower and middle
classes. Unpublished master’s thesis, Kansas State
University, Manhattan, Kansas.

Cromwell, R. E.; Vaughan, C. 3.; & Mindel, C. H.
(1975, August) Ethnic minority family research in an
urban setting: A process of exchange. The American
Sociologist. _12, 141-150.

Delgado, M. (1980, September) Providing child care
for Hispanic families. Young Children.

Durrett, M. 13.; O’Bryant, S.; & Pennebacker, J. W. (1975)
Child-rearing reports of White, Black and Mexican American
families. Developmental PsychomQy. _l_l_(6), 871.

Elkind, D. (1981) The hurried child: Growing @too
fast too soon. Reading, Massachusetts: Addison-Wesley

Publishing Company .

Ferguson, L. R. (1970) Personalitydevelogment.
Belmont, California: Wadsworth Publishing Co., Inc.

Field, T. M.; Widmeyer, S. M.; Stinger, S.; & Ignatoff, E.
(1980) Teenage, lower—class Black mothers and their pre-
term infants: An intervention and developmental follow—up.
Child Development. _S_l: 426-436.

Freeman, L. C. (1965) Elementary applied statistics. New
York: John Wiley and Sons, Inc.

129

Griswold, D. L. (1975) An assessment of the child-
rearing information needs and attitudes of Anglo, Black, and
Mexican-American mothers . Unpublished doctoral
dissertation, Arizona State University.

Harshbarger, T. R. (1977) Introductory statistics: A
decision map. morallan Publishing Company, Inc.

Hess, R. D. & Shipmen, V. C. (1972) Parents as teachers:
How lower class and middle class mothers teach. In C. S.
Lavatelli & F. Stendler (Eds.), Readings in child
behavior and development . New York: Harcourt Brace
Jovanovich.

Hess, R. D. (1970) Social class and ethnic influences
upon socialization. In P. H. Mussen (Ed.), Camuchael’s
manual of child psychology. Volume 2. NY: John Wiley
and Sons, 457-557.

Hollingshead, A. B. (1957) The two factor index of
social position. New Haven, Connecticut: Yale University
Press.

Hull, C. H. & Nie, N. H. (1981) SPSS update 7-9: New
procedures and facilities for release 7-9. McGraw-Hill.

Kagan, S. & Ender, P. B. (1975) Maternal response to
success and failure of Anglo-American, Micah-American,
and Mexican children. Child Developmelt. 4_6_, 452-458.

Kantor, D. & Iehr, W. (1975) Inside the family.
San Francisco, California: Harper & Row, Publishers.

Kearns, B. J. (1970) Child-rearing practices among selected
culturally deprived minorities. Journal of Genetic
Psychology. 116(2), 149-155.

Kerlinger, F. N. (1964 or 1974) Fundamentals of behavioral
research. New York: Holt, Rinehart and Winston.

Iaosa, L. M. (1977) Socialization, education, and
continuity: ' The importance of sociocultural context.
Young Children. 12(5), 21-27.

Iaosa, L. M. (1978) Maternal teaching strategies in Chicano
families of varied educational and socioeconomic levels.
Child Development. _4_9_, 1129-1135.

Iaosa, L. M. (1980a) Maternal teaching strategies and
cognitive styles in Chicano families. Journal of Educa-

130
tional Psychology. 7_2(l), 45-54.

Iaosa, L. M. (1980b) Maternal teaching strategies in Chicano
and Anglo-American families: The influence of culture and
education on maternal behavior. Child Development. _5_l_,
759-765.

Iaosa, L.M. (1980c) Measures for the study of maternal
teaching strategies . Applied Psychological Measurement.
3(3), 355-366;

Laosa, L. M. (1981) Maternal behavior: Sociocultural
diversity in modes of family interaction. In R. W. Henderson
(Ed.) Parent-child interaction; theory, research, and prospects,
pp. 125-167.

Leonard, W. M. II. (1976) Basic social statistics. New
York: West Publishing Co.

IeVine, E. S. & Bartz, K. W. (1979) Comparative child-
rearing attitudes among Chicano, Anglo, and Black parents.
Hispanic Journal of Behavioral Sciences. yz), 165-178.

Iindholm, K. J. & Padilla, A. M. (1981) Socialization via
communication: Patterns used by Hispanic mothers and children
in mastery skill communication. In R. Duran (Ed),
Discourse processes: Advances in research and theory. Volume
III. Norwood, NJ: Ablex Publishing Corp..

Martinez Byrd, E. (1980) The effects of a Dreikurs parent
education prpgram with Hispanic and Anglo HeadStart
mothers. Unpublished master ’3 thesis, Colorado State
University, Fort Collins, CO.

Mirande’, A. (1980) The Chicano family: A reanalysis of
conflicting views. In Arlene and Jerome Sholneck’s
Family in Transition Little, Brown and Co.

Mirande’, A. & Enriquez, E. (1979) La Chicane: The
Mexican American woman. The University of Chicago Press.

Mussel, P. H.; Conger, J. J.; & Kagan, J. (1974) Child
development and personality. Harper and Row, Publishers.

Nie, N. H.; Hull, C. H.; Jenkins, J. G.; Steinbremer, K.; &
Bent, D. H. (1975) SPSS statistical package for the
Social Sciences. NbGraw-Hill Book Company.

Osofsky, H. J. & Osofsky, J. D. (1970) Adolescents as
mothers: Results of a program for low-income pregnant
teenagers with some emphasis upon infants’ development.
American Journal of Orthopsychiatry. fl: 825-834.

l3l

Padilla, A. 'M. (1980) The role of awareness and ethnic
loyalty in acculturation. In A. M. Padilla (Ed),
Acculturation: Theog, models and some new findings
(pp. 47-83) . Boulder, Colorado: Westview Press.

Parra, E. & Henderson, R. W. (1982) Mexican-American
perceptions of parent and teacher roles in child
development. In J.A. Fishman and G.D. Keller (Eds.),
Bilingual education for Hispanic students in the U.S. '
(pp. 289-299) New York: Teachers College, Columbia
University.

Rosenblatt, A. & Wiggins, L. M. (1967) Characteristics of
the parents served. Social Casework. 48(10), 639-645.

Schaefer, E. S. & Bell, R. Q. (1958) Developrent
of a parental attitude research instrument. Child Devel-
m. 29(Septamber), 339-361.

Sears, R. R.: Rau, L.; & Alpert, R. (1965) Idmtification
fl child rearing. Stanford University Press.

Sears, R. R.; Macooby, E. E.; & Levin, H. (1957) Patterns
of child rearing. Evanston, Illinois: Row, Peterson, and
Co.

Spence, J. T.; Cotton, J. W.; Underwood, B. J.; &- mncan,
C. P. (1976) Elementary statistics. Prentice-Hall, Inc.

Stallings, J. A. & Porter, A. (1980, September). A national day
care hare study: Observation component. Final report of the
Stanford Research Institute. Menlo Park, ax: SRI
International.

Staton, R. D. (1972) A comparison of Mexioan and
Mexican-American families. The Family Coordinator. a,
325-330.

Stramen, E. A.; McKinney, J. P.: Fitzgerald, H. E. (1983)
Developmental psychology: The school-aged child.
Hatewood, Illinois: The Dorsey Press.

U.S. Depart-mat of Camerce. (1980) Bureau of the Census.

U.S. Department of Camerce: Bureau of the Census. (1983)
Condition of Hispanics in America today.

Vigil, J. D. (1980) Fran Indians to Chicanos: A
sociocultural history. The C. V. Mosby Co.

132

Weigert, A. J. (1968) Parent-child interaction patterns
and adolescent religiosity: A cross-nation study.
Dissertation Abstracts International , _2_9_, 3691 .