- '\ val)!
‘J
"I: x. .,

“an :-

 

«in f E {1,424

' 1 V
’7 LA“
can,»

1 .m i

4. 31.1”".

“.1 a

“(a M
«1:.I

.4
. mu: 1-11.:
n

”w-

41-7122“

,.

‘J—
" “£3.
-\.;.x=.:

I f
Imnnaief‘
a —.
21. y: a}:

"3:3“?5'5 9Q

q; x

*‘2‘ fi“’§‘§‘
“’3

11%;- gm;

“’1“:
1:” Max-1
.1" ' ‘

M
.33.“.441‘J -

u"
223‘s x 1 u “fizz?

,
£111:
3' x
' an: M,

..4
h
\.
11.3473“ «3.0:
-‘ ‘ 1 .-
u.

Ar
m.»
r .

wag:
1..
-‘:;'M‘v'~’fi

*u

7 14. . at. £125
‘ . .isis‘xtz‘ima‘s 13:. {f a.
“9.?

§- {31:17.

‘r m
13‘

.v

0

.13.].

.37: hr
a.

 

 

 

_ A ,
, "ZHRT

;
1'2"" .9

1-3.;9

LY.
:- :.,‘::.r: u.
v

 

 

‘r_ ~
" v5.3: 'Zn
‘3': 1.:- 1

”.34.; 14w:-

'71:; :::~
ji’LLrfl»
“MEL: :1. ii;.,
4

2::
2..

qua- 'novflwq .
.:1.--ummu—-c

“ 414*? «$1.,

-Ao..'.

N
-K .11.-..7. , I. .-
:v-t- #Qrwo'r

v“ ” Woo—w «.-
' W"? i‘...g/L

.3911

:3: ’13:}: -
:5- .gmm

~,,..

f-‘m «n: ”H!”

”"35.

‘ 'wvaw...?;....m .
1...

:wmvu A'tcfi-m:

’I‘bonnu:tm
m...” we“
-——'*1IO~O
on...
on. ...~~rm.«

fins-2.1:...“

cw-.-
a W»

M32...” '

:v—n.

 

‘ Hl_\1lk ‘

llllllllllllllllllllllllllllllHHIJIHI'IIllllllllll

31293 00913 4788

This is to certify that the

thesis entitled
The Relationship Between Risk Factors, Mothers'

Level of Social Support and Stress Among Users

of Respite Child Care
presented by

Maria Zeglen Townsend‘

has been accepted towards fulfillment
of the requirements for

M.A. Child Development

degree in

 

 

    

Major professor

Date M

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

LEERARY
Michigan Start.
University

i
J
z

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU In An Affirmative ActlorVEquel Opportunity Institution
cMMan-M

THE RELATIONSHIP BETWEEN RISK FACTORS,
MOTHERS’ LEVEL OF SOCIAL SUPPORT AND STRESS
AMONG USERS OF RESPITE CHILD CARE

by

MARIA ZEGLEN TOWNSEND

ATHESIS

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

MASTER OFARTS

Department of Family and Child Ecology

1991

Abstract

THE RELATIONSHIP BETWEEN RISK FACTORS.
MOTHERS’ LEVEL OF STRESS, AND SOCIAL SUPPORT
AMONG USERS OF RESPITE CHILD CARE

by

Maria Zeglen Townsend

There is relatively little empirical data on parents who use respite
child care and the relationship between the use of these services and
reduction in parental stress. A sample of two sub-groups; American and
foreign mothers who used respite child care in a midwestern town was used
to examine the relationships between identified risk factors and maternal
levels of stress and social support. The majority of the mothers were married,
high school graduates and living below the poverty threshold.

The results include relationships between some of the identified risk
factors and maternal levels of stress and social support. These relationships
were often difi‘erent in magnitude and direction for the sub-groups. An
interesting finding within the sub-groups was that the American mothers
who used the respite child care service more Muently had higher levels of
stress and the foreign mothers who use the service more frequently had lower
levels of stress.

DEDICATION

I'd like to take this opportunity to dedicate this thesis to my mother,
Joanne Zeglen. Though widowed in her forties with a teenage son and two
schooled-aged daughters, she reared us in a home filled with love and
taught each of us the importance of a college education. For your
encouragement and all those Financial Aid Forms you complewd for me,
thank you and I love you very much.

ACKNOWLEDGEMENTS

I would like to thank Dr. J eanne Brown for her guidance and
friendship throughout my graduate career. You have been a wonderful
friend and mentor. I would also like to thank the other members of my
committee; Dr. Tom Luster for his comments and guidance, especially
with chapter four, and Dr. Margaret Bubolz for her comments. Special
thanks is extended to Judy Pfafi' for her patience and knowledge of
statistics. I could not have done the data analysis as quickly as I did (and
needed to) without your assistance. I would also like to thank a dear friend,
Tracy, for helping me with the data collection on Thursday afternoons.

I would like to thank the Family Growth Centers and the staff.
Merry Stanford, for allowing me to conduct my study, along with Maureen
and Delores for their assistance. Most especially, I would like to thank the
mothers who participated in this study. Thank you for your time and
honesty; without your help this thesis would not have been possible. You
ARE the backbone of social research.

I would also like to thank my husband, Ed, for the many meals you
cooked, your editorial comments, your inspirational screen savers, and
most especially your unwavering confidence in my ability to complete this
thesis. It also deserves mentioning that you survived my roller coaster ride
I of emotions throughout the writing of my thesis.

I wish to thank all the members of my support network. I could not
have survived all the “crises” of graduate school and research without you.
I would like to thank my special FCE friends, Connie (for handling my '
long distance cries), along with Joy and Gloria. Thanks also to the “Chem

iv

Geeks” and their significant others; Kurt, Kris, Laura, Art, Mary, Allen,
Jason, Sue, Berly, Gilbey, Gary, and all the rest, for all those great and
sometimes humiliating times.

TABLE OF CONIENTS

P809
List of Tables ............................................................................ VIII
List of Figures ........................................................................... IX
Chapter 1. Identification of the Problem ...................................... 1
A . Introduction ................................................... 1
B. Purpose of Study .............................................. 2
C. Research Objectives ......................................... 2
D. Research Questions ......................................... 3
E. Rationale ........................................................ 3
Chapter 2. Review of Literature .................................................. 5
A . Introduction .................................................... 5
B. Parental Stress ................................................ 6
C. Risk Factors Associated with
Parental Stress ................................................ 8
D. Parental Social Support ..................................... 17
E. Respite Child Care ........................................... 23
Chapter 3. Methodology ............................................................. so
A. Research Objectives and Questions .................... 3)
B. Conceptual and Operational Definitions .............. 3)
C. Instrumentation ............................................. 33
1. Parenting Stress Index-Short Form .......... 33
2. Maternal Social Support Index ................. 35
3. Family Growth Center Questionnaire ........ 37
D. Sample Design ................................................. 38
E. Techniques of Data Collection ............................ $
F. Statistical Analysis ........................................... 41

vi

G. Limitations of the Sampling Design and Data

Collection Techniques ...................................... 42
Chapter 4. Results of Statistical Analysis ..................................... 44
A . Introduction .................................................... 44
B. Cleaning of the Data ......................................... 44
C. Reliability of the Instruments ............................ 45
D. Description of the Sample .................................. 47
E. Difi‘erences between the American and Foreign
Mothers .......................................................... 49
F. Results of the Study ...................... . .................... 53
1. Question 1 .............................................. 53
2. Question 2 .............................................. 56
3. Question 3 .............................................. $
G. Chapter. Summary ........................................... 64
Chapter 5. Discussion and Implications ...................................... 67
A . Introduction .................................................... 67
B. Differences within the Sample ........................... 67
C. Discussion of the Research Objectives ................. a9
1. Research Objective 1 ................................ as
2. Research Objective 2 ................................ 72
3. Research Objective 3 ................................ 74
D. Summary ........................................................ 76
E. Future Research .............................................. 78
References .................................................................................. so
Appendix A. Questionnaires ...................................................... $
Appendix B. Letters and Forms .................................................. Q

LIST OF TABLES

Pace

Frequency distribution of income for the study sample ............ 48
Difi‘erences between the American and foreign mothers ......... 51
Correlations between the risk factors and the mothers’

PSI-SF scores by sub-groups and total sample ........................ 54
Correlations between the risk factors and the mothers’

MSSI scores by sub-groups and total sample .......................... 57
The relationship between the social support measures and

level of stress for both sub-groups ......................................... 6)

LIST OF FIGURES

Flam A P386

41 Description of the study sample by ethnicity and race ........ 47
4-2 Scatterplot of the relationship between the MSSI scores
and level of education for the foreign sub-group. .............. 58

1

Chapter 1
Identification of the Problem

Introduction

Respite care is a support service that ofi‘ers parents a break from the
demands of caring for a dependent person. It is most often associated with
care for handicapped or elderly pe0ple, and its primary intent is to provide
support to the family, more specifically the primary caregiver (Salisbury &
Intagliata, 1986). However, not all respite care programs target families
with special needs children or elderly members; some programs target
children who are at-risk of maltreatment. "This is a large population of the
country's children, according to the Feb/March 1990 Children's Defense
Fund Report fi-om Washington, D.C., every day in America 1,849 children i
are abuse and neglected" (Peterson, 1990, September 9).

Despite the fact that respite child care was developed to relieve
parental stress and lessen the possibility of maltreatment, there have been
few nonempirical research attempts to study the efi‘ects of respite child
care. In fact, only three studies that have examined the efi'ects of respite
care on parental stress are consistently cited in the literature
(Rimmerman, 1989; Subramanian, 1986; Wikler, Hanusa, 8: Stoychefi',
1986). Consequently, funding for respite child care programs is difiicult to
obtain because of the lack of evidence in the research literature that certifies
its merit. Although the ideal study would be to determine if respite child
care reduces maternal levels of stress through the use of pre- and post-test
on mothers who use respite care, it was not possible at this time to conduct

this type of comprehensive study. Because of restrictions imposed by the

2

Family Growth Centers (FGCs), the researcher chose a study that only

required a one time participation of the mothers.

Purposeof Study

The purpose of this study was to examine the relationship of the
identified risk factors on levels of social support and stress among those
mothers who use respite child care. The measurement of the variables
included two standardized questionnaires, one original questionnaire and
the use of FGCs' records on the families. The three FGCs sites used for
data collection are located in a midwestern town. They offer free, drop-in
respite child care.

Reseathhjecfives
In order to achieve this purpose, more specific objectives were

developed and were used as guides for this research:

1) To determine if the identified risk factors were related to the

maternal level of stress in this sample;

2) To determine if the identified risk factors were related to the
maternal level of social support in this sample;

3) To determine if the maternal levels of social support were

related to the levels of maternal stress in this sample.

Research W
In order to accomplish the objectives of this study, several research

question were addressed.

1. * What is the relationship between m and/or number of

l ' identified risk factors the mothers had and their level of
stress? These risk factors are: (a) low level of mcome, (b) low
level of education, (c) high number of children, ((1) low number
of adults residing at home, (e) high number of young children
residing at home, (0 minority race status, (g) non-marital
status. (h) teen parenthood status, (i) inadequate space in the
home, and (j) low number of amenities in the home.

.i r 5 ‘4'” P“ *
2’ What 18 the relationship between the type ‘and/drr‘n’umber of

identified risk factors the mothers had and their level of social?
support? These risk factors are: (a) low level of mcome, (b) low"M
level of education, (c) high number of children, (d) low number
of adults residing at home, (a) high number of young children
residing at home, (i) minority race status, (g) non-marital
status, and (h) teen parenthood status?

3. What is the relationship between the levels of social support

and the levels of stress?

Rationale

This is an important area of research because there is little available,
data on respite child care as a support for mothers (Joyce & Singer, 1983;
Joyce, Singer, & Isralowitz, 1983; Wikler, Hanusa, & Stoychefi', 1986) and no
data on respite child care as a moderating efi‘ect between identified risk
factors and maternal levels of stress. By examining the moderating effects
of respite child care on the maternal levels of stress, the FGCs may be able

4

to use the results of this study as evidence of the need for respite care
applying for future grant proposals. This study will also add to the

- literature in the area of respite child care. Since the sample was drawn
utilizing a non-random sampling procedure, generalizability of the study
will be limited to the parents using the three Family Growth Centers.

5

Chapter 2
Review of Literature

Introduction

The literature review is presented in four sections. In the first
section, theoretical models of stress and how stress is related to the
parental role will be presented to provide information needed to understand
the relationship between the risk factors, maternal social support, and
maternal stress. The second section discusses those studies focusing on
risk factors associated with parental stress and support social. The third
and fourth sections examine the moderating efi‘ects of parental social
supports and use of respite child care services on the levels of parental
stress.

Most of the research in the area of respite child care focuses on
physically and mentally handicapped children. The researcher believes
that although there are specific problems faced by parents of handicapped
children, there are also similarities between parents of handicapped
children and parents who are at-risk for maltreatment of their children.
Both sets of parents are experiencing stress in their parental role and both
could benefit from time away from their children. Except for section four,
respite child care services, the focus of this review of literature is on the
identified variables' influence on high-risk populations. The rationale for
this approach is that the study sample was drawn from mothers who use
respite care, a program developed to serve a high risk population, i.e.
mothers at-risk for child maltreatment.

Parental Stress

The ABCX Crisis Model was created by Hill, a sociologist, to explain
why some families are able to c0pe with crises while others are not
(McCubbin & Patterson, 1983). In this model, A is defined as the stressor
event, B is defined as the family’s resources for meeting the stressor event,
C is defined as the family’s perception of the stressor event, and X is defined
as the crisis. Examples of stressors are transition to parenthood, teen
pregnancy and parenthood, single-parenthood, and economic hardships
related to job loss, divorce, and low levels of education. These stressors
could originate outside the family, such as an economic depression, or
inside the family, such as a divorce.

How the family copes with the stressor depends on the resources the
family has available to prevent the stressor from precipitating into a crisis
event. Possible resources for a family include: familial cohesion, social
support. collective goals of the family, agreement on role structure in the
family, and members’ satisfaction with the family’s past experiences in
meeting the members’ needs. Social support could come from within the
family, the extended family, friends, agencies, or organizations. An
example of an agency or organization could be the Family Growth Centers.

In order for the family to decide to use their available resources, they
must identify an event as a stressor event. The family’s perception of a
stressor event is a reflection of the family's values and previous experiences
in dealing with stressor events and crises (McCubbin & Patterson, 1983).

According to the ABCX model, family stress is defined as an
imbalance between the demands on the family, stressor events, and the
family’s capability (resources) to overcome these events (McCubbin &
Patterson, 1983). Whether the family experiences stress depends on many

7

things. These might include: the type of stressor event, the families
resources, familial characteristics, and characteristics of individual family
members. A black, single-parent mother may define an event as stressful,
whereas a dual-career, white family may define the same event as not
stressful.

The ABCX Model focuses on what the family does when a stressful
situation occurs and defines familial stress as an imbalance between. the
family's demands and capabilities. The Parenting Stress Model focuses on
sources of stress for parents that lead to dysfunctional parenting (Abidin,
1990). “It is posited that the total stress a parent experiences is a function of
certain salient child characteristics, parent characteristics, and situations
which are directly related to the role of being a parent’(Abidin, 1990, p. 2).

The pertinent child characteristics include: the child’s ability to
adapt to changes, the demands the child makes on the parent, being prone
to excessive crying or withdrawal, hyperactivity, how close the child
resembles the parents’ idealized child, and how much the child reinforces
the parents’ in their role as parents. The parental characteristics include:
depression, sense of competence in parenting role, and the investment of
the parent in the parenting role. The parents’ relationship with each other,
the amount of social support available to the parents, the parents’ health,
and the restrictions placed on the parents due to the parenting role are all
situational variables in this model.

The Parenting Stress Index is based on the Parenting Stress Model.
The short form of the Parenting Stress Index (PSI-SF) is a simplified
version of the Parenting Stress Index. In the PSI-SF, the stress a parent is
experiencing is a function of three things. They are: the distress parents
feel in their role as parents, the parents’ perceptions of the child as not

8

meeting their expectations, and those characteristics of the child that make
him/her either dificult or easy to parent. The amount of distress parents
feel in their parenting role is related to the amount of available social
support, restrictions placed on the parents due to their parenting roles, and
conflict with spouse or significant other.

These two models ofi'er some conceptual background on the variable
of parental level of stress. The next three sections of this chapter will
examine the relationship between risk factors (some of which could be
- defined as stressor events), social support (an element in each of these two

models), and parental level of stress.

Risk Factors Associated with Parental Stress

In this section, the relationship of certain risk factors found in high-
risk families to the level of stresses experienced by these families is
examined; These risk factors are income, education, number of children
in the home, ages of children in the home, single parent status, minority
status, parental age at birth of first child, amount of space in the home,
and number of amenities in the home. Only those studies, that focus on
the identified risk factors as they relate to the level of stress experienced by
the parent are utilized to ensure that the information gained is closely
related to the purpose of this study.

Low level of income has been found to increase the level of stress
experienced by parents (Andrews 8: Withey, 1979; Lindblad-Goldberg,
Dukes, & Lasely, 1988; McLanahan, 1983; Makosky, 1982; Pearlin &
Johnson, 1977). The financial strain of raising children can be stressful for
any parent regardless of Social Economic Status (SES) (Miller & Myers-

9

Wall, 1983). However, this stressor can be even more pronounced when the
parent is in the low SES classification.

In a national study examining the social indicators of well-being,
people with lower SES expressed more negative evaluations of the following
concerns: economic, family and marriage, health, and fun and enjoyment
(Andrews & Withey, 1979). Also, this group experienced more stress in
the areas of economic, family, and marriage concerns (Andrews & Withey,
1979). Among those people who reported low level of income, there was a
disproportionate number of blacks and single-parents.

The finding in the previous study that many of the people who are low
SES were single-parents is not unusual. In fact, the average single-parent
family income is half the average income for a two-parent family (Hogan,
Bushler, & Robinson, 1983). This is the result of economic changes
associated with divorce and lower levels of income for women in general.
Therefore, it is not surprising that several studies have found that low level
of income increases parental stress were focused on single-parents
(Lindblad-Goldberg, Dukes, & Lasely, 1988; McLanahan, 1983).

In a study utilizing longitudinal data from the Panel Study of Income
Dynamics, single-parent households were rated as experiencing more
stress because of their increased likelihood of experiencing the following
life strains: poor education, low income, and being black (McLanahan,
1983). It was also found that single-parent families had significantly
higher scores on the Life Events Scale, with more disruptions in the areas
of income, household composition, and work.

By examining an all black, single-parent sample, researchers found
that the mothers' demographic variables of income and education
influenced their level of stress (Lindblad-Goldberg, Dukes, & Lasely, 1988).

10

The sample included 70 functions] and 56 dysfunctional, low-income
families. It was found that a mother's dissatisfaction with her income
positively relawd to her summary stress score and both subscale stress
scores. The subscales are the number of events requiring the most
adjustment and the number of negative events reported by the families
(rs=.19, .19, .22, respectively).

In another study, 43 low-income mothers, who were not necessarily
single-parents, were found to experience many stressful life events
(Makosky, 1982). “These mothers reported from 5 to 33 events, with a
median of 13.5 in one year. This compares to the average of 1 to 2 events per
year found in most community surveys” (Makosky, 1982, p. 37). In a
ranking of the areas of stressfulness completed by the mothers, concern or
worry for money ranked the highest (60.1 of a possible 100) and was followed
by the quality of living conditions, which was ranked 51.7.

Based on data collected from 2,300 people in the Chicago area, the
relationship between depression and both marital status and life strains
were examined (Pearlin 8: Johnson, 1977). Though the study did not
examine the relationship between economic strain and depression for
single-parents, 38% of unmarried people in the sample were parents. It
was found that the unmarried people in the sample experienced greater
economic strain. In fact, 13% of the unmarried people, compared to 5% of
the married people, reported experiencing severe economic strain. In this
study, economic strain was defined as not having enough money to feed,
clothe, or provide medical care for the family. Also, a relationship was
found between the intensity of economic strain and level of depression for
the general sample. Fifty-four percent of the 92 single people experiencing

11

severe economic strain scored the highest on the adopted measure of
depre ssionr

Low level of income is associated with increased levels of stress
among parents. However, this relationship does not exist within a
vacuum. Those parents that are experiencing stress due to economic
strain are. experiencing stress from being a single-parent and/or being
black (Lindblad-Goldberg, Dukes, 82. Lasely, 1988; McLanahan, 1983;
Andrews & Withey, 1979). These two risk factors, marital status and race,
are examined in more detail below.

When comparing single-parents with married parents, single
parents are found to be experiencing higher levels of stress (McLanahan,
1985; Weinraub & Wolf, 1983). A small sample of 28 mothers, halfof whom
were single—parents, were interviewed to determine the efl'ects of stress on
mother-child interaction (W einraub & Wolf, 1983). The single mothers
reported an average of 2.6 more life changes than the married mothers. In
another study involving over 8,000 parents, it was found that the well-being
of single mothers declined by 11% over a one year period (McLanahan,
1985). This same study found that being a single mother was less stressful
for black women than white women (McLanahan, 1985, p. 262).

This last finding is contrary to other studies that examine the
relationship between race and level of stress experienwd by the parents
(Andrews & Withey, 1976; Conger et al., 1984; Pearlin & Johnson, 1977).
Black families are subject to different types of stressors than white families
because of racism (McAdoo, 1983). Blacks are overrepresented in the ranks
of the unemployed and underrepresented on the enrollment lists of colleges
and universities. These trends lead to the increased likelihood of blacks

experiencing more stress due to lower levels of education and income. In

12

some studies, the most stressed families were black. This was determined
by measures of well-being (Andrews & Withey, 1976) and the number of
chronic stressors such as family income, number of parents, number of
children, and education of mother (Conger et al., 1984).

The interrelationship of race, number of children, and level of
depression was examined in a previously mentioned study (Pearlin &
Johnson, 1977). In this study, 2,300 people from the Chicago area were
interviewed to determine the relationship between marital status, life
strains, and depression. The results of the study indicated a positive
relationship between the number of children in the home and level of
depression of the parents. On the average, the married parents in the
sample had more children than the single-parents. Among single-parents,
black parents had a larger number of children than white parents and had
higher levels of depression.

The risk factors of single-parenthood and race were found to increase
the level of stress for parents. Being black increases the chances of
experiencing other risk factors such as being a single-parent, having a
lower level of income, and having more children. Evidence of parenting
stress is confirmed by the large decrease in the marital satisfaction and
well-being of parents after the birth of the first child (Miller 8r Myers-Wall,
1983; Straus, Gelles, & Steinmetz, 1980). The relationship of marital
satisfaction and children follows a U-shaped curve with the lowest point
occurring when the oldest child is an adolescent and rises as the children
leave home (Lamanna & Riedmann, 1988). The downward slope of this
U-shsped curve may be related to the increase in the homework load and
the decrease in the parent’s free time, which occurs with the addition of
each new child to the family.

13

As mentioned in an earlier study, there is a positive relationship
between the number of children in the home and the parents’ level of
depression (Pearlin & J ohnson, 1977). This is true for both single-parents
and married parents. However, single-parents experienced higher levels of
depression as the number of children increases. Thirty-eight percent of the
single-parents with three or more children scored the highest on the
adapted depression scale compared to only 2.3% of the married parents.

In a study conducted to determine the relationship between family
~ size, parent»child interaction, affect, and stress, it was found that as the
number of children increased the parents experienced more stress in their
role as parents (Nye, Carlson, & Garrett, 1970). The fathers in the study
experienced an increase in strain as measured by disposition after the birth
of the fourth child. The fathers were described as being more moody,
having frequent displays of temper, and lacking cheerfulness. According
to the authors, mothers experienced an increase in stress after the birth of
the second child. The evidence for this trend was found in the decrease in
the mother’s Satisfaction Index. There was no linear relationship between
the number of children and parental stress. In fact, the authors feel that
these findings are fragmentary and only suggestive of a relationship (Nye,
Carlson, & Garrett, 1970). Instead, they suggest that the increase in
parental stress may be partially caused by the decrease in family income
level, which is a direct result of the increase in the number of children.

In a study conducted by McLanahan (1985), the number of children
had an inverse efi'ect on the changes in well-being of the parent. Parents
with a large number of children experienced a decrease in their scores of
well-being from Time 1 (1968), to Time 2 (1971). It was found that single
mothers were less afi‘ected by the number of children than married fathers.

14

This result is similar to the finding in the Nye, Carlson, and Garrett study
(1970). That is, the decrease in the fathers’ disposition was greater than the
- decrease in either the" mothers’ disposition or Satisfaction Index scores.

Not only does the number of children affect the levels of stress
experienced by the parents, but the ages of the children also afi'ect the level
of stress. For example, physical stressors of parenting are greater on the
parents of young children compared to the parents of adolescents (Miller 8:
Myers-Walls, 1983). This is because of the greater dependence of the
younger child on the parent to meet all of his/her needs. Some studies do
not directly examine the relationship between ages of children and parental
levels of stress. Instead, they define this risk factor as a stressor
(McLanahan, 1983; Pittman, Wright, & Lloyd, 1989).

In a study utilizing data from the General Social Surveys, an inverse
relationship was found between the ages of children and the measures of
well-being for the adults (Glenn & Weaver, 1979). The presence of children
in the home was found to have a negative efi’ect on the parents’ well-being
as measured by the global happiness question. This question asks the
respondent to rate how they think things are going these days. Having
children between the ages of birth through five years was negatively related
to the fathers’ well-being but not for mothers. Mothers’ well-being was
negatively related to having children who were between the ages of six to
twelve years. Both fathers’ and mothers’ well-being was negatively related
to having adolescent children. More significant findings could have been
found if a less global and more specific measure of well-being was utilized.

A relationship between age of children, marital status, and
depression was found in a study conducted by Pearlin and Johnson (1977).
Single-parents of children five years and younger had higher levels of

l5

depression than those parents with children who are six years or older.
“The gamma coeficient for this association among parent whose youngest
child is five years or less is .48; for those whose youngest is from 6 to 17, the
coeficient is .39; and it declines further to .13 among parents having a
youngest child over 18" (Pearlin & Johnson, 1977, p. 712).

Just as the age of the children can afi‘ect the parents’ level of stress,
so can the age ofthe parent at the birth ofthe first child. In a study of 107
families drawn from three samples in two difl‘erent states, the relationship
between the mother’s age and maternal behavior was examined (Conger et
al., 1984). The three samples varied in levels of risk, from high risk to
middle class families. In one of the two models-the chronic stress model
beingexsminedinthestudy, theauthors defineageatfirstbirthasa
stressor. This is consistent with another study that defined age at first
birth as a chronic life strain (McLanahan, 1983). This definition is logical
because teen parenting is associated with chronic environmental and
psychological stressors, such as low levels of education and income, and
single-parenthood.

Testing of the second model (life experiences model), revealed a
relationship between the chronic-stress measure and age at birth of first
child (Conger et al., 1984). The chronic-stress was a combination of four
demographic variables: family income, number of parents, number of
children, and education of mother. This measure was negatively related to
the age at first birth for all three samples, but significant for only two
samples. It was not significant for the highest-risk sample. The authors
suggested that this may be a result of the sample’s homogeneity.

Age at first birth influences the levels of stress experienced by the
parents. This risk factor is often accompanied by the risk factors of low

16

income, single-parenthood, and low education. When a person becomes a
parent at a young age, their education is often interrupted in order to
assume this- new parental role and to handle the added responsibilities.
Social scientists recognize the efi‘ect of low levels of education on the lives of
people and often define low level of education as a stressor in research
studies (Conger et al., 1984; McLanahan, 1983; Zuravin, 1988). f

In the study conducted by Lindblad-Goldberg, Dukes, and Lasley
(1988), it was found that the “nonclinic single-parents reported significantly
more positive educational changes than did the clinic single-parents”

(p. 111). Not only did the nonclinic single-parents report significantly more
positive educational changes, but they also reported fewer stressful life
events. Thus, the clinic families experienced more negative educational
changes and more stressful life events.

Several of the studies listed the family’s living conditions as being
stressful to the parents (Makosky, 1982; McAdoo, 1983). McAdoo stated that
an indirect result of racism’s denial of economic opportunities is
inadequate housing. Many blacks live in houses or apartments that are too
small and are located in high-crime neighborhoods. In Makesst (1982)
study of 43, low-income mothers, she found that only 14% of the sample
lived in the more spacious single-family dwellings. *

Though not researched in the articles dealing with parental stress,
lack of space and lack of amenities in the home can be stressful to the
parent. A parent who has no phone may worry about what to do if her child
needs immediate medical help or she may be socially isolated. The lack of
the basic amenities, such as running water and electricity can increase the

work load of a parent and her level of stress.

17

The literature has shown a relationship between parental level of
stress and the following identified risk factors: income, single-parenthood,
race, number of children, ages of children, age at first birth, and education
(Andrews & Withey, 1976; Conger et al., 1984; Glenn & Weaver, 1979;
Lindblad—Goldberg, Dukes, & Lasely, 1988; Makosky, 1982; McAdoo, 1983;
McLanahan, 1983 and 1985; Nye, Carlson, and Garrett, 1970; Pearlin and
Johnson, 1977; Weinraub 8: Wolf, 1983). Some of the evidence ofl‘ered here to
support this relationship is weak, especially for the risk factors of ages and
number of children, education, and age at birth of first child (Andrews &
Withey, 1976; Conger et al., 1984; Lindblad-Goldberg, Dukes, & Lasely, 1988;
Nye, Carlson, & Garrett, 1970). However, the fact that many of these risk
factors are cited as stressors in the research suggests a common definition
of these risk factors as stressors for parents (McLanahan, 1983; Pittman,
Wright, & Lloyd, 1989; Zuravin, 1988).

Parental Social firppm't

In this section, studies examining the moderating efi‘ect of social
support on parental levels of stress are summarized. Social support is one
of the four characteristics of social networks; the other three are structure,
interaction, and quality. ”Its function is defined as the behavior of 1"
providing information and feedback, comfort and cheer, material aid,
advice, help in constructive problem solving, and so forth” (Heller &
Swindle, 1983, p 92). «

According to the Buffering Hypothesis, social support protects people
against the efi‘ects of stressful life events on the person’s well-being (Wilcox,
1981). However, this hypothesis found limited support in a study conducted
by Wilcox (1981) involving 500 people from a large Southwestern city.

18

Multiple regression analysis of the data revealed that social support, as
measured by the the types of support received from the social network,
explained only 12% of the variance of the interaction between Life Event
scores and the Langner Symptom Checklist. This finding suggests that
social support is somewhat important as a bufl'er for stressful life events,
but it does not explain the whole relationship. This could be a result of this
hypothesis’ failure to account for the individual difl‘erences in the sample’s
networks and utilization of the support ofi'ered (Heller & Swindle, 1983).

In order to examine the relationship between social support and
parental stress as reported in the literature, the following studies will be
grouped according to similarities in the characteristics of the sample. This
will allow for the examination of differences according to groups of
mothers. Those studies whose sample are low-income women will be
examined first, followed by studies whose focus are single-parents, and
finally, the studies examining teen parents.

The first study examined the mediating efi'ects of social support on
low income women and found that the level of support ofi‘ered by one’s
spouse or significant other was positively related to the mother’s level of
stress (Zur-Szpiro & Longfellow, 1982). Mothers’ parenting stress scores
were correlated with the type of suppert the husbands or significant others
provided. The types of support that were significantly correlated were child
care support, help with household tasks, and emotional support. These
three types of support were also significantly correlated to the mothers’
depression scores.

Utilizing data from the same study, Belle (1982) found that women
with large social networks did not experience less depressive symptoms or

have lower stress scores. They also found no relation to the amount of

19

contact the mothers had with a member of their social networks or their
proximity to them. Instead, these were significantly related to the mothers
' increased worry level about her friends and relatives. The author suggests
that for low income mothers, the support they receive from their network
members is both a relief and a burden. This “double-edged sword” may be
the reason for the non-significant findings between the correlation of social
_ support from family and friends with levels of stress.

The ”double-edged sword” is not just a characteristic of this one study
sample. In a paper examining the role of family social networks in
moderating the effects of familial stress, Unger and Powell (1980) cited
several studies that have found both positive and negative effects of social
networks on familial stress. The reason for this dichotomy is the reciprocal
nature of support among network members. Though the members received
needed support, they are also responsible for providing support to other
members of the network.

In a study conducted by Lindblad-Goldberg, Dukes, and Lasely (1988 ),
it was found that not all support or contact with family members have a
positive influence on the mothers’ level of stress. This study was mentioned
in the first section of this review. The sample included 126 black, low-
income, single-parents. The study found that the family stress score was
higher if the single-parent mother had more contacts with relatives and if
her support network members were younger (Lindblad-Goldberg, Dukes, &
Lasely, 1988).

In contrast to the findings supporting the “doubleoedged sword”
characteristic of social networksYa/study of low-income, single-parents _
found that the support of friends and relatives was significantly related to
the measures of well-being (Gladow & Ray, 1986). The study measured the

/.
/

2)

support received from friends, relatives, neighbors, and the community for
sample of 63, single-parent mothers. Total support and its two' subscales of
friend and relative support were negatively related to the severity of the
problems faced by these single-parent mothers‘/7A stepwise regression
revealed that the total support score explained/42% of the variance of total
problems, the study measure of the everyday stresses experienced by the
mothers.

One possible reason for this last study’s contrary findings is that the
support received fi'om those peeple who are similar to the receiver is more
efl’ective (Thoits, 1986). People who are similar have experienced similar
stressors and are able to ofl‘er more efi‘ective support in the form of problem-
solving advice and emotional support. This support can be ofi'ered by
family, friends, and co-workers.

r: " In another study of low-income, single-parent mothers, the social «
support provided by friends and co-workers was found to contribute to the X"
mothers’ level of well-being (D'Ercole, 1988). Eighty-three single mothers 1‘
from New York city were administered objective measures of level of stregg,
strain. and well-being. Social support of friends and co-workers was
negatively related to a global measure of psychological f symptoms
experienced by the mothers. Level of social support was found to bufi‘er
mothers against the stress of low SES. Those mothers with high levels of =
support fi'om fiiends and co-workers were lggsfllikely to experience
”financialstr-ain due to the stressor of low-income. ‘7 ,,

From the above studies of single-parents, it was found that those
women who received support from friends, family, and co-workers scored
higher on measures of well-being and lower on measures of stress. Even

though single-parent mothers seem to benefit from the support they receive

21

from their social networks, they are less likely to receive this support. This
is a result of their smaller social networks.

Two studies cited in section one found that single-parent families had
smaller social networks (McLanahan, 1983; Weinraub & Wolf, 1983). When
compared to married families, female-headed families were less likely to
know their neighbors and lived farther away from relatives (McLanahan,
1983). In addition, they were less likely to belong to social organizations and
to receive free help with household responsibilities.

In the other study, 14 single mothers experienced more stressful life
events, were more socially isolated, and had less support than the married
mothers (Weinraub & Wolf, 1983). Frequency of social contacts were less
for single-parent mothers compared to the married mother. Single
mothers also reported receiving less emotional support and tended to rate
their family and friends as less supportive. However, single mothers did
receive more support than married mothers in area of practical help with
child care.

Not only do older single-parents have smaller networks, but so do
teen mothers. Teenagers who become mothers are out of step with the rest
of their peers. They are experiencing parental stressors that most of their
peers and friends are not experiencing. Their network members are not
very characteristically similar to them, and they may even lose network
members due to these dissimilarities. However, these teens may now find
themselves more similar to their family members. This is because some of
the teens that find themselves pregnant are themselves children of teen
parents (Hayes, 1987). Therefore, it is not surprising to find that the teen
mothers received more effective support from family members than from
their peers.

22

In a study of teen mothers, a correlation between the level of overall
satisfaction and family support between the first and second measurement
was found (Unger & Wandersman, 1988). The majority of the mothers were
black (89%), single (93%), and lived at home with one or two of their parents
(74%). The teens were interviewed during their second trimester and again
when their child was eight months old. Family support, but not partner
support, was significantly related to fewer concerns over financial and
residential issues at both the prenatal and second interviews.

In another study, it was found that the greatest amount (30.75%) of
variance in the teen mothers’ emotional stress score was accounted for by
her total support scale (Colletta & Gregg, 1981). The sample consisted of
three groups of pregnant teens: 25 who were in a Baltimore City Public
Schools Infant-Parent program, 25 mothers who were in school, and 14
mothers who had dropped out of school. The people who ofi‘ered the social
support were not defined. Another finding of this study was that the level of
emotional stress experienced by the adolescent mother was negatively
related to level of social support for all three groups.

In a later reporting of the data from the same study, a significant
relationship between total support and emotional stress experienced by
adolescent mothers was found for the program and dropout mothers .
(Colletta & Lee, 1983). For drop-out mothers, the specific types of support
that were negatively related to overall emotional stress were living
arrangements, housework, and support during illness. For the program
mothers, care and living arrangements, peer group support, support
during illness, child care information, and support from individuals were
all significantly related to their overall emotional distress. The findings

support the hypothesis that adolescent mothers are not a homogeneous

23

group of mothers, but rather have different levels of support available to
them.

The articles reviewed in this section suggest that there are
difi'erences among the groups of mothers in the moderating effect of social
support on parental stress. Some of the studies provide support for the
“doubleedged sword” of social support received from one’s social network
with the mothers experiencing both positive and negative efi'ects (Belle,
1982; Lindblad-Goldberg, Dukes, & Lasely, 1988). However, other studies
found support for the idea that increased efi'ectiveness of social support is
because of the increased similarity of social network members (D’Ercole,
1988; Gladow & Ray, 1986; Zur-Szpiro & Longfellow, 1982). There were
differences among the teen mothers in terms of types of support received
(Colletta & Lee, 1983) and who gave the support (Unger 8r Wandersman,
1988).

Respite Child Care

In this section, the moderating effects of respite child care on
parental level of stress are examined. “Respite child care is child care 1
provided on a short-term basis for families who are experiencing stress due
to medical or emotional problems, or who are going through dificult
transitions and need relief” (Weinstein & Siegel, 1989, p 4). These services
can be offered for several hours or for a few weeks in the home of the family
or at a center.

Respite care ofi'ered at the Wisconsin Respite Center has been shown
to reduce parental levels of stress (Subramanian, 1985). This center serves
a high-risk population who are at-risk for child abuse. A pro-and post-test

24

methodology was utilized with a sample of thirty-six parents who had used
the center over a six month period. The study found that the overall stress
. level of the mothers decreased from Time I to Time II. The biggest
decrease was found in the subscale of Tension/Anxiety. All the other
subscales, Depression/Dejection, Anger/Hostility, and
Confirsion/Bewilderment, showed significant decreases. The only
exception was for Fatigue/Inertia, which increased. The increase in this
scale could be a result of the mothers accomplishing necessary tasks while
their child was in respite child care. /

A study by Subramsnian (1985) was the only empirical study
conducted on a respite child care center whose main purpose is preventing
child abuse of non-handicapped children. What is found in the literature
are summaries of the services ofi‘ered by respite child care programs that
serve a non-handicapped child population, but with no empirical data on
program outcomes (Allison, Kilgallon, & Reilly, 1988; Franz, 1980,

J anuary/February; Hansenfeld, Murphy, & Olson, 1981; O’Connor, Davis,
& Sahlien, 1984). The article by Franz (1980, January/February), explains
in more detail the Wisconsin Respite Center that was used as a data
collection site in Subramsnian’s (1985) article. The goals, services, and
problems of the center are discussed in detail and helpful hints are given
for those people who are considering developing a similar program.

Two respite care centers that serve children 5 years and younger
were discussed in two articles (Allison, Kilgallon, & Reilly, 1988; O’Connor,
Davis, 8: Sahlien, 1984). Parents Place Incorporated’s main goal is to
reduce parental stress caused by the demands of young children
(O’Connor, Davis, & Sahlien, 1984). Anecdotal examples were employed to
explain the techniques used by the stafi‘ of social workers and early

%

childhood professionals to reach the goals of the program. Children’s
Cottage is a 24 hour crisis nursery utilized by battered women who have left
their homes (Allison, Kilgallon, & Reilly, 1988). The children are cared for
by trained stafl', while the mothers deal with such crises as finding suitable
housing or a job.

Another article explains the programs of three respite care centers
in Michigan (Hansenfeld, Murphy, 8: Olson, 1981). These centers are the
Family Growth Center, Pooh’s Place and the Washtenaw County Centers.
The article discusses the child care program, level of parental involvement,
funding, and agency support for all three centers. Though no empirical
data were collected, the article does state that parental involvement in the
programs was varied.

Those articles that provide basic information about existing respite
care centers do so because the authors feel there is a need for this type of
programing. The day-care system in the United States does not provide
child care services to the population who needs them most, the low-income,
at-risk mothers (Straus, Gelles, & Steinmetz, 1980). What is needed are
respite child care centers that offer free, quality child-care. More empirical
studies involving these centers are needed to show that these centers are
making a difl'erence. For more information on the moderating efi‘ect of
respite child care on parental levels of stress, the only literature available is
in the area of respite care for handicapped children.

The literature on respite care for handicapped children is more
extensive than the literature on respite child care for non-handicapped
children. The respite care programs for both populations have similar
goals. Two main goals of respite care for handicapped children are to
reduce emotional costs to the family and to help prevent burnout that might

%

precipitate neglect, family disruptions, or institutional placement (Butler &
Friensen, 1988). These goals are similar to the goals of respite child care
for parents at-risk for abusing their children. Therefore, the findings of
studies utilizing parents of handicapped children could provide additional
insight into the moderating effects of respite child care on parental levels of
stress.

Mothers who used respite care experienced a decrease in their levels
of stress (Rimmerman, 1989). A pre- and post-test methodology was
utilized to compare 32 mothers who used home-based, respite care to 25
mothers who were non-users. The short form of the Questionnaire on
Resources and Stress was used to measure the maternal levels of stress. It
was found that the users of respite child care experienced a significant
reduction in their scores from Time I to Time II. The study found changes
over time in the level of stress and coping resources for both groups. These
changes could have been a result of the mothers dealing with the handicap
and learning new ways to cope. However, only these mothers who used the
home-based, respite care experienced a significant decrease in level of
stress.

In another study on respite care, families experienced a reduction in
their levels of stress and burdens of care (Wikler, Hanusa & Stoychefi',
1986). Two projects were conducted in which home-based respite care was
provided by college students for parents of children with handicaps. A pre—
and post-test methodology was utilized. The parents level of stress was
significantly reduced from Time I to Time II. Many of the parents also
experienced an increase in social relations. This increase was probably
caused by the increase in the parents’ free time, which they used to
cultivate new friendships.

27

This increase in the size of the parents’ social network is probably not
unusual. Because of the increased amount of care necessary for a
handicapped child, the parents often have smaller networks and higher
levels of stress compared to parents of non-handicapped children
(Salisbury, 1986). The decrease in the levels of stress may also be a result of
the increase in the parents’ social networks.

In 1983, it was found that the parents who used respite care
experienced emotional relief and an increase in their social activities (Joyce
. & Singer, 1983; Joyce, Singer, & Isralowitz, 1983). Twenty-four families
completed questionnaires designed to determine their perceptions of respite
care. The families indicated that the service was helpful in relieving the
emotional and physical strain associated with caring for a special needs
child. They also reported an increase in their social activities, which is a
direct result of the fi'ee time they had when the child was in respite care.

The literature in this area, though extensive, faces the problem of
little methodological sophistication and small sample sizes (Intagliata,
1986). Few empirical studies have been conducted and the results that are
reported are'about the satisfaction levels of parents regarding the program
(Fagan & Gurry, 1981: Frisbie & Slater, 1979; Stenson, 1989). The studies
that report parents’ satisfaction with respite care are discussed below.
Though these studies ofi‘er little pertinent information, the fact that parents
are satisfied with the services suggests a decrease in their level of parental
stress related to child care.

In a study conducted by Frisbie and Slater (1979), the majority of the
parents reported being very satisfied with the respite care services. The
study consisted of only 35 parents which represented a 20% return rate.
Sixty-one percent of the parents reported being very satisfied with the

8

emergency respite care service. The study also found that the parents were
very satisfied with the other respite care services: Family Friend (81%),
Foster (71%), and Home-Based (57%). Only 12% of the parents reported that
they were very dissatisfied with the Foster Respite Care. Though this study
offers some information, the small sample size, low response rate, and lack
of data on other measures decreases the importance of the findings.

'le0 other studies found the parents to be satisfied with respite care
services (Fagan & Gurry, 1981; Stenson, 1989). In a pilot study, ten parents
who utilized home-based respite care offered by Temporary Care Services,
Inc. found the service to be very satisfactory (Fagan & Gurry, 1981). The
parents mentioned that the program’s flexible hours and quality workers
were the most appreciated elements of the program. Saturday Afternoon
Free is a respite care program that is offered on weekends for families with
moderate to severely handicapped children (Stenson, 1989). The program
serves 15 children each semester. An evaluation conducted over two
semesters polled 30 parents. On a scale of one to five with five being the
highest, all of the parents rated the program a five. What the scale was
measuring was not mentioned in the artic1e.

Several studies have found that respite child care reduces parental
levels of stress (Wikler, Hanusa & Stoychefi', 1986; Joyce & Singer, 1983;
Joyce, Singer, 8; Isralowitz, 1983; Rimmerman, 1989; Subramsnian, 1985).
Many of the studies sufi'ered from poor methodology, small sample sizes,
and lack of additional variables. Three of the studies claimed respite child
care was efl‘ective because the parents expressed satisfaction in the services
(Fagan & Gurry, 1981; Frisbie & Slater, 1979; Stenson, 1989).

There is a need for more empirical studies that examine the

outcomes of respite child care. Of interest to this researcher was the role of

3

respite care as a moderator of the relationship between the risk factors
identified in section one and the mothers’ level of stress. In this study, the
_ usage of respite child care was defined as one measure of social support.
Therefore, the purpose of this study was to examine the influence of the
identified risk factors on the maternal levels of stress and social support

among those mothers who use respite child care.

1!)
Chapter 3

Methodology

Research Otrjectives and Questions
The research objectives and questions were stated in Chapter one .
' pages 2 and 3. These objectives and questions define the focus of the study:

Conceptual and W Definitions

In this section, the major variables of the study will be defined. The
major variables are the mother's level of stress and social support and the
identified risk factors. The format of this section will be: conceptual
definition first, followed by the operational definition for both the dependent

and independent variables.

We is the level of stress experienced by the
mothers as a result of their children's behavioral characteristics,

parental distress, and stresses derived fi-om the interaction between
parent and child (Touliatos, Perlmutter, & Straus, 1990).
Operationally, this was measured using the Parenting Stress Index
Short Form (PSI-SF) (Abidin, 1990) a questionnaire in which the

. mother responded to 36 items on a five-point Likert scale ranging
from strongly agree to strongly disagree (Appendix A, pages 95 to 97,
items 1- 36). The short form includes three subscales: Parental
Distress (PD), Parent-Child Dysfunctional Interaction (P-CDI), and
the Dificult Child (DC). The mothers recorded their responses
directly on the questionnaire. The researcher reverse scored all of
the responses to obtain subscale scores and then added all three
subscale scores to obtain the Total Stress Score.

W is the amount of instrumsntal and

functional support the mothers receive fi-om their social network and
formal social agencies. Operationally, this was measmd by using

31

the Maternal Social Support Index (MSSI) (Pascoe, Loda, Jefi'ries, &
Earp, 1981) and a rating of the mothers use of the respite child care
services at the Family Growth Centers (FGCs). The mothers
responded to the 21-item MSSI questionnaire by checking the
appropriate column or circling the appropriate answer (Appendix A,
pages 91 and 93, items 1-21). The index has two identified clusters of
items; Child Care Tasks (items 3, 8, and 9) and Non-Child Care
Tasks (items 1, 2, 4, 5, 6, 7, and 10). A score was obtained by adding
item weights and summing the results to obtain a total score. The
maximum score for the MSSI is 39. A high score indicates higher
support from the social network. The literature does not suggest that
a score be used as a cut-ofi‘ point between low and high social support
scores. A second variable of low and high maternal social support
was created by placing the scores into two categories: 1) low
maternal social support and 2) high maternal social support. High
support was defined as having a score above the mean for this
sample. The mean MSSI score for the sample was 21.6. This follows
the procedure used by Pascoe and Earp (1984) in their study to
determine the efi‘ect of mothers’ social support and life changes on
the the home environment of their children. A measure of the
maternal social support received fiom the FGCs was determined by
both the length of time and the average monthly usage of the respite
child care services. The mothers were asked to indicate the month
and the year they began using the respite care services and
approximately how many times a month they used these services
(Appendix A, page 87, items 5 and 6).

Winning is the total take home income of the family. It
was determined from the files at the Family Growth Centers. It was
coded by placing them into the following categories based on family
size: 0) above poverty threshold, 1) poverty threshold, 2) 20% below
poverty threshold, 3) 40% below poverty threshold, 4) 60% below
poverty threshold, and 5) 80% below poverty threshold. The poverty
threshold by family size is based on the results of the 1990 census and
is currently being used by the Michigan State Department of Social
Services to determine eligibility for services (Tellar, 1991).

32

Won is the total number of years completed in
school by the mother. The mothers were asked to list the last year of

school they completed (Appendix A, page 89, item 14). It was scored
by placing the number into the appropriate category: 1) 8, 2) 9, 3) 10,
4) 11, 5) 12, 6) 13, 7) 14, 8) 15, 9) 16, 10) 17 or more.

ThaNumbsLnflAdulmLinnainflmz The mothers were asked to
indicate the first names, ages, and relationship for all the people

living in their home. Those responses that fit the criterion of being
over the age of 18 years old were utilized. The number of adults was
used as the code for the response (Appendix A, page 90, item 17).

Wm: The mothers were
asked to indicate the first names, ages, and relationship for all the

people living in the home (Appendix A, page 90, item 17). These
responses that fit the criterion of being under the age of 18 were
utilized. The number of children was used as the code for the
response.

WWW: The mothers were asked
to indicate the age of each child (Appendix A, page 90, item 17). The
responses were then placed into the following categories:

1) Birth to 5 years, 2) 6 to 11 years, and 3) 12 to 17 years. The number
assigned to each response was used as the code for the response.

Wm: Each subjects’ race/ethnicity was determined from
the files at the Family Growth Centers. It was coded by placing them
into the following categories: 1) Caucasian, 2) Asian, 3) Black,

4) Hispanic, 5) Middle—Eastern. A second variable of minority status
was created by placing the scores into two categories: 1) Caucasian
and 2) minority. This second category, minority includes the race
categories of 2 (Asian) through 5 ( Middle Eastern).

We: The mothers were asked to indicate their marital
status (Appendix A, page 89, item 13). The response categories were:
1) Never married, 2) Married, 3) Separated, 4) Divorced,

33

5) Widowed. A second variable was created by placing the scores into
two categories: 0) Married and 1) Single.

We: Teen Parent status was determined from the
files at the Family Growth Centers. A teen parent status was
assigned to those mothers who were under the age of twenty when
they gave birth to their first child. It was coded by placing them into
the following categories: 1) under the age of 20 years and 2) 20 years
or older.

W: The mothers were asked to rank the
amount of space they have in their home on a 7-point Likert Scale

ranging fi-om not enough room (one) to more than enough room
(seven) (Appendix A, page 93, items 22). The number assigned to
each response was used as the code for the response.

Wanna: The mothers were asked to
circle all of the following amenities they have in their home

(Appendix A. page 94, items 23): 1) Running water, 2) Electricity,
3) Heat, 4) Telephone, 5) Refiigerator, 6) Stove/oven, 7) Radio,

8) Television, 9) Washer, and 10) Dryer. The total number of
amenities circled was used as the code for the total number of
amenities in the home. In order to add the ownership of a car to the
total number of amenities, a one was added to the total number of
amenities if the mother choose 2, 3, or 4 as her response for item ten
in the MSSI (Appendix A, page 91). Item ten reads: Who takes care
of car problems on short notice (if appropriate)? Answer one is no
car.

Instrumentation
EarsnfinLStresLIndexzfihomEom

One standardized instrument, the Parenting Stress Index Short
Form (PSI-SF) was used in this study to measure the variable of maternal

34

level of stress. The PSI-SF is based on the 101-item Parenting Stress Index
(PSI) developed by Richard R. Abidin in 1976. The PSI measures the stress
_ experienced by parents of children ten years of age and younger.
Originally, a letter was sent asking for the PSI and manual. However, the
PSI-SF was chosen because of pressures from the Family Growth Center
for a shorter questionnaire and a telephone conversation with Dr. Abidin
about the present research. It was during this conversation that Dr. Abidin
suggested that the short form of the PSI would work better with this
particular sample population of high-risk, possibly low-education mothers
because it would ofi'er the needed information to answer the proposed
research questions in a shorter questionnaire (Abidin, personal
communication, 11/9/90).

Since the PSI-SF was only recently developed (1990), independent data
on reliability and validity for this scale was unavailable. However, the
items that comprise the 36-item, short form are original items from the 101-
item PSI. The concurrent validity between the two scales is shown in the
high correlation of .94 for the total stress scores of the PSI and the PSI-SF
(Abidin, 1990). Two of the three subscale scores of the PSLSF are highly
correlated with the domain scores of the PSI subscales. The Parental
Distress and Dificult Child Subscales of the PSI-SF have Pearson
correlations of .92 and .87 with the analogous areas of the PSI (Abidin,
1990). The third subscale of the PSI-SF, the Parent-Child Dysfunctional
Interaction subscale, has a lower correlation with both the Parent Domain
(r: .50) and the Child Domain (rs .73) on the PSI (Abidin. 1990).

W

The maternal level of social support was assessed by using the
Maternal Social Support Index (MSSI). (Pascoe, Leda, J efi'ries, & Earp,
1981). It was selemd because it specifically measures subjective and
objective aspects of maternal support in everyday activities. This is
accomplished through a 21-item questionnaire assessing seven areas:

1) help with daily tasks, 2) satisfaction with visits fi'om kin or relatives,
3) help with crisis, 4) emergency child care, 5) satisfaction with
communication fi'om a male support figure, 6) satisfaction with
communication fi-om another adult, and 7) community involvement
(Pascoe et al., 1987). There are several studies that determine the validity
and reliability of the MSSI with difi'erent types of samples.

The MSSI has been used to explain the variance measured by the
Csldwell's Inventory of Home Stimulation (IHS) in a study involving
parents of infants who had been in a Neonatal Intensive Care Unit in North
Carolina (Pascoe, Loda, Jefi‘ries, & Earp, 1981). The sample was
predominantly rural and working class; 59% of the mothers were black and
41% were white. The average level of education for the sample was tenth
grade. The MSSI was found to explain between 12% and 22% of the
variance on three of the IHS subscales. In the same study, it was found
that these mothers who perceived more social support scored higher on the
IHS scale regardless of the amount of life change experienced by them as
assessed by the Rahe’s Schedule of Recent Life Events (Pascoe & Earp,
1984). In another study involving 219 predominantly white women, the
MSSI was slightly correlamd with the quality of child rearing by the mother
when the infant was four months of age (rs .30) and the mother's score on
the Cranley's Maternal/Fetal Attachment Scale (r- .30) (French, 1990).

3

MSSI scores were found to be predictive of child maltreatment and
low birth weight infants (Pascoe et al., 1987; Pascoe, Walsh-Clifi'ord, &
Earp, 1982). In a study to determine construct validity of the MSSI, the
scores for two groups of mothers, protective service referred and non-
protective service referred, were compared. The means for the protective-
service mothers was 7.4 and for the non-protective-service mothers the
mean was 10.5 (Pascoe, Walsh-Clifi'ord, 8: Earp, 1982). The difi‘erence
between the means was significant. In another study involving 222
mothers fi-om a depressed area in Toledo, Ohio, there was a significant
association between certain items on the MSSI and low birth weight (Pascoe
et al., 1987). It was found that these mothers who did not receive help with
grocery shopping, fixing things, and paying bills were more likely to give
birth to a low birth weight infant.

The MSSI was found to have acceptable test-retest reliability in two
studies. In the first study, one hundred mothers in a mdiatric clinic
subgroup were retested six to eight weeks after the initial interview; the _
test-retest reliability coeficient was .72 (Pascoe et al., 1988). Another study
examining the social support of primiparous mothers found a correlation of
.75 for retest scores after a nine month period (Pascoe & French, 1990). This
same study found an inverse relationship between the scores on the Center
for Epidemiologic Studies Depression Scale and MSSI at both six weeks and
nine months postpartum.

The internal consistency of the two clusters, Child Care Tasks and
Non-Child Care Tasks, was determined in a study with 488 mothers fi-orn
three clinics: prenatal, pediatric, and psychology (Pascoe et al., 1988). A
Cronbach’s Alpha was calculated for all three subgroups and a range of .58
to .72 was found for the Child Care Cluster and .49 to .78 for the Non-Child

37

Care Cluster. The authors suggest that the low coeflicients are probably
due to the small number of items in each cluster.

E'lGIlGIQI"

The Family Growth Center Questionnaire is a questionnaire created
by the researcher to assess the mothers’ usage of the respite child care
services and to obtain additional demographic information not available in
the mothers’ files at the FGCs (Appendix A, pp 86-90). The seventeen-item
- questionnaire is divided into two parts. The first part consists of eleven
items that ask the mothers how they learned about the FGCs and its
services, when they began to use the respite child care, what they do when
their children are at the FGCs, and if they have ever used any of the other
services ofi'ered by the FGCs. The purpose of these eleven questions is to
obtain information for the FGCs and to ensure eligibility of the mothers by
determining that they are not a member of either the FGCs’ support groups
or parent classes. The second part of the questionnaire obtains needed
demographic information on the mothers and members of their household
that cannot be found in the FGCs files. This part consists of six items.

During the pretest, the mothers had no trouble completing this .
questionnaire. After the pretest was” conducted, an additional question was
added to bring the total items to seventeen. This item asks the mothers how
they learned about the services at the FGCs. The services were listed in
column A and five possible sources including “other” were liswd under
column B. The mothers were asked to match the services in column A to
the sources in column B. Although the structure of matching was chosen
to simplify the answering of the item for the mothers, it was often
incorrectly completed by the mothers during the study. Many of the

38

mothers underlined services in column A and sources in column B without
drawing a line to connect the two. This item was added because of a request
by the stafi‘ at the FGCs; therefore, incorrect or incomplete responses will
not sfl'ect the ability to answer the proposed research questions.

SampleDesisn

A non-random sampling procedure was used in this study. The
sample for this study was drawn from those mothers who used only the
respite child care service at three Family Growth Centers in a midwestem
town fi-om February 11, 1991 to May 10, 1991. At first, one of the centers was
not included in the sample frame because of the large number of foreign
parents fi'om a nearby university who use the facility. It was felt that the
inclusion of the foreign mothers, who have difl'erent levels of assimilation
into our culture, could affect the results of the study. However, due to the
small number of mothers using the other two centers, the third center was
added. Separate analyses were condumd with the data fiom the foreign
mothers to determine if they were significantly difi'erent fi'om the other
mothers on the variables of interest.

All mothers who were using the respite child care at the three
centers fi'om February 11, 1991 to May 10, 1991 were asked by the researcher
to participate in the study. A list of eligible mothers was compiled fi-om the
files at the three FGCs. The criteria used to select the eligible mothers
were: (1) the mothers were using only the respite child care services and
(2) were not presently members of any of the support groups or attending
parent education classes offered at the FGCs. Because this sample fi'ame
consisted of approximately 90 mothers, every efi‘ort was made by the
researcher to contact all eligible mothers to ask them to participate. Due to

19

the non—random sampling design of the study, the generalizability of the L
results of this study will be limited to the mothers who use respite care at
the three Family Growth Centers.

Techniques of Data Collection

A meeting between the researcher, the director, and the two
receptionists at the Family Growth Centers was held prior to data collection
to inform the stafi' about the study and provide instruction on recruiting
mothers for the study. At this meeting, a time and place for a pro-test of the
questionnaires was determined. A pro-test was conducted to determine any
possible problems with the selected instruments and the researchers’
original questionnaire. One of the existing support groups, with no
mothers who were eligible for this study, was chosen to complete the
questionnaires and to answer questions about the ease of completing and
understanding the questionnaires (Appendix B, pp 104-105).

Data were collected on the mothers fi-om February 11, 1991 until May
10, 1991. Posters advertising the study were displayed at the centers to
ensure that every efi'ort was made to reach eligible mothers (Appendix B,

p 106). The receptionists asked the mothers if they would like to participate
in a research study. An incentive of one session of guaranteed child care
was ofi’ered. The mothers were asked to complete the questionnaires while
at the centers.

The packet given to the mothers included a letter fi'om the stafi' at the
Family Growth Centers introducing the researcher to the mothers, a letter
from the researcher explaining the study, a consent form, a release form
that allows the researcher access to the participants' files at the FGCs, the
Parenting Stress Index-Short Form, the Maternal Social Support Index, the

._\_.

40

Family Growth Center Questionnaire, and a 3x5 index card (Appendix B,
pp 98-102 for letters, consent form, release form, and drawings of the
reserved care coupon and the 3x5 card). An identification number was
placed on all the questionnaires and the 3x5 card. The index card had a
place for the parents to sign their names and to request a summary of the
research findings. The completed index card was then placed by the
participants in a locked file card box in order to preserve their
confidentiality and to allow the researcher to determine which mothers
participated in the study when additional information was needed fi-om the
mothers’ files at the FGCs. The outside of the questionnaire envelope had a
checklist of items that the mother were asked to check to ensure the
completeness of the questionnaires (Appendix B, p 103). The envelopes
were sealed by the mothers and placed in the box.

The researcher attended each respite care session and personally
talked to every eligible mother present. It was at this time that the mothers
were given a flyer (Appendix B, p 107) explaining the study and were told
about the child care coupon. The researcher thoroughly discussed the
questionnaire packet with each mother. If the mothers decided to
participate, they were asked to sign out a questionnaire packet and were
asked if the researcher could contact them by phone or mail, as a reminder,
if the packet was not returned in two to three weeks. It was discovemd that
the mothers preferred to take home the packet instead of filling it out at the
FGCs, hence the sign-out policy.

In order to reach all the eligible mothers who had used the centers
during the period January 1 to March 14, 1991 but not since then, a greeting
card was designed inviting those mothers to try the FGCs again, and to tell
them about the study and the reserved care coupon (Appendix B, p 108).

41

These cards were mailed to twenty-one mothers who had used the FGCs
since January 1, 1991 and to twenty-eight mothers who had used the FGCs
in the Fall of 1990 but had not been back in 1991. Of the forty-nine cards that
were mailed, two came back with an incorrect addresses and a no-
forwarding address stamp and three were brought in by mothers who
asked to participate in the study.

At the end of data collection, there was a total of 89 mothers who were .
eligible. Of these 89, only six mothers were not contacted either in person or
by a greeting card. This represents a contact rate of 93% of all eligible
mothers. Sixty-four mothers took a questionnaire packet home. Of these 64
mothers, three mothers declined to complete the questionnaire. One
mother felt the questions were too personal, another had just returned fi'om
the Gulf War and did not feel she had the time, and the third did not want to
complete it for undisclosed reasons. For the other mothers who took a
questionnaire home but failed to return them completed, an efi‘ort was
made to contact them at least once, but no more than three times. Fifty
mothers returned completed questionnaires. This translates to a response
rate for all eligible mothers of 56%, with 78% of all mothers who took a

questionnaire home returning it completed.

Statistical Anubis
The data were prepared, cleaned, and entered into the Macintosh

statistical program, StatView II (v 1.02, Abacus Concepts, Inc.). In this
stage, missing data and unclear responses were examined and decisions
on coding them were made. Descriptive statistics of frequency, mean,
mode, and variance were used to describe the sample. Zero-order

correlations were utilized to answer all three research questions. Trends at

42

.20 to .01 are significant for this exploratory study. The research questions

are:

1. What is the relationship between the type and/or number of
identified risk factors the mothers had and their level of
stress? These risk factors are: (a) low level of income, (1)) low
level of education, (c) high number of children, (d) low number
of adults residing at home, (e) high number of young children
residing at home, (0 minority race status, (g) non-marital
status, (h) teen parenthood status, (i) inadequate space in the
home, and 6) low number of amenities in the home.

2. What is the relationship between the type and/or number of
identified risk factors the mothers had and their level of social
support? These risk factors are: (a) low level of income, (b) low
level of education, (c) high number of children, ((1) low number
of adults residing at home, (e) high number of young children
residing at home, (i) minority race status, (g) non-marital
status, and (h) teen parenthood status?

3. What is the relationship between the levels of social support
and the levels of stress?

IimitationsottheSamplingDesignandDataCollectionTechniques

The relatively small sample size of fifty mothers does afl‘ect the
statistical analysis. With such a small N, power to detect relationships
between variables is limited. Even though the sample size is small, it
represents 56% of all eligible mothers. This is actually an acceptable
response rate for social science research and even exceptional for the high-

risk population of mothers who were sampled for this study. The

43

researcher is confident that every efi‘ort was made to ensure that all the
mothers who were eligible for the study were included in the sample.

The inclusion of those mothers who are foreign students or married
to a student may afi‘ect the results of the study. These mothers may score
difi'erently on the measures of level of stress and level of social support as a
result of their difi‘erent cultures and beliefs as well as their varying levels of
acculturation into American culture. Also, many of these mothers live in
university housing, where many amenities are provided and the space is
only slightly more than adequate. Therefore, it is expected that they will
score difi'erently on the PSI-SF and the MSSI and that they will have
difi'erent types of risk factors. The researcher separated the foreign
mothers from American mothers in the analysis to determine any
differences and to answer the research questions. The results will be
discussed in Chapter 4.

The use of an instrument, the Parenting Stress Index-Short Form,
that has no prior use in research and no independent data on reliability and
validity may afi‘ect the quality of the data collected. However, this shortened
form of the PSI is highly correlated with the PSI and is expected by its
author to have similar reliability and validity characteristics.

44
Chapter 4

Results of Statistical Analysis

Introduction

The results of the statistical analysis of the data will be discussed in
this chapter. A description of the sample is given first to provide the
necessary information on the demographic characteristics of the sample.
The sample was then split into two groups, American and foreign mothers,
and Chi-Square and AN OVAs were run to determine the differences
between the two groups. Results pertaining to each of the research
questions in chapter three will be discussed in the order in which the
questions were presented. Each research question will be discussed for
both the American and foreign mothers.

Cleaning of the Data
The sample consisted of fifty mothers who used respite child care

during the period of February 11, 1991 to May 10, 1991. Two mothers were
dropped fi'om the study due to missing data on the Parenting Stress Index-
Short Form (PSI-SF). One mother failed to complete the first page of the
PSI-SF and the other mother did not respond to 5 of the 36 items or 14% of
”the questionnaire.

There were five other mothers who did not respond to one item and
one mother who did not respond to two items on the PSI-SF. A statistical
consultant suggested the missing datum for each item be replaced with the
mean value of the item for all the participants. This method was chosen

45

because the added values did not affect the placement of the PSI-SF score in
the continuum of all PSI-SF scores.

All missing values for the risk factors, such as income and number
of amenities, rating of the usage of respite child care service, such as
length of time and average times a month, and items in the Maternal Social
Support Index that were not utilized in the scoring of the instrument were
coded as missing data.

Reliability of the Imtruments

Cronbach’s Alpha using Mean Inter-item Correlation (p) was
utiliud to determine the reliability of the PSI-SF with this study’s sample.
The equation of Cronbach’s Alpha is:

(1: ENE
[1+ p(N- 1)]

where N is the total number of items in the instrument, 5 is the Mean
Inter-item Correlation, and a is the reliability coeficient (Carmines &
Zeller, 1979). The range for a is 0 to 1 with a higher (1 indicating higher
reliability. The Mean Inter-item Correlation is defined as

where n is the total number of items in the instrument, r is the correlation,
N is the total number of items in the correlation matrix, and l5 is the mean

of all the items in the correlation matrix.

46

The Cronbach’s Alpha for the PSI-SF for this sample was .90. The
reliability for each of the three subscales was also computed using the
Cronbach’s Alpha. The reliability coefl'rcients for the subscales: Parental
Distress, Parent-Child Dysfirnctional Interaction, and Dificult Child were
.87, .88, and .78 respectively. The reliabilities of the PSI-SF total score and
each of the three subscales for this sample are acceptable.

The reliability for the MSSI was not computed due to the difi‘erent
levels of measurement used in the instrument. The M881 utilizes nominal,
ordinal, and ratio measures. Because the Cronbach’s Alpha determines
how reliable maple are in answering the questions, the levels of
measurement need to be similar in order for the reliability coeficient to be
an accurate measure of the index’s reliability. In the literature, only test-
retest reliabilities are listed for this instrument (Pascoe et al., 1988; Pascoe
& French, 1990).

However, a Cronbach’s Alpha was computed for the two clusters:
Child Care Tasks and Non-Child Care Tasks. The reliability coemcients for
the clusters were .60 for the Child Care Tasks and .45 for the the Non-Child
Care Tasks. Though these are low, the alpha for the Child Care Cluster is
in the range reported in the literature and the alpha for the N on-Child Care
Cluster is slightly less than the range reported (Pascoe et al., 1988).
Possible reasons for these low correlations may be due to the small sample
size and very small number of items in the correlation matrix. Only three
items comprise the Child Care Tasks cluster and seven items comprise the

second cluster.

47

Description of the Sample

The sample consisted of 48 mothers who used respite child care
during the spring of 1991. The mean age for the mothers was 30.5 years
(SD = 5.29) with a range of 20 to 42 years. The sample was highly educated
with a mean of 14.4 years/(SD = 2.45) and a mode of 16 to 17+ years. As
illustrated in Figure 4-1, the sample represents an array of ethnic groups
and races, with a majority of the women being Caucasian. The sample is
mostly poor, with 59.4% of the women classified as being below the poverty
. threshold (Table 4-1). A large portion (75%) of the women in the study are
married and the same percentage of women cite two adults in the
household. 'leenty-three percent of the mothers listed themselves as the
only adult in the home.

a Caucasian 58.3%

In Asian 16.7%

Black 10.4%

I Hispanic 6.25%

I Middle Eastern 8.3%

 

Figure 4-1. Description of the study sample by ethnicity and race.

48

The average number of children for this sample was 2.2 (SD = 1.2),
with 75% of the mothers having either one or two children. All/of the
mothers had children who were 5 years or younger, with 50% of the
mothers having one child and 39.6% of the mothers having two children in
this age category. Men mothers had children between the ages of six and
eleven years and only thme mothers had children who were thiMen years

or older.

Table 4-1. Frequency distribution of income for the study sample.

Gatorade: Humbeufmpls fiercest
Non-poverty 15 4054
Poverty threshold 4 10.81
20% below poverty 2 5.41
40% below poverty 10 27.03
60% below poverty 5 13.51
80% below poverty 1 2.70

The study asked the mothers about the amount of space and the -
number of amenities in their home. Forty-eight percent of the mothers
classified the amount of space in their home as below the “just enough”
category on a seven point Likert scale. All of the mothers have running
water, electricity, heat, a refiigerator, and a stove/oven. Three mother did
not have telephones and seven women did not have cars. Slightly more
than half of the mothers did not have a washer or dryer.

49

In determining the mothers’ usage of the respite child care service, it
was found that 50% of the mothers have been using the center for less than
one year. The average length of time the mothers have been using the
service is between seven and nine months. Forty-one percent of the
mothers use the centers an average of four times a month and 30.1% use
the centers only 1 to 3 times a month.

The mothers’ levels of social support and stress were measured.

The sample’s mean score on the MSSI was 21.6 (SD = 6.5) with a range of 7
to 34. The maximum score on this instrument is 39. It was found that the
mothers were experiencing high levels of stress. The mean for the Total
Stress score was 81.4 (SD = 16.8) with a range of 42 to 118. A score of 90 or
above represents a mother who is experiencing clinically high level of
stress. Thirteen mothers in this sample had scores of 90 or above.

A mother in this sample is most likely to be Caucasian, married, and
living below the poverty threshold. She has a high school diploma and has
possibly completed a year or two of college. She has two children with at
least one who is 5 years or younger. In her home, she has all the amenities
except for a washer and dryer and she has slightly less than just enough
space for her family. She is experiencing a high level stress and a
moderate level of social support.

Diflermrces between the American and Foreign Mothers

The inclusion of the foreign mothers who use the third center
introduwd the possibility that there were significant difl‘erences between
these mothers and the American mothers in the sample. Therefore, the
sub-groups were compared to determine any difi'erences. The sample of 48
mothers was split into two groups: American and foreign. Chi-square and

50

AN OVAs were run to determine any significant difi'erences between the
two groups.

The American sub-sample consisted of 32 mothers and the foreign
sub-sample had 16 mothers. Between the two groups, there were eleven
significant difi‘erences below the .05 level and three between the .10 and .05
probability level. They are liswd in Table 4-2. The foreign mothers tended to
have lower social support scores. A difi‘erence of 3.3 points exists between
the means of the two groups. The foreign mothers average score was 19.5.
However, these mothers were also likely to have higher scores on the two
MSSI clusters: Child Care Tasks and N on-Child Care Tasks. The average
score was 1.9 for the Child Care Tasks cluster and 4.0 for the Non-Child
Care Tasks. The mean difi'erences were .781 and .969, respectively.

All sixteen foreign mothers were married compared to only 62.5% of
the American mothers. The foreign mothers all had one other adult
besides themselves in the household. Only 21 or 62.5% of' the Americans
listed two adults in the home. None of the foreign mothers had children
before the age of twenty, compared to seven American mothers who were
teen parents.

There are difi'erences in the number and ages of the children between
less two groups of mothers. Based on the mean difi'erences between the
/:o sub-groups, the American mothers had on the average one more child
( than the foreign mothers and .344 more children in the age group of five

l

l

\,\ years or younger.
Besides difi'erences in the number of people in the home, there are
difi‘erences in the number of amenities and amount of space. It was found
that the foreign mothers have fewer amenities than the American mothers.
Also, the foreign mothers tended to rate their amount of space as not being

51

.8. v a:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

no. v a .

3. v a I.
.24. SE a 8.3
re 8 o 5%
SE 2..” 8d :3.
13d 3 a: a:
83:5
as a
tame EN 84 new cote—Eu
no a
tame m3 m3 2: 8:8ch
| .8 a
:43. alone Br 88 3.33 e e
+53 8.3 3.2 ES 938
an:
In}. new 8... 8.» 2:3 28
6:50-:0:
am:
:Se «3 x: m: sacs
«.30 2mg“.
.mmz
.....§.om 5: 3.x lemme secs:
In; 3.3 8.3 q: 558:3
tame Re are 8e 82: e
unnoE<
38,—. m seesaw ~38. system "83.88471. :33. ammouom canton}. $33.29.
.30 3:8qu aces:

 

6.85:2 sunfish e5 deer—23w 23 8253 80:98me

.er 03am.

 

 

52

enough for their family. There was a mean difi’erence of 1.25 between the
sub-groups in the variable of amount of space.

The foreign mothers on the whole, were more educated. They
received 1.3 more years of schooling then the American mothers. Whereas
the American mothers had an average level of education of 14 years, the
foreign mothers went to school an average of 15.3 years.

Fifty percent of the foreign mothers are classified as Asian, 25% as
Middle Eastern, 12.5% as black and 12.5% as Caucasian. This compares to
the American sample which is 81.2% Caucasian, 9.3% Black, and 9.3%
Hispanic.

There was a difference in the length of time that the two groups of
mothers have used the respite child care service. The American mothers
were likely to have used the center longeruan average of eight months
longer. The average for the American mothers was 16 months. There
were no difi‘erences in the number of times a month the mothers used the
centers. Both groups’ average was 4.4 times a month.

The foreign mothers in this sample had higher levels of education
when compared to the American mothers. All of the foreign mothers were
married and tended to have fewer children. These mothers tended to have
less space and fewer amenities in their homes. Though they receive more
help with every day tasks including child care, they have lower social
support scores.

As a result of the findings of significant difi‘erences between the
American and foreign mothers, the sample was split into two sub-groups
and separate analyses were done for each group. The findings of the study
are discussed by the research questions posed m Chapter 3 and by the two

sub-groups of American and foreign mothers.

Results of the Stark

W: What is the relationship between the type and/or number of
identified risk factors the mothers had and their level of stress? Theisfloflrisk ..
factors are: (a) low level of mcome, (b) low level of education, (c) high
number of children, (d) low number of adults residing at home, (e) high
number of young children residing at home, (0 minority race status,

(g) non-marital status, (h) teen parenthood status, (i) inadequate space in
the home, and 6) low number of amenities in the home.

Zero-order correlations were used to answer this question. The
identified risk factors were correlated with the mothers’ scores on the PSI-
SF (Table 4.3). The results of the correlational analyses are discussed for
the American sub-group first followed by the foreign sub-group.

TW‘
American Sub-group

The results mostly show only slight relationships between the risk
factors and the mothers’ level of stress for the American sub-group. A
positive relationship was found for the risksfgoooréofi ngzrber of children and
the mothers’ scores on the PSI-SF (r- — 27; p <. 20). The m thers with more
children had higher levels of stress. Another positive relationship was
found for the numberyd of adults m the home and the mothers' PSI-SF scores
(r = .1;? p < ..20) Las wyfmoihers with fewer amenities were experiencing
higher levels of stress (r = .21; p < .20).

The most significant finding for this sub-group was that the number
of children the mother had that were five years or younger was related to
the mothers' scores on the PSI-SF (r = .45; p < .05). The more young

children the mother had, the higher her score on the PSI-SF.

54

Table 4-3. Correlations between the Risk Factors and the Mothers’ PSI-
SF Scores by Sub-groups and Total Sample.

Income - .016 -.26+ -.087
Education -.081 -.23+ -.099
# of children .27+ -.34+ .13-t-
# of adults ’ .17+ —- .15+
# of children
s 5 years .45" -.36+ .21+
Minority Status .14 .27+ .18+
Single-parent
- Status -.O76 -- -.086
Teen parent Status .066 - .029
Amount of space .038 .053 .011
# of amenities .21+ .17 .15+
""" + '52?2'o"""""""""'-"'"""""""""'"
"' p < .05

There were no significant findings for the following risk factors: income,

mmofity status, single-parent status, teen-parent status, and

amount of space.

Foreign Sub-group h;

Correlations were run for the risk factors and the mothers’ scores on
the PSI-SF for the foreign sub-group. There were three risk factors: single-
parent, number of adults, and teen parent, which did not apply for this sub-
group. The mothers were all married; therefore there are two adults in
every home and none of the foreign mothers were under the age of twenty at
the birth of their first child.

Two of the risk factors had unexpected correlations with the stress

scores at the .20 level. Both number of children and nomber of children
6‘. 7”,. L455, / WI} {QM
Ni 0 (:1 C27

55

who are S 5 years of age were negatively related to the foreign mothers’
scores on the PSI-SF (rs = -.34 and -.36, respectively). For these mothers,
. the more children they had, the lower their stress scores.

All other risk factors were related to the mothers’ stress scores and
were in the emcted direction. These risk factors are: income (r = -.26;
p < .20), education (r = .24; p < .20), and minority status (r = .27; p < .20).
Foreign mothers who were either poorer or less educated had higher stress
scores. The fourteen foreign mothers who classified themselves as
minorities had higher PSI-SF scores. The amount of, spgogrggd number of

 

amenth ggmfiwfly related to the foreign mothers’ levels of

stress.

Summary
The data indicate that all but one of the significant correlations are at

the .20 level for both sub-groups. For the American sub-group, the number
of children, adults, and amenities were all positively related to the mothers’
levels of stress. Those American mothers who had more children, more
adults and more amenities in the home had higher stress scores. It is
surprising to find that the American mothers who were single—parents
were experiencing lower levels of stress. The most significant finding for
the American mothers was that the number of children they had that were
five years of age or younger was positively related to their levels of stress
(r s .445; p < .05).

For the foreign mothers sub-group, all significant correlations were
at the .20 level. These risk factors that were negatively related to the foreign
mothers’ PSI-SF scores were income, education, number of children, and

number of children five years or younger. The only risk factor that was

56

positively related to the mothers’ levels of stress was minority status.
Number of adults, single-parent status, and teen parent status did not
apply to this sub-group, since all the mothers were married and none were

teen parents.

W: What is the relationship between the type and/or number of
. identified risk factors the mothers had and their level of social support?
These risk factors are: (a) low level of income, (b) low level of education,
(c) high number of children, (d) low number of adults residing at home,
(e) high number of young children residing at home, (0 minority race
status, (g) non-marital status, and (h) teen parenthood status?
Correlations] analyses of the risk factors and the mothers’ scores on
the MSSI were used to answer the second research question. The results of
the analyses are discussed by sub-groups beginning with the American
mothers, followed by the foreign mothers. The results are liswd in Table
4-4.

American Sub-group

A significant finding of this study is that the American mothers with
lower levels of social support were found to have lower levels of income
(r s -.56, p < .01). This is a negative relationship because the higher the
number assigned to the category of income, the farther they are from the
poverty threshold. Another negative relationship. was found for the risk
factor of single-parent status and the mothers’ scores on the MSSI. The
single-parent mothers in this subgroup had lower levels of support
compared to the married mothers (r = -.36; p < .05).

57

Table 4-4. Correlations between the Risk Factors and the Mothers’ MSSI
Scores by Sub-groups and Total Sample.

 

 

Income - .56" -.13 -.47**
Education .27+ .36+ .21
if of children .031 -.32+ .079
# of adults .044 -- -.01
# of children
s 5 years -.081 .009 .001
Minority status -.19+ .17 -.23
Single-parent
Status -.36* -- -.18
Teen parent Status -.20+ -- -.088
+p<20
* p < .05
*" p < .01

The other significant findings for the American sub—group were all
at the .20 level. There was a positive relationship between the risk factor of
education and the mothers’ scores on the MSSI (r a: .27). The mothers with
the fewest years of schooling had lower levels of social support. A mother
who listed herself as a minority or was a teen parent tended to have lower
levels of social support (rs = -.19 and -.20, respectively). the number of
adults or children in the home regardless of their ages were not
significantly related to the American mothers’ levels of social support.

Foreign Sub-group

The foreign mothers were similar to the American mothers on the
risk factor of education (r = .36; p < .20). The scatter graph for level of
education and mothers’ scores on the MSSI shows a curvilinear

58

relationship, as illustrated in Figure 4-2. Those mothers with low or high
levels of education had higher levels of social support compared to those
mothers with moderate levels of education. For this sample, moderate

levels of education is approximately 13 to 16 years of schooling.

0000

A

L

MSSSITOTALSCORE
SEKSEBE§888
O

 

 

0
1'2 125 T4 1'5 1'6 1'7 133
EDUCATION

H
H

Figure 4-2. Scatterplot of the relationship between the MSSI scores and
level of education for the foreign subogroup.

The only other significant finding for this sub-group was the negative
relationship between the foreign mothers’ scores on the MSSI and the risk
factor, number of children. The foreign mothers who had more children
had lower social support scores (r = -.36; p < .20). Income, number of
children five years or younger, and minority status were not significantly
related to the foreign mothers’ levels of social support. The risk factors;
number of adults, single-parent status, and teen parent status did not apply

59

to this sub-group, since all the mothers were married and none were teen

parents.

Summary

The results of this study indicate that the majority of the risk factors
that were relawd to the mothers’ score on the MSSI for both the American
and foreign sub-groups were at the .20 level. Education, minority status,
and teen parent status were significantly related to the American mothers’
levels of social support (rs = .27, -.19, and -.36, respectively). For the foreign
mothers, the risk factors of education and number of children were
significantly related to their levels of social support (rs = .36 and -.32,
respectively).

There were two risk factors that were highly related to the American
mothers’ scores on the MSSI. They are single-parent status (r = -.36;
p < .05) and income (r = -.56; p < .01). The We risk factors: number of
adults, single-parent, and teen parent, did not apply for the foreign sub-

group, because there were no differences in these risk factors for them.

W: What is the relationship between the levels of social support
and the levels of stress? ’

Zero-order correlations were used to answer question three. The
mothers’ scores on the PSI-SF and its three sub-scales were correlated
with the mother’s scores on the MSSI, their scores on the MSSI clusters,
and two variables of mothers’ usage of the respite child care service;
number of months and times a month (Table 4-5). The results of the
correlations! analyses are discussed for the American sub-group first,
followed by the foreign sub—group.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8. v Am...
3. v Ar...

8. v .r
*3. +3. +3.- Km. wmor EN. +3.- ++mmz 5.82
a 385.
2. ma. 3.. vo. S.- Ec m7 $6. 3362 we a
“mo.- mfi. 8o. m8. +3".- Er Na: «8.- 00.202
Hmmz
+5.- H. 9.- v8.- 2: S.- +wmc m8.- 00 mag
Eb. +5. 30.- mmov ..omr ..er +3.- m8.- ~wm2
:mmfioh 58th< ammouoh swam-8:2 camps-m user—25‘ ago-m 5.33:5. fiommam
l r Emeom

Savanna 0Q oiomnsm HOD-m eigensm Gm ram-Em

 

 

 

 

 

anew-paw 5e: .8.“ 38.5 mo .33 was $5302 tandem 38m 23 5953 nEuqeflaEm 25. .m-v Snag.

 

American Sub-group

In this sub-group, the PSI-SF was positively related to the number of
times a month the mother used the respite child care service (r=.33; p<.10).
Those mothers who had higher stress scores used the service more times
on the average during the month.

There was a significant relationship between the mothers’ scores on
the PSI-SF subscale, Parental Distress and their scores on the MSSI
(r a -.42, p < .05). The lower the mother's level of social support, the higher
her level of distress as a parent. The Parental Distress subscale was
positively related to the number of times a month she used the service
(r = .25; p <.20). The more stress a mother was experiencing in her
parental role, the more often on average that she used the service.

The mothers scores on the Parent-Child Dysfunctional Interaction
(P-CDI) subscale were positively related to the number of times \a month the
mother used the respite child care service (r = .37; p < .05). Those mothers
who had a child that did not measure up to their idealized child or did not
positively reinforce them in their parenting role used the center more often
in a months’ time.

The mothers’ scores on the PSI-SF subscale, Dificult Child, was
positively related to the mothers’ scores on the MSSI and the number of
times a month the mothers used the service at the .20 level (rs :- .27 and .17,
respectively). A mother who was raising a child who was dimcult had a
higher level of social support and she used the service more often in a
months’ time than a mother who was not raising a dificult child.

Foreign Sub-group

The mothers’ scores on the PSI-SF were negatively related to three of
the measures of social support. Those mothers who had used the respite
child care services more times during a month’s time had lower levels of
stress (r = -.34; p < .20). The mothers who had higher levels of social
support as measured by the MSSI, had lower levels of stress (r s -.29;

p < .20 ). Lastly, those mothers who received more help with child care
tasks as measured by the MSSI Child Care Cluster had lower scores on the
PSI-SF (r = -.28; p < .20).

The mothers’ scores on the PSI-SF subscale, Parental Distress (PD)
was negatively related to their levels of social support as measured by the
MSSI (r a -.50, p < .05). Those mothers who scored higher on the PD
subscale received less help with household tasks as measured by the MSSI
Non-Child Care Cluster (r a -.34; p < .20). Those mothers who had lower
levels of social support and who received less help with household tasks had
higher levels of distress as a parent.

A negative relationship was also found for the mothers’ scores on the
subscale, Parent-Child Dysfimctional Interaction (P-CDI) and the number
of times a month the mothers’ used the respite child care service (r = -.32:

p < .20). Those mothers who had a child that did not measure up to their
idealized child or did not positively reinforce them in their parenting role
used the service less often than those mothers who were receiving positive
reinforcement in their parent role.

As with the American sub-group, there was a positive relationship
between the mothers’ scores on the subscale, Dificult Child (DC) and the
number of times a month the mothers used the service (r a .49; p < .05).

Mothers who rated their children as more difficult used the service more

w

often. The mothers’ scores on the DC were negatively related to. the
mothers’ scores on the MSSI Child Care Tasks Cluster (r = -.27; p < .20).
Those mothers who rated their children as more dificult received less help
with child care tasks than those mothers who rated their children as being

easy.

Summary

For the American sub-group, those mothers who used the respite
child care service more frequently were experiencing higher levels of stress
(r s .33; p < .20). The opposite relationship was found for the foreign sub-
group. Those foreign mothers who used the service more frequently were
experiencing lower levels of stress (r a -.34; p < .10). For only the foreign
mothers, there were significant relationships between the mothers’ scores
on the PSI-SF and her scores on the MSSI and the MSSI Child Care
Clusters (rs - -.29 and -.28, respectively both at the .20 level).

Another negative relationship was found between both sub-groups’
scores on the PSI-SF subscale, Parental Distress and their scores on the
MSSI. This finding was true for both sub-groups at the .05 level. American
mothers who were experiencing distress in their parental role were using
the center more mquently (r . .25; p < .20). The foreign mothers’ scores on
the PD were also negatively related to the mothers’ scores on the MSSI Non-
Child Care Cluster.

For both sub-groups, there was a significant relationship between the
mothers’ scores on the P-CDI Subscale and the number of times a month
they used the service. For the American mothers, the relationship was
positive and it was a negative relationship for the foreign mothers (r a .37;

p < .05 and r s -.32; p < .20, respectively).

64

The American mothers’ scores on the Dificult Child Subscale were
positively correlated with the mothers scores on the MSSI and the number
of times a month they used the service (rs = .27 and .17, respectively). Both
relationships were significant at the .20 level. The foreign mothers’ scores
on the DC were positively correlated with the number of times a month they
used the service (r = .49; p < .05). The foreign mothers’ scores on the DC
were negatively related to their scores on the MSSI Child Care Cluster
(r = -.27; p < .20).

Chapter Summary ,

Statistically significant difi'erences were found between the
Amefimmtheflfogign‘ mothers diagram-i: variables: MSSI total
score, MSSI Child Care Tasks cluster, MSSI Non-Child Care Task cluster,
number of adults, number of amenities, number of children, number of
children s five years of age, education, marital status, single-parent,
number of months using the respite child care, minority status, amount of
space, and men parent status. Thflothem were separated into two
groups and analyses were run to answer the research questions.

Statistically significant findings for this study between the type or
number of risk factors and the mothers’ score on the PSI-SF and MSSI
were found for both the American and foreign sub-groups. In answering
research question one, a positive relationship was found for the American
mothers’ scores on the PSI-SF and the following risk factors; number of
children, number of children who were five years of age or younger,
number of adults, and number of amenities. For the foreign mothers,
their scores on the PSI-SF were negatively related to the following risk

factors; income, education, number of children, and number of children

65

who are five years of age or younger. The risk factor of ethnidty/race was
positively relawd to the foreign mothers’ scores on the PSI-SF.

In answering research question two, it was found that the American
and foreign mothers’ scores on the MSSI was positively related to the risk
factor of education. Negative relationships were found between the
American mothers’ scores on the MSSI and the following risk factors;
single-parent status, teen parent status, ethnidty/race, and income. For
the foreign mothers, there were negative relationships between the
mothers’ scores on the MSSI and two risk factors; number of children,
and number of children five years of age or younger.

Research question three examined the relationship between the
mothers’ levels of stress and their levels of soda] support . For the
American mothers, the number of times a month the mother used the
service was positively related to the mothers’ scores on the PSI-SF and all
three of its subscales. The American mothers’ scores on the MSSI were
negatively related to the mothers’ scores on the Parental Distress subscale
and positively related to the mothers’ scores on the Dificult Child subscale.

For the foreign mothers, times a month they used the service. was
negatively related to the mothers’ scores on the PSI-SF and the Parent-
Child Dysfunctional Interaction subscale. The variable, times a month,

i was positively related to the foreign mothers’ scores on the Dificult Child
subscale. The foreign mothers’ scores on the MSSI were negatively related
to the mothers scores on the PSI-SF and the Parental Distress subscale.
There were negative relationships between the foreign mothers’ scores on
the MSSI Child Care Cluster and their scores on the PSI-SF and the
Dificult Child subscale. Lastly, a negative relationship between the

$

foreign mothers’ scores on the MSSI N on-Child Care Cluster and the
Parental Distress subscale was found.

67
Chapter 5

Discussion and Implications

Introduction

The main objectives of this study were: 1) to determine if the
identified risk factors were related to the sample's maternal level of stress,
2) to determine if the identified risk factors were related to the sample's
maternal level of soda] support, and 3) to determine if the maternal level of
soda] support was related to the sample's level of maternal stress. In this
chapter, the results of this study are discussed in terms of these research
objectives. This is followed by suggestions for future research.

Dlflsrences within the Sample

The sample was composed of two sub-groups; American mothers
and foreign mothers who were living in married, student housing at a
midwestern university. The American sub-group included 32 mothers and
the foreign sub-group consisted of 16 mothers.

Chi-square and AN OVAs revealed significant difi‘erences between
the two groups. The foreign mothers had lower levels of soda] support as
measured by the Maternal Sodal Support Index (MSSI). Even though the
foreign mothers scored lower on the MSSI, they scored higher on its two
clusters, Child Care Tasks and Non-Child Care tasks. The foreign mothers
received more help with everyday tasks, which may be a result of the fact
that all of the foreign mothers were married compared to only 20 of the
American mothers. Their lower scores on the MSSI are probably a result of
their isolation from relatives and lack of involvement in groups or

organizations.

E

The foreign mothers had higher levels of stress as measured by the
Parenting Stress Index-Short Form (PSI-SF). These mothers also scored
- higher on two of the three subscales: Parental Distress and Parent-Child
Interaction. “Parental Distress subscale (PD) measures the stress the
mother feels in her role as a parent as a function of personal factors that
are positively related to parenting" (Abidin, 1990, p. 19). One of these
personal factors is lack of soda] support. A possible explanation for their
higher scores on the PD is that the foreign mothers have younger children
and as such, are still making the transition into the parenting role. They
are making this transition without support from their extended family and
with little or no outside support in the form of groups or organizations.

The foreign mothers had fewer risk factors than the American
mothers. The few risk factors they had were: fewer amenities in the home,
lack of adequate space, more likely to be classified as a minority, and lower
levels of income. All of these foreign mothers live in married student
housing. Even though the small apartments have many amenities, they do
not have washers and dryers. The American mothers had more room in
their homes and possessed washers and dryers. All mothers in the sample
had running water, electridty, and kitchen appliances, and only three
American mothers had no telephones. The foreign mothers were most
likely to be Asian, Black, or Middle Eastern. The majority of the foreign
mothers had income levels between 40 to 80% below the poverty threshold,
whereas the majority of the American mothers had incomes that were
either at poverty threshold or above it.

Despite having a higher number of risk factors (six out of ten)- the
American mothers had lower levels of stress and higher levels of support.

The American mothers were more likely to be involved in outside

69

organizations and groups, frequently see relatives, and have been using the
respite child care services for a longer period of time. Though the majority
of the American mothers were married, twelve mothers were single-
parents and there were seven American teen mothers in the study. The
American mothers had lower levels of education, a higher number of
children, and more children who were S 5 years of age.

The research objectives are discuswd together for each subcsamme of
this study. When there are difi‘erences in the findings for the sub-groups,
'they will be discussed separately.

Disarssion of the Research Olfiectives
Wm: To determine if the identified risk factors were
related to the sample's maternal level of stress.

For the American sub-sample, the number of children s 5 years of
age was related to the amount of stress the mothers were experiendng.
The mothers with high levels of stress had more young children. This
finding supports the observations of Miller and Myers-Walls (1983) who feel
that mothers with young children experience more stress. It also supports
the definition of Pittman, Wright, and Llyod (1989) for this variable as a
stressor in their research. They defined the presence of preschoolers as a
relevant background factor that contributes to the parents’ levels of stress.

However, this result is not supportive of the results found by Glenn
and Weaver (1979). In their study, they found that the number of children
between the ages of 0-5 years was negatively related to the father’s well-
being but not to the mothers. One possible reason for the discrepancy is that
these researchers examined the relationship between the ages of children

and parents’ measure of well-being, whereas, this study examined the

7O

relationship between the ages of children and the mothers’ level of stress.
Well-being and stress are measuring two difi‘erent ends of a continuum and
a decrease in a person’s score of well-being is not indicative of an increase
in their level of stress.

Contrary to the findings for the American sub—group, the foreign
mothers who had more children between the ages of infancy through five
years had lower levels of stress. This finding is supportive of the results
found by Glenn and Weaver (1979). However, it is not supportive of the
literature in which this risk factor is defined as a stressor (Pittman,
Wright, & Lloyd, 1989).

The risk factors of low level of education and income were both
negatively related to the foreign mothers’ levels of stress. The finding that
mothers with lower levels of education had higher levels of stress is
supportive of the definition of this risk factor as a stressor in the literature
(Conger et al., 1984; McLanahan, 1983; Zuravin, 1988). It is also supportive
of a study involving black, single-parent mothers. The researchers found
that low levels of education were related to higher levels of stress (Lindblad-
Goldberg, Dukes, & Lasely, 1988).

The finding that mothers with low levels of income had higher levels
of stress is also supportive of the literature (Andrews & Withey, 1979;
McLanahan, 1983; Makosky, 1982; Pearlin & Johnson, 1977). McLanahan
(1983) found a negative relationship between level of income and number of
stressful life events experienced by the family. In another study, 43 low-
income mothers were found to experience more stressful life events
(Makosky, 1982). In a study of people from the Chicago area, economic
strain was positively related to the level of depression (Pearlin & Johnson,
1977). Lastly, Andrews and Withey (1979) found that people with lower

71

levels of income expressed more negative evaluations of the following
concerns: economic, family and marriage, health, and fun and enjoyment.

In this study, other identified risk factors such as number of
children, number of adults, and number of amenities were positively
related to the American mothers’ scores on the PSI-SF. Mothers with more
children had higher levels of stress. This finding is supported in the
literature (McLanahan, 1985; Nye, Carlson, & Garrett, 1970; Pearlin and
Johnson, 1977). McLanahan (1985) found that parents who had more
children experienwd a decrease in their level of well-being over a three year
period. Nye, Carlson, and Garrett (1970) found an increase in mothers’
levels of stress after the birth of the second child. Lastly, Pearlin and
Johnson (197 7) found a positive relationship between the number of
children and the parents’ levels of stress.

A finding of this study that is contradictive of findings reported in the
literature, is that the number of adults in the home was positively related to
the American mothers’ levels of stress. In the literature, those mothers
that are single-parents (the only adult in the home) experience higher
levels of stress (McLanahan, 1985; Weinraub & Wolf, 1983). McLanahan
(1985) found that the well-being of single-mothers declined by 11% over a one
year period. In a study conducted by Weinraub and Wolf (1983), it was
found that the single-parents experienced more life changes than the
parents who were married.

One of the reasons for the positive relationship between single-
parenthood and level of stress is that single-parents are more likely to also
be experiendng the following risk factors; income, education, and minority
status. In this study, all but one of the mothers who were single-parents
listed themselves as the only adult in the home. Therefore, when

72

discussing the risk factor, number of adults, all but one of the mothers who
listed two adults or more in the home were married. These single-parent
mothers had lower levels of education compared to the married mothers by
an average of two years. Their average income level was 50% below poverty
level compared to 10% below for the married mothers. These single
mothers were not all minorities. In fact, all but one of the mothers were
Caucasian. McLanahan (1985) found that single-parenthood was more
stressful for whit mothers than for black mothers. Even though the single-
parent mothers had more risk factors they were experiendng lower levels
of stress.

For the foreign mothers, a positive relationship between
ethnidty/race and level of stress was found. This relationship is supported .
in the literature (Andrews & Withey, 1976; Conger et al., 1984; McAdoo,
1983; Pearlin & Johnson, 1977). As with single—parenthood, race is
assodated with an increased likelihood of having more risk factors. These
risk factors are; education, income, single-parent status, and number of

children.

W: To determine if the identified risk factors were
related to the sample's maternal level of soda] support. *

For the American mothers, level of income, ethnidty/race, single-
parent status, and teen parent status were all negatively related to the
mothers’ level of soda] support. Those mothers who were either poor
and/or a single-parent had lower levels of support. This is suppomd in the
literature. Both McLanahan (1983) and Weinraub and Wolf (1983) found
that single-parent mothers had less contact with their extended families,
reported receiving less support from them, and were less likely to know

73

their neighbors. D’Ercole (1988) found that single mothers who had higher
levels of support experienced less finanda] strain due to the stressor of low

- income. American mothers who were classified as minority were more
likely to be experiendng lower levels of sodal support.

American mothers who were teen parents had lower levels of soda]
support. These mothers who were teen parents are probably younger than
their children’s fiiends parents and are out of step with women their own
age. As a result, they may find themselves with a support network
composed of members that are not very similar to themselves. Studies have
found that sodal support fi-om network members is more efi‘ective if the
members are similar to the receiver of support (D'Ercole, 1988; Gladow &
Ray, 1986; Zur-Szpiro & Lonflellow, 1982).

For both sub-groups there was a positive relationship between the
mothers’ levels of soda] support and their level of education. For the
American mothers, the more schooling she completed, the more support
she was receiving. These more educated mothers may be more aware of
the importance of soda] support and of the services available for them and
may therefore actively seek them. The foreign mothers, who had only
moderate levels of education compared to the other foreign mothers, were
receiving a lower amount of soda] support.

The foreign mothers who had only moderate levels of education may
be dissimilar to those mothers who have either low or high levels of
education. Those with the low levels are probably in Michigan while their
husbands go to school and there may be a large number of mothers in this
similar position. The same trend may exist for the mothers with high
levels of education. That is, these mothers represent a large group of

women who are enrolled in school and pursuing a Ph.d. This would be

74

supportive of the work done by D'Ercole (1988), Gladow and Ray (1986). and
Zur-Szpiro and Longfellow (1982).

For only the foreign mothers, a negative relationship was found
between the number of children and level of soda] support. The
explanation that the more children a parent has, the less time he/she has to
maintain and develop his/her sodal networks may be plausible (Miller &
Myer-Walls, 1983). The foreign mothers had younger children compared to
the American mothers. Therefore, they may be experiendng higher
'demands on their time and energy because of this lower age and thus have

less time with member of their sodal network.

W: To determine if the sample’s level of soda] support
was related to the sample's maternal level of stress.

For the foreign sub-group, the mothers’ scores on the MSSI total and
the MSSI Child Care Tasks Cluster were negatively related to the mothers’
scores on the PSI-SF. Also for this sub-group the mothers’ scores on the
Parental Distress subscale were negatively related to her scores on the Non-
child Care Tasks Cluster. Mothers with higher levels of soda] support had
lower levels of stress. The foreign mothers scores on the Dificult child
subscale were negatively related to her scores on the MSSI Child Care
Tasks. Those mothers who were experiendng stress due to raising a
difficult child received less help with child care tasks.

These findings are supported in the literature by the Bufi'ering
Hypothesis (Wilcox, 1981) and research on the mediating efi'ects of soda]
support (D’Ercole, 1988; Gladow & Ray, 1986; Zur-Szpiro & Longfellow,
1982). According to the Bufi'ering Hypothesis, sodal support protects people
against the efi‘ects of stressful life events on the person’s well-being (Wilcox,

75

1981). Zur-Szpiro and Longfellow ( 1982) found that the help with child care
tasks, household tasks, and emotional support were negatively related to
the mothers’ levels of stress. Gladow and Ray (1986) and D’Ercole (1988)
found that the support a single-parent mother receives from her family and
friends was positively related to her level of well-being. The negative
relationship between the American and foreign mothers’ scores on the
subscale, Parental Distress and their scores on the MSSI is also supported
by this literature.

It was surprising to find contradicting findings for the two sub-
groups on the relationship between the mothers’ usage of respite child care
services and their levels of stress as measured by the PSI-SF and two of its
three subscales, Parental Distress, and Parent-Child Dysfunctional
Interaction. The American mothers who had used the service longer or
more frequently had higher levels of stress. This is not supported in the
research literature focusing on respite child care. What is supported is the
negative relationship between usage of the service and level of stress for the
foreign sub-group as measured by the PSI-SF and the Parent-child
Dysfunctional Interaction subscale. In three empirical research studies,
there was a decrease in the mothers’ level of stress among users of respite
child care (Rimmerman, 1989; Subramanian, 1985; Wikler, Hanusa, &
Stoychefi', 1986).

These three studies used a pre- and post-test methodology that was
not utilized in this study. Where causation can be inferred for those three
studies, it can not be inferred from this study. There is a positive
relationship between the frequency of American mothers’ usage of the
service and their levels of stress as measured by the PSI-SF and its three
subscales and the foreign mothers frequency of usage of the service and the

76

Dificult child subscale. It could be that those American mothers with the
higher stress scores were aware of being under a high level of stress and
intentionally sought the respite child care services. As for the foreign
mothers, those mothers who are raising a dimcult child may be more
willing to use the service than those mothers who rate their children easier
to rear.

This may also be an explanation for the positive relationship between
the American mothers’ scores on the MSSI and the Dificult Child
subscale. American mothers who were raising diffith children were
receiving more support from their network members. Not only are
American mothers possibly more willing to use support services but also to
receive support from friends and family members.

Summary
The general conclusion of this study was that there were

relationships among the identified risk factors, mothers’ levels of stress,
and their levels of soda] support. Due to statistically significant
difi'erences between the American and the foreign mothers on the soda]
support measure, usage of respite child care service, and several risk-
factors, separate analyses were run to meet the research objectives. Those
American mothers who had higher levels of stress had a higher number
of; children, children who were five years of age or younger, adults in the
home, and amenities in the home. Little or no relationship was found for I
the American mothers’ levels of stress and the following risk factors; level
of income, level of education, minority status, single-parent status, teen

parent status, and amount of space.

77

The foreign mothers who had higher levels of stress had lower levels
of income and education and fewer number of children and number of
children who are five years of age or younger. Those foreign mothers who
were Asian or black had higher levels of stress. There was little or no
relationship found for the foreign mothers’ levels of stress and the risk
factors of number of amenities and amount of space. There was no
relationship between their levels of stress or soda] support and the risk
factors of single-parent status, teen parent status, and number of adults.

American mothers who had higher levels of education had higher
levels of soda] support . For the foreign mothers, lower and higher levels
of education were related to higher levels of sodal support. American
mothers who had lower levels of sodal support had the following risk
factors; single-parent status, teen parent status, ethnidty/race, and
income. Little or no relationship was found between the American
mothers’ levels of soda] support and the following risk factors; number of
children, number of children S five years of age, and number of adults.
The foreign mothers who had lower levels of soda] support had a higher
number of children and number of children five years of age or younger.
The risk factor, level of income, was only slightly related to the foreign
mothers’ levels of soda] support.

Relationships were found between the measures of soda] support
and stress for both sub-groups. For the American mothers, the more
stress a mother was experiendng as measured by the PSI-SF and all three
of its subscales, the more frequently the mother used the service during a
months’ time. The American mothers who received more support
reported that they were experiendng less stress in their parent role and
rated their child as being difficult. There was little or no relationship

78

found between the mothers’ scores on the PSI-SF and its three subscales
for either the two MSSI clusters and the number of months the mothers
used the respite child care service.

The more fi'equently the foreign mothers used the service, the less
stress they were experiendng as measured by the PSI-SF and the Parent-
Child Dysfunctional Interaction subscale. These same mothers rated their
child as being dificult to raise. The foreign mothers who were
experiendng higher levels of stress and parental distress had less soda]
support available to them. Those mothers who received less help with child
care tasks were experiendng more stress and rated their child as being
dificult. Lastly, those mothers who received less help with non-child care
tasks reported experiendng distress in their parental role. Little or no
relationship was found between the PSI-SF and its three subscales and the

number of months the foreign mothers used the respite child care service.

Future Research

Because of the significant difi'erences found between the American
and foreign mothers, a research study with more foreign mothers would
provide additional information that would provide greater detail on those
differences. A limitation of this study was the small number of foreign
mothers, which resulted in combining both the American and foreign
mothers into one group. Consequently, Asian, Black, and Middle Eastern
mothers were all placed into one sub-group. A study with a larger number
of Asian and Middle Eastern mothers may provide data on the differences
within the foreign sub-group population.

The finding that the American mothers who use the respite child

care service more are experiendng higher levels of stress is contrary to the

79

research in this area. A pre-and post-test methodological study with a
sample from the same centers would provide the additional information
needed to determine why these mothers are experiendng higher levels of
stress. In order to increase the sample size, the pre-test and the data
collection time period should be at least one year. This should increase the
sample size and possibly include those mothers who are only seasonal
users of the service.

Lastly, the respite child care service examined in this study was
' deve10ped to reduce the possibility and the occurrence of child abuse.
Therefore, a longitudinal study that would begin when the mothers start to
use the service and follow them for several years after their last visit would
provide valuable information on whether or not the service meets its
purpose. This study should include interviews with the parents, other
family members and child care providers as well as standardized tests and
possibly protective service reports and records.

REFERENCES

REFERENCES

Abidin, R. (1990) °
Charlottesville, Virginia: Pediatric Psychology Press.

Abidin- R. (1986). W Charlottesville-
Virginia: Pediatric Psychology Press.

Allison, J., Kilgallon, P. & Reilly, S. (1988). “The Children’ s Cottage. A
Respr te Service for Children and Parents rn Crisis.” W

Qarc- 3(5). 1- 12

Andrews, F. & Withey, S. (1976).
' ’ ' ° New York: Plenum Press.

Belle, D. (1982). “Sodal Ties and Sodal Support.” In D. Belle (Ed) Limit]
Winn. Beverly Hills: Sage Publications

anins- J- &Kintz- B. (1968). QamtafionalfiandboaknflSt-afistics.
Glenville, Illinois: Scott Foreman Company.

Butler, T. 8: Friesen, B. (1988) WWII. Portland,
Oregon. Portland Sate University. (ERIC Document Reproduction
Service No. ED 303 986).

Carmines- E. & Zeller- R. (1979). WWW
Beverly Hills: Sage Publication.

Collette, N. 8: Lee, D. (1983). "The Impact of Support for Black Adolescent
Mothers” Wanner-4(1). 127- 143.

Collette, N. & Gregg, C. (1981). "Adolescent Mothers' Vulnerability to
Stress.” WW 169(1)- 50-54.

Conger, R. McCarty, J. Yang, R., Lahey, B., & Burgess, R. (1984)
“Mother’s Age as a Predictor of Observed Maternal Behavior 1n Three
Independent Samples of Families.” WW
Earnilx- $5.- 411-423-

D’Ercole, A. (1988). “Single Mothers: Stress, Coping, and Sodal Support. "
W 16- 41-54.

Fasan C &Gurry- S. (1981). Excennonalfihildranandfiasmtcfiam.
W Cambridge, Massachusetts. (ERIC Docurrlent
Reproduction Service No. 245 459).

81

Franz, J. (1980, J anuary/Februaruy) “Being There: A 24-Hour Emergency
Crisis Care Center.” W, 7-10.

French, J. (1990).

 
 

Unpublished doctoral dissertation, Michigan State University, East
Lansing, MI.

Frisbie, D. & Slater. K (1979). W

Lansrng,
Michigan: Family Support Project. (ERIC Document Reproduction
Service No. ED 219 919).

Gladow, N. 8: Ray, M. (1986). “The Impact of Informal Support Systems on
theWellBeingofLow Income Single Parents.” W35,
113-123.

Glenn, N .& Weaver,C. (197 9). “A Note on Family Situation and Global
Happiness.” Socialism. 51(3). 960-967.

Hansenfeld, H., Murphy, 8., 8: Olson, K. (1981). “The Child Parent Drop-In
Center: Community Based Primary Prevention.” W
W. 2(3)- 155-158.

Hayes, C. (1987) “Determinants of Adolescent Sexual Behavior and
Dedsion Making.” In C. Hayes (Ed)
. (pp. 95-139).

Washington D.C.: National Academy.

Heller, K. & Swindle, R. (1983). "Sodal Networks, Perceived Soda]
Support- and Coping with Stress." In R. Felner, L. Jason, J.

Moritsugu, & s. Faber (Eds.). WW (pp. 87-103).
New York: Pergamon Press.

Hogan, M., Bushler, C., & Robinson, B. (1983). “Single Parenting:
Transitioning Alene”. In H. McCubbin & C. Figely (Eds) Streams]

(pp. 116-
132). New York: Brunner/Mazel, Inc.

Itagliata, J. (1986). “Assessing the Impact of Respite Care Services: A
Review of Outcome Evaluation Studies.” In C. Salisbury & J.

Intagliata (Eds. )..
(pp. 263-288).
Baltimore, Maryland. Paul H. Brooks.

Joyce, K. 8; Singer, M. (1983). “Respite Care Services: An Evaluation of the

Perception of Parents and Workers.” Wm, 44(9-
10): 270-274.

Joyce, K., Singer, M., & Isralowitz, R. (1983). “Impact of Respite Care on
Parents’ Perceptions of Quality of Life.” Wyn, 21(4):
153-156.

. Lamanna, M. &Riedmann, A. (1988). ' '
theeunifiadnsflranae (3rd ed). Belmont California:
Wadsworth Publishing Company.

Lindblad-Goldberg, M., Dukes, J, & Lasley, J. (1988). ”Stress in Black,
Low-Income Single-Parent Families. Normative and Dysfunctional

Patterns " Americanianmalnmrthonmhiatrx- 55(1): 104-120.

Lindblad-Goldberg, M., Dukes, J. ,& Lasley, J. (1985). "Sodal Support in
Black, Low-Income Single-Parent Families: Normative and

Dysfunctional Patterns." WWII!-
55(1) 42-58.

Makosky, V. (1982). “ Sources of Stress: Events or Conditions?” In D. Belle
(Ed.-) WWW (pp. 35-53). Beverly

Hills: Sage Publications.

Mandel], C. (1984).Dexelanins.1ndirddnalized£arent_2artidnatian
Bowling Green, Ohio; Bowling Green State University,
Department of Speda] Education. (ERIC Document Reproduction
Service No. ED 257 244).

McAdoo, H. (1983). 'Sodetal Stress: The Black Family. In H. McCubbin
8: C. Figely (Eds. ). ' '
(pp. 178-187). New York: Brunner/Mazel, Inc.

McCubbin, H. 8: Patterson, J. (1983). “Family Transitions: Adaptation to
Stress” In H. McCubbin & C. .Figely (Eds. ),
(pp. 5-25). New York:
Brunner/Mazel, Inc.

McLanahan, S. (1985). “Single Mothers and Psychological Well-Being: A
Test of the Stress and Vulnerability Hypotheses” W

Wealth- 5: 253-266.

McLanahan, S. (1983). "Family Structure and Stress: A Longitudinal
Comparison of Two-parent and Female-headed Families. Journal
ofMarriaaeandtheEamilr. 45: 347-357.

Miller, B. & Myers-Walls, J. (1983). “Parenthood: Stresses and Coping
Strategies.” In H. McCubbin & C. Figely (Eds)
(pp. 54-73). New York:
Brunner/Mazel, Inc.

83

\/ Nye, E. Carlson, J., 8: Garrett, G. (1970). “Family Size, Interaction, Afi‘ect,
and Stress. llanmalnflliarriaseandthefiamrlx- 32: 216-222

Pascoe, J. & Earp, J. (1984). “The Effects of Mothers’ Sodal Support and
Life Changes on the Stimulation of their children in the Home.”

Americanianmalnflhrbliefiealth- 14(4): 358-360.

wPascoe, J., Ialongo, N., Horn, W, Reinhart, M.A., & Perradatto, D. ( 1988)
35‘ “The Reliability and Validity of the Maternal Sodal Support Index.

EarnibLMedidne 29(4): 271-276.

<67 Pascoe, J., Chessare, J., Baugh, E. Urich, L. & Ialongo, N. (1987). “Help
with Prenatal Household Tasks and newborn Birth Weight: Is there
an Association?” alonmalntDelelamnentaLandBehaxiara]

Pediatrics, 8(4): 207-212.

Visas-roe, J. a French, J. (1990). “The Reliability and Validity of the
Maternal Sodal Support Index for Primiparo us Mothers: A Brief

Report.” Wane): 228-230pm

\/Pascoe, J., Loda, F., Jefiiies, V., & Earp, J. (1981). "The Assodation
Between Mothers' Sodal Support and Provision of Stimulation to
their Children.” WWW

W95- 2(1): 15-19.

Pearlin, L. 8: Johnson, J. (1977). “Marital Status, Life-Strains, and
Depression.” Ameficanfiadalodcalkxim- 42: 704-715.

Peterson, E. (1990, September, 9). "Yes, It's Harder to be a Child." Detroit
Freehan- p. 6H-

Pittman, J. Wright, C., & Lloyd, S. (1989). "Predicting Parenting
MIMIIY- WWW 19(2) 267-236.

Rimmerman, A. (1989). “Provision of Respite Care for Children with
Developmental Disabilities: Changes in Maternal Coping and Stress

over Time.” WM]:- 21(2): 99-103.

Salisbury,C (1986). “Generic Community Services as Sources of Respite.”
In C Salisbury _& J Intagliata (Eds.) W

~ . ‘ = -= : . pp
195-216). Baltimore,Maryland: Paul H. Brooks Publishing Company.

Straus, M. Gelles, R., & Steinmetz, S. (1980). W
Garden City, New York. Anchor
Press/Doubleday.

 

 

Stenson, C. (1989).

Pocateno. Idaho.
Idaho State University. (ERIC Document Reproduction Service No.
ED 306 909).

Subramanian, K. (1986). "Redudng Child Abuse through Respite Care
Intervention." W MG): 501-509.

Tellar, C., (personal communication, 1991). Michigan State Department of
Soda] Services, Lansing, MI.

Thoits, P. (1986). “Sodal Support as coping Assistance.” W
hannselinaandfllinieallsxchalosx. 54(4): 416-423.

flouliatos, J., Perlmutter, B., & Straus, M. (1990). '
MeasnrernenLIechnianes. Newbury Park: Sage Publications

Unger, D. (1987). ' ' ° ' '
Columbia, South Carolina; South Carolina
State Department of Health and Environmental Control. (ERIC
Document Reproduction Service No. ED 294 083).

Unger, D. & Powell, D. (1980). “Supporting Families Under Stress: the
Role of Sodal Networks.” Eamilyjelamms, 22: 566-574.

Unger D. & Wandersman, L. (1988). "The Relation of Family and Partner
Support to the Adjustment of Adolescent Mothers. Child

W52: 1056-1060.

Weinraub, M. & Wolf, B. (1983). "Efi'ects of Stress and soda] Supports on
Mother-Child Interactions rn Single- and Two-Parent Families.

Wayne-nt- 54:1297-1311
Weinstein, V. & Siegel, P. (1.988) '
Northridge, California.

California Children' s Coundl. (ERIC Document Reproduction
Service No. ED 307 959).

Wikler, L., Hanusa,D., 8r Stoychefi‘, J. (1986). "Home-Based Respite Care,
the Child with Developmental Disabilities, and Family Stress. Some
Theoretical and Pragmatic Aspects of Process Evaluation. "In C.

Salisbury & J. Intagliata (Eds. )-
(pp. 243-261).
Baltimore, Maryland. Paul H. Brooks Publishing Company.

Wilcox,B (1981). "Sodal Support, Life Stress, and Psychological
Adjustment: A Test of the Buffering Hypothesis. American

Wale-hologr- 9(4): 371-386.

85

Zuravin, S. (1988). "Child Maltreatment and Teenage First Births: A
Relationship Mediated by Chronic Sododenographjc Stress?"
Americaninmalnfflrthansxchiatnr. 53 1 91-103

Zur-Szpiro, S. & Longfellow, C. (1982). “Fathers’ Support to Mothers and
Children.” InD Belle (Ed) Wand
Dimension. (pp. 145-153). Beverly Hills; Sage Publications.

Appendices

Appendix A

$
Appendix A

Family Growth Center Quesionairre:

FAMILY GROWTH CENTER QUESTIONNAIRE

We would like to know how you learned about the Family Growth
Centers (FGC). Please CIRCLE the appropriate number or FILL
IN the blank for each question.

1. How did you first learn about the Family Growth Centers and
its services? Circle all that apply.

Friend

Poster/brochure

Newspaper (Which one)

Referred by a caseworker/professional

Church

Presentation

Other

 

$999!”.N!‘

 

2. What was the first service you used at the Family Growth
Centers? Circle the number of your answer.
1. Drop-in child care
Attended a support group
Parent education classes
Crisis Intervention
Other

9‘99.”

 

87

 

 

 

 

3. How did you learn about the other services at FGC? Please
drawalinefromeachserviceyouknow (ColumnA) tothe
wayts) you found out about the service (column B). Ifyou do
notknowofthese servicespleasegodirectiytoquestian4.

Comm (2de
Drop-in child care Referral by professional

Recommended by a FGC stafl' member
Support Groups Recommended by a FGC parent

Read about it in a FGC poster/brochure
Parent Education classes Other

4. When did you learn about the Family Growth Center's (FGC)

drop-in child care? Circle the number of your answer.
1. When I first heard of the FGC
2. At my first visit to the FGC
3. After I had visited the FGC several times
4. Other

5. When did you first start using drop-in child care? Fill in the
blanks.

Month year

6. About how many times a month do you use drop-in child care
at any center? Circle the correct number.

1 2 3 4 5 6 7 8 9 10

7. At which center(s) do you use drop-in child care? Circle all

that apple'-
1. Downtown Center
2. East Lansing Center

3. Mt. Hope Center

3

8. How do you plan to or do you use the time you have while your
child is in drop-in child care? Circle all that apply.

Rest '

Shap/family errands

Visit with own fiiends

Job interview

Appointments

Go to school

Go to work

Other

9°$9°9‘:“9’!°P

 

 

9. Have you ever attended a support group at FGC? Circle the

numberofyouranswer.
1. Yes
2. No

10. Have you ever attended parenting classes at FGC? Circle the

number-cumm-
1. Yes
2. No

11. Ifyouhaveonlyuseddmp-inchildcarcatFGC,couldyou
please explain why you choose not to use the other services,
support groups and parent education classes?

$

We would like to know a little more about you and your family. Please
CIRCIEtheappmpr-iatemnnherm-FIILIN theblankfioreachqnesflon.

12. Your age

13.

14.

15.

16.

 

Your marital status

1. Never married
2. Married

3. Separated

4. Divorced

5. Widowed

The number of years you have completed in school. Circle the

8 9 10 11 12 13 14 15 16 17 +
How many children do you have? Circle the correct number.
1 2 3 4 5 6 7 8 9 10+

How many people are living with you at this time? (Please
include spouse, children, relatives, and friends.) Circle the

 

so

17. Pleaselistthefimtnamesofallthepeoplelivingwithyou,
theirsextheiragewndthdrrelafionshiptoyou.

Thank you for your time and cooperation in completing this questionnaire.
Please continue to the next questionnaire, The Maternal Social Support
Index (MSSI).

Maternal Social Support Index:

MATERNAL SOCIAL SUPPORT INDEX (MSSI)
Please share with us the things you do in your home as a mother by

answering the questions below. Please use the following scale and CIRCLE
the answer you feel is m for you.

10.

11. If no car, can you get one in a few hours if needed?

12. How many relatives do you see once a week or more often ?

o 1 2 3 4 5

6

7

I Generally
Generally Someone Else
No one Do It Does It

1 2 3
Who fixes the meals? 1 2
Who does the grocery shopping? 1 2
Who lets your children know what

is right or wrong? 1 2
Who fixes things around the house

or apartment? 1 2
Who does the inside cleaning? 1 2
Who works outside around the house

or apartment? 1 2
Who pays the bills? 1 2
Who takes your child to the doctor
if he/she is sick? 1 2
Who sees to it that your children
go to bed? 1 2
Who takes care of car problems No

on short notice (if appropriate)? Car 2

8

3

Yes

9

Someone
Else and
I Do It

4

No

10+

92

Please CIRCLE the answer that is me for you.

13.

14.

15.

16.

17.

18.

19.

Would you like to see relatives:
More often Less often It's about right

How many people can you count on in times of need?
0 1 2 3 4 5 6 7 8 9 10 +
How many people would be able to take care of your children for
several hours if needed;
0 1 2 3 4 5 6 7 8 9 10 +
Do you have a boyfriend or husband? Yes No (If an, go to ll18)
If yes, how happy are you in the way your boyfriend or husband lets
you know what he feels or thinks?
Very Very
Happy Happy Unhappy Unhappy
Are there adults, NOT including your boyfriend or husband, with
whom you have regular talks?

Ye s No (If no, go to M)
If yes, think about the person you talk with the most. Are you happy
with the talks that you have with this person?

Very Very
Happy Happy Unhappy Unhappy

 

33

For the next question, please use the following scale and CIRCLE the

answer that is m for you:
Don't Attend less Attend Attend more
Belong than 1x a mo. 1x a mo. than 1x a mo.
1 2 3 4

H.999”?

21.

How often do you attend meetings of the following groups?

Religious (e. g. church) 1
Educational (e.g school, parent groups) 1
Social (e.g. bowling groups, scouting groups)1
Political (e.g. work for a local candidate) 1
Other: ______________________ 1

NNNNN
mammal
hubthnhfi

Are you a member of any committee or do you have any other duties
in any of your groups?

Yes No

How would you describe the amount of room your family has in your
home? Use the scale below.

1 2 3 4 5 6 7
Not Enough More than
Enough Room Enough

Room Room

94

23. Please circle all of the things you have in your home.

599°:49’9‘P9’N!‘

H
9

Running water
Electricity
Heat
Telephone
Refrigerator
Stove/oven
Radio
Television
Washer

Dryer

PARENTING STRESS INDEX
(Short Form)

 

 

Richard R. Abidin
University of Virginia
Directions:
in answering the following questions. please think about the child you are most
concerned about.

The questions on the following pages ask you to marl: an answer which best describes
your feelings. While you may not find an answer which exactly states your feelings. please
marl: the answer which comes closest to describing how you feel.

 

Pluumrkthedegmtowhichyouagreewdhagreewiththefououdngsnmby
circlingthenumberwhlchbestmatcheshowyoufeel. Uyouannotmpleasecirclen.

.1 2 3 4 5
Strongly Agree Agree Not Sure Disagree Strongly Disagree .
Example:
Ienloy golngtothemoviee. (lfyouaunetimeeenjoygolngtothemovies. you would
circle .2.) - 1 ® 3 g 5

 

Copyrighted 1990 - Abidin
Not to be duplicated

l . 2 3 ‘ g. 5

1. loftmhavethefeelingtluticannothandlethingsverywell.

z Ifindmyseifgivingupmoreofmylifetomeetmychildren’sneedsthanl
evcrcxpected.

3.1feeltrappedbymyresponsibilitiesasaparent.
4. Sincehavingthischildlhavebeenunabletodonewanddifferurtthings.

S. Sincehavingachildlfeelthatlamalmcstneverabletodothingsthat!
liketodo.

6. Iamunhappywiththelastpurchaseofclothinglmadeformyself.
7. Therearequiteafewthingsthatbothermeaboutmylife.

8. Havingachildhascaused moreproblenuthanlexpectedinmyrelationslup
with my spouse (male/ female friend).

9. lfeelaloneand without friends.
10.Whenlgotoapartyiumallyexpectnottoenjoymyself.
11.1mnotasintereetedinpeopleaslusedtohe.
'lZIdon'tmjoythingsasiusedto.

l3. Mychildrarelydomthinpformethatmakemefeelgood.

14. Mosttimeslfeeithatmychilddoeenotlikemeanddoesnotwanttobe
doeetome.

15. Mychildsmileeatmenmdtlesthanlexpected.

l6. Whenldothingsformycluldlgetthefedingthatmyeffortsarenot
apprschtedveryaudt.

17. Whenphyhpmydtilddomn'toftengiggleorlaugh.

18. Mydtilddoon'tseemtoleamasquicklyasmcstdtildren.
19.Mydulddoesn'tseemtosmileunnrchasmostd1ildren

20. Mycluldisnotabletodoasmchaslexpected.

l

l

21.1thkesalongdnuanditisveryhardformychildtogetusedtonewthings. i

Copyrightedlm-Afldin
Nottoheduplicaud

2345

. 1 2 3 4 5
Strongly Agree Agree Not Sure Dingree Smgly Disagree

22- ”RUIN“!!! 1. notverygoodatbeingaparent.
2. a person who has some trouble being a parent.
3. an average parent. 1 2 3 4 5 .
4. a better than average parent.
5. a very good parent.
2. Iexpectedtohavecioserandwarmerfeelingsformychildthanldoandthis
bothers me. 1 2 3 4 S
24. Sometimesmychilddoestlungsthatbothermejusttobemean. 1 2 3 4 5
econ:
25. Mychildseemstoayorhissmoreoftenthanmostchildren 1 2 3 4 s
26. My child generally wakes up in a bad mood. 1 2 3 4 S
27.1feelthatmychildisverymoodyandeasilyupeet. 1 2 3 4 5
28. My child does a few things which bother me a great deal. 1 2 3 4 5
29. My child reacts very strongly when something happens that my child
doesn't like. 1 2 3 4 5
30. Mychildgetsupeeteasilyoverthesmallestthing. 1 2 3 4 5
31. My child's sleeping or eating schedule was much harder to establish than i
expected 1 2 3 4 s

32.1havefoundthatgettingmychildtodoeomethingorstopdoing
mhinsis 1. muchharderthanl
2. somewhatharderthanlexpected.
3. aboutashardaslexpected. 1 2 3 4 s
4. somewhatmsierthanl
5. muchmsierthanlexpected.

33. lhinkarefullyandcountthemmberoftlfingswhichymrdulddoestlut
botheryou. Formmple: dawdlarehuestolistenmveractivncries.
interrupufighnwhinethc. Pleasedrdethenumberwhichinciudesthe
mimberofthingsymianmtsd.

1.104- 2.” 3.67 4.4-5 5.1-3 1 2 3 4 5-
34.1'herearesomethingsmychilddoesthatreallybothermealot. 1 2 3 4 s
35. Mychildnirnedouttobemoreofaproblemthanlhadexpected. 1 2 3 4 5
36. Mychildmakesmoredemandsonmethanmostchildren. 1 2 3 4 s

Copyrighud 1990-am
Nottobeduplicand “NW"

 

Appendix B

$

Appendix B

Letter from Director to participants:

Family Growth Center

A Program of Child Abuse Prevention Services

January 28, 1991

Dear Parents:

Maria Townsend is a friend of the Family Growth Center. She is Helping
us learn more about the needs of parents who use the center.

Thank you for taking part in this study. It will help the Family Growth
Center provide a better service to you.

YOUR ANSWERS ARE TOTALLY CONFIDENTIAL. THE FGC STAFF
WILL NEVER SEE TEI-IM. Please be as frank as possible, so that Maria’s
research can give us a true picture of the needs of families who use the
center.

We so appreciate your help with this, that we want to thank you in a special
way. When you complete the forms, bring them back to the desk. M11

-:u ttt o It-:‘::yttt ': ‘ s 5‘ t t t o ‘l

Thank you again for participating in this project. And please don’t forget to
ask at the desk for a Child Care Coupon.

Sincerely,

Merry Stanford
Director
Family Strengthening Programs

Letter to participants:
Dear Mother:

We would like you to tell us how you use the Family Growth Centers (FGC).
’ We would also like to know if you find being a mother stressful and how
much other people help you in your role as a mother. The results of this
study will be used by the staff of FGC to better serve your needs as a parent
and by me to complete a requirement for my college classes.

The questionnaire packet, which includes a green consent form, a blue
release form, three questionnaires, and a 3x5 card, will only take 20 to 35
minutes to complete. All the materials that you complete will be identified
by an identification number only. Your name will not be linked to your
answers on any of the materials. This ensures that your answers will be
_ totally confidential. To participate in this study, you must read and sign
both the green consent form and the blue release form found in your

questionnairepacket.

IN NO WAY ARE YOU REQUIRED TO PARTICIPATE IN THIS STUDY,
NOR WILL NOT PARTICIPATING AFFECT YOUR BEING ABLE TO
USE THE SERVICES. YOU MAY CHOOSE TO ANSWER ONLY THOSE
QUESTIONS YOU FEEL COMFORTABLE ANSWERING. IF YOU DO
NOT WISH TO PARTICIPATE IN ANY WAY, PLEASE RETURN YOUR
QUESTIONNAIRE PACKET.

I will be happy to answer any questions you may have about this study.
Please feel free to write to me at the address below or call me at (517) 351-
7617. If you wish to receive a summary of the results of this study, please
check the box on the 3x5 card. Thank you for your time and assistance.

Sincerely

Maria Zeglen Townsend
Institute for Child, Youth &
Families

Paolucci Building, Unit 2
Michigan State University
East Lansing, MI 48823

 

100

Consent Form:
CONSENT FORM

1. I, the undersigned, understand that the purpose of this study is to
understand the effects of the stress and support I am
experiencing as a mother on my participation in drop-in child
care offered at the Family Growth Centers.

2. My participation in this study will involve my completion of a
packet of questionnaires, which will take approximately 20 to 35
minutes.

3. I understand that my answers will in no way be linked to my
name and that they are confidential.

4. I understand that I can choose not to participate in this study at
any time and this decision will not afi‘ect my being able to use the
services at the Family Growth Centers.

5. I understand that I can obtain a summary of the study's findings
by checking the box on the 3x5 card in the questionnaire packet.

 

6. I understand that by participating in this study, I will receive a
reserved care coupon which entitles me to one session of reserved
grop-in child care for all of my children, ages six weeks through

ve years. '

7. If I have any questions or concerns about my participation in this
study, I know that I am encouraged to contact Mrs. Maria Zeglen
Townsend at 351-7617 to discuss my concerns. V '

8. I have read the above and agree to participate in this study.

A. I, the researcher, guarantee confidentiality to the person who
signed the form. Your name will not be linked to any of your
answers on the questionnaires. I will be happy to address any
questions or concerns you may have about the questionnaires and
your participation in this study.

Maria Zeglen Townsend
Graduate Student

101

Release Form:

RELEASE FORM

I give my permission to the Family Growth Centers to release
general information from the forms that I have completed for the Center.
This will include information such as, my age at the birth of my first child,
income, and the number of times I have used drop-in child care for my
children to the following person:

Maria Zeglen Townsend

Institute for Child, Youth 8: Families
Paolucci Building, Unit 2

Michigan State University

East Lansing, MI 48824

“‘1.

This permission will end as of December 31, 1991.

 

 

Signature Date

Reserved Care Coupon & 3x5 Card:
ReservedCamCoupon

The reserved care coupon was created as an incentive to encourage
participation in the study. It will be given to the mothers after they have
completed the questionnaires. The coupon will be copied onto bright, heavy
paper. The following will appear on the front of the coupon:

 

N Reserved care for your children, ages six weeks throum years,
for one session of care at the Family Growth Center of your choice.

You must call before 8:15 AM to reserve care for the mornings
session or before 12:45 PM to reserve care for the afternoon session. Spaces

may be limiwd on any given day.
484-2610 E : June 1991

 

          

3x5 Card

The 3x5 card placed in the questionnaire envelope will allow the
researcher to track the mother's participation in the parent education
classes or support groups and still ensure the mother's confidentiality. The
3x5 card will have the following information on it.

 

Ember __

Please clearly print your first and last name:

[ ] Please check the box if you would like a
ofthis stud 's fin ' .

 

     

f .

Checklist for outside of envelope:

STOP

Before you hand in your completed questionnaires at the desk, please make
sure you did these things.

1.

Did you mad the green CONSENT form and sign your name
at the bottom?

Did you mad the blue RELEASE form and sign your we at
the bottom?

Have you completed the:

___FGC Questionnaire,
Maternal Social Support Index, and
Parenting Stress Index?

Have you printed your name on the 3x5 card and checked the
box if you would like to receive a summary report of this
study's findings?

Please man fiom the envelope the:
3x5 card
__ Signed green CONSENT form
__ Signed blue RELEASE form
and seal the envelope shut.

Slam place in the locked, wooden box on or near the desk
e:

 

__Completed 3x5 card
__Signed green, CONSENT form
__Signed blue, RELEASE form

Please place your sealed envelope in the designated box next
to the locked, wooden box.

Don't forget to ask at the desk for your Reserved Drop-in
Child Care Coupon.

Thank oykou for your time and careful consideration in completing this
packet questionnaires.

 

104

Pretest Questionairre:

mmasr QUESTIONNAIRE

' We would like to know how you would rate the questionnaire packet you
just completed. Please read the questions carefully and CIRCLE the answer
that best describes your experience in completing the questionnaire packet.

1. Was the consent form understandable and clear?

very somewhat average somewhat very
clear clear unclear unclear

2. Was the release form understandable and clear?

very somewhat average somewhat very
clear clear unclear unclear

3. How easy were the instructions to follow for the Demographic
Questionnaire ?

very somewhat average somewhat very
easy easy difl‘icult diflicult

4. Was the Demographic Questionnaire understandable and clear?

very somewhat average somewhat very
clear clear unclear unclear

5. How easy were the instructions to follow for the Parenting Stress Index ?

very somewhat average somewhat very
easy easy dimcult dificult

 

105

6. Was the Parenting Stress Index understandable and clear?

very somewhat average somewhat very
clear clear unclear unclear

7. How easy were the instructions to follow for the Maternal Social Support
Index ?

very somewhat average somewhat very
easy easy diflicult difiicult

8. Was the Maternal Social Support Index understandable and clear?

very somewhat average somewhat very
clear clear unclear unclear

9. Do you have any additional comments that may help us to make the
questionnaire packet more understandable and easy to complete?

10. Was the checklist on the envelope helpful?

very somewhat average somewhat very
helpful helpful useless useless

Poster:

If you are not attending our
parent classes or support groups
and are interested in
one guaranteered session

of care for your child,

Please see the receptionist

or other staff members.

Flyer:

Attention Mothers

A reserved child care coupon is being offered to all
participants of a research study.

The study will examine the influences of stress and
social support on mothers who use drop-in child
care offered by the Family Growth Centers.

To participate in the study, you must be using drop-
in child care at the downtown, East Lansing, or Mt.
Hope centers.

Please ask the receptionist or other staff members
for more information.

The study is being conducted by Mrs. Maria
Zeglen Townsend, a graduate student at
Michigan State University, (who can . be
reached at 351-7617) in cooperation with the
staff of the Mt. Hope and downtown Family
Growth Centers.

 

Greeting card to mothers:

mm mm 'm
mm aims Mum n at m u no mu

use 5? ueieess sue sq sesame
am pm shell us so; secede cusses q ndi
nous mA Ml ounce use mess: e sauces

use not 'iauiut cm ili animals! us

inm-
want new no 'm-Iul m w
M Mama: Nut ti m Jams «we Ilium
an he would: ti union cm 'ma Inna
lit-“JP In on one 040mm 1' hue-l Mud
«um-01mm»! alumna-m

CHIN“ Print 3““ I991

0 I 93-99? ”001W UUOUd

Ind oo=v - M 0m
Jams mum in: annual
«Id oo=v - «It can
mm «mm limo-M
an era l - m on
mm mil in 0mm
«Id oo=tv - M can
mm mu in 0mm

”N33 J'A [M96

 

You and your children, ages six
weeks through five users, are
invited to try the Family Growth
Centers' drop-in child care.

 

Euchris Letter:

MICHIGAN STATE UNIVERSITY

 

mammalian-cam MW'MM'Cfl-i“
”MOMMA?!“

November 12, 1990

RE: IRE! 90-472

Maria Zeglen Townsend
Fanily and Child Ecology
Unit 2, Paolucci bldg.

RE: IRE FACTORS OE HATERNAL STRESS AND SOCIAL SUPPORT RELATED TO HOTEERS'
PAliICIPATIOR IN PARENT PROGRAMS AMONG USERS OE RESPITE CHILD CARE.
IRE 90-472

Dear as. Townsend:

the above project is axe-pt free full 081188 review. The proposed research
protocol has been reviewed by another connittee eenber. The rights and
welfare of hunen subjects appear to be protected and you have approval to
conduct the research.

You are relinded that 0611!! approval is valid for one calendar year. If you
plan to continue this project beyond one year. please sake provisions for
obtaining appropriate UCIIIS approval one nonth prior to hove-bar 4, 1991.

Any changes in procedures involving hunan subjects nust be reviewed by ucrxss
prior to initiation of the change. OCIIIS nuet also be notified pronptly of
any problees (unexpected side effects. couplainte. etc.) involving hunan
subjects during the course of the work.

thank you for bringing this project to ny attention. If I can be of any
future help. please do not hesitate to let us know.

Sincerely,

< .l 2122'.“

o'me 3. Wright.
Chair. scars:

 
  

DIV/doc

cc: Dr. Jeanne Brown

NSU'IQMWWW

”’11111111111313