‘n?I5‘IMI‘t‘\i‘ .1- vi J1:

., n‘fil

k‘w

u'gfiu as;

s

l

 

 

 

 

 

. L.

“an:
.

J.JL.n-..

 

"J

\ :3” if». ‘37:?
1:: f} :‘-‘XIE>1
.17 ‘
Affiniu

;36‘:

u
.n.....x-..(Z
u..,,...u.....x
if

”-1

 

L‘
‘21:: ‘1“

'41.
x-E‘Cih‘ afifg‘:
; 23:1,:

.r ' 1‘ R
v.13}?
325%5:
"n“: _

 

 

 

:r‘é ‘-

.' ‘ 1M
’ {x ‘5,
gr!!!

:vr

. ‘P
h‘

'l
1 .

1", if 1

1" .
‘zr: : "“12
§::-:§~I‘:*t::=‘**

»

1‘1'

:1
1|!

1 ‘

.. 1%,}‘3':
"£17“;- “'5'; ‘.

.,.

’3.“ \. ‘ill
5.. j“:— -‘ --..p—a...,—nu:. .
, ...-. ,.....-
:rr- .4. -a« "4
' no ro‘

.nO—I .

, .r
.-,:...-!..;"f
" ' .u

, Jur-f r r
sa- '0; 1.... ~p..n.~ Eu, _ “(M
.1” .,

”fix”... 1‘"
.,
,;:....! ’1» ."

a...
4’. 2.;‘7.

 

1.1501“. u) I!
”“331.-.” ....
1. :-.....~'..:.

I
.-
. . a. . _
3- n;::'.
c
J:

v M
.‘ . 1 I}. .-. n..- ”W
. . < - - .. .M

l 7 : n *“":W

'l' : J. . u ' u. r u -, _. ‘ n. on: '"wafl-‘hspoZf-z Da’m J
.l..-W.~A .-A-u« 'v.‘ ' 'lravI-O- “0"

:C l v Iv cog-4 -»
w v

 

MICHIGAN ATE UNIVERSWY U

llwillmull\mxmllll

L 1293 00876 1482

 

 

W I

This is to certify that the

thesis entitled
Gender Differences In The Perceived Thinking Styles Of
College Students And Their Parents

presented by

Helene R. Lensky

has been accepted towards fulfillment
of the requirements for

M.A. degree in Clinical Esychology

\ 1

Major professor

 

Date 4/2/91

 

 

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

 

 

 

LIBRARY
Michigan State
University

 

 

 

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 Is An Affirmative Action/Equal Opportunity lnflitution
cmmms-q

GENDER DIFFERENCES IN THE PERCEIVED THINKING STYLES
0F COLLEGE STUDENTS AND THEIR PARENTS

By

Helene R. Lensky

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

MASTER OF ARTS

Department of Psychology

1991

ABSTRACT

GENDER DIFFERENCES IN THE PERCEIVED THINKING STYLES OF
COLLEGE STUDENTS AND THEIR PARENTS

By

Helene R. Lensky

Hypothesized gender differences in cognitive style
were explored using Epstein and Meier's (1989)
Constructive Thinking Inventory (CTI). Adressing both
constructive and nonconstructive cognitive forms and
reported to correlate with successes in love, physical
health, and work, the CTI was administered to 118
undergraduates and to their mothers and fathers by mail.
In simulations of their parents' CTI performance,
students depicted each parent as thinking more
constructively than themselves but less constructively
than parental self-reports. Hales' higher Emotional
Coping scores was the salient gender difference.
Daughters' simulated parental CTI responses were
generally more favorable than sons. 0n CTI scales of
both Naive Optimism and Categorical Thinking, however,
parents own response were significantly more favorable
than sons' and daughters’ simulations. Noteworthy
differences also held among the various family dyads
(mother-daughter, mother-son, father-daughter, father-
son). These findings were discussed in the context of

myths about gender-linked differences in thinking.

T0 DAVE

iii

ACKNOWLEDGMENTS

I would like to thank Dr. John Hurley, who enthusiastically
shared his ideas and his time. I am grateful to him for the
his encouragement and direction throughout the development of
this work. I would also like to express my appreciation to
Dr. Norman Abeles and Dr. Bertram Karon for their guidance in
the completion of this thesis.

iv

TABLE OF CONTENTS

List of Tables ....................................... vii
Introduction ........................................... 1
Constructive Thinking ............................. l
Cognitive-experiential Self-theory ................ 3
The Constructive Thinking Inventory ............... 5

Family Relationship Factors and Thinking Style....8

Hypotheses ............................................ 17
Method ................................................ 19
Participants ..................................... 19
Measures ......................................... 19
Procedures ....................................... 21
Data Analyses ..................................... 23
Results ............................................... 24
Discussion ............................................ 38
Sample Considerations ............................ 38
Ratings of Parents ............................... 39
Sex Differences on the CTI..................;....44
Parent-Child Conflict ............................ 45

The Relationship Between Conflict and
Ratings of Parents .......................... 47
Future Research..................; ............... 48

Appendix A: Standard deviations of self-reported
and parent-simulated responses to CTI and
relationship indicators .......................... 50

Appendix B: Letter to Parents ......................... 51

Appendix C: Demographic Items ......................... 52
Appendix D: Student's Informed Consent Form ........... 55
Appendix E: Debriefing ................................ 56
Appendix F: Parent's Informed Consent Form ............ 57
List of References .................................... 58

vi

Table

Table

Table

Table

Table

Table

Table

Table

Table

LIST OF TABLES

Means of Self-reported and Parent-simulated
Responses to CTI and Relationship
Indicators .............................. 25

Means and t-tests of Differences Between CTI
Discrepancy Scores (parent's actual score
minus student's simulation) for Mothers
and Fathers ............................. 28

Corelated trtests of Differences Between the
Means of Parents' Self-reports and Their
Responses as Projected by Their Children
on the CTI .............................. 27

Means, Standard Deviations, and trtests of
CTI Differences Between Daughters' and
Sons' Discrepancy Scores (father's
actual score minus student's simulation)
for Mothers ............................. 28

Means, Standard Deviations, and trtests of
CTI Differences Between Daughters' and
Sons' Discrepancy Scores (father's
actual score minus student's simulation)
for Fathers ............................. 29

Means, Standard Deviations, and trtests of
Differences Between Sons' and Daughters'
Mother-simulated CTI Scores ............. 30

Means, Standard Deviations, and trtests of
Differences Between Sons' and Daughters'
Father-simulated CTI scores ............. 30

Correlations Among CTI Discrepancy Scores
(between student-simulated and parents'
actual scores) and Parent Conflict ...... 33

Correlations Among Students"Scores, Parent
Scores, and Student-Simulated Scores for
Parents on the CTI's Constructive Thinking
Scale ................................... 34

vii

INTRODUCTION

This paper addresses young adults' perceptions of
parental thinking style and the influences of parent
and child gender on these perceptions. The effects of
gender on various aspects of family interaction have
been frequently reported. Childrens' adjustment seems
importantly influenced by their perceptions of parents.
(Whitbeck, 1987). There is, however, a dearth of
research in the area of childrens' perceptions of
parents cognitive style and it is precisely on this that
the present paper will focus. In addition, the concept

of constructive thinking will be both reviewed and

investigated.

WW

Constructive thinking is a relatively new construct
that can be described as “life smarts" - a thinking
style that enables one to function effectively at life
tasks. Epstein and Meier (1989) used the phrase
"practical intelligence" to describe this construct.
This name underscores their finding that constructive

thinking is conceptually distinct from intelligence as

we usually define it.

Intelligence has traditionally been defined by
school-related skills and is generally thought to
include such skills as vocabulary, arithmetic, and
spatial ability. At present, rather than measuring
specific skills, there is a new emphasis on the
processes that are necessary to acquire such skills.

A person high in intelligence would process information
in a way that facilitates intellectual learning
(Kaufman, 1979). Thompson and Melancon (1987), in their
investigation of critical thinking skills, measure such
skills as hypothesis testing and deduction.

Constructive thinking, unlike intelligence, does
not refer to intellectual functioning. Constructive
thinking is a broad concept that encompasses diverse
areas of functioning and adjustment. High constructive
thinkers should be relatively successful in life--
career, love, friendships, etc. However, high
constructive thinkers would not necessarily have high
I.Q.'s or do well in school. In fact, constructive
thinking, as measured by the Constructive Thinking
Inventory (CTI), was found uncorrelated with academic
achievement or with 1.9. (Epstein & Meier, 1989).

How might one come to think constructively as
opposed to nonconstructively? Several cognitive
theorists have linked destructive or negative thinking

to various forms of psychopathology. For example, Beck

(1978) implicates illogical thinking as a causal agent
in depression. According to Beck, people who view
themselves, the world, and the future in an
unrealistically negative fashion would be more likely to
become depressed and remain depressed. Several studies
have provided empirical support for this theory of
depression (e.g. Burns, Shaw, & Crooker, 1987; Rizley,
1978).

Ellis (1987) points to "irrational beliefs" as the
cause for emotional disturbance. According to Ellis,
humans often think in terms of "musts", "shoulds", and
other unrealistic beliefs; and it is this type of
thinking that leads to neuroticism (Ellis, 1985; Ellis,
1987). In accordance with Ellis' theory, Thyer,
Papsdorf, and Kilgore (1983) found a correlation between
irrational thinking and psychological disturbance. In
fact, irrational thinking styles have been associated
with various psychological problems ranging from eating

disorders (Mizes, 1988) to wife abuse (Edleson, 1984).

WWW

Epstein and Meier's (1989) CTI was developed to
assess constructive versus nonconstructive thinking
styles. The CTI scales were designed with adherence to
cognitive-experiential self-theory (CEST). This theory
defines three separate “conceptual systems“: the

rational, the experiential, and the associationistic

(Epstein, in press). The rational system operates
consciously and is guided by reason and logic. The
experiential system operates at the preconscious level
and is strongly influenced by feelings (both conscious
and unconscious). The associationistic system
"corresponds to a state of altered consciousness and is
viewed as similar to Freud's unconscious system...“
(Epstein, in press, p. 8).

The experiential system provides the passion that
the rational system lacks. Conflicts "between the heart
and the mind" are examples of interaction between the
experiential and rational systems. Irrational fears
provide another example of interaction between the
systems (Epstein, in press). For example, we may know
in our minds, while watching a horror movie, that we are
safe, yet we may feel frightened and scream. According
to Epstein, it is the experiential system that responds
rapidly while the rational system experiences a delay.
Therefore, it is the experiential system that guides
much of our behavior. If we are to understand human
motivation and behavior, we must tap into the
experiential conceptual system. And it is with this
purpose that the CTI was devised (Epstein & Meier,
1989). It is a 84-item measure that is designed to

assess constructive thinking as defined above.

The CTI's prospective item pool was gathered from

three sources: books and articles on cognitive theory;

reports from acquaintances of the author on various
constructive and destructive thoughts they experienced;
and daily logs of students enrolled in a class on
"emotions and self-concept". From these items Epstein
constructed statements which could be responded to on.a
five point scale--the extreme anchors for which are
"definitely false" and "definitely true". Epstein then
sorted the items were sorted into 18 categories based on
face validity. A confirmatory factor analysis
identified six "interpretable factors“. The six scales
of the CTI were formed from these factors after slight
modifications. Twelve was the maximum number of items
in any scale. Each scale included only those items
which correlated most highly with each other.

These six basic scales are: Emotional Coping,
Behavioral Coping, Superstitions Thinking, Categorical
Thinking, Negative Thinking, and Naive Optimism. Added
was a composite set of 28 items drawn from the six basic
scales and labelled the global index of Constructive
Thinking. Brief descriptions of each scale will further
elucidate these constructs.

Emotional Coping taps such areas as not taking
things too personally or not being overly sensitive, not

being afraid of failure, and not being needlessly

worried. Behavioral Coping, on the other hand, deals
with the ability to act efficiently. People who are
effective behavioral copers are optimistic and focus on
future action rather than past disappointments or
injustices. Emotional Coping is tied to the absence of
negative thinking while Behavioral Coping involves the
presence of positive thinking (Epstein & Meier, 1989).

Superstitions Thinking includes items on personal
superstitions (e.g., I must eat a candy bar before each
exam or I will fail) as well as items on institutional
superstitions (e.g., it is bad luck to walk under a
ladder). Categorical thinking includes items that refer
to rigid thinking patterns. A person who thinks
categorically would tend to see people as good or bad,
for them or against them (usually against), and see
things as right or wrong. Naive Optimism includes items
that refer to "seeing the world through rose colored
glasses" and is not viewed as a constructive thinking
style. It is characterized by overgeneralizing after
positive events occur. For example, "I passed this exam
therefore I will be successful on every exam I ever take
no matter what it entails."

Negative thinking is characterized by excessive
pessimism. A high scorer on negative thinking would
tend to see the worst aspects of any situation.

Finally, the global CTI scale is a selected composite of

all the other scales except Naive Optimism. Its items

were excluded from the global index because Epstein and
Meier (1989) found that all basic CTI scales except
Naive Optimism intercorrelated significantly. More
recently however, Hurley's (1990) study found that the
Naive Optimism scale was significantly associated

positively with Superstitions and Negative Thinking but
negatively with coping both Behaviorally and

Emotionally, as well as with the global index.

Since the CTI instrument is so new (1989), there

is little other published research about it.
Preliminary findings (Epstein & Meier, 1989) reported
correlations of the global Constructive Thinking scale
with such variables as job success, success in love,
mental well being, and physical well being.

Other research (Epstein, in press) reported
relationships between CTI scores and scores on the
Primary Emotions and Trait Scales (PETS). Global
Constructive Thinking also correlated positively with
ego-strength and self-esteem, but negatively with anger,
depression, and neuroticism. In addition, the various
CTI subscales correlated with the PETS variables
divergently, providing evidence that they do address
different aspects of thinking. For example, anxiety
correlated highest with Emotional Coping, whereas anger
correlated most highly with Categorical Thinking.

Hurley (1990) investigated the relationship between

CTI scores and one's ratings of self and others within

small interpersonal groups. All members rated
themselves and all other same-group members for self-
acceptance and other-acceptance. CTI scores were found
to correlate with the discrepancy between self ratings
and ratings one gave to others for self-acceptance.
Those higher on the subscales of constructive thinking
were significantly more likely to rate self higher than
they rated their group peers, while those higher in
nonconstructive forms of thinking were significantly
less likely to overrate themselves.

For acceptance of others, smaller discrepancies
between ratings of self and ratings given to others was
found to be associated with Categorical Thinking,
Superstitions Thinking, and Negative Thinking. In
summary, a high degree of self-acceptance was associated
with constructive thinking, as would be expected. In
addition, as was suggested by the research of Taylor and
Brown (1988), a self-serving bias appears to be

predictive of constructive thought.

Family Relatiggghip Eagtggg and Thinking Style
Constructive Thinking is also believed to be.
influenced by parental behavior. Epstein (1987)
investigated the effects of parental independence-
enconragement and parental acceptance-rejection on the
development of constructive thinking in children.

Independence-enconragement was found to be significantly

correlated with constructive thinking, however, the
relationship between constructive thinking and parental
acceptance-rejection was not found significant.
According to these findings, encouraging children to be
independent appears to be more important in deveIOpment
of their thinking style than is providing warmth and
acceptance. Later, Epstein (in press) reported that the
situation most conducive to censtructive thinking in
children is a father who encourages independence and a
mother who is accepting. What is most apparent is that
parent-child interaction some impact on the development
of constructive thinking in children.

Another early CTI study by Epstein and Lee (cited
in Epstein, in press) yielded some expected and some
puzzling findings. In a study of 79 college students
and their parents, it was found that these parents
scored more favorably on a revised version of the CTI
than did their children. This was expected because
constructive thinking should, unlike intelligence,
increase with experience. There were no sex
differences.

Some unexpected correlations were found between the
CTI scores of these parents and their children. The
global Constructive Thinking scores of both sons and
daughters were found to correlate significantly with
those of their fathers, yet not with their mothers.

Dangters' CTI responses, however, did have significant

10

positive correlations with those of their mothers on two
scales--Categorical Thinking and Esoteric Thinking.
Esoteric thinking, a scale which measured a belief in
superstitions and supernatural phenomena, was later
incorporated into the Superstitions Thinking scale.

Thus, it appears that daughters acquired some negative

aspects of their mother's thinking style and some
positive aspects of their father's. In contrast, these
sons' thinking style modelled only that of their
fathers. What might account for this unexpected
finding?

These college-student children had been asked to
simulate CTI responses separately for each parent.
Daughters consistently depicted their mothers as below
their fathers for global Constructive Thinking than
fathers despite the fact that these fathers did not
score higher than mothers on the CTI. Sons, on the
other hand, described their mothers and fathers as equal
for Constructive Thinking. The authors offered no
explanation for these findings, but acknowledged the
need for replication. Thus, Epstein's findings appear
to conflict with the work of O'Neil and Reiss (1984) who
found no sex differences in the way sons and daughters
(aged 20-64) perceived their mothers' competence.

O'Neil and Reiss examined adult childrens' perceptions
of their mothers across the life-span using a new

measure called the Parent Perception Inventory and

ll

concluded that "by age twenty men and women perceive
their mothers in basically similar ways on a number of
important dimensions" (p. 341).

The present study seeks both to replicate Epstein's
puzzling findings and to uncover some possible
explanations. Several questions remain unanswered. th
do sons and daughters have different perceptions of
their mothers? An extension of this question is: If
sons and daughters do have different perceptions of
their mothers, then why are there not parallel
differences in the way that they viewed their fathers?
Why do daughters see their mothers' Constructive
Thinking as inferior to their fathers' despite the
absence of differences between the parents? why do sons
fail to use their mothers as models of constructive
thinking even though they rate them as highly as their
fathers?

The way children view their parents is an important
factor in children's development. Hhitbeck (1987) found
that a child's self-adjustment is better predicted by
the child's perception of parental efficacy than actual
parental self efficacy as recorded by self-report.
Therefore, the way in which children perceive their
parents' cognitive style seems an important area of
research which merits further attention.

There also seem to be gender differences in the

salience of parental efficacy to sons and to daughters.

12

Girls appear more affected by parental behaviors and
parental interaction, while boys appear more affected by
the perceived efficacy of their parents (Whitbeck,
1987). Thus, it seems that having parents believed to
be inefficacious would be more detrimental to boys than

to girls. Girls could view their mother as less

efficacious but still have high efficacy themselves.
This is precisely what was reported by Epstein and Lee
(cited in Epstein, in press). Although daughters rated
their mothers below fathers for Constructive Thinking,
they rated their own Constructive Thinking as equivalent
to the male peer group.

When discussing parent-child relationships an
important fact to consider is that, in our society, the
mother is the primary caretaker. As Chodorow (1987)

describes,

"fathers are from the outset separate people and
special...As a result, representations of the
father relationship do not become so internalized
and subject to ambivalence, repression, and
splitting of good and bad aspects..." (p.97).
Thus, both son and daughter are more likely to have
closer and more emotionally intense relationships with
mother than with father. Fathers and sons are believed
to have a stronger bond than exists between fathers and
daughters (Radin, 1988), although the mother is still in
the position to be closest with children of both sexes.

The majority of children, whether male or female, tend

to see fathers as "separate people and special".

13

Perhaps, this provides partial explanation for the
absence of differences between sons and daughters in the
way they view father's Constructive Thinking. Because
neither child has the opportunity to be extremely elose
with father, there may be less room for variance in
their impressions.

The remainder of this paper will focus on the
mother-child relationship for two reasons. First, it is
this relationship that takes up most of the child's
time. Second, the inconsistent prior findings
concerrned only mothers only and it is these findings
that the present study addresses.

Time spent with mother can often be less enjoyable
than time spent with father. Chores and care-giving
type activities occur more often with the mother, while
proportionally, play time more often occurs with father
(Montemayor & Brownice, 1987). Hazzard, Christensen,
and Margolin (1983) found that both sons and daughters
perceived their mothers as performing more negative
behaviors than father. In a study by Smith and Forehand
(1988), daughters reported significantly more conflict
with mothers than with fathers. And Montemaynor (1982)
reported that girls have more conflict with their
mothers than do boys. Thus, mother's increased
involvement with children may bring with it increased
opportunity for conflict and "ambivalence" (Hazzard et

al., 1983; Smith & Forehand, 1988); and this difference

14

may be greater for girls than for boys (Montemaynor,
1982).

In addition, it seems apparent that the more
intimate we are with people, the more likely we are to
perceive their less logical and less rational sides.

Mothers, therefore, might well be regarded by their

children as less constructive thinkers. Epstein and Lee
reported that only daughters have this misperception of
their mothers’ constructive thinking. This gender
difference may be partly attributed to conflict in the‘
mother-daughter relationship. Are there other reasons
why only daughters would see their mothers' as less
constructive in their thinking? As was mentioned above,
girls appear more affected than boys'by the qualitative
aspects of interaction with their parents. Perceived
parental competence may be more salient to boys.
Hetherington and Frankie (1987) use the terms "warmth"
and "dominance" to refer (respectively) to the
characteristics most salient to girls and to boys.
The adjustment and competence of boys, therefore, would
be expected to be more highly influenced by, and more
similar to, their perceptions of parental efficacy than
‘would the adjustment and/or competence of girls.
Quality of interaction would seem more likely to
fluctuate than would perception of parental efficacy or
dominance. Thus, the ”mother" that daughters see may

appear less stable and secure than the “mother" that

15

sons see. Following from this, it appears that
daughters would be more likely to see the less logical
and less rational side of their mothers than would boys.
This is because daughters are more likely than sons to
attend to the fluctuating qualitative aspects of
parental interaction. Thus, one may predict that
daughters would be more likely to underrate their
mother's constructive thinking to a greater extent than
would sons.

Why do sons and daughters have different
perceptions of their mothers? Chodorow (1987) describes
an intensity in the mother-daughter relationship that is
not present in the mother-son relationship. Since
mother and daughter are of the same gender, they tend to
share a more instense bond than is usually developed
between mothers and sons. There is also a diminished
sense of boundaries between mothers and daughters--a
fuller sense of oneness. In addition, boys are
socialized to be independent, while girls are socialized
to be nurturant and caring of others (Block, 1983;
Gilligan, 1980). This would further promote attachment
to mother on the part of the daughter and separation on
the part of the son.'

Into adulthood, daughters struggle with the
conflict between maintaining and diminishing the close
ties to mothers. Sons likely experience less of this

ambivalence as their movement toward manhood includes

16

greater separation from mothers (Gilligan, 1980).
College students are in the process of achieving
independence from family and emerging into adulthood.
Thus, unlike in the adult sample of O'Neil and Reiss
(1984), the time is ripe for increased ambivalence on

the part of the daughter toward her mother. This may

partially explain the discrepancies between the CTI
responses which daughters attributed to their mothers
and fathers in the study by Epstein and Lee.

Conversely, sons are achieving the expected separation
from mother that is a necessary part of manhood. Mother
and father now have more comparable roles in their son's
life. Therefore, one would expect lesser discrepancies
between the CTI scores that sons give to their mothers
and fathers. The late adolescent period, marked by
ambivalence, may be unique in its capacity to foster in
daughters a distorted view of mother. Recall that
O'Neil and Reiss (1984) found no differences in the
maternal perceptions of adult daughters and sons.

In summary, it appears that there are several
family system factors that may account for the
discrepancies between the ways in which daughters and
sons view their mothers. However, the first task here
is to replicate the findings of Epstein and Lee, either
empirically supporting or disconfirming these
discrepancies.

Similarities and differences between parents and

17

their children in Constructive Thinking will be
examined. All configurations of the parent-child dyad
will be compared, e.g., mother-daughter, mother-son,
father-daughter, father-son. Parent-child conflict is
hypothesized to play a role in the relationship between

gender and perception of parents and in the degree of

parent-child similarity.in constructive thinking.
Hypotheses

The following hypotheses will be tested:

Hypothesis 1. With reference to parents' actual CTI
responses, students will underrate their
mothers more than their fathers for all
modes of constructive thinking addressed

by the Constructive Thinking Inventory.

Hypothesis 2. Daughters will underrate their mothers'
global Constructive Thinking scale

performance more than will sons.

Hypothesis 3. Daughters will report significantly more
conflict with their mothers than will
sons as measured by the Conflict Behavior

Questionnaire.

Hypothesis 4. Student-reported parent-child conflict
will correlate positively with the amount

of discrepancy between the simulated and

Hypothesis 5.

Hypothesis 8.

Hypothesis 7.

Hypothesis 8.

Hypothesis 9.

18

actual performance of parents on the

CTI's global Constructive Thinking scale.

Global scale scores of both sons and
daughters will correlate significantly

more highly with the parallel scores of

their fathers than with those of their

mothers.

Both daughters and sons will perceive
their mothers as performing more negative
behaviors than fathers as measured by the

Conflict Behavior Questionnaire.

Sons' CTI global Constructive Thinking
will correlate significantly more highly
with their parent—simulated scores than

will daughters' parallel scores.

There will be no gender differences in

CTI scores.

Students of both sexes will report
spending more time with their mothers

than with their fathers.

19

Method

B I' . i |
Participants were 147 college students (ages 18-20)
and 118 of the 147 sets of parents. These students (95
female, 52 male) were enrolled in an introductory
psychology course at a large university and chose to
participate in the study in exchange for class credit.
Both parents of 76 female and 42 male students also
contributed their CTI responses. According to student-
report, 782 of mothers and 95% of fathers were employed

outside of the home.

Measures

The CTI was administered to all participants. For
124 University of Massachusetts undergraduates, Epstein
(1989) reported a range in reliability for the scales of
the CTI from a high internal consistency (coefficient
alpha) of .87 for the global scale to .87 for Naive
Optimism, with a mean alpha of .77 for the basic set of
seven scales.

To demonstrate these measures' construct validity,
Epstein and Meier (1989) calculated correlations between
three CTI scales (Global, Emotional Coping, and
Behavioral Coping) with several criterion variables.
Each of these three scales correlated positively and

significantly with success in work, love, and social

20

relationships. Information on these criteria was
gathered using The General Information Form, a four-item
questionnaire on work success, a four-item scale on
success in love, and an eight-item scale on success in
social relationships. The same three CTI scales were
also found to be negatively and significantly correlated
with physical and psychological symptoms as well as with
the abuse of alcohol and drugs as measured by the
Medical Checklist. Independent findings using Michigan
State University students (Hurley, 1990) indicated that
Constructive Thinking correlated positively, and
nonconstructive modes of thinking negatively, with self-
acceptance.

The present study assessed “Time Spent" with each
parent using the following eight questions, "0n the
average, how many hours did you spend with your
(mother/father) per week when you lived at home during
high school?" "Currently, how many times do you speak to
your (mother/father) on the phone each month?"
"Currently, how often do you see your (mother/father)
each month?" "Approximately how many letters per month
do you write to-yonr (mother/father)?"

The variables of “parent-child conflict“ and
"perceptions.of parents“ were assessed with the 44-item
version of the Conflict Behavior Questionnaire (CBQ;
Prinz, Foster, Kent, & O'Leary, 1979) which uses a true-

false format. This instrument yields scores for two

21

content areas: "Appraisal of the Dyad" and ”Appraisal of
the Other". These two scores serve, respectively, as
measures of "parent-child conflict" and "perceptions of
parents".

Foster and Robin (1988) reported a correlation of
.98 between the 44-item CBQ and the original 73-item
measure. The CBQ (long version) was also shown to
distinguish troubled from untroubled families (Robin &
Weiss, 1980). Foster and Robin (1988) reported a
correlation of .55 between the 44-item CBQ and a subtest
of Snyder's Marital Satisfaction Inventory measuring
dissatisfaction with child rearing practices; and a
correlation of -.52 between the CBQ and a problem
solving ability score on Nayar’s (cited in Foster &
Robin, 1988) Parent-Adolescent Interaction Coding
System--an observational rating of conflict. Test-
retest stability was reported at .37 - .85 over six to
eight weeks (Robin & Foster, 1988). Alpha coefficients
of .94 and .95 were reported for adolescent's report on
dyad ("Appraisal of the Dyad”) and adolescent's report

on mother ("Appraisal of the Other") respectively.

Wares

Participants were recruited via a standard
Departmental sign-up sheet requesting male and female
participants between the ages of 18 and 20 whose

biological parents are currently married. Participants

22

could sign up for one of four scheduled research
sessions. The study was entitled "Thinking Styles of
Parents and their Children". Students were given three
credits for their participation and six credits (total)
if both of their parents also participated. These
credits can be used to fulfil requirements in the
introductory psychology classes in which these students
were enrolled.

Students were given a research packet consisting of
all previously mentioned scales in the following order:
the Constructive Thinking Inventory, the Conflict
Behavior Questionnaire, demographic questions, and the
"Time Spent" scale. Theycompleted all instruments
during a two hour session in a classroom setting with
other participants and an experimenter present. All
students finished within the two hours allotted with
many students finishing in approximately one and a half
hours.

Participation for students also included writing a
standardized letter to their parents and requesting that
they each fill out the Constructive Thinking Inventory
(CTI) separately and mail it back (see Appendix A).
These handwritten letters were collected and mailed by
the experimenter with two informed consent forms in
envelopes addressed by the students. Students were
given a "debriefing" form, which provided general

information about the study, after they returned all

23

materials to the experimenter.

Parents were contacted only by this written
communication. They received the following materials:
their student-child's letter, two informed consent
forms, two copies of the Constructive Thinking
Inventory, and a pre-stamped addressed envelope in which
to return the completed CTI's. Approximately half of
the students elected to hand-deliver the forms; thus no
stamp was included. All instructions were included in

the letter written by their child.

Qata_Analxses

Correlations were computed to assess the
relationship between gender (of both parent and
adolescent child) and CTI scores. Tests of mean
differences (Lrtests) were calculated to investigate the
differences between mothers and fathers for hypotheses
1, 5, 8, and 9; also for the differences between
daughters and sons for hypotheses 2, 3, and 5. The
discrepancy between the scores simulated for parents and
actual performance of parents on the CTI were correlated
with the variable of parent-child conflict to test
hypothesis 4. Students' scores were correlated with
both the simulated and actual CTI scores of parents to

test hypotheses 5 and 7.

24

Results

Precautions were taken to limit errors in this data
set. ‘All response forms which omitted the last item
were excluded out of concern that other responses might
be flawed due to the participant having lost track of
the questions. In total, materials from eight
participants (sets of student-child and parents) were
excluded. In addition, postal returns were analyzed
separately from those which students returned by hand
because students motivated to receive credit might
simply complete each parent questionnaire and return
them without involvement of either parent. To help
ensure that the data which students hand-delivered were,
in fact, completed by parents, a comparison of parents'
mean scores on all scales was performed. There were no
significant differences between mean CTI scores of
parents' postal returns versus those which students
hand-delivered. Therefore, both data sets were pooled
and all subsequent analyses were performed on this pool.

Table 1 provides an overview of student responses
to the several measures as well as their parents' data.
The conservative two-tailed test of statistical
significance was used to assess all differences among
these means and also among the pertinent correlations.
The subsequent paragraphs deal sequentially with this

study's hypotheses.

25

m.o~ m.m~
s.s~ o.o~
n.m~ n.e~
e.s~ o.o~
s.- H.n~
s.sn o.s~
m.sv «.ev
o.om «.mm
neon use:
amaaquHm

uzom

Icoaueaom use H80 ou.non:omnmm nousnsawmlucouem use ocuuonouruaom no memo:

n.bN
m.v~
«.mm
N.mN
N.nN
o.Nn
m.bv
h.Nm

Ame
mean

v.0"
m.n~
n.mN
N.vn
m.NN
Hymn
h.bv

o.om

Ian—v

mac:

m.20m

o.hN n.QN
H.wN a.v~
w.v~ m.m~
o.hfl 5.QN
O.NN m.HN
H.nn b.0N
h.mv n.mv
o.nm v.ao
mono nso:
DNB¢ADZHQ

ImmBEUDCQ

m.s~ n.a~_
m.v~ s.n~
s.e~ e.m~
s.m~ o.w~
o.n~ o.n~
.s.sn n.s~
n.oe e.sv
n.em e.oo
.es a so

mean use:
m.mma=o=¢o

m.»
w.o
v.n
m.n
«.0
m.o
v.0n
v.m~
m.mn
b.0N
h.HN
o.hN
wsmv

v.hm

m.a
m.m
o.v
Q.N
".5
0.0
H.Hn

v.wN

can no Henweunmd
no: no Hemwenmng
can nuas vownucoo
no: and: vodducoo

can saw: on“?

so: mad: ends
messages o>auaouz
.uxea sensuoeouao.

o.o~.uxsa nsoauqumuomsm

m.hN
0.HN
H.mN

o.vv

H.vo.

smwmwuno o>aez
asses~u>
Heeoduosu
Heu0a>eaom

o>duosuum=oo

aflfl U CV “W5 I fly
meow mumunmsen

mBmomwmlhflum

mmbmcfiz

muonsu«o:H mama
.H canes

26

Mildly supporting the initial hypothesis that
mothers would be viewed less favorably than fathers,
this sample of 118 students with cooperating parents
viewed mothers as differing significantly from fathers,
relative to each parent's actual performance, only on
Naive Optimism. Contrary to expectations, however,
Naive Optimism was the sole CTI scale on which these
students rated their fathers significantly more
favorably than mothers (see Table 2). In total, mothers
were rated less favorably on five of the eight CTI
scales and more favorably than fathers on three. A
comparison of daughters' and sons' direct ratings of
parents, independent of parents' performance, will be
reviewed later.

Table 2. Means and trtests of Differences Between
CTI Discrepancy Scores (parent's actual

score minus student's simulation) for
Mothers and Fathers.

CTI Scales : Mothers : Fathers :trvalue

:Discrepancleiscrepancy:

: n = 118 l n = 118 i
------------ +-------—---+-----------+------------
Constructive: 2.01 l .99 l .68

I I I

I I I
Behavioral : -.04 l -.20 I 17

I I I

I I I
Emotional : -.10 l -.51 l .40

I I I

I I I
Validity :. .98 : 1 18 : - 81

i l i
Naive Optim : -3.37 i -2.32 :-2.05*

I I I

I I I
Superstitions -.54 i -.11 :-1.00

I I I

I I I
Categorical : -l.73 : -1.81 : .13

I I I

I I I
Negative : -.39 : -.87 : .48

* p g .05, two tailed test

27

Clearly refuting the second hypothesis, the present
findings directly conflict with Epstein and Lee's report
(Epstein, in press) that college daughters, but not
sons, described their fathers as significantly above
their mothers for global Constructive Thinking.

As is shown in Table 3, here it was sons, not daughters,
who depicted their mothers as significantly below their

mothers' actual performance on Constructive Thinking.

Table 3. Correlated trtests of Differences Between
the Means of Parents' Self-reports and Their Responses
as Projected by their Children on the CTI.

CTI :Father vs. :Father vs. :Mother vs. :Mother vs.
Scales :Dau-Projec :Son-Projec :Dau—Projec :Son-Projec
ESQEEEZ‘E-"TE ----- inf]; ----- 3”"?13 ----- EHETEI;"-
Behavior g -.42 g .00 E -.87 g 1.21
Emotional? -1.59 g 1.05 i -1.37 E 1.61
Validity é e.14*** g 1 24 g 6.09*** i - 58
Naive Op i -3.Bs*** E -3.60*** E -5.34*** g -5.77***
Supersti g -.21 i -.30 i -.57 g -1.81
Categoricé -2.02* E -3.98*** i -1.87 i -3.1e**
Negative 3 -.30 E -2.33# E .02 E -1.60

*** 2,; .001
** p_$ .01
x n g .05

28

Tables 4 and 5 compare daughters' and sons' CTI
parent simulations for their mothers and fathers with
reference to each parents’ selfereports. Further
conflicting with Epstein and Lee, these discrepancy
scores revealed that the present sons depicted their
mothers less favorably than did daughters on Emotional
Coping (t = —2.11, 2,; .05) and Naive Optimism (L =
2.02, 2 g .05) and as almost significantly lower on
total Constructive Thinking (L = -1.93, 2,; .06). Only
on the brief Validity scale were the sons' discrepancy
scores (mothers 1 = 4.04, 2,; .01; fathers L = 2.44, 2,;
.05) more favorable than daughters'.

Table 4. Means, Standard Deviations, and t—tests of
CTI Differences Between Daughters' and Sons'

Discrepancy Scores (mother's actual score
minus student's simulation) for Mothers.

CTI Scales IDaughters I Sons IL-value
' n = 76 I n = 42 I
........................ +--_--_---_--+------------
Constructive 24(13.86)I 5 21(12.96)I-1 93
I

Behavioral -.83 (8.31): 1.38 (7.42)I-1.44

Emotional 1.32 (8.40)§ 2.10 (8.44)I—2.11*
1.64 (2.35): -.21 (2.47): 4.04::
Naive Optim -2.72 <4.45)§-4.55 (5.11): 2.02*
Superstition -.28 (4.26)§-1.02 (3.67): .96

Categorical-

I.

l

l

Validity I

E

i

s

I .

:-1.18 (5.53)I-2.71 (5.56): 1.44

: I
i

I
Negative .01 (5.38):-1.12 (4.53): 1.16
** a g .01, two-tailed test I
** 2,; .05, two-tailed test

+ Standard deviations in parentheses.

' Emotional

29

Table 5. Means, Standard Deviations, and trtests of CTI
Differences Between Daughters' and Sons‘
Discrepancy Scores (parent's actual score
minus student's simulation) for Fathers.

CTI Scales :Daughters
' n = 76

Sons :L-value
n = 42 I

63;;2;3321;;
Behavioral -.32 (8.63): .00 (6.18)§-.25
-1.40 (7.64): 1.10 (6.74)i-1.77
Naive Optim -2.18 (5.20)I-2.57 (4.83)§ .40
Superstition -.09 (3.90):-1.43 (3.13)§ .07

Categorical

i
I
i
Validity E 1.57 (2.22): .47 (2.49)§ 2.44*
i
5
:-1.26 (5.45):-2.79 (4.54)! 1.54

I
Negative -.16 (4.53)I-1.60 (4.44): 1.86

** p_$ .01, two-tailed test
* 2,3 .05, two-tailed test

+ Standard deviations in parentheses.

Tables 6 and 7 show comparisons of daughters' and
sons' simulations of their mothers and fathers,
independent of parent scores, a comparison that is
similar to Epstein and Lee's (Epstein, in press). As
can be seen, the apparent differences listed above
between sons and daughters dissipated. The only
significant difference that remained between sons' and
daughters' ratings was the sons' higher Validity

ratings of their mothers.

30

Table 6. Means, Standard Deviations, and Lrtests of
Differences Between Sons' and Daughters'
Mother-Simulated CTI Scores.

CTI Scales :Daughters : Sons -:t-value

I n = 76 I n = 42 I
------------ +--—---------+------------+---—--------
Constructive: 88.38(12.22) 85.43(10.72) 1.31

I I I

I I I
Behavioral I 48.38 (6.97) 46.29 (5.82) 1.66

I I I
Emotional : 28.66 (8.74) 26.95 (6.83) 1.09

I I I
Validity I 21.90 (2.20) 23 12 (2.04)-2.94**

I I I

I I I
Naive Optim I 28.71 (3.16) 28 76 (2.80) -.09

I I I
Superstitions 25.92 (3.16) 26 29 (3.52) -.58

I I I
Categorical I 24.88 (4.89) 26.62 (4.32)-1.93

I I I

I I I
Negative I 28.26 (4.45) 29.48 (3.40)-l.54

** 2,; .01

+ Standard deviations in parentheses.

Table 7. Means, Standard Deviations, and Lrtests
of Differences Between Sons' and Daughters'
Father-Simulated CTI Scores.

Sons It-value

CTI Scales :Daughters
' n = 42 I

n = 76

Constructive

Behavioral

Emotional

I

I

I

I

I

I

I

I

I

Validity I 22.03 (1.9

I

I
Naive Optim I
I

s

I

I

I

I

I

Superstition 24.77 (2.5 24 31 (2.82)-1 O5

Categorical 28.13 (5.6 27 89 (4.50)-1 55
I

Negative 27.80 (3.3 28 90 (3.58)-1 89

+ Standard deviations in parentheses.

31

Overall, the present student sample, and especially
the sons, tended to underestimate their parents'
Constructive Thinking capabilities according to the
data of Table 1. Among 16 comparisons of son-simulated
parental CTI mean scores with parents' own responses,
only sons' mean mother-simulated Validity scores
favorably exceeded the parents' self-descriptions, and
then nonsignificantly. In contrast, daughters' parent-
simnlated responses were more favorable (us) than
parents' own responses in 7 of 16 comparisons.

As shown earlier in Table 3, daughter-simulated
responses were significantly less favorable than each
parent's actual responses for Naive Optimism, Validity
and, for fathers only, on Categorical Thinking. Son-
simulated responses were significantly less favorable
than the parents' actual score in the following six
cases: Naive Optimism, Categorical Thinking, mothers'
global Constructive Thinking, and fathers' Negative
Thinking.

The pertinent Table 1 means indicate that sons
reported more conflict with mothers than did daughters
(L = -2.18, 2,; .05), contrary to hypothesis 3. Sons
and daughters reported about equal amounts of conflict
with father and their evaluative perceptions of each
parent on the "Appraisal of the Other" scale did not
differ significantly. Daughters reported significantly
less conflict (1 = 2.64, 2,: .01) with mothers than

32

fathers, consistent with their report of a more negative
view of fathers than mothers (1 = -2.02, 2,; .05). In
contrast, son's reported no interparent distinction on
either “Appraisal of the Dyad" or "Appraisal of the
Other". For sons, there were no significant differences
in the amount of conflict with each parent, or in their
related evaluative perceptions of each parent.

The correlations between student-reported conflict
with parents and discrepancies between student-simulated
and parents' actual CTI scores are shown in Table 8.
Largely supporting hypothesis four, these findings show
that mother-student conflict correlated positively and
significantly (1 = .36, 2,; .01) with the direction and
amount of discrepancy between mothers' actual and
student-simulated CTI responses for daughters, while
sons' parallel correlation also approached significance
(1 = .30). Similar discrepancies also correlated
positively but somewhat less firmly with father-student
conflict (daughters' 2 = .21, 2:; sons' 1_= .25, as).
Daughter-reported conflict with their mothers and
fathers also correlated positively (; = .23, 2:; .05),
similar to sons' parallel correlation (1': .29).

Separately for sons and daughters, all correlations
between students' scores, parents' scores, and parent-
simulated scores on the Constructive Thinking scale are
shown in Table 9. Sons' global Constructive Thinking

correlated positively with the parallel scores of both

33

their fathers (1 = .40, 2 ; .01) and mothers (2 = .24, 2
; .07), partially supporting the fifth hypothesis.
Surprisingly, however, daughters' global Constructive
Thinking responses failed to associate with either

parent's parallel scores (fathers' ; = -.02; mothers' 1

= .03)

. Table 8. Correlations Among CTI Discrepancy Scores
(between student-simulations and parents‘ actual scores)
and Parent—conflict.

Discrepancy Discrepancy FATHER MOTHER
with mothers with fathers Confl. Confl.

DISCREP .05 .02 .38**
MOTHER

DISCREP .11 .21 .17
FATHER

FATHER .11 .25 .23*
CONFL.

MOTHER .30 .05 .29
CONFL.

** 2,; .01, two-tailed test

* 2 ; .05, two-tailed test

Note: Correlations for 78 daughters above the diagonal;
correlations for 42 sons below it.

34

Table 9. Correlations Among Students' Scores, Parent
Scores, and Student-Simulated Scores for Parents on the
CTI's Constructive Thinking Scale.

Sons Simulat. Simulat. Fathers Mothers
Fathers Mothers Actual Actual

Simulat. .33* .30** .25* .17
Fathers

Simulat. .25 .22 .22 .32**
Mothers -

Fathers .40** .47** .05 .12
Actual

Mothers .24 .00 .21 -.07
Actual

** 2,; .01, two-tailed test

* 2,; .05, two-tailed test

Note: Correlations for 76 daughters given above the
diagonal; correlations for 42 sons given below it.

35

Daughters' global Constructive Thinking scores did
correlate positively and significantly with their
simulated responses for each parent (fathers g = .39, 2'
; .01; mothers 1,: .40, 2 ; .01); the parallel
correlations for sons' (fathers 2 = .33; mothers 2 =

.25, as) were somewhat lower. Daughter-simulated mother

and father responses also correlated positively and
significantly with parents' actual scores (fathers ; =
.25, 2M; .05; mothers 2 = .32, 2,; .01), much like son-
simulated mother and father responses (fathers n = .47,
2 ; .01; mothers 2 = .21, as). The .30 correlation
between daughters simulated responses for each parent
was significant (2,; .01), but the sons’ parallel value
was not (1 = .22). Overall, the CTI responses of these
daughters were significantly more similar (according to
2 to a transformation) to their parent-simulations

.40) than to each parent's

(mothers 2 = .39, fathers ;
actual responses (mothers 1 = .03, fathers ; = -.02).

This pattern did not occur among sons, however, as only
small differences were found between the correlations of
sons' scores with their parents' actual scores (fathers

.40; mothers 2 = .24) and the parallel correlation

I.
with son-simulated parent scores (fathers z = .33;
mothers 2 = .25).

When parent-student correlations were examined
across the set of eight scales, students' own responses

were closer to those of the same-sex than the opposite-

 

36

sex parent. Thus, daughters' responses correlated
positively and significantly with their mothers'
reaponses on four CTI scales (Emotional Coping, 2 =
.22, 2,; .05); Superstitions Thinking, 2,: .25, 2,; .05;
Validity, 2 = .27, 2 ; .01; and Categorical Thinking, 2

= .36, 2 ; .01). However, merely one parallel

mother/son correlation was significant (Behavioral
Coping 2 = .28, 2,; .05). On the other hand, sons' CTI
responses resembled their fathers' more than did
daughters'. Sons' scores correlated positively and
significantly with fathers' scores on five scales:
Constructive Thinking (2 = .40, 2M; .01), Behavioral
Coping (2 = .55, 2 ; .01), Emotional Coping (2 = .31, 2
; .05), Negative Thinking (2 = .30, 2,;,.05), and
Validity (2,: .34, 2M; .05). Across sexes, merely one
parallel father/daughter correlation was significant
(Validity 2 = .34, 2,; .01).

On the Conflict Behavior Questionnaire, daughters
depicted their fathers as behaving more negatively than
their mothers (2 = 2.02, 2,; .05). Sons made no
parallel interparent distinction, although they tended
to view each parent more negatively than did daughters.
However, this difference between sons' and daughters'
perceptions of parents did not attain statistical
significance, requiring rejection of hypothesis 6.

Both sons' and daughters' global Constructive

Thinking scores correlated significantly with their

37

father-simulations. For mothers, the correlation
between daughters' scores and daughter-simulated
performance was statistically significant (2,; .05)
and sons' parallel correlation approached significance.
Daughters' responses tended to correlate more highly

with those which they simulated for parents (mean 2 =
.40) than did sons' (mean 2 = .29), although the

difference between these correlations was not
significant. Thus, hypothesis #7 was rejected.

The following significant sex differences partially
supported hypothesis #8:

a. Fathers scored significantly more favorably than
mothers for global Constructive Thinking (2 =
3.83, 2 ; .05) and Emotional Coping (1,: 5.14, 2,;
.01).

b. Mothers scored significantly less favorably than
fathers on Categorical Thinking (L = 2.15, 2 ;
.05).

c. Male students scored significantly more favorably
than female students on Emotional Coping (L =
2.63, 2,; .01).

Daughters, but not sons, reported spending
significantly more time with their mothers than their
fathers (2,: 8.31, 2,;,.01). In a comparison of sons
and daughters, daughters reported spending significantly
(2 < .01) more time with each parent than did sons.

These findings supported hypothesis 9.

38

Discussion

WW
Often there are problems in generalizing to the
broader population from research featuring college

students. The present study is relatively broad in

including both college students and parents. However,
this sample is also quite distinct. All families had a
child in college and were apparently intact at the time
of this study. In addition, the students were asked to
enlist their parent's participation. Thus, this sample
likely represents a relatively cooperative, less
conflictual, and economically advantaged set of American
families. Mean scores for the present sample on the
Conflict Behavior Questionnaire (CBQ) and Constructive
Thinking Inventory (CTI) were compared with previously
published means. On all scales the mean scores of both
sons and daughters, as well as those of their mothers
and fathers, were within one standard deviation of the
parallel means for Epstein's (1987) sample of 124
Massachusetts undergraduate volunteers. Additionally,
the present students' mean CTI scores were also within
one standard deviation (in 15 of 16 comparisons) of
Hurley's (1990) sample of 103 Michigan State University
undergraduates enrolled in a psychology course featuring
small group experience. The lone exception was that all

present participant groups other than fathers scored

39

more than one standard deviation above Hurley's (1990)
sample for Superstitions Thinking. In summary, the
present data are consistent with those reported in
earlier research and, therefore, appear reasonably

representative of contemporary college samples.

Eatingscfmants

Somewhat unexpected was the finding that these
college-student children had relatively unfavorable
views of the thinking of both their fathers and mothers,
in the sense that these parents were underrated on
several constructive modes of thinking yet were never
significantly overrated. Students did, however, rate
parents more favorably than students' own self-reports
on all scales. Fathers were underrated (depicted less
favorably than their actual performance) by both
children on Naive Optimism and Categorical Thinking, by
daughters only for Validity, and by sons only for
Negative Thinking. Mothers were rated below their
actual performance by both daughters and sons for Naive
Optimism and Categorical Thinking, by daughters only for
Validity, and by sons only for global Constructive
Thinking. Thus, both sons and daughters underestimated
their mothers and fathers about equally.

It had been hypothesized that, as in Epstein's
report (in press), mothers' cognitive responses would be

depicted less favorably depicted than fathers'

40

(hypothesis 1) and, because daughters are closer than
sons to mothers, that this negative bias would be more
pronounced among daughters than among sons (hypothesis
2). It was further thought that the recent separation
of these college-age daughters from their families would

heighten mother-daughter conflict. This transition

period was considered likely to involve a higher degree
of conflict in the close and more intense mother-
daughter dyad than in the other parent-child dyads
because the mother-daughter unit requires the most
change as daughters exit from home. The present
daughters, however, did not show this negative bias
toward their mothers.

Methodology differences between the present study
and that of Epstein and Lee (Epstein, in press) merit
attention. Epstein and Lee examined the CTI scores
which students simulated for their parents and found
that students' father-simulated responses were
significantly more favorable than their mother—
simulations. The present sample of daughters and sons
also rated their fathers more favorably than their
mothers on virtually all scales. However, when the
scores of the parents are considered by comparing
simulations with the parents actual performance, the
consistent underrating of mothers tend to dissipate. As
stated earlier, only for Naive Optimism were mothers

rated significantly less favorably than fathers,

41

relative to each parents actual performance.

Perhaps these daughters were not actually
separating from their mothers in ways that heightened
mother-daughter conflict. The average amount of time
that daughters reported spending with mothers each month

(via phone calls and visits) was quite high--three to

five contacts in addition to one letter. Upon
examination of the relationship between age of the
student and degree to which that student underrated
parents, it was found that the older the student, the
more likely the father was to be underrated. This
relationship did not, however, hold true for mothers.
The correlation between age and rating of father was
significant for both daughters (2,: .22, 2 ;,.05) and
sons (2 = .44, 2,; .01). In addition, as the reported
“Time Spent" with father increased, the student-
simulated parental Constructive Thinking responses
tended to become more favorable (son's 2 = .35,
daughter's 2 = .11).

Perhaps separation of the mother and daughter
becomes more pronounced at ages later than the present
sample's mean of 19 years, with the subsequent change in
the daughter's perception of mother occurring
concurrently. This sample did not include a sufficient
number of participants of different ages to permit a
meaningful comparison of groups by age and sex.

Another possible explanation for the absence of low

42

mother—simulated CTI scores by daughters is that the
mother-daughter bond explicated in the introduction
serves to hinder, rather than to facilitate, a negative
view of mother on the part of daughter. Daughters,
through their stronger identification with mother, may

come to perceive mother as very similar to themselves.

It is likely beneficial, therefore, that they perceive
this parent in a favorable light since self-serving
biases have been linked with both constructive thinking
(Hurley, 1990) and mental health (Taylor & Brown, 1988).

The present sons underrated their mothers for
Constructive Thinking as much as daughters did.
Stereotypes regarding women may partially explain these
generally lowered simulated scores for mothers. Perhaps
these students developed a negatively-biased attitude
toward women with regard to rationality and constructive
thinking. Women have traditionally been viewed as less
rational and more emotional than men (Broverman, 1972;
James, 1973).

Sons underrated mothers on Emotional Coping and
Naive Optimism more than did daughters as shown in Table
2. These scales may tap areas in which women have been
perceived to be less efficient. Perhaps the male
students were more prone to perceive their mothers in a
stereotyped manner. Only on the Validity scale did
daughters underrate parents more than did sons. The

Validity scale items seem clearly associated with

43

attention and comprehension as illustrated by the
following items: "I have never seen anyone with blue
eyes." and ”I feel that people who wear glasses usually
can see better without their glasses.“ Daughters seemed
to believe, more than did sons, that their parents would

respond less carefully.

Sons', but not daughters' global Constructive
Thinking scores correlated significantly with fathers'
aetual performance (sons' 2 = .40; daughters' 2 = -.02)
Table 8). A similar pattern held for mothers' (sons' 2
= .24; daughters' 2 = .03). These are consistent with
Whitbeck's (1987) finding that parental efficacy was
more salient to sons than to daughters. These sons had
apparently acquired the thinking style of their parents
more fully than did daughters. Daughters, however, may
be modelling their parents' behavior by emulating the
thinking style which they perceive their parents to
hold, as daughters' own repsonses consistently
correlated higher with their parent-simulated responses
than with their parents' actual responses.

Also notable was the finding that the global
Constructive Thinking responses which both daughters and
sons simulated for their parents correlated with the
students' own global responses (daughter-mother 2,: .40,
daughter-father 2 = .39; son-mother 2 2 .25, son-father
2 = .33) as highly as with the parents' actual scores.

Thus, these students may be attributing to their parents

44

a style that is closer to their self-image than to an
independent and objective view of their parents'

behavior.

Sexlliffarsncascntham

It was unexpected to find that the present sample

of fathers scored higher than mothers for Constructive
Thinking and Emotional Coping because gender differences
on the CTI had received little prior attention (Epstein
& Meier, 1989). As shown by their parent-simulated
responses, this difference was also perceived by
students. Both daughters and sons generally rated their
mothers less favorably than fathers on these scales.
Unexpected also were the findings that sons scored
significantly higher on Emotional Coping than daughters
and that mothers scored significantly higher than
fathers on Categorical Thinking. No sex differences
were,reported in Epstein and Lee's similar study
of 79 college students and their parents. Epstein and
Meier's (1989) study of 122 college students reported
only one sex difference, that men were significantly
higher than women for Categorical Thinking. In the
present study the opposite was found; women scored
significantly higher than men for Categorical Thinking.
These sex differences defy a satisfying explanation and
may be due to sampling issues.

It might be assumed that these gender differences

45

are unique to this sample. However, the consistency
with which both fathers and sons scored higher than the
participating mothers and daughters for Emotional Coping
suggests that more than chance was involved.
Parggt-Child Qonfiligt

Wholly unexpected was the significantly higher
amount of mother-child conflict reported by sons than
daughters. These college-age sons and daughters
reported about equal amounts of father-child conflict
and their overall evaluative perceptions (positive-
negative) of each parent were also comparable. Why
would sons, rather than daughters, report a higher
amount of conflict with mother? Since daughters spend
more time with mothers and are generally believed to
have a more intense bond with mothers, it is puzzling
that daughters did not report more maternal conflict
during this age in which separation from parents is a
notable issue. These daughters did report spending
significantly more time with their mothers than did the
sons. Perhaps, as suggested earlier, daughters spent
sufficient time with mother to allay uncertainties
arising during this transition period. Sons, on the
other hand, reported spending significantly less time
with mother than did daughters. Perhaps the high degree
of conflict between sons and mothers reflects this
lowered degree of contact. Perhaps sons separate from

the family at an earlier age than daughters, with

46

increased conflict ensuing.

In addition to reporting less conflict with mothers
than did sons, the present daughters reported less
conflict with their mothers than with their fathers.
Again, perhaps the closeness with mother precludes the

display of overt conflict during this transition period.

The father-daughter relationship, conversely, has been
reported to be particularly bland and distant
(Steinberg, 1987). Perhaps this less intense
relationship is less able to withstand the inevitable
changes in the parent-child relationship than the more
intense mother-daughter relationship.

The present daughters appeared to hold more
distinct views of mothers and fathers and also of their
parent-child relationships than did sons. Sons depicted
the thinking of their father and mother as more similar.
The differences between daughters' simulations of the
thinking of their fathers and mothers surpassed sons'
parallel differences. This seems as expected, since
relationship variables have been found more salient to
daughters than to sons (Whitbeck, 1987). Hazzard,
Christian, and Margolin (1983) theorized that perceived
similarity among family members would be most beneficial
because it signifies parental harmony. They found that
children from distressed families perceived their
parents as more dissimilar than did children from

nondistressed families.

47

mmmmammum
Discrepancy scores were calculated by subtracting
the students' parent-simulated scores from the parents'
actual scores. Such discrepancies could be either
positive or negative, the meaning of the sign depended

on whether the pertinent scale addressed a thinking

modality which was constructive or nonconstructive. For
example, a positive discrepancy score resulted if the
student had depicted a parent 1225.222222212 than that
parents' actual performance on any of the four scales
featuring constructive modes of thinking (global
Constructive Thinking, Emotional Coping, Behavioral
Coping, and Validity) or 2222 {21223211,than the
parents' actual performance on a scale from the
nonconstructive quartet (Categorical Thinking,
Superstitions Thinking, Naive Optimism, and Negative
Thinking).

Only mother—student conflict was found
significantly associated with discrepancies between
student-simulated and parents’ actual CTI scores. The
more conflict that students reported experiencing with
their mothers, the more likely they were to underrate
her for global Constructive Thinking (sons 2 = .30, 2,;
.06; daughters 2 = .36, 2,; .01). Father-child
conflict, in contrast, appeared less related to such
discrepancies (sons 2 = .25, 2:; daughters 2 = .21, 22).

Although the differences between these correlations for

48

mothers and fathers were not significant, one might ask,
what is it about mother-child conflict that selectively
influences the child's perception of mother's thinking
style? Why were father-perceptions not similarly .
affected?

Perhaps mothers show more emotion during conflicts

with their children. Displays of emotion are generally
more socially acceptable for women. Taking this one
step further, emotionality may be seen as the opposite
of rationality. Thus, in conflicts with children,
mothers may come to be viewed as less rational.

Fathers, on the other hand, may show less emotion in
such conflicts. Thus, increasing conflict might be less
likely to interfere with father's appearance as the

more rational parent.

mm

The family is an evolving and dynamic system.
Thus, the age at which one taps into this ever-changing
system would likely affect the findings. That the work
of O'Neil and Reiss (1980) on attitudes toward mothers
found no differences in adult childrens' perceptions of
mother, while the present study found numerous
differences in such perceptions, suggests the need to
study these differences with varied age groups. For
example, the present students underrated their parents

in a majority of cases; would this be found in a sample

49

of adult children; of young children? There is, in fact
a children's version of the CTI that could be used in
investigation of such questions.

Gender differences on the Emotional Coping scale
appeared with some consistency in the present study.
Item analysis of the CTI comparing responses of males
and females seems necessary to ascertain if certain
items on the scale are gender—dependent.

Consistent with Epstein and Lee's research
(Epstein, in press), the present daughters resembled
their mothers on Superstitions and Categorical Thinking.
The absence of similar correlations on four of the CTI's
eight scales suggests that this replicated finding
may be reflective of an interactive style unique to the
mother-daughter dyad. Further research comparing
parents and children on the scales of the CTI will help
to elucidate whether this pattern of performance was

simply a chance occurrence, or whether mothers and

daughters do in fact differ from the father-son dyad in
transmission of these aspects of thought.

Finally, the differences between sons' and
daughters' perceptions of each parent were numerous. Do
gender stereotypes play a role in these perceptions?
Research using the CTI and measures of attitudes on
masculinity/femininity and gender roles would provide

further information on this important issue.

APPENDIX A

50

APPENDIX A

Standard deviations of self-reported and parent-simulated
responses to CTI and Relationship indicators.

—---+ ----------- + ----- + ----- + ----------- + ----------- + ----------- +
MEA-ISelf—reports DAUGHTER'SIDAU-SIHULATI SON'S I SON-SIM I
SUREI DAU I SON I HOHSI DADSI HOKSI DADSIHOHS I DADSINOHS DADS I
-—--+ ----- + ----- + ----- + ----- + ----- + ----- + ----- + ----- + ----- + ----- +
ConsI10.57I 8.23:11.21: 7.37:12.22: 9.99: 9.90: 9.03:1o.72:1o.99:
I I I I I I I I I I I
BehaI 6.31I 5.58I 6.34I 4.83I 3.97I 5.83I 6.01I 3.21I .82I 7.29I
I I I I I I I I I I I
I I I I I I I I I I I
Emo I 7.15I 6.30I 7.85I 5 31I 8.74I 8 70I 6.77I 5.55I 3.83I 5.53I
I I I I I I I I I I I
I I I I I I I I I I I
Val I 1.88I 2.96I 1.68I 1 70I 2.20I 1.94I 1.96I 2.20I 2.04I 2.03I
I I I I I I I I I I I
I I I I I I I I I I I
Nai I 3.52I 3.21I 3.80I 3.80I 3.13I 3.37I 4 39I 4.08I 2.80I 2.53I
I I I I I I I I I ' I I
I I I I I I I I I I I
Sup I 3.41I 3.73I 3.37I 3 27I 3.18I 2.53I 3 III 3.33I 3.52I 2.82I
I I I I I I I I I I I
I I I I I I I I I I I
Cat I 3.77I 4.88I 5.07I 4 30I 4.89I 5.61I 3.85I 4.53I 4.32I 4.50I
I I I I I I I I I I I
I I I I I I I I I I I
Neg : 4.14: 3.90: 4.39: 3.34: 4.45: 3.31: 3.64: 3.44: 3.40: 3.56:
I I I I I I I I I I I
I I I I I I I I I
McflI 3.47: 2.63: : : : : I I : :
. : : : : : : : : : :
FcflI 3.75: 2.02: : : : : : : : :
I I I I I I I I I I I
I I I I I I I
MperI 9.74: 7.15: I I I : : I : :
: : : : : : : : : : :
Fper: 9.28I 7.09: I I I I I I I I
----+ ----- + ----- + ----- + ----- + ----- + ----- + ----- + ----- + ----- + ----- +

APPENDIX B

51

APPENDIX B

Letter to Parents

Dear Mom and Dad,

I am participating in a research study as one way
to gain additional credit toward the grade in a
psychology course. I chose this option rather than the
other alternatives available to me. This study compares
the way parents and children repond to questions about
ways of thinking. I have filled out this questionnaire
and the study will compare my responses with yours on
it. Please complete the brief 64-item questionnaire
separately and return the answer sheets to the
researcher in the enclosed envelope (Please don't mail
back the actual questionnaire, just the answer sheet).
My participation and yours is voluntary. You may choose
not to participate. We have been requested to not
discuss our responses to these questions until after
everyone has completed it, although some of the answers
may seem humorous. Also, each of you need to read the
"informed consent" sheet attached to the questionnaire.

Thanks for your help with this.

APPENDIX C

Your year

52

APPENDIX C

Demographic Items

in school

A = Freshman
B = Sophomore
C = Junior
D 2 Senior
Your age
A = 18
B = 19
C = 20
Sex
A = female
8 = male

Your mother's educational attainment

recount»
II" "II"

grammar school
junior high school
high school
college

graduate school

Your father's educational attainment

A = grammar school
B = junior high school
C = high school
D = college
E = graduate school
Number of siblings
A = 0
B = 1
C = 2
D = 3
E = 4 or more
Birth order
A = you are an only child
8 = you are the first born child
C = you are the second born child
D = you are the third born child -
E = you are the fourth (or later born) child

Does your
A yes
8 no

mother work outside the home?

53

9 Does your father work outside the home?
. A = yes
3:

10 On the average, how many hours did you spend with
your mother per week when you lived at home during high
school?

0

1-2 hours
3-4 hours
5-7 hours
8 or more hours

NUOWD
II II II II

11 On the average, how many hours did you spend with

your father per week when you lived at home during high
school?

A = 0

B = 1-2 hours

C = 3-4 hours

D = 5-7 hours

E = 8 or more hours

12 Currently, how many times do you speak to your
mother on the phone each month?

A = 0

B = 1-2

C = 3—5

D = 6-8

E = 9 or more

13 Currently, how many times do you speak to your father
on the phone each month?

A = 0
B = 1-2
C = 3-5
D = 6-8
B = 9 or more
14 Currently, how often do you see your mother each
month?
A = 0
B = 1 to 2 times
C = 3 to 5 times
D = 6 to 8 times
E = 9 or more times
15 Currently, how often do you see your father each
month?
A = 0
B = 1 to 2 times
C = 3 to 5 times
D = 6 to 8 times
E = 9 or more times

54

16 Approximately how many letters per month do you write
to your mother?
0

NCO!”
II II II II II

1
2
3-5
6 or more

17 Approximately how many letters per month do you write
to your father?

0

WOOD!
II II II II I

1
2
3-5
6 or more

APPENDIX D

55

APPENDIX D

Student's Informed Consent Form

The purpose and procedures involved in this study on
"Thinking Styles of Parents and Their Children" have
been verbally explained to me. I agree to participate
in this study and to write a letter to my parents
requesting their participation. I understand that no
identifying information will be attached to the answers
I provide on any questionnaires. I realize that my
participation is voluntary and that I may refuse to
answer certain questions and/or withdraw from the study
at any time without penalty.

APPENDIX E

56

APPENDIX E

Debriefing Sheet

As you are aware, although your participation is now
complete, this study is not yet concluded. After
responses from parents are collected we will compare
your responses on the Constructive Thinking Inventory
(CTI) with those of your parents. We are interested in
investigating the similarities and differences between
the CTI performance of parents and their children. This
study is also designed to assess the perceptions that
late adolescents have of their parents, and it is for
this reason that you were asked to simulate your
parent's responses on the measure. Constructive
thinking is a relatively new construct that is sometimes
referred to as "practical intelligence". If interested,
you may find information on this topic in:

Epstein, S. & Meier, P. (1989). Constructive thinking: A
broad coping variable with specific components.

Wflwmmmfilm).
332-3 .

Principal investigator: Helene Lensky
485-0944

APPENDIX F

57

APPENDIX F

Parent's Informed Consent Form

The purpose and procedures involved in this study on
thinking styles have been explained to me by my child.

I agree to participate in this study. I understand that
no identifying information will be attached to the

answers I provide on any questionnaires. I realize that
my participation is voluntary and that I may refuse to

answer certain questions and/or withdraw from the study
at any time without penalty to myself or my child.

I indicate my voluntary agreement to participate by

completing and returning this questionnaire. (This form
is for you to keep.)

Principal Investigator: Helene Lensky
(517) 485-0944

LI ST 0F REFERENCES

58

LIST OF REFERENCES

Back A '1‘ (1976) mmmmammai
2122222222,New York: International Universities
Press.

Burns, D. D., Shaw, B. F., & Crooker, W. (1987). Case
histories and shorter communications. 22221122

Research and Ingram 25(3). 223-225.

Block, J. H. (1983). Differential premises arising from
differential socialization of the sexes: Some
conjectures. 92112,Q21212222222,§12 1355-1382.

Broverman, I. R., Vogel, S. R., Broverman, D. N.,
Clarkson, F. E., & Rosenkrantz, P. S. (1972). Sex-
role stereotypes: A current appraisal. {222221 22
529.13.]. Issues... 2.8... 59-78-

Chodorow, N. (1978). The. Emmdncticn of. WI...
Berkely and Los Angeles: University of California
Press.

Edleson, J. L. (1984). Working with men who batter.
3.93.111 Earls. 29(3). 237-242.

Ellis A. (1985). Cognition and affect in emotional
disturbance. American Esrahalcgistt 40(4). 471-472.

Ellis A. (1987). The impossibility of achieving
consistently good mental health. Amszican
Esmhalcaist... 42... 364- 375.

Epstein. S. (1987). The Canstructin
Unpublished manuscript, University of Massachusetts
at Amherst.

Epstein, S. (in press). Cognitive experiential self-

theory. In L. Pervin (Ed.). Handback cf. 29.259.931.112
Luann! and,2ssaa22h2 New York: Guilford.

Epstein, S. & Meier, P. (1989). Constructive thinking: A
broad coping variable with specific components.

wuwmmmwzx

59

Foster, S. L. & Robin, A. L. (1988). Family conflict and
communication in adolescence. In E. J. Mash & L. G.
Terdal (Eds. ). Eehexieral assessment cf.

shildheed
diserders Land ed... unabridgedl (pp 717- -775) New
York: Guilford Press.

Gilligan, C. (1980). Restoring the missing text of
women's development to life cycle theories. In D. C.

McGuigan (Ed. ). Meals 11293.3. Nee theerL. researeh a
seller (pp 17- -31)

Hazzard, A. ,Christensen, A., & Margolin, G. (1983).
Children' s perceptions of parental behaviors.

efAhnermalChlldEsrehelesLllu). 49- 60.

Hetherington E. M. & Frankie G. (1967). Effects of
parental dominance, warmth, and conflict on imitation

in children. learn]. at. Eamnelitx and Seen].
Efilfihfllma. 5(2). 119-125-

Hurley, J. R. (1990). Constructive Thinking and Elevated
Ratings of Self in Interpersonal Groups. l222221 22
Esmhelegx. 12.4.. 563-573.

James, M. (1973). The down-scripting of women for 115
generations: A historical kaleidoscope. 1222221 21
WWLls- 22.

Kaufman, A. S. (1979). Intglligent testigg with the
ELSQ;B; New York: John Wiley 9 Sons.

Mizes, J. S. (1988). Personality characteristics of
bulimic and non-eating disordered female controls: A
cognitive behavioral perspective. 1222222212221
Jeanna]. of. Eating Ciserdets... 11(4). 541-550.

Montemayor, R. (1982). The relationship between parent-
adolescent conflict and the amount of time
adolescents spend alone with parents and peers.

Child Queleement... 5.3... 1512-1519.

Montemayor, R. & Brownice, J. R. (1987). Fathers,
mothers, and adolescents: Gender-based differences in

parental roles during adolescence. l222221,2£ 12212
and Adeleseanec... 1.6.(3). 281-291.

0' Neil, M. & Reiss, S. (1984). Adult perceptions of
their mothers: A life span analysis. 122

Esmhelezieal Resend... 3.4... 333- 342.

60

Prinz, R, J., Foster, 8., Kent, R. N., & O'Leary, K. D.
(1979). Multi-variant assessment of conflict in
distressed and non-distressed mother-adolescent

dyads. Janinel 21 AEElied.§ehe£iQ£el Anelzeiei 121
691-700.

Radin, N. (1986). The influence of fathers on their sons
and daughters. Seeial Kerk in.Ednealieni 3(2). 77-91.

Rizley, R. (1978). Depression and distortion in the

attribution of causality. 1222221 21,22222221
Esxehelegxi.811 32-48-

Robin, A. L. & Weiss, J. G. (1980). Criterion-related
validity of behavioral and self-report measures of
problem-solving communication skills in distressed
and non-distressed parent- -adolescent dyads.

Behaxieral Assessmentl.21 339- 352.

Smith, K. A. & Forehand, R. (1986). Parent-adolescent
conflict: Comparison and prediction of the
perceptions of mothers, fathers, and daughters.

Cenrhel.ef £2111 Adeleseehee1.§(4). 353-337.

Taylor, S. E. & Brown, J. D. (1988). Illusion and well—
being: A social psychological perspective on mental

health Esxehelegieal,Bulletin1.1C31 193- 210

Thompson, 8. & Melancon, J. G. (1987). Validity of a
measure of critical thinking skills. Esx2hgl2g12al
Reeertsi 5C1 1223-1230.

Thyer, B. A., Papsdorf, J. D., & Kilgore, S. A. (1983).
Relationship between irrational thinking and
psychiatric symptomatolosy. learnal.ef.Esxehelegxi
1131.31-34.

Whitbeck, L. B. (1987). Modeling efficacy: The effect of
perceived parental efficacy on the self-efficacy of

early adolescents. leurnal.ef.Earlx.Adeleseanee1
1(2). 165-177.

MICHIGAN STATE UNIV. LIBRARIES
IHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIHI
31293008761482