ENERGY CONSERVAHON: FAMHY VALUES, HOUSEHOLD
PRACTICES, AND CONTEXIUAL VARIABLES

Dissertation for the Degree of Ph. D.
MICHIGAN STATE UNIVERSITY
MARY JAMCE HOGAN
1976

 

 

\|\\|\\\\\|\l\ 'i“.'.‘“.“' ‘3" '“ “' “$.31“. ”N .‘

3:129:31” 1

 

 

This is to certify that the

thesis entitled

ENERGY CONSERVATION: FAMILY VALUES, HOUSEHOLD
PRACTICES, AND CONT EXT UAL VARIABLES \\

\

\.

presented by

Many Janice. Hogan

has been acceptedtowards fulfillment
of the requirements for '

PH.D. degree in Famdygozogy

 

/’ 4"" I" A.
w “Ii/W/ (28%.,

Major professor

 

Date Decembu 8, 1975

0—7639

 

‘ PLACE IN RETURN BOX to remove this checkout from your record. ,
’ To AVOID FINES return on or before due due.

DATE DUE DATE DUE DATE DUE ,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

*

 

 

“\Jmm

I MSU lsAnAffirmative Action/Equal Opportunity Institution
0 Minna-9.1

fl

 

 

 

 

 

 

 

 

 

 

 

t' "~

\ \
\t /

ABSTRACT

ENERGY CONSERVATION: FAMILY VALUES. HOUSEHOLD
PRACTICES. AND CONTEXTUAL VARIABLES

By

Mary Janice Hogan

Adoption of energy conservation practices was viewed as
desirable family managerial behavior in an era when finite resources
were assessed to be incapable of meeting high consumption levels on
a sustained basis. The family, a principal social unit in which
values are internalized and patterns of energy consumption are
learned, was the unit of analysis. The primary objective of the
research was to determine if there were differences in the adoption
rate of household energy conservation practices among families with
varying husband-wife patterns of congruency and commitment to values.

The values included in the study were self3esteem3mfamilism.
social_rg§ponsivenesslfiand‘eco-consciousness. Nine types of family
value configurations were conceptualized based on combinations of
husband-wife positions in the distribution of sample scores: above
the mean, around the mean, or below the mean. The purpose of the
intrafamilial value typology was to capture the interaction effects
of family members’ behavior, i.e., synergetic behavior of family.

A scale of 14 practices was used to measure the adoption

rate of household energy conservation practices. The family score

Mary Janice Hogan

for practices was basically a percentage of practices adopted based
on husband and wife responses. Contextual variables were also
studied in respect to adoption of practices and intrafamilial value
patterns. The sets of relationships among the variables with mean-
ingful correlations were used in building a conceptual model of
family management using an ecosystem perspective.

Survey data collected during May-June l974 in a multistage
probability sample of the Lansing Standard Metropolitan Statistical
Area were used. A subsample of l57 families was selected from the
larger study, Michigan Agricultural Experiment Station Project 3152,
"Functioning of a Family Ecosystem in a World of Changing Energy
Availability." Reliability was established for each of the value
scales and the energy conservation scale with alpha coefficients.
The coefficients, which ranged from .74 to .89, were a measure of
internal consistency of the scales. In a test of the difference
between two dependent means, the husbands' and wives' scores were
not found to be significantly different on the four value scales,
i.e., the scales were not biased by sex role.

The results of the analysis indicated that the value of
eco-consciousness is a meaningful predictor of energy conservation
behavior. In families where the husband and wife had a high commit-
ment to ecu-consciousness, they were high adopters of energy con-
servation practices. The eco-consciousness value was conceptualized
to measure the interrelationship of human beings and the physical

environment and contained an explicit "scarcity of energy resources"

Mary Janice Hogan

component. There was no systematic relationship between conserva-
tion behavior and contextual variables--education, occupation,
employment status of wife, age, family size, income, and urban-
rural residence. The intrafamilial value patterns of social
responsivehess and eco-consciousness were related to selected con-
textual variables. Social responsiveness had a positive relation-
ship with husband's education, wife's education, and family income.
The commitment and congruency levels of eco-consciousness were
positively related to husband's education, wife's education, and
husband's occupational prestige.
‘ The values of self-esteem and familism were not significantly
related to adoption of conservation practices or contextual varia-
bles. The major statistical procedures implemented to test the
hypothesized relationships and differences included one-way analysis
of variance, analysis of covariance, multiple classification analy-
sis, and multiple regression.

Based on the finding that eco-consciousness value is posi-
tively related to the adoption of energy conservation practices
and the linkage between education and ecoaconsciousness, it is
recommended that educational programs explicitly examine this value.
For example, the capacity of the earth to furnish resources and to
assimilate wastes could be assessed in respect to the demand brought
about by alternative family consumption patterns. And, the costs
and benefits of high and low commitment to eco-consciousness could

be estimated and technical information included which would assist

Mary Janice Hogan

family members in adjustment of consumption practices and rescaling
of values.

Further research was recommended to study the effect of
intrafamilial value patterns on outcome behavior. For example, is
consensus-dissensus of social responsiveness in families related to
their level of participation in public policy decisions at the com-
munity level? The exploration of changes in indirect energy con-
sumption patterns is another recommended area of research. Do
families perceive the relationship between scarce energy resources
used for production, transportation, and distribution costs of food,
appliances, etc., and consumer demand? How are the values of thrift
and ace-consciousness related to adoption of indirect energy con-

sumption patterns?

ENERGY CONSERVATION: FAMILY VALUES.
HOUSEHOLD PRACTICES. AND
CONTEXTUAL VARIABLES

By

Mary Janice Hogan

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Family Ecology

1976

© Copyright by
MARY JANICE HOGAN

1976

ACKNOWLEDGMENTS

Many of the concepts developed in this research were stimu-
lated in formal and informal courses of study with Dr. Beatrice
Paolucci, dissertation director and model teacher. Her infinite
support and intellectual curiosity have contributed greatly to my
professional development and to the completion of this study.

Dr. Peter M. Gladhart, a member of my committee and principal
investigator of the larger energy research project, provided valuable
assistance in every step of the research process. Numerous statis-
tical procedures and analyses were initiated as the result of his
guidance.

Drs. Carol Shaffer, Vera Borosage, and Allan Schmid served
as supportive committee members. Furthermore, they were inspiring
teachers who helped me explore alternative family futures in a
changing world of resources and values.

Special appreciation goes to the members of the energy
interdisciplinary research team--Professors Bonnie Morrison, James
Zuiches, Dennis Keefe, Mary Zabik, Larry Schiamberg, David Imig,
and Peter Gladhart. They extended guidance and economic support
throughout the study. And, numerous ideas and impressions were
refined with team members. I am very grateful to my colleague,
Mrs. Joanne Keith, for her introduction to the world of computer

science and her suggestions for data analysis.

ii

A number of people at the Institute for Family and Child
Studies contributed important support. Dr. Mary Andrews, data unit
consultant, assisted in the design of the study and analysis of the
data. Dr. Robert Boger, Institute Director, encouraged an environ-
ment for graduate student-faculty interchange. Colleagues Sandra
Evers, Jane Lott, Tracey Feild, and Terry Boyd provided constant
encouragement and shared their knowledge of stress-reducing strate-
gies.

A special contribution was made by Arizona State University
in granting a sabbatical leave and subsequent study leaves. My
co-workers and the students in the Home Economics Department made
numerous adjustments to enable me to complete my doctoral program.
Finally, the College of Human Ecology and Family Ecology Department
at Michigan State University made assistantships, research grants,
and a fellowship available. All of the support was deeply appre-

ciated.

TABLE OF CONTENTS

LIST OF TABLES
LIST OF FIGURES .
Chapter

I. INTRODUCTION

Conceptual Framework . .
Statement of the Problem .
Research Objectives
Assumptions

Definitions

II. REVIEW OF LITERATURE .

Approaches to the Study of Values . . .
Relationship Between Values and Practices
Energy Conservation PracticeS' . .
Related Energy Research .

III. DESIGN OF THE STUDY

Procedures for Sampling and Data Collection
Procedures for AESP Study . .
Selection of the Subsample .
Description of the Subsample .

Comparison of Chosen and Rejected Subsample

Families . .
Development of Measures
Measure: Energy Conservation Practices
Measure: Self-Esteem Value
Measure: Familism Value .
Measure: Social Responsiveness Value .
Measure: Eco- Consciousness Value . .
Typology of Intrafamilial Value Patterns .
Measure: Contextual Variables .
Research Hypotheses . .
Data Analysis Procedures .
Statistical Models
Computer Programs

iv

Page

'vi

ix

'Chapter
IV. RESULTS OF DATA ANALYSIS

Family Values and Energy Conservation Practices

Contextual Variables and Energy Conservation
Practices . .

Family Values and Contextual Variables

Summary . . . . . . .

V. SUMMARY, CONCLUSIONS, AND IMPLICATIONS .

Overview of the Study .
Discussion of Findings
Limitations of the Study.
Implications of the Study
Research .
Educational Programs

APPENDICES
A. Accepted and Rejected Subsample Families
B. Scales and Coding Rules .
C. Supplementary Findings

BIBLIOGRAPHY .

Page

Table

0101-wa

\I

10.
ll.
12.

l3.

T4.

15.

LIST OF TABLES

Stage of the Family Life Cycle of Subsample

Size of Household of Subsample .

Age of Husbands and Hives in Subsample

Family Income Distribution for l973 of Subsample .
Educational Attainment of Husbands and Hives

Occupation Classification as Defined by Husbands and
Hives in Subsample . . . . . . . . .

Employment Status of Hives

Classification of Accepted and Rejected Subsample
Married Families by Belief in Reality of the Energy
Problem .

Classification of Accepted and Rejected Subsample
Married Families by Actions Taken in Response to the
Energy Problem . . . . . . . . .

Energy Conservation Practices Adopted by Families
Descriptive Statistics for Energy Conservation Scale

Distribution of Value Scores Above, Around, and
Below the Mean by Husbands and Hives . .

Classification and Frequency of Family Value Patterns
of Values--Self-esteem, Familism, Eco-consciousness,
and Social Responsiveness . . . . . .

Summary of One-Hay ANOVA Tests for Differences in
Adoption of Energy Conservation Practices by Family
Values--Self—esteem, Familism, Social Responsive-
ness, and Eco- consciousness . . . . . .

Means and Standard Deviations of Energy Conserva-
tion Scores Based on Eco-consciousness Family Value
Patterns . . . . . . .

vi

Page
4O
41
42
42
43

44
44

47

47

51
53

59

61

70

71

Table Page

16. Post Hoc Comparisons of A Priori Contrasts of
Eco-Consciousness Family Value Patterns in Adoption
of Energy Conservation Practices . . . . . . . 71

17. Analysis of Covariance Test of Differences in
Adoption of Energy Conservation Practices Among
Intrafamilial Patterns of Eco-Consciousness with
Control Variables . . . . . . . . . . . . 73

18. Multiple Classification Analysis of Relationship
Between Eco-consciousness Value and Energy Conserva-
tion Practices with Covariates: Wife's Education
and Husband's Occupation . . . . . . . . . . 73

19. Multiple Regression Analysis Predicting Adoption of
Energy Conservation Practices From Contextual
Variables . . . . . . . . . . . . . . . 75

20. One- -Hay ANOVA Test for Differences Among the Stages
of Family Life Cycle Relationship in Adoption of
Energy Conservation Practices . . . . . 75

21. Summary of ANOVA and MCA Tests for Differences Among
Intrafamilial Patterns of Social Responsiveness
Value in Respect to Contextual Variables . . . . 76

22. Means and Standard Deviations.of Wife's Education,
Husband's Education, and Family Income Based on
Social Responsiveness Intrafamilial Value Patterns . 78

23. A Priori Contrasts of Wife's Education, Husband's
Education, and Family Income Based on Family Value
of Social Responsiveness . . . . . . . . . . 79

24. Summary of ANOVA and MCA Tests for Differences Among
Intrafamilial Patterns of Eco-consciousness Value in
Respect to Contextual Variables . . . . . . . 80

25. Means and Standard Deviations of Husband's Occupation,
Husband's Education, and Wife's Education Based on
Family Value of Eco-consciousness . . . . . . . 81

26. Post Hoc Comparisons of Differences Among Family
Patterns of Eco-consciousness Value in Respect to
Wife's Education and Husband's Education and Occupa- 82
tion . . . . . . . . . . . . . . . .

vii

Table Page

27. Summary of Lambda and Uncertainty Coefficient Tests
of Relationship between Stages of Family Life Cycle
and Intrafamilial Patterns of Values . . . . 83

28. Summary of Findings by Hypothesis with Significance
Level and Correlation Coefficient. . . . 84

A-1. Classification of Accepted and Rejected Subsample
Families by Stage of Family Life Cycle . . . . . 103

A-2. Classification of Accepted and Rejected Subsample
Families by Size of Household . . . . . . . . 103

A-3. Classification of Accepted and Rejected Subsample
Families by Role and Age . . . . . . . . . . 104

A-4. Classification of Accepted and Rejected Subsample
Families by Educational Attainment . . . . . . 104

A-5. Classification of Accepted and Rejected Subsample
Families'1973 Income . . . . . 105

A-6. Occupational Classification of Accepted and Rejected
Subsample Families as Defined by Husbands and Hives . 105

A-7. Employment Status of Hives in Accepted and Rejected
Subsample Families . . . . 106

A-8. Classification of Accepted and Rejected Subsample
Families by Geographic Location . . . . 106

8-1. Use of Rule 2 in Family Energy Conservation Practice
Scores . . . . . . . . . . . . . . 118

C-1. Summary of ANOVA and MCA Tests for Differences Among
Intrafamilial Patterns of Self-Esteem Value in
Respect to Contextual Variables . . . . . . . 121

C-2. Summary of ANOVA and MCA Tests for Differences Among
Intrafamilial Patterns of Familism Value in Respect
to Contextual Variables . . . . . . . . . . 121

viii

LIST OF FIGURES

Figure Page
1. Model of Variables for Research Study . . . . . 9
2. Model of Empirically Supported Relationships:

Family Values and Household Energy Conservation
Practices, Family Values and Contextual Variables . 95

ix

CHAPTER I

INTRODUCTION

The energy shortage attracted the attention of United
States families during the winter of 1973-74 as they experienced
gasoline lines and increased utility prices, and debated the effects
of the Arab oil embargo. These events signaled the termination of
cheap and abundant energy from fossil fuels and for the first time
many families were aware of the linkage between mechanical energy
supplies and levels of living.

Present consumption patterns in the household sector of
the economy use two-thirds of all U. S. energy consumed (Hannon,
1975). In direct use, most families consume energy for a central
heating system, water heater, range, refrigerator, washing machine,
lighting, and automobiles. Though electric toothbrushes, stereos,
lawnmowers, and hair dryers do not use large amounts of energy in
operation, they require energy for manufacturing and distribution
processes, i.e., indirect energy. The combined direct and indirect
energy consumption data reveal that the major portion of energy is
ultimately consumed by the family unit. Thus, any significant move-
ment toward decreased energy supplies nationally will necessitate

shifts in family consumption patterns.

The basic constraints on the present high consumption life
style include: the finite supply of renewable energy resources,
the capacity of the earth to assimilate wastes, and the net cost of '
substituting nechanical energy for human and solar forms of energy.
Analysis of the physical principles that govern the universe gives
evidence that the existing patterns of high energy consumption can-
not be sustained indefinitely (Odum, 1971; Koenig and Edens, 1975).

While the U. S. contains 6 percent of the world population,
it consumes 30 percent of the world's annual energy budget (Energy
Policy Project of the Ford Foundation, 1974:5). Calculated in
energy units per capita, the U. S. consumes 9500 units in compari-
son to 1072 for Mexico, 8 in Nepal, 5140 in Sweden, and 7870 in
Canada (Fritsch and Castleman, 1974). The U. S. family is the
highest user of mechanical energy of any in the world.

The U. 5. level of consumption and standard of living, i.e.,
desired level of living which includes consumption as one component,
will need to be reassessed and adjusted to reflect physical princi-
ples that govern our universe as well as issues of equity and equal-
ity. Are the societal values of individualism, independence, and
mastery-over-nature linked to current consumption levels?

Twenty years ago Cottrell (1955) recognized that the preser-
vation of values in social systems is related to availability of
energy. He stated:

The preservation of a system of values requires a continuous
supply of energy equal to the demands imposed by that sys-
tem of values. Conversely, . . . changes in the amount or

form of energy available give rise to conditions likely to
result in changes in values, for men who share conmon values

make similar changes in choice when faced with similar
changes in the consequences of their acts (p. 4).

Given the current awareness of finite energy resources, there is
need to reassess values and resource use.

The family is an energy driven system. It is regarded as
a principal environment in which values are internalized and patterns
of energy consumption are learned. Families are a critical mana-
gerial unit because they make decisions about energy expenditures
and participate in the public debate about energy conservation
policy. Because the family is a mediating link tying the individual
to the larger ecosystem and because the aggregate of families form
the structure of society from which all other social institutions
draw their component units, alternative futures are linked to family

socialization and consumption processes.

Conceptual Framework

 

A management conceptual framework will be used in this
study of families. The managerial theory focuses on patterns of
decision making motivated by scarce resources and interdependent
utility functions. The focus will be on the family as a unit rather
than individual family members.

Values and their function as motivators of conscious
behavior are basic concepts in the managerial framework. According
to Paolucci (1966):

Management seems to be motivated by the perception of a
conflict situation: when home members note that the values
held by individuals within the household are in conflict

with those held by individuals both inside and outside the
home; when what is being accomplished is not cons1stent w1th

what the group believes it ought to be accomplishing; when
there is disparity between goals and goal achievement
(9 339)

Some leaders in the field of family management have classi-
fied values as part of a psychosocial or personal conceptual organi-
zation and defined decisions regarding resources as part of the
managerial framework (Gross, Crandall, and Knoll, 1973; Deacon and
Firebaugh, 1975). These leaders have, however, included values as
a key concept in management theory. According to Deacon and Fire-
baugh (1975:140), "Values provide the underlying meanings that give
continuity to all decisions and actions."

Values are held to be a directive element in decision
making. Gross, Crandall, and Knoll (1973) state:

Motivation for management comes fundamentally from a
variety of sources both without and within the family
. . Key concepts in understanding motivation within
the person and the family are values and standards and
the closely related concepts of goals (p. 112).
While values are assumed to guide choice, there is little empirical
work to support this generalization.

Decisions motivated by multiple goals and scarce resources
are accepted by authorities in family management as central concepts
in the management framework. According to Deacon and Firebaugh (1975:
157), "Resources are the supply reservoir for use in the system's
specific actions, and are necessary in some form to solve every
management problem."

In discussing the kinds of decisions motivated by normal

conflicts in the family, Paolucci linked values to social decisions

and scarce resources to economic decisions (1966). She stated:

Social decisions are made when there is a conflict in
values, goals, or "duties" (roles) among family members
or between those held by the family and by other social
groups, such as peers, occupation groups, the school, or
the neighborhood. Economic decisions are evoked when the
goal§3gf the home members compete for scarce resources

p. .

Recently, the ecological approach has been proposed to
study family management behavior (Steidl, 1969; Hook and Paolucci,
1970). From an ecological perspective, all parts of the family
system are linked to each other; values are interrelated to
resources through energy flows. A change in one part of the system
will produce a change in all other parts, given enough time for the
processes to transmit signals and enough magnitude for the change
to transmit itself through the linkage network.

This approach raises questions about the directive force
of values in decision making. From the ecosystem perspective,
value change may be precipitated by availability of resources and
decisions about resource use may be directed by dynamic value

systems.

The system concept of causation was examined by Bates and
Harvey (1975). They state:

In the systems approach there are no dependent and inde-
pendent variables. There are no causes or sets of causa-
tive factors. There is, instead, simultaneous universal
responsiveness among parts, so that they act together as
an unfolding operation of parts functioning in relation
to one another as a whole. . . . In the broadest sense,
the systems perspective takes the view that it is the way
the elements of a system are organized in relation to
each other that accounts for the events occurring within
the system, and between it and its environment (p. 31).

6

This point of view suggests that if the hierarchy of values of theg
family system was to change or alter in organization, then use of A
resources would change. Conversely, if change in resource avail-
ability was of sufficient magnitude, vaers would be reorganized :
once enough time had lapsed for the system to process the effect.

It is the movement in this set of relationships that is presumed to
account for changes in the family system.

According to Williams (1970), the stability or continuance
of a family, in part, rests upon consensus of values. He stated
that family values "are not independent of the societal context of
support or stress, abundant or scanty resources." And, he proposes
that values will change as basic life-sustaining and life-threatening
conditions change. Thus, the family is viewed as a relatively open, .l
adaptive system of interacting personalities, linked to other
social systems and the environment through energy flows.

In an effort to contribute to building a symbolic model of
family management, relationships among values, contextual variables,
and adoption of energy conservation practices will be studied.
Intrafamilial value patterns and household energy conservation
indices will be developed. Rather than isolate the demographic
variables and values which will predict variation in practiCes,
from the systems perspective, linked relationships will be identified.
According to Bates and Harvey (1975):

. the systems perspective focuses on trying to isolate

the means by which the behavior within the system and
between it and its environment are produced, rather than on

isolating the variables which will predict or "statistically
account for" the variation in one or another variable (p. 32).

The principal contribution of this study will be in theory
building. The model of family ecosystem management can not be
tested without developing linkages between system concepts. Although
no set of linked concepts can be completely understood separate from
other linked system concepts, the holist model can only be built
with empirical support from the existence of relationships among
variables.

In their role as educators, home economists need information
about decisions requiring changing amounts of energy for implementa-
tion. They are charged with responsibility to assist families in
assessing alternative managerial patterns regarding scarce resources
and resolving value conflicts which emerge from new resource con-
straints (Paolucci and Hogan, 1973). This research contributes base-
line data on energy conservation practice adoption, value commit-
ments and congruency,and the linkage of values, practices, and con-

textual variables to assist in developing educational programs.

Statement of the Problem
Energy conservation is viewed as necessary management
behavior in a country where finite energy resources are assessed as
incapable of meeting consumer demand on a sustained basis. Since
energy is a fundamental resource used by families to reach goals,“
changes in the amount of energy available will affect family manage-

ment. Remarkably little is known about families' adoption rate of

energy conservation practices. What values are linked to energy
conservation? What is the relationship between incongruent family
values and adoption rate of energy conservation practices? Are
education levels, stages of family life cycle, occupation, and
other contextual variables closely linked to values and energy con-
servation practices?

The primary purpose of this research is to determine if
there are any differences in the adoption rate of household energy
conservation practices among various intrafamilial patterns of
values. The values to be investigated are self-esteem, familism,
eco-consciousness, and social responsiveness.

Secondary purposes include testing the rate of adoption of
energy conservation practices in relation to contextual variables
and investigating the differences in levels of education, occupation
prestige, income, and other contextual variables among the intra-
familial value patterns. The sets of relationships among these
variables with the greatest magnitude will be identified to assist
in building a conceptual model of family management using an eco-
system perspective. The model is schematically presented in

Figure 1.

Research Objectives
1. To determine if there is a difference among the intra-
familial patterns of values--self—esteem, familism, social respon-
siveness, and eco-consciousness--in adoption of energy conservation

practices.

mmanmem> szpxmpcou can

.xvaum cucmmmmm Loy

mapnawga> Lo _auo=--._ aaamwa

 

 

mm< mpuzu zpwsmm
«NPm apwsmu cowumaauuo
msoucn compmuaom

mm4m<~m<> 4<=hxmhzou

 

 

meowuumga compm>gwmcoo amcmco

mo cowuaovo cmmZumn
awsmcowpmpmc m mews» mm

Nmmpnmpgm> pazuxmpcou

op pomamme cam: mccmppaa
m=Pm> paw—wanemgucp ecu
ocean mucocmmewu a mngu mm

 

mmuHhu<mm
onh<>mmmzou

>wmmzm.
opozmmsoz

 

 

 

wmmUVuomLa coppm>cmm

icou sneeze Co cowuaoum cw
mccmuuma oapm> pmwpwsammcucw
ocean mucmcowewu a mcmca mH

mmmcm>wmcoammc pmwuom
mmmcm=o_umcoUioom
Emwpmsmm

ammummtwpmm

mm24<> >4Hz<u

 

 

 

10

2. To determine if adoption of energy conservation prac-
tices varies with contextual variables--occupation, edUcation, and
age of husband; employment status, education, and age of wife;
family size, income, stage of family life cycle, and urban-rural
residence.

3. To determine if there is a difference among the intra—
familial patterns of values--self—esteem, familism, social respon-

siveness, and eco-consiousness--with respect to contextual variables.

Assumptions

 

1. Values of family respondents can be identified using
self-administered scales.

2. Husbands and wives can accurately assess and record
the changes in household energy conservation practices adopted by

members of the family.

Definitions

 

Values; A generalized and organized conception of a pref-
erence field, resulting from transactional interplay of physical-
environment, nature of man, and interhuman relations; explicit or
implicit concepts of the desirable. The basis of the definition is
elaborated upon in Chapter II. The specific values are defined as:

l. Self-esteem: Respect that a person has for self,

 

includes recognition of self-limitation and growth
potential, overall judgment of worth or competence as

a person;

11

2.. Familism: Perception a family member has of his

 

family unit and level of integration of its members within
the unit; family as a social group;

3. Social responsiveness: Bonding with larger social

 

structure, interdependence of person-society, degree of
integration into society;

4. Eco-consciousness: Interrelationship of man-

 

nature, interlinked with earth's capacity to sustain

life style of man.

Energy conservation practices: Activities which directly
use mechanical energy in the household. Excluded are indirect uses

of mechanical energy and use of solar and human energy.

CHAPTER II

REVIEW OF LITERATURE

Research and related literature will be reviewed in three
major sections-~approaches to the study of values in family manage-
ment, the relationship between values and practices, and contextual
variables and energy conservation practices related to functioning

of a family ecosystem.

Approaches to the Study of Values

 

Philosophers and social psychologists have written exten-
sively about values in an attempt to clarify the role of values in
directing human behavior. According to C. Kluckhohn (1954:390),
"the only general agreement is that values somehow have to do with
normative as opposed to existential pr0positions."

In a comprehensive study of operational definitions of
values by researchers, Handy (1970) supported Kluckhohn's observa-
tion of conceptual diversity. He proposed that there is not one
valid measure of values or one scale of universal importance to be
discovered, but rather various aspects of value transactions to be

measured using different techniques.

Several researchers have included the transactional property

in their approach to studying values. F. Kluckhohn and Strodtbeck

12

13

(1961) defined value orientations as patterned principles resulting
from transactional interplay of cognitive, affective, and directive
elements of the valuation process. C. Kluckhohn (1954) linked
values to nature, i.e., man-nature transactions, cultural conceptual-
izations of nature. He defined values as concepts of the desirable,
normative criteria by which human beings are guided in their choice
among alternative courses of action.
It is the transactional property of values that brings
about changes in values over time. According to participants
attending the National Institute of Child Health and Human Develop-
ment Conference (1968:50):
Protean man seeks values not because he lacks values but
because he must find new values suitable for the changed
and changing world in which he lives. Renaissance man was
a kind of protean man, yet perhaps the reason he was dif-
ferent from today's protean man, and his values were dif-
ferent, is because values are always in context. Essentially
values mean relationships to things and to people.
Rescher(l969b) theorized that value change is linked to
the finiteness of resources and identified conditions stimulating
value change. He stated:
. . values can come into conflict with one another, not of
course in the abstract, but in the competing demands their
realization and pursuit make upon man's finite resources
of goods, time, effort, attention, etc. Thus when a change
occurs in the operating rationale that constitute the
operative framework within which a value is pursued in a
given society, we may expect a series of stresses upon our
scale of values militating from a rescaling in their order-
ing or a change of the value standard, etc. (p. 76).

Rescher proposed that stresses brought about from reduced supplies of.

resources will influence value changes.

14

Williams (1970) also proposed that patterns of values in
families are dependent on stocks of available resources and will
change as basic life-sustaining and life-threatening conditions
change. The characteristic of interdependence implies that the
family system is acted upon by the environment, acts on the environ—
ment.and reacts to the environment.

Two approaches to measuring values can be distinguished in
reviewing the literature: expressed and revealed concepts of the
desirable. Some scholars focus their study of values on the
expressed measure of values by analyzing verbal or written state-
ments of "what ought to be." In discussing this method of measure-
ment, Rescher (1969a:2) indicated that it is assumed that the respon-
dent would take the value into proper account in making decisions.

The second approach to measuring values, revealed pref-
erence, is derived from economic theory and is based on analysis of
choice or decision outcome in the marketplace. According to Boulding
(1969), the preference field of a person is made up of a set of
possible exchanges with known resource constraints in the form of
prices and from observing responses to different choice situations,
values are revealed. Thus, it is assumed that values can be
inferred from observed behavior.

Values as revealed preferences may be adequate for appli-
cation to some problems and some types of data. However, the ques-
tion of whether the researcher is observing a concept of the

desirable or heuristic decision making creates certain problems.

15

Jacob and Flink (1962:20) critique this approach to measuring values
and suggest that inferring values from short-term overt behavior
masks the property of stability inherent in basic values. They
state:

Values can and do change, though they have a strong hold

upon most human beings and constitute a relatively stable

component of the personality (p. 15).

Rokeach (1974) used the expressed preference approach to
measure value stability in a national survey research project over
a three year period, 1968-71. He classified values as terminal,
i.e., ideal end-states of existence, and instrumental values, i.e.,
ideal modes of behavior. He reported:

The composite ranking of the terminal value a world at peace
was first in 1968 and first again in 1971; family security_
was second and freedom third on both occasions; an exciting
life, p1easure,_sociai recognition, and a world of beauty
were at the bottom of fhe nationSl sample's terminal vilfie
hierarchy in both 1968 and 1971. For both years, the most
important instrumental values were honest, ambitious, and

responsible; the least important were imaginative, logical,
obedient, and intellectual (p. 225).

 

 

 

 

Rokeach proposed that values specifically related to societal issues
undergo measurable change and other values remain relatively stable.
For example, he found that for American women family security became
significantly less important in 1971, and equality and being_logjcal
significantly more important. Changes in women's values may be
related to the national issue of women's societal role.

The extent to which the observed value changes were a
function of education, income, or age were also tested in the

Rokeach study. Only age was a significant determinant; the younger

16

adult respondents showed considerably more changes than older adults.
According to Rokeach (1974:231), the significant value changes found
for persons beyond the thirties are few in number and could easily
have arisen by chance.

The importance of the Rokeach study is its quantitative
monitoring of value change and stability in a national area proba-
bility sample. Survey research has been used to assess attitude
changes over time, but not with the measurement of values. While
attitudes may be used as predictors, values encompass more basic
objectives and are general predictors of behavior (Nye, 1967).

A question critical in investigating values is the unit of
analysis. While it is very appropriate for some disciplines to
study values of individuals to predict economic, social, and politi-
cal trends, scholars of the family give empirical attention to
familial values.

Can either the husband or wife accurately describe the
value base of family members or the "family" value hierarchy? The
adequacy of interviewing a single family member has been debated in
recent literature (Granbois and Willett, 1970; Safilios, 1969;

Turk and Bell, 1972). In general, authors conclude that there are
few differences_when aggregated responses for husbands and wives
are compared, but important differences are revealed when spouses'
answers are compared.

Williams (1970) proposed that the family as a social unit

does not have one value system which commits each of its members to

17

a single concept of the desirable. Rather, the family has a pattern
of values which reflects variability of commitment among members.

He suggested that diverse value systems may result in conflict,
active dissent, or covert opposition in decision making behavior of
family members. Williams stated ". . . value-concensus does not
guarantee family harmony, nor does dissensus necessarily lead to a
high level of conflict." He cencluded that family scholars have
little systematic information to predict the effects of congruency
levels in families.

One of the early research studies to investigate intra-
familial value patterns was conducted by Cutler (1947), a home
economist. Included in the sample were all persons over 10 years
of age living in the household. She used a measure of expressed
values-~beauty, comfort, convenience, location, health, personal
interests, privacy, safety, friendship, activities, and economy--
and administered the instrument to the family members independently.

Cutler found that family members were more likely to agree
on the most important and the least important values than on values
of intermediate importance. The correlations between values for
the husbands and wives were .36, .38, and .55 for the upper class,
middle class, and lower class respectively. The correlations were .
higher for mothers and children than for fathers and children. The
higher the socio-economic group, the less similarity between fathers
and children--.ll, .35, and .31 respectively. There was no signifi-

cant difference for mothers and children among the socio-economic

18

levels--.39, .42, and .37 for high, medium, and low status groups
respectively. The sample size was too small to generalize the
findings of the heterogeneous group of families.

In a more recent study, Martin (1965) studied values in 51
families to construct profiles reflecting value similarities. The
families were comprised of two parents and children between the
ages of 12 and 18. A value typology developed by Engebretson (1965)
was used to classify individual values according to four types:
social, change-prone, traditional, and autonomous. Martin reported
that husbands' profiles were about equally distributed between
autonomous--emphasizes growth and development, fairness, impartial-
ity, and responsible inner-direction--and traditional-—emphasis on
production, duty, rights and responsibilities, and security. Wives'
profiles were largely autonomous. Father-daughter and mother-
daughter had the most profiles alike; mother-son pairs had the fewest
similar profiles.

The major contribution of the Martin study was in assessing
the value hierarchies of family members to determine similarities
and to empirically support the margin of error when one family member
is used as the spokesperson for the family value system. She con-
cluded that no one person can be the indicator of values for the
entire family. She recommended research to study decision outcome
of different combinations of values in the family. Because the
study had a nonprobability sample, generalizing the findings was

limited.

19

Strodtbeck (1958) studied intrafamilial values of Jewish
and Italian families in relation to social mobility using a revealed
differences instrument. The family members individually completed
responses to nine value stimuli items and the revealed differences
were resolved collectively. The final choices became the "family"
value hierarchy. The Strodtbeck study found three values in the
hierarchy related to socio-economic achievement in the United
States: mastery-over-future, individualism, and low commitment to
familism. He found that the Italian stress on "familism" contrib-
uted to the lower occupational achievement of Italians as compared
with Jews.

The Strodtbeck study broke new ground both theoretically
and empirically in the study of values in the family. The values
tying the family to the larger social system were empirically
supported and the role of family in the socialization process was
made explicit. Finally, the study provided an alternative method
of testing family theory regarding the production of human capital
in relation to the valuing process.

A number of studies of homemaker's values have been con-
ducted with the wife as respondent. Ketchum (1961) compared
wives' values, measured by rank order and forced choice tests of
expressed preference, to a one-day recall of activities to determine
the level of awareness of value content in their role as homemaker.
A low correlation between expressed values in tests and revealed

values in activity recall gave empirical support to the theoretical

20

difference in measuring values. As discussed earlier, there is a
limitation to identifying values from short-term revealed behavior.

Dyer (1962) conducted a study of three techniques to rank
values: projective stories, rank order test, and Q-sort of activity
recall. The correlations between tests varied from .51 to .80,
with the highest correlation reported between the projective stories
and Q-sort of expressed preference in activities. The researcher
questioned the validity of the rank order test in identifying
values. It may be that Dyer's measures identified both revealed and
expressed preferences.

Reflecting on the studies reviewed, there are alternative
methods in the measurement of values. The contrast between measur-
ing overt or revealed behavior and expressed preference need not be
viewed as an "either-or" situation, but determining what aspect of
valuing behavior is to be measured. As Handy (1970) so clearly
stated:

All the various forms of behavior may be worth studying.
For example, assume that a given individual responds to a
questionnaire in terms of the "self-image" he has and that
such responses diverge considerably from the behaVior he
shows in other settings. For some inquiries, it might be
precisely such "self-images" that are of interest and quite
possibly reliable information about them will help predict
some types of behavior. So in that instance disparity
between verbalized statements of preference and some other
evincing of preference will not matter. Sometimes, indeed,
what is marked on a paper may be exactly what is most
important; in an election, who is elected depends on what
choices are made on the ballot, not the "inner feelings" of
the voters (p. 200-201).

In sum, the basic statement of theory from studies of

intra-familial values is that there appears to be a network of

21

common values and also dissimilar values. 00 families sharing
similar values advocate similar responses in managerial situations?
And conversely, do members with generally dissimilar values advocate
different responses? If family behavior is a function of the con-
tent and the degree of similarity of value systems, the direction
and congruency of intra-familial values will produce variance in

decision outcome.

Relationship Between Values and Practices

The study of values in family management has received
continued empirical attention from scientists since the benchmark
conference, "Values and Decision-Making in Home Management," nearly
two decades ago. In an introductory statement at the conference,
Gross (1955) stated: "Values, if not synonymous with motivation of
management, underlie it." Many of the studies which followed have
focused on the linkage of values with decision-making, planning,
resource use, and implementation processes (Eigsti, 1973; Forten-
berry, 1963; Johnson, 1962; Meeks and Deacon, 1972).

According to Gross, Crandall, and Knoll (1973:346—351),
most of the family management studies have focused on describing
processes, not the results or consequences of managerial action.
They suggest that one of the output units that could be measured is
adoption of practices or activity analysis.

Decision implementation was supported as an area of manage-
ment which should receive attention by Schlater (1967). She stated

that the goal-directed activities include carrying out a series of

22

decisions, i.e., decision implementation, under conditions of uncer-
tainty and with limited resources.

Adoption of practices is classified as output related to
values in most managerial models. However, concepts of decision
making and goal achievement are intervening variables. According
to Deacon and Firebaugh (1975):

Management processes lead to extrinsic values by meeting as

nearly as possible the goals that are value based . . . A

managerial activity may serve one or more values at any

level of generality or specificity. A decision may be

complex because more general or underlying values influence

the goal choices (pp. 144-145).
They indicate that values that are held to be generally important
may not be important in every decision situation. The more basic
or general the values, the more they are linked to general rather
than specific action situations. Thus, the content of values varies
from general to specific, and may require different actions over
time to fulfill.

The guesion of which values and managerial practices to
investigate may be related to the conceptual framework used for
research. Using an ecosystem framework, Bredemeier (1973) cited
characteristics from which value content can be inferred. He noted
that human beings are interdependent with one another, both in the
development of human resources and the division of material
resources. Values and cognitions that underlie choice are capable

of great variation and the consequences of differential predisposi-

tions can be measured from Bredemeier's perspective. He suggested

23

that the purpose of social science research is to study distributive
justice, i.e., the sharing of scarce resources.

In a study of family consumption of water, Field (1973)
used an ecosystem perspective and gave attention to predicting
adoption of practices from values. She collected water use data
from 100 middle class families living in a Michigan urban subdivi-
sion. Field hypothesized that conscious attempts to conserve water
were related to the beliefs in fate held by the husband and wife.
The measure of beliefs developed by Brim, et a1. (1962:54-57)
included value orientations similar to F. Kluckhohn and Stodtbeck's
(1961) concepts: control over destiny, future-present time orienta-
tion, and optimism-pessimism reference.

Field found no support for the relationship between values
and practices. She concluded:

Belief in fate did not appear to be related to the amount of

water used nor to attempts to conserve water. It may be that

those who believe, not in fate but rather in their own ability

to control what happens in their lives, do not view the use

of water as an area in which they could or should try to exer-

cise control (p. 69).
The results of the Field study suggested that families did not
perceive water as a scarce resource, and therefore were not moti-
vated to adjust their practices to conserve water. The amount of
water that families used was related to their socio-economic status
--income, education, occupation, and area of residence--and to their
stage in the family life cycle.

Kimball (1960) studied the relationship between husbands'

and wives' value patterns and the adoption of innovative household

24

and farm practices. Data were collected in a stratified sample of
farm families. The 53 couples in the sample completed a forced
choice value test used by Cutler (1947) and responded to questions
regarding the adoption of role related practices. Kimball reported
one value, family life, was positively related to the husband's
adoption of farm practices; the values of recognition and religion
were negatively related to practice adoption of farm practices. For
the wives, he found the value of helpfulness positively related to
adoption of innovative household practices; recognition and freedom
were negatively related values to practice adoption for the aggre-
gate of wives.

Kimball hypothesized that the more similar the intra-
familial value patterns, the more practices they would have adopted.
He found a curvilinear relationship: very low practice adoption
was associated with very dissimilar value patterns for the husband
and wife; as intrafamilial value similarity increased, the total
practices increased, but only to a point, beyond which greater
similarity appeared to have a negative effect on adoption. The
Kimball study's major contribution was in testing the intra-
familial values in relation to behavioral output.

The Field and Kimball studies provided insight into the
family's adoption of innovative practices. In sum, there was support
in the Kimball study and no support in the Field study for linkages
between values and practices.

The work by Rogers and Shoemaker (1971) is related to the

adoption of practices cross-culturally and provides additional

25

insight for the current study. Their model includes four processes:
knowledge, persuasion, decision, and confirmation. While values
could be classified as an antecedent variable and adoption of
practices a consequence of the processes, the researchers point out
that nonadoption or rejection of the practices may also be the out-
come of the decision process.

Some of the generalizations that Rogers and Shoemaker
(1971:Appx.) reported with empirical data were: early adopters
have more education, high social status, more social participation,
and are more highly integrated with the social system than later
adopters; early adopters are no different in age from later adopters.
While the authors do not inclUde the study of values specifically
as one of the antecedent variables, several of their measures of
behavioral phenomena approximate the transactional properties of
values. For example, participation and integration into the larger
social system are indicators of the value of social responsiveness.
In sum, the Rogers and Shoemaker comprehensive review of the
research revealed empirical support for some of the contextual and

behavioral variables linked to adoption of practices.

Energy Conservation Practices
Conservation of reSources has remained an important concept
in family management over time. However, the particular resources
that have been perceived as scarce have changed. Gross, Crandall,
and Knoll (1973:172-173) traced historically the scarcity of material

goods in the early stages of industrialization. They identified the

26

limited time and human energy emphasis in an era of affluence and
raised questions about the outcome of "use-and-throw-away" consump-
tion patterns.

A number of research studies have focused on the conserva-
tion of mechanical energy in the past two years. Those studies
which investigated family consumption patterns will be discussed in
terms of implications for managerial behavior change.

The Warren (1974) study reported that most households had
adopted some practices to conserve energy during the energy crisis
winter of 1973-74. In an interview survey of 766 household respon-
dents in Detroit area suburbs, approximately 83 percent of the
respondents said they had lowered home thermostats, turned out
lights, or some similar household energy conservation practice.
About 20 percent of the sample reported they were not using their
car as frequently to go to work and 65 percent had reduced
mileage for leisure purposes. Additional home insulation was
installed by 9 percent of the families. Only 2 percent of the
respondents reported no adoption of energy conservation practices. .

Warren concluded that it is necessary to distinguish between
energy conservation of gasoline for the car and direct energy for
the household. He stated:

. our findings suggest that the short term experience of
the energy crisis of 1973-74 was fundamentally a gasoline
shortage and that it would be possible to distinguish in
the attitudes of individuals their view of the conspiratorial
character of those gasoline shortages apart from the view

that, in fact, there is a broader question of energy resources
and their future availability (p. 84).

27

Because many of the respondents indicated that the energy crisis

was a form of distraction from the political problems of Watergate
and equated "crisis" with gasoline shortages, Warren questioned how
much control families had in adoption of gasoline conservation prac-
tices.

Tests of significance were not reported for any of the find-
ings in the Warren study. This study, an exploratory phase of a
proposed four-year longitudinal study, was analyzed with frequency
data and some relational coefficients. The insights that Warren
reported are helpful in designing future research where probability
of error will be tested.

The analysis of socio-economic factors related to practices
in the Warren study suggested that the higher the income, the higher
the adoption of energy conservation practices-~particularly those
related to the use of the car. Warren proposed that families with
higher incomes were better able to make adjustments in their life-
style because they had more discretionary resources. The ability of
the lower income families to conserve energy was found to be a func-
tion of the range of household appliances and transportation
resources available. Lower income groups reported energy conserva-
tion practices, but the number of practices adopted were fewer than
higher income families adopted.

The high adoption rate of energy conservation behavior of
families in the Warren study was reported to exceed the expectation

of government policy makers. Warren proposed that the explanation.

28

for high adoption of energy conservation practices is linked to

values and states:
The sense that Americans have strayed from the values of
the Protestant ethic is something which has been widely
discussed and reviewed in the media. We believe its appli—
cation to the energy crisis takes on the following character:
that individuals in the more affluent income groups viewed
the problems of the energy crisis as an opportunity to
restore a balance in their lives between their fundamental
values of self-regulation and restraint and the existing
lifestyle of which they had become accustomed (p. 87).

The type of neighborhood--integral, parochial, diffuse,
stepping-stone, transitory, and anomic--was used as an intervening
variable linked to attitudes and practice adoption. Warren reported
that attitude is only a critical predictor of household energy con-
servation behavior in the presence of the neighborhood variable. He
concluded:

Thus, if people have great distrust in the reality of the
energy crisis but many individuals around them are taking
action regarding conservation, . . . to the extent that
they identify themselves with that neighborhood or commu-
nity, their behavior may move more in line with these
publicly defined norms . . . (p. 88).

Thus, it appears that the family's integration into the
neighborhood is a factor in their adoption of practices and may be
an important influence in the rate of vaer change.

Two communities in New York City were randomly sampled dur-
ing the summer of 1974 to determine the characteristics of families
who had adopted energy conservation practices and high energy knowl-
edge. The Kilkeary and Thompson (1975) study selected the commu-
nities based on a "blackout" variable. Half of the 602 families
resided in Queens and had experienced a weekend without electricity

the previous summer.

29

Kilkeary and Thompson reported the following variables
were significantly related to adoption of energy conservation prac-
tices: exposure to blackouts, direct payment of utility bills,
car ownership, belief in family effort to produce change affecting
the energy crisis, income, and family composition. There were nine
practices in the conservation measure, four of which concerned use
of car and home improvements through insulation and storm windows.
Since only 18 percent of the families owned their housing unit and
about half of the families owned a car, the limitation of the con-
servation practices measure is a concern.

The household respondent's knowledge about energy conserva-
tion was tested in a 16-item questionnaire administered by the
interviewer. Kilkeary and Thompson reported a positive relationship
between energy knowledge and car ownership, income, education, and
family composition. They concluded that the higher socio-economic
families who can afford to pay higher bills are not conserving as
much as the moderate income families.

Direct energy use in households was found to be positively
related to family income in the Newman and Wachtel study (1974).
They reported from a national survey of 1,455 households that the
poor use less energy in maintaining their level of living and allo-
cate a greater proportion of income to direct energy costs than
higher income groups. The poor families spend about 15 percent of
their income for natural gas, electricity and gasoline in comparison
to 7 percent for lower-middle, 6 percent for upper-middle, and

4 percent for well-off families. There was little reported

30

difference in natural gas by income group, greater difference in
electricity consumption, and the largest difference in gasoline
consumption by family income.

Most of the poor families lived in smaller homes than higher
income families, were more likely to live in multi-unit dwellings
rather than single structures, and paid proportionately higher
energy costs for space heating. Newman and Wachtel reported that
about 50 percent of the poor have insulation in their dwellings
compared to 95 percent of the well-off; 31 percent of the poor have
storm windows compared to 63 percent of the well-off families'
homes. Thus, it was not surprising that the study reported high
energy costs for low income families one of the most important prob-
lems uncovered.

Newman and Wachtel recommended subsidized loans for home
improvements in energy efficiency and a system of "energy stamps"
to assist in the short-run. Furthermore, they perceived the unequal
energy costs as an issue related to the larger problem of poverty.

Studies by Heberlein (1974) and Winett and Nietzel (1975) o
tested the effects of incentives on energy conservation behavior.
Heberlein selected 96 apartment residents for his experimental design
research model and used three treatments: information given to
respondents on energy conservation techniques; information presented
on the negative effects of high consumption to individuals and
society; and, the positive effects of high consumption to individuals,

thus discrediting conservation behavior. The control group received

31

no information input from the researchers. No significant differ-
ence was found among the four groups in their direct energy con-
sumption behavior.

Winett and Nietzel offered monetary incentives contingent
on energy conservation of direct energy to 16 families and only
information on energy conservation techniques to 15 families. The
families volunteered for the study in response to newspaper articles
soliciting respondents but information about the monetary incentives
was given only to the one treatment group at the time the study was
implemented. The researchers reported significantly greater reduc-
tions in electricity and no reduction in natural gas for the families
receiving monetary rewards than for the families receiving informa-
tion. The self-selecting sampling procedure used may have attracted
high adopters of conServation practices that were unwilling to
further reduce their consumption of natural gas. Winett and Nietzel
monitored the energy consumption levels at two-week and eight-week
periods after the treatments and reported that the differences in
electricity consumption were maintained between the two groups.

In a study of townhouses in an upper middle class planned
community development, Grot and Socolow (l974) hypothesized that
building materials, exterior exposure, appliance design, space
heating, cooling systems, and other aspects of household technology
would account for the variance in gas and electric consumption.
After exploring the relations between housing structure variables,
considerable variance was left unexplained. Grot and Socolow con-

cluded that life style decisions accounted for the difference

32

between 800 and 1300 therms consumed by families in like physical
household structures. They recommended that behavioral variables
be included as predictors of energy consumption. In sum, Grot and
Socolow controlled for structural variables and thereby isolated

the role of managerial behavior in energy consumption.

Related Energy Research

 

Several research studies have been reported from the Michi-
gan Agricultural Experiment Station Project, "Functioning of a
Family Ecosystem in a World of Changing Energy Availability" (Glad-
hart, 1975; Morrison, 1975; Eichenberger, 1975; Zuiches, 1975). Each
of the researchers chose their sample from a multi-stage area proba-
bility sample of 217 families.

Family income was found to be the strongest single predictor
of home energy use. Gladhart (1975) reported that families in the
upper third of the income distribution used 17 percent more direct
energy than did those in the lowest third. While the consumption
differential varies in absolute amounts from study to study, in
general Gladhart's findings corroborate the Newman and Wachtel study
reported earlier. A '

Morrison (1975) studied a subsample of families living in
single family dwellings to predict energy consumption levels. While
she included such physical housing variables as number of windows
and doors, type of construction materials, and presence of insulation
in walls, floors, and ceiling, she found that number of rooms was

the best structural housing predictor of consumption levels. Number

33

of persons in the household was a meaningful family factor in pre-
dicting levels of energy consumption; stage of family life cycle
was not a significant factor. Number of major appliances was also
one of the important variables in consumption.

Family income was not found to be a meaningful predictor of
energy consumption in the Morrison study. She suggested that number
of rooms and number of major appliances are positively related to
income levels. However, the third important predictor, number of
persons in the household, may not lend support to the income theory.

A study of household energy consumption and appliance owner-
ship was conducted by Eichenberger (1975). She reported no signifi-
cant differences in ownership of major appliances among the families
with employment of wife as a variable. Literature proposing
causal factors for the increasing levels of household energy consump-
tion had suggested that employed wives were substituting mechanical
energy fer their human energy and time inputs into household work.
The Eichenberger study rejects employment status of women as a predic-
tor of energy consumption levels.

Acceptability of energy policies was the focus of the Zuiches
study (1975). In general, women supported more policies aimed at
energy conservation than men and were more likely to accept energy
shortages as a real problem. The acceptability of policies varied
according to.the degree of coercion and mode of implementation
inherent in the policy. Voluntary policies such as more home
food preparation and home gardening, received the highest overall

support from all respondents. Noncoercive redistribution policies

34

such as tax deductions for home improvements, were acceptable to
over 43 percent of the men and women. Further, policies with low
acceptance were viewed as coercive regulations and included tax
deduction for sterilization, reverse school year seasons, tax deduc-
tion for apartment living, and added taxes for large families.
Zuiches found that energy awareness was the crucial predictor of
policy support, irrespective of level of coercion.

In summary, the effects of family income were reported to
be the best predictor of energy consumption levels in the general
population. However, among single family dwellers number of rooms,
major appliances, and persons in the household explained more of
the variance than other physical and socio-economic factors. Employ-
ment status of the wife is rejected as a determinant of consumption
levels. Families who were most supportive of public policies,
irrespective of the cohersive nature of the proposed policy, had a

high level of energy consumption awareness.

CHAPTER III

DESIGN OF THE STUDY

This study was undertaken to determine the relationships
among family values, conservation practices regarding energy, and
contextual variables. Data collected in the Michigan Agricultural
Experiment Station Project (AESP) 3152, “Functioning of a Family
Ecosystem in a World of Changing Energy Availability," were used to
answer research questions about the interrelationships. These data
were collected from families during May-Jone 1974. The current
study was conceived subsequent to data collection and used a sub-
sample from the larger study. A

Discussion in this chapter will focus on: (1) sampling and
data collection procedures used in the larger study and selection
of the subsample, (2) development of measures for values and prac-
tices and identification of contextual measures, and (3) data analy-

sis procedures.

Procedures for Sampling and Data Collection

Procedures for AESP Studyf

 

The interview population selected for the AESP was the

Lansing Standard Metropolitan Statistical Area (SMSA), a three

 

*

Complete information on sampling procedure and sample
description is presented in Agricultural Experiment Station Project
3152 report (Zuiches, Morrison, and Gladhart, 1975).

35

36

county area which contains the state government complex, a large
university (Michigan State University), commercial enterprises relat-
ing to the automotive industry, and a diversified crop and livestock
agricultural sector. The Lansing SMSA provided a heterogeneous
population of urban suburban and rural households. The family was
defined as two or more related individuals living together, one

who was 18 years of age or older.

A multistage probability sample was selected based on census
tract infbrmation from the 1970 U. S. Bureau of the Census and the
1973 City Directory for Lansing. Random selection of ten tracts out
of the possible 62 tracts which were enumerated in the City Directory
was made. Each tract had a probability of selection proportionate to
the number of households therein. From the ten tracts, a random
sample of 39 blocks out of 478 blocks was selected. Later, five
blocks which encompassed the University population were dropped
because residents did not meet the family criterion. About 20 per-
cent of the population, 615 households was selected. This number
included two extra addresses per needed household to allow for screen-
ing out of ineligible persons and refusals. Using this procedure,
160 families were selected to represent the urban population.

In the rural area of LanSing SMSA, two townships were ran-
domly drawn from the 12 townships within the three counties which
contained no incorporated city or village. The latter criterion for
the rural portion of the sample was implemented to maximize the

rural character of the respondents at the expense of geographical

37

completeness. The primary sampling frame in each township was 36
square mile sections with one section being randomly chosen from
each row of six. Research assistants mapped and numbered the resi-
dences in each of the 12 sections selected. Every second residence
was randomly drawn which allowed for replacement addresses. The
final sample contained 57 rural families and 160 urban families, a
total of 217 families.

Interviewers screened the sample households to determine if
the residents met the AESP criteria of a family. Eligible families
who completed the self-adninistered questionnaires and answered the
interviewer-administered questions received a stipend of $10.

The self-administered sections of the instrument were given
to eligible family members: wife-parent, husband-parent, and oldest
child over 12 years living in the household. All eligible respon-
dents completed a series of questions regarding their knowledge,
belief, and behavior in regard to energy, food, and interpersonal
relationships.

The question of validity of measurement with self-administered
portions of the instrument was recognized by the AESP researchers.

To minimize collaboration between family members,separate color coded
packets of materials were given to each family by the interviewer and
independence of response was verbally stressed. Cursory checks were
made by the interviewer and AESP director before payment was made to
the family. After the data had been coded, the three self— .
administered sections of the instrument were again evaluated and

data from 35 families (16 percent) were found to have handwriting

38

and answer similarity which suggested possible collaboration on one
or more sections.

In order to assess the representativeness of the sample, a
comparison was made between the census data and the sample. Compari-
sons of educational attainment, income level and occupational
classification revealed only minimal divergence from 1970 census
distributions. Further, the sampling procedure did not underrepre-
sent any significant type of housing or family composition. In
sum, the selected sample was found to be representative on a set of
contextual variables deemed significant for generalizing to the

population (Zuiches, et al., 1975).

Selection of the Subsample

 

Criteria for selection of the subsample used in this study
are: (1) presence of a husband and wife respondent in the family,
(2) completeness of data on value and managerial practice variables
and, (3) independence of respondent response on value variables.

The first criterion, husband and wife respondents in each
family, reduced the sample from 217 to 195 families. Omitted from
the sample were 22 single parent households, approximately 5 percent
of the AESP sample. Although seven of the single parent families
had a child over 12 years old as a second family respondent and
therefore value congruence could be measured, the advisability of
treating them as a separate group was negated because of the small
number and lower socio-economic skewed distribution.

The second criterion, completeness of data, eliminated any

family that (l) omitted the managerial practice measure or (2)

39

if either the husband or wife omitted approximately 10 percent of
the items on any one of the value measures.* No families were
eliminated on the basis of missing managerial practice data. How-
ever, 26 families, 13 percent of the husband-wife families in the
subsample, were eliminated because of incomplete value data.

The third criterion, independence of response, further
reduced the subsample by 12 families. Rules adopted for excluding
families were: if husband and wife had identical responses to all
items in a section of the self-administered questionnaire from
which one or more value scales were derived; or if the husband and
wife had handwriting which appeared to be identical and the same
response to 90 percent or more of the items on value scales. Due to
a pattern of erasures and checked answers, it appeared that some
couples worked together in arriving at a "family" response for the
behavioral variables. In Other families, one respondent may have
lacked the time, interest or literacy level needed to complete
his/her questionnaire; the spouse or children may have intervened
in the data collection process.

The missing data and collaboration criteria were not always
mutually exclusive; five families had missing data and patterns of
possible collaboration in their responses. Since the missing data
criterion superseded the collaboration criterion, the five dual
criteria cases were rejected for missing data. In sum, the subsample

of husband-wife families was reduced from the larger study sample as

 

*Specific rules for missing data were developed for each of
the value scales and are listed in Appendix B.

40

follows: 26 families rejected due to incomplete data, five of which
also had collaboration characteristics; 12 families rejected due to
collaboration criterion. The sample used for this study is com-

prised of 157 husband-wife families.

Description of the Subsample

 

The composition and developmental stage of families varied
widely (Table l). About three-fourths of the families had children
living at home, the majority of whom were school age dependents. Of
the families without children living at home (25 percent), an equal
number were younger and older families. Stages of the family life
cycle are presumed to approximate different developmental stages

based on a childbearing and rearing criteria.

TABLE 1.--Stage of the Family Life Cycle of Subsample.

 

 

Stage of Family Life Cyclea Number Percent
Couple, wife under 40 years, no children living

at home 20 12.7
Couple, oldest child six years or less 35 22.3
Couple, oldest child seven to 12 years - 31 19.7
Couple, oldest child 13 to 19 years 35 22.3
Couple, oldest child 20 years and over 16 10.2
Couple, wife over 40 years, no children

living at home _gg_ _12;Z_

TOTAL 157 100.0%

 

aReflects composition of family living in household at time
of survey.

41

Only one family reported presence of an extended family
member and one family listed a nonrelative present in the household.
The most frequent household size was four members: wife, husband,
and two children. However, the number of persons varies from two

to nine (Table 2).

TABLE 2.--Size of Household of Subsample.

 

 

Number of Members in Household Number Percent
2 38 24.2
3 33 21.0
4 45 28.7
5 21 13.4
6 10 6.4
7 4 2.5
8 1 .6
9 __ ___3_-e
TOTAL 157 100.0%

 

About three-fourths of the respondents were in the 18 to 44
age range (Table 3). Family heads in the 65 years and over age
category were under-represented. While 10 percent of the males and
6.6 percent of the females in the Lansing SMSA married population
are 65 years and over, only about 2 percent of the subsample respon-
dents are categorized in this age group. The AESP sample reported
about twice the representation in the forementioned age group as
the subsample. An analysis of rejected subsample families, discussed

later in the chapter, revealed a dominance of older couples.

42

TABLE 3.--Age of Husbands and Wives in Subsample.

 

 

Age Categories Husbaagssgercent Wingssgercent
Under 18 years 0 O
18 - 29 years 35.0 40.8
30 - 44 years 37.6 35.7
45 - 64 years 24.8 21.7
65 years and over __g;§_ __;ng
TOTAL 100.0% 100.0%

 

Family income distribution, as categorized in Table 4,
closely approximates the AESP sample families. The median income
of $13,400 is the same as the AESP reported. A wide range of levels

of economic well-being are represented in the subsample.

TABLE 4.--Family Income Distribution for 1973 of Subsample.

 

 

Gross Family Income Categories Number Percent
Less than $4,999 13 8.3
$ 5,000 to $ 9,999 25 15.9
$10,000 to $14,999 ‘ 51 32.8
$15,000 to $24,999 45 28.7
$25,000 or more _2§__ _14;§_
TOTAL 157 100.0%

Median Income $13,400

 

43

Over 80 percent of the respondents have completed a minimum
of a high school education (Table 5). While the median educational
level of the population and the subsample are approximately the
_same, i.e., high school graduate, the proportion of husbands and
wives with less than a high school education is less in the sub-
sample. Approximately half of the husbands and one-third of the

wives have attended college.

TABLE 5.—-Educational Attainment of Husbands and Wives.

 

Husbands Percent Wives Percent

 

Years of School Completed N=157 N=157
0-11 years, less than high school

graduate 17.4 16.6
12 years, high school completed 33.8 47.1
1-3 years college or vocational

school 21.3 18.5
4 years or more college, profesa

sional training 27.1 17.8

TOTAL 100.0% 100.0%

 

Occupational characteristics by sex role are presented in
Table 6. In comparison to the AESP sample, the subsample under-
represents husbands employed in managerial occupations by about
3 percent and blue collar jobs by 4 percent. This occupational
deviation is confirmed in an analysis of the rejected subsample and
will be discussed later in the chapter.

An analysis of the data for occupation of wives revealed that

this concept may be ambiguous. While about 50 percent of the wives

44

TABLE 6.--Occupation Classification as Defined by Husbands and

Wives in Subsample.

 

Husbands Percent Wives Percent

 

 

 

 

Occupation Category N=157 N=157
Professional 22.3 11.5
Managerial 7.6 4.5
Clerical-Sales 12.1 33.8
Blue Collar 51.6 8.9
Service Workers and Private
Household Workers 3.8 11.5
Homemaker 20.4
No Response 2.5 9.4
TOTAL 100.0% 100.0%
TABLE 7.--Employment Status of WiVes.
Employment Category Number Percent
Fulltime: 35 hours minimum 57 36.3
Parttime: 34 hours or less 22 14.0
Not Employed 78 49.7
TOTAL 157 100.0

 

reported that they were employed (Table 7), 70 percent gave a labor

market category as their occupation (Table 6).

About one-fifth of

the subsample wives indicated that their occupation was homemaker or

housewife, and about 10 percent did not report an occupation.

The most frequent occupation reported by wives was clerical-

sales; professional and service worker occupations both ranked

45

second (Table 6). Wives in the subsample are underrepresented in the
professional and blue collar categories by 5 and 3 percent respec-
tively in comparison to the AESP sample. The single parent families
ineligible for the subsample account for most of the shift in
occupational distribution for wives.

Because of the frequency with which wives are employed
parttime, data were classified as parttime and fulltime employment.
However, fulltime employment occurred 2.5 times as frequently as
parttime employment (Table 7).

Comparison of Chosen and
Rejected Subsample Families

 

 

Since the potential for bias could exist in the findings as
the result of ineligibility of 19 percent of the husband-wife
families in the AESP sample, a number of contextual variables were
compared for the chosen subsample and the rejected subsample
families. These comparisons provided information on a set of measur-
able demographic variables that assisted in defining limits for
generalizing the findings. Supportive data are given in a series of
tables in Appendix A.

In general, the subsample is comparable in demographic
variance to the AESP sample. However, there are some distributional
shifts due to missing data and/or collaboration criteria which
eliminated 38 husband-wife families. The majority of rejected
families were two-member families in the latter stages of the family

life cycle with high school or less education. Although husbands in

46

both the rejected and accepted subsample were most frequently in
blue collar occupations, husbands in the rejected subsample were
overrepresented in the managerial occupation category and under-
represented in the professional category.

The rejected and accepted subsamples were similar in income
and wives' employment status variance. The accepted subsample was
found to be representative of the urban-rural ratio (3:1) in the
AESP sample.

About 25 percent of the rejected families had missing data
on the eco-consciousness scale. Since bias in the findings could
exist if this nonresponse represented a covert response to issues
surrounding the energy problem, measures of attitudes and action
regarding the energy situation were analyzed.

Belief in the reality of the energy problem and actions
taken in response to the energy problem were similar in the rejected
and accepted subsamples (Table 8 and 9). Over 85 percent of the
respondents reported attempting to conserve energy in both sub-
samples. Most families did not respond with an unqualified “yes"
to the reality of the energy problem. However, wives in the accepted
subsample deviated from other respondents in their more frequent
belief in the energy problem. Also, wives in the accepted subsample
were more likely to have persuaded others to conserve energy than
other groups of respondents. In general, the accepted subsample
families are not dissimilar from the rejected families on attitude

and action regarding the energy situation.

47

TABLE 8.--Classification of Accepted and Rejected Subsample Married
Families by Belief in Reality of the Energy Problem.

 

Accepted Families Rejected Families
Husbands Wives Husbands WiVes

 

Is Energy Problem Real?

 

N=157 N=157 N=38 N=38

Yes 42.7 57.3 44.7 44.7
No, but there might be in

the near future 21.0 21.7 10.5 21.1
No, but there might be in

the distant future 14.6 11.5 15.8 13.2
Yes, but it will be solved

by ending Arab Embargo 1.9 1.3 2.6 .
No 19.7 8.3 23.7. 15.8
No Response '0. O. 2.6

100.0% 100.0% 100.0% 100.0%

 

TABLE 9.--C1assification of Accepted and Rejected Subsample Married
Families by Actions Taken in Response to the Energy
Problem.

 

Accepted Families Rejected Families

 

Action in Response to
the Energy Problema Husbands Wives Husbands Wives
N=157 N=157 N=38 N=38

 

Done little or nothing 10.3 ' 7.6 10.5 7.9
Read, studied 50.3 49.7 42.0 44.7
Talked to others about it 52.9 62.4 47.4 50.0
Tried to conserve energy 87.3 90.4 86.8 89.5
Tried to persuade others

to conserve energy 22.9 31.8 23.7 18.4
Taken political action 2.5 3.2 5.3 O.
No response 0. O. O. 2.6

 

aColumns not cumulative due to possible multiple responses
by each respondent.

48

Development of Measures

 

The research model included values, energy conservation
practices, and contextual variables. Individual husband and wife
measures were transformed into family scores for energy conserva-
tion practices and a family typology for each of the four values--
self-esteem, familism, social responsiveness, and eco-consciousness.

The inherent properties of the two variables, values and
practices, suggested different procedures for forming family scores.
Values are psychological constructs, concepts of the desirable, and 'I
not directly observable. Energy conservation practices are overt
activities, have physical properties, and can be directly observed.
The family score for practices is basically a percentage of prac-
tices adopted based on the husband and wife responses.

Nine types of family value configurations were conceptual-
ized based on the husband-wife degree of congruency and commitment
for each value. For example, husband and wife may both hold the
valuetyfeco-consciousness with high, medium, or low commitment; the
husband may hold the value with high commitment and the wife with
low commitment and thus have an incongruent value commitment. Each
of the values was studied independently rather than combined into a
composite of the fOur values. The nominal scale properties of value
patterns and sample size prevented the latter approach.

Discussion which follows will center on describing each of
the measures and the scoring procedures. The energy conservation

measure is a dependent or criterion variable in the parametric models

49

and therefore the measures of central tendency and dispersion will
be discussed in relation to the assumptions of the statistical
models. Each of the four value scales will be described and the .
reliability coefficient reported for the aggregate of husbands and
wives separately. The coefficient reported for reliability repre-
sents a measure of internal consistency, i.e., unity of scale items
(Mehrens and Lehmann, 1973). Since the procedure for developing the
nine/types of intrafamilial value configurations was the same for
each of the values, family value scores will follow scale descrip-'
tions.

Contextual variables will be briefly described. Each of
the scales and coding rules for all variables is displayed in
Appendix 8.
Measure: Energy Conservation
Practices

A scale of 14 practices was used to measure the rate of
adoption of energy conservation practices within the household over
the past year. The measure included practices which met the follow-
ing criteria: practices widely promoted for the one-year period
prior to the survey and during the survey to minimize the possi-
bility of families being unfamiliar with it; appropriate for families
without major capital outlay; and, independent so adoption of one
practice would not necessitate the adoption of other practices.
Respondents were directed to indicate if the practice was implemented

more during the winter of 1973-74 than the previous season, the same

50

as the previous season, or if the practice was not physically possi-
ble in their home.

A family energy conservation index was developed to represent
the increase of practices families adopted within the past year. One
point was given for each practice adopted on the l4-item scale and
the sum divided by the total possible; a percentage score was the
statistic. For example, if the husband and wife indicated it was
"not physically possible" to close the fireplace damper nor control
water heater temperatures, and had adopted six of the remaining 12
practices during 1973-74, their score was 50 percent.

The response of family respondents was not the same for all
scale items. It is assumed that lack of intrafamilial agreement
occurred because the spouses were not equally informed about prac-
tices adopted and/or they used a personal rather than a family frame
of reference for their response. Rules for arriving at a family
score were developed for cases in which the husband and wife gave
different answers. They were:

Rule 1: If one spouse indicated "this season more than

last season," the scale item was coded as a
practice adopted. '

Rule 2: If one spouse indicated "not physically possible"

and the other spouse recorded "this season the
same as last season," additional information
from the AESP data bank was used to validate
the answer.

The frequency of adoption for each practice in families
(Table 10) is based on the use of Rule 1. In about 85 percent of

the families, spouses differed in reporting practice adoption. For

51

TABLE lO.--Energy Conservation Practices Adapted by Families.

 

 

Families
Energy Conservation Practice Percent

N=157
Turn off lights not . . . in use 83
Maintain daytime . . . 68° 80
Turn down thermostat . . . night 76
Wear heavier clothing at home 76
Turn down thermostat . . . vacation 74
Close off . . . unused areas of home 61
Close drapes . . . night 55
Open drapes . . . sunny days 54
Replace bulbs . . . lower wattage 54
Use only warm-cold cycles . . . washing 44
Change furnace air filter . . . 43
Have heating equipment checked . . . 28
Turn down temperature . . . water heater 26
Close fireplace damper . . . 15

 

example, the husband may have reported that the heating equipment
was checked and the wife indicated that warm-cold wash cycle was an
adopted practice. The frequency with which Rule 1 was used in cod-
ing family scores suggested that sex role differences were reflected
in this measure.

The most frequently adopted practices by families were turn-
ingiyffunused lights, adjusting the thermostat to use less energy,
and wearing heavier clothing. About 4 out of 5 families reported
that they maintain daytime temperatures of 68° or less and turned
off all lights not actually in use. The low adoption rate of

‘\

52

furnace maintenance, water temperature efficiency, and fireplace
adjustments reflects the presence of dwellings where these practices
were "not physically possible" in the sample. So these families
were not penalized, the family practice score was calculated with
the number of practices physically possible as the denominator.

The determination of "not physically possible" required
the use of Rule 2. Most frequently the husband and wife did not
agree on the physical possibility of closing doors and turning off
heat to unused areas of the house, having heating equipment adjusted
for efficiency by a serviceman, and turning down temperature control
of hot water heater. There were 66 occurrences of Rule 2 and it
affected 44 family scores (Appendix B, Table B-1, page 118).

To implement Rule 2, additional family data were used to
validate the physical possiblity of the practice being adopted.
Most frequently data on the existence of a thermostat, ownership of
a water heater and washing machine, and type of furnace if family
did not live in an apartment-type dwelling were used to code the
family response. The lack of agreement on closing off unused areas
of the home was perceived too subjective to be determined by data on
number of rooms and size of family. Therefore, in the ten families
where this disagreement occurred, the practice was coded "not
physically possible."

The analysis strategies used to test the association 0f
independent variables with energy conservation practices, the

dependent variable, assumed some level of normality. Multiple

53

regression, analysis of variance,and multiple classification analy-
sis (the statistical models used) are based on the assumption that
the sample has been drawn from a population that is normally dis-
tributed. However, most research reported on the effects of sample
nonnormality and heterogeneity of variance conclude that these
parametric measures are robust, i.e., the ordinary t and F tests are
nearly immune to violation of these assumptions (Boneau, 1971;
Lindquist, 1953; Glass and Stanley, 1970). Lindquist stated:
. unless the heterogeneity of either form or variance
is so extreme as to be readily apparent upon inspection

of the data, the effect upon the F distribution will proba-
bly be negligible (p. 86).

TABLE 11.--Descriptive Statistics for Energy Conservation Scale.

 

 

Descriptive Statistic Faniiges
Measure of central tendency
Mean 60.8
Median 63.9
Measures of dispersion
Variance , 839.3
Standard Deviation 28.9
Standard Error 2.3
Kurtosis -.8
Skewness -.4

Measure of reliability
Alpha coefficient .85

 

54

The descriptive statistics for the distribution of the
conservation adoption scores (Table 11) indicate that the scores
deviated slightly from normality. The mean and median would both
be 50 with a symmetric distribution; greater numbers of families
have a higher adoption rate than would be expected from the popula-
tion. The negative skewness, -.4, revealed a distribution with
more values clustered to the right and some extreme values to the
left. A flatter than normal distribution is indicated by the nega-
tive kurtosis.

The effects of nonnormality in this sample of scores were
viewed as within the tolerable limits of parametric statistics,
with some minor reservations. The probability of a type one error,
i.e., the null hypothesis is rejected when in fact it should not be
rejected, may be increased. According to Lindquist (1971):

. . . the F-distribution is practically unaffected by lack
of symmetry, per se, in the distributions of criterion
measures, but is slightly affected if the distribution of
criterion measures is roughly symmetrical but either very
flat or very peaked. In the latter cases, the probabilities
read from the normal-theory F-table are too small to repre-
sent the true risk of a Type I error, and due allowances
should be made for this in the interpretation of results.
In such cases, . . . when the "apparent" risk (as read from
the F-table) of a Type I error is 5%, the true risk may be
as large as 8%, and when the apparent level of significance
of an F-test is the 1% level, the actual level of signifi-
cance may be the 2% level (approximately) (p. 351).
In summary, the reported tests of association and differences using
parametric models, will be discussed with the probability of Type

I error due to distribution nonnormality included.

55

Reliability was established for the subsample on the prac-
tice adoption scale. The reliability coefficient, .85, represents

a measure of high internal consistency for the scale (Table 11).

Measure: Self-Esteem Value

 

The self-esteem test was developed by Rosenberg (1965) to
measure attitudes toward the self along a favorable to unfavorable
dimension. It consisted of ten statements on a Likert-type summated
rating scale (Appendix 8). Robinson and Shaver (1973:98-110) docu-
mented construct validity and reliability for the scale.

Self-esteem is defined as the respect that a person has for
self and included recognition of self-limitation and growth poten-
tial. This value is the conception of self as an overall judgment
of worth as a person.

Reliability was established for the subsample. The relia-
bility coefficients, .74 for husbands and .79 for wives, represent
a measure of internal consistency for the scale.

Measures of central tendency included: 3.19, 3.18 means for
husbands, wives; 3.09 median for both husbands and wives. In a test
of differences between the mean of husbands and wives, they were

found not to be significantly different.

Measure: Familism Value
The measure of familism value was originally developed by
van der Veen, et al., (1964; 1974) to measure the real and the ideal

concepts of family as perceived by family members. The AESP adapted

56

the measure to be used as a self-administered test by selecting 20
of the 80 items based on item discrimination scores reported by
Imig (1971). The scale, Family Concept Inventory, was also converted
from a Q—sort to a Likert-type scale so it could be self-administered
with a large sample.

The value of familism is defined as the perception that a
family member has of his/her family unit and the level of integra-
tion of its membership. For this study, one item was omitted from
the 20-item scale in an effort to maintain independence of value .
scales. The item, "we get along very well in the community," was
inconsistent with the conceptual definition of familism and over-
lapped with the social responsiveness value. The scale items used
to measure familism are listed in Appendix B.

The alpha coefficients for reliability were calculated for
the l9-item scale used to measure familism. The coefficients, .89
for wives and .86 for husbands, were the highest of any of the
scales.

The means were similar for husbands and wives and in a test
of difference, the husband and wife scores did not differ signifi—
cantly. (Mean 3.74 for husbands, 3.80 for wives; median 3.79 for
husbands, 3.89 for wives.)

Measure: Social Responsiveness
Value

Social responsiveness value is defined as the conception of

the interdependence of person-society, the degree of integration of

57

family members into larger social structure. The eight-item measure
is listed in Appendix B.

This measure included the five-item anomia scale developed
by Srole (1965) and three items from a test of internal versus
external control by Rotter (1966). The anomia scale was developed
to measure integration of the individual into society. Robinson
and Shaver (1973:173) reported the Likert-type scales met the
criteria of unidimensionality. Since only three of the 23 forced
choice pairs of items developed by Rotter were used to measure
societal connectedness, statements of support for validity from the
widely used index are inappropriate.

The social responsiveness scale was analyzed to determine
reliability for the subsample. The alpha coefficients which measure
the unidimensionality of the scale, were .79 for husbands and .77
for wives.

In a test of the difference between two dependent means, the
husbands'and wives'scores were not found to be significantly differ-
ent. Means were 3.01, 3.07 for husbands and wives and medians 2.88,
3.13 for husbands and wives, respectively.

Measure: Eco-Consciousness
Value

 

The eco-consciousness measure was adapted from two Likert-
type scales used as indices of a conceptualization of ecological
linkages to the energy situation. The value of eco-consciousness is

defined as the perception of the interrelationship of man-nature.

58

The scale items used to measure the value were selected from scales
in the larger study to measure this unitary concept (Appendix 8).
Several scales were constructed and the scale item correlations
examined. The final eight-item measure reflects an overall alpha
correlation of .75 for husbands and .81 for wives.

The mean and median scores for husbands and wives were
similar: 3.62 mean and median for husbands; 3.77 mean and 3.87
median for wives. In a test of difference between the means of
the two groups, no significant difference was found due to sex role.

Typologyiof Intrafamilial
Value Patterns

 

Family value patterns were conceptualized based on the
intrafamilial patterns of congruency and commitment. The basic
procedure for developing the family typology was the same for each
of the four values and was developed in three steps. First, an
average score was calculated for each respondent from the raw score
items.

Second, the distribution of means for the aggregate of
husbands and wives was calculated separately to test for sex role
bias in each of the value scales. This test of two dependent means
was a prerequisite to determining the procedure for partitioning
the scores into high, medium, and low commitment categories. If the
test had indicated that the distribution of the means was signifi-
cantly different for wives and husbands, the partitioning would have

been done separately for husbands and wives scores. As reported in

59

the description of each value scale, there was no significant dif-

ferences between scale means.

TABLE 12.--Distribution of Value Scores Above, Around, and Below
the Mean by Husbands and Wives.

 

 

. . Scores to be Husbands Wives
Classif1cation Of Scale Included Percent Percent
Self-esteem

Above mean 3.33 - 4.00 31 32
Around mean 3.50 3.88 26 22
Below mean 2.20 - 3.00 _1E1 _4§_
100% 100%
Familism
Above mean 3.95 - 4.84 34 38
Around mean 3.77 3.95 28 26
Below mean 2.20 - 3.78 _;§1 _§§_
100% 100%
Social Responsiveness
Above mean 3.38 - 4.50 38 34
Around mean 2.75 3.25 26 25
Below mean 1.38 - 2.74 _§§_ _51.
100% 100%
Eco-consciousness
Above mean 3.87 - 5.00. 36 34
Around mean 3.50 - 3.86 22 23
Below mean 1.50 - 3.49 _J§; _J£§
100% 100%

 

The last step in developing the family typology was classi-

fication of husband-wife scores into one of nine intrafamilial types

based on the partitioned distribution (Table 12). Approximately

60

one-fourth of the scores around the mean comprised the "medium
commitment" classification. Scores higher than this group were
defined as "high commitment" and scores lower than around the mean
were termed "low commitment." The family typology was as follows:
1. Husband-wife have high commitment to the value.
Husband has high commitment and wife medium commitment.
Wife has high commitment and husband medium commitment.
Husband-wife have medium commitment.
Wife has high commitment and husband low commitment.
Husband has high commitment and wife low commitment.
Wife has medium commitment and husband low commitment.

Husband has medium commitment and wife low commitment.

«amnesia-boom

Husband-wife have low commitment.

Three of the patterns reflect high levels of value congru-
ency between husband and wife: husband—wife above the mean, around
the mean,and below the mean (patterns 1, 4, and 9 respectively).
About half of the 157 families were classified in one of these con-
gruent patterns for familism, eco-consciousness, and social respon-
siveness (Table 13). Only 39 percent of the couples held a congru-
ent self-esteem value.

The maximum incongruency between husband and wife on a
given value is categorized in patterns 5 and 6, husband below the
nean and Wife above the mean and visa versa. Most frequently this

incongruency was found in the self-esteem value, about 22 percent;

61

.chos PmUPumwumum mucmmco> mo mmmxpmcm cw pm>mp genome Fpmom

 

 

 

~m_ ~m_ um, um, o<eoe

mm mm on em zoos sopoo ace: .ocoamax .m

up 5 NF mm cows zopmn mew: .uczosm ucmamzz .m

NF m _F _P zoos :oPoo ocoamag .ocaooo ace: .N

_F o_ a RF cooe zopoa oat: .o>ooa econmzz .o

m m_ m_ N_ cooe zoFoo oceans; .o>ooo act: .m

mp mp op u come ucaocm mew: .ucoamzz .e

op op a, S_ cows ocsoca oceans; .o>ooo out: .m

N_ op op mp cams ucaogo mew; .m>onm ucmamzz .N

mm mm on mp some m>onm mew: .ccmnmaz .p
lemmas: Lonszz Lmnszz Longsz

upwpwnmmcoammm mmmcmzowumcou EmmeEmm smmumm mmccmuuma >Fw5mu
_owoom -oom -LLom

 

.mmwcm>wmcoammm Pewuom one .mmmcmaomuwcoUtoum .Emmmemm
.Ewmummimmeitmm=_m> we mccmuuma mapm> zfiwsmu mo mocmacwcu ucm cowumumwwmmmFoit.mF mpm<k

62

eco-consciousness, familism, and social responsibility had declining
proportions of families, l8, l4, and 13 percent respectively.

The highest commitment to a given value is categorized in
pattern 1, husband and wife above the mean,and the lowest commit-
ment in pattern 9, husband and wife below the mean. Families tabu-
lated as patterns 2 and 3 reflect a higher commitment than families
who were pattern 7 and 8.

While there is some hierarchial order to the patterns, the
typology is a nominal measurement, i.e., all the numbers assigned
to the patterns cannot be ordered or added. For example, there is
no conceptual basis for weighing pattern 5 over pattern 6 or visa
versa. Statistical tests to determine significance among different
patterns in relation to energy conservation practices and contextual
variables were conducted using analysis of variance and multiple
classification analysis. The nine types of families with respect to
direction and congruency of values were cell factor levels in the

analysis of variance model.

Measures: Contextual Variables

The contextual variables included in the study were: educa-
tion, occupation, and age of husband; education, employment status,
and age of wife; and family income, size, stage of family life
cycle, and urban-rural residency. The distribution of these vari-
ables was discussed in describing the sample. The measures are
standardized indices; the coding for contextual variables is listed

in Appendix B.

63

Occupation of husbands was recoded from census categories
into prestige scores. The research methodology employed for the
prestige scores was implemented and reported by Siegel (1975) and
is an update of the 1963 North-Hatt-NORC study. The National
Opinion Research Center, University of Chicago, sponsored the studies
to measure the social status associated with occupations.

Since occupation of wives was reported earlier in this
chapter as an ambiguous concept, employment status was operational-
ized as an alternative variable. A dichotomous employment status
variable, employed or not employed, divided the sample in equal

proportions.

Research Hypotheses

 

The data analysis will be reported around general research
questions because of the large number of variables tested in this
exploratory study. Differences among the types of intrafamilial
value patterns in the adoption of household energy conservation
practices were specifically hypothesized and other linkages were
stated in general hypothesizes.

General Research Question I: Is there a difference among the

intrafamilial'value patterns in adoption of energy conserva-
tion practices?

 

Ho 1: There is a difference among the nine intrafamilial
value patterns of self-esteem in adoption of energy
conservation practices.

Ho 2: There is a difference among the nine intrafamilial
value patterns of familism in adoption of energy
conservation practices.

64

Ho 3: There is a difference among the nine intrafamilial
value patterns of social responsiveness in adoption
of energy conservation practices.

Ho 4: There is a difference among the nine intrafamilial
value patterns of eco-conséiousness in adoption of
energy conservation practices.

 

General Research Question II: Is there a relationship between
adoption of energy conservation practices and contextual
variables?

 

Ho 5: Adoption of energy conservation practices vary with
contextual variables--occupation, education, and
age of husband; employment status, education, and
age of wife; family size, income, stage of family
life cycle, and urban-rural residence.

General Research Question III: Is there a difference among the
intrafamilial value patterns with respect to contextual varia-
bles?

 

Ho 6: There is a difference among the nine intrafamilial
value patterns of self-esteem, familism, social
responsiveness, and eco-consciousness with respect
to contextual variables--occupation, education,
and age of husband; employment status, education,
and age of wife; family size, income, and stage
of family life cycle.

Data Analysis Procedures

 

Parametric and nonparametric statistical models were used
in hypotheses testing. An alpha level of .05 was used in two-
tailed tests to determine the probability of a Type I error, i.e.,
the null hypothesis is rejected and in fact it is true.

The statistics (t,F) used to generate alpha levels of
significance were not the sole criteria for evaluating the hypothe-
ses. Rather, the test of significance and coefficients of correla-
tion both assisted in determining the degree to which the relation-

ships were meaningful pr0positions (Rozeboom, 1971).

65

Statistical Models

 

The parametric models that were used to test the hypotheses
included analysis of variance and covariance, multiple classification
analysis, and multiple regression. The underlying assumptions of
the parametric models, normality and continuous variables, were
reviewed first. The assumption of normality evaluated in describing
the measures of dispersion for household energy conservation prac-
tices was of primary concern. It was concluded that the effects of
nonnormality were within the tolerable limits of parametric sta-
tistics.

One-way analysis of variance was used to test the differences
among intrafamilial value patterns in adoption of energy conserva-
tion practices and with respect to contextual variables (hypotheses
1 to 4 and 6). Each family value typology was an independent varia-
ble, i.e., factor with nine cells, a nonorthogonal design due to
unequal cell frequencies.

Post hoc tests were implemented on contrasts developed from
a priori contrasts to investigate which intrafamilial value patterns
contributed to the tested difference in adoption of energy conserva-
tion practices. This technique was implemented only if a significant
overall difference among intrafamilial patterns was determined with
the criterion variables in an analysis of variance test (Glass and
Stanley, 1970).

Analysis of covariance was used to test the differences in

practice adoption rate among intrafamilial value patterns while

66

controlling for the effect of contextual variables. Covariates,
the contextual variables, were selected to remove extraneous varia-
tion from the criterion variable, energy conservation practices.

Multiple classification analysis (MCA) is a statistical
technique for examining the relationship between independent varia-
bles, with no better than nominal measurement, and an intervally
scaled dependent variables (Andrews, et. al., 1973). MCA, a multi-
variate technique, calculated the coefficients of the relationship
between the patterns of family values and adoption of energy con-
servation practices before and after adjusting for the effects of
covariates. This parametric model is designed to handle nominal
level independent variables, correlated independent variables, and
nonlinear relationships. Tests of significance (F,t) generated in
the analysis of variance procedures were used to determine the
probability of the relationships.

Multiple regression, a statistical technique by which the
linear dependence of one variable on others is summarized and
decomposed so relationships in the population can be evaluated,
was used to predict the adoption of energy conservation practices
from contextual variables. Since interval measurement is required
for multivariate regression, family values and stages of the family
life cycle variables were not included. Forward stepwise regression
was used to order the contextual variables to provide the best
prediction equation to predict energy conservation practices.

Lambda and uncertainty coefficient, nonparametric statistics,

were used to measure the relationship between stages of the family

67

life cycle and each family value pattern. Both of the variables
were nominal in scale; statistics were calculated from crosstabula-
tions. Lambda measures the percentage of improvement in predicting
the dependent variable based on the modal value of the independent
variable (Babbie, 1973). The uncertainty coefficient is similar to
lamdba, except that the uncertainty coefficient considers the entire
distribution, not just the mode (Nie, et al., 1975). The uncertainty
coefficient is the proportion by which "uncertainty" in the depend-
ent variable is reduced by knowledge of the distribution of the
independent variable. The concept of uncertainty is related to the
ambiguity of data distributions and was developed from literature on
information theory.

Chi-square, a nonparametric test of statistical significance,
could not be used to determine whether a systematic relationship
existed between family values and stages of family life cycle because
the sample was not sufficiently large to meet the cell frequency
requirements. The six by nine design would require a sample of 270
to meet the minimal 5 per cell requirement. Collapsing categories
in levels of variables to achieve a five by six design was consid-
ered inappropriate due to information loss and problems of interpret-

ing the results.

Computer Programs
The Control Data Corporation 6500 model computer was used to
perform all of the analyses. The programs to compute the statistics

were available through the 6.0 version of the Statistical Package

68

for Social Sciences (Nie, et al., 1970). All of the computations
were implemented at the Michigan State University Computer Labora-

tory.

CHAPTER IV

RESULTS OF DATA ANALYSIS

Results of the data analyses are reported in relation to six
hypotheses and are presented under the following section headings:
(1) family values and energy conservation practices, (2) contextual
variables and energy conservation practices, (3) family values and

contextual variables, and (4) summary.

FamilyAValues and Energy_Conservation Practices

 

Each of the values were entered separately as factors* in a
one-way analysis of variance with adoption of energy conservation
practices as the criterion variable. The first three hypotheses will
be discussed as a group because of similar test outcome.

Ho 1: There is a difference among the nine intrafamilial
value patterns of self-esteem in adoption of energy
conservation practices.
Ho 2: There is a difference among the nine intrafamilial
value patterns of familism in adoption of energy
conservation practices.
Ho 3: There is a difference among the nine intrafamilial
value patterns of social responsiveness in adoption
of energy conservation practices.
The magnitude of the F-statistic suggested that there was no

difference among the intrafamilial patterns of self-esteem,

 

*The nine cell factor levels are described on page 60.

69

7O

familism and social responsiveness in the adoption of energy conser-
vation practices (Table 14). Thus,Ho 1, Ho 2, and Ho 3 were

rejected.

TABLE l4.--Summary of One-Way ANOVA Tests for Differences in
Adoption of Energy Conservation Practices by Family
Values--Self—esteem, Familism, Social Responsiveness,
and Eco-consciousness.

 

One-Way ANOVA Testsa

Family Value Between Group Within Group F-Statistic
Mean Squares Mean Squares (Probability)

 

 

Self-esteem .1222 .0817 1.496 (.163)
Familism .0538 .0854 .630 (.752)
Social responsiveness .0490 .0856 .573 (.799)
Eco-consciousness .1886 .0781 2.415 (.018)b

 

as and 148 df
bSignificant level: p < .05.

Ho 4: There is a difference among the nine intrafamilial
value patterns of eco-consciousness in adoption of
energy conservation practices.

There was support for the hypothesized difference (Table 14).
An examination of means and standard deviations of energy conserva4
tion practices by pattern of eco-consciousness in families revealed
that families with high commitment to eco-consciousness adopted 72

percent of the practices in comparison to 46 percent by families

with low commitment to the value (Table 15).

71

TABLE 15.--Means and Standard Deviaions of Energy Conservation

Scores Based on Eco-consciousness Family Value Patterns.

 

 

 

 

 

 

Family Typologya N Mean Std. Dev.
1. Husband, wife above mean 36 .72 .21
2. Husband above, wife around mean 16 .56 .24
3. Wife above, husband around mean 16 .53 .35
4. Husband, wife around mean 13 .56 .27
5. Wife above, husband below mean 15 .69 .25
6. Husband above, wife below mean 14 .65 .25
7. Wife around, husband below mean 8 .63 .35
8. Husband around, wife below mean 7 .74 .30
9. Husband, wife below mean _§g. .46 .31
TOTAL (Grand Mean) 157 (.61)
aCell factor levels in analysis of variance statistical
model.
TABLE 16.--Post Hoc Comparisons of A Priori Contrasts 0f Eco-
Consciousness Family Value Patterns in Adoption of
Energy Conservation Practices.
Se arate Variance
c a Pagled Varianceb p
ontrast - tatist c T-Statistic
' f
(Probabil1ty) (Probability) d
Family type 1 and 9 3.710 (.000)c . 3.800 (.000)c 54.0
Family type 1 and 4 1.744 (.083) 1.860 (.079) 17.6
Family type 4 and 9 1.025 (.307) .994 (.330) 25.2
Family type 1 and 5/6 .655 (.514) .771 (.446) 34.7
Family type 1/4 and 5/6 -.478 (.633) -.519 (.606) 44.0

 

aFamily types classified in Table 15.
b148 df

cSignificant level: p < .05.

72

The location of the differences between means was calculated
using post hoc tests (Table 16). Families with high and low
commitment to eco-consciousness value had significantly different
adoption of energy conservation practices (contrast: family type 1
and 9). The t-statistic was in the area of rejection for contrasts
involving congruency differences (contrasts: family type 1 and 5/6;
family type 1/4 and 5/6).

Analysis of covariance was computed with husband's occupa-
tion, wife's education, and husband's education as covariates. The
covariates were selected because of their supported variance with
family eco-consciousness. However, husband's education as a covar-
iate was dropped becauSe it was in the area of rejection in the
analysis of covariance test (F-statistic .713; p > .99). With
husband's occupation and wife's education as covariates, the differ—
ences among the nine intrafamilial value patterns of eco-
consciousness in adoption of energy conservation practices was
supported (Table 17).

Multiple classification analysis was computed to determine
the magnitude of the relationship between'eco-consciousness and
practices, with and without the covariates (Table 18). In general,
there were small differences in the adoption of energy conservation
practices due to the confbunding effects of wife's education and
husband's occupation. The combined effects of family eco-
consciousness, wife's education, and husband's occupation increased

prediction of practices by about 4 percent. Approximately 11.5

73

TABLE l7.--Analysis of Covariance Test of Differences in Adoption of
Energy Conservation Practices Among Intrafamilial
Patterns of Eco-Consciousness with Control Variables.

 

 

. Mean F-Statistic

Source of Variat1on Squares df (Probability)
Covariates .221 2.925 (.05)a

Husband's occupation .336 1 4.452 (.03)a

Wife's education .306 1 4.053 (.04)a
Main effects .200 8 2.553 (.01)a
Residual .075 l_ji

TOTAL .084 156

 

aSignificant level: p < .05.

TABLE 18.--Multip1e Classification Analysis of Relationship Between
Eco-consciousness Value and Energy Conservation Practices
with Covariates: Wife's Education and Husband's Occupa-

 

 

 

 

tion.

Deviation from Grand Meana

Eco-consciousness Value N

Family Typology ‘. Adjusted for
Unadjusted Covariates

l. Husband, wife above mean 36 .ll .13

2. Husband above, wife around mean 16 -.05 -.05

3. Wife above, husband around mean 16 ' -.O7 -.06

4. Husband, wife around mean 13 -.05 -.06

5. Wife above, husband below mean 15 .09 .06

6. Husband above, wife below mean 14 .04 .05

7. Wife around, husband below mean 8 .02 .OO

8. Husband around, wife below mean 7 .14 .17

9. Husband, wife below mean _3g, —.14 -.15

TOTAL 157
Multiple R .340 .396
Multiple R2 .115 .156

aGrand mean - .61 (61 percent of practices adopted).

74

percent of the variation in adoption of energy conservation prac-
tices can be predicted by family eco-consciousness value. Thus,

Ho 4 is accepted.

Contextual Variables and Energy Conservation Practices

 

The relationship between eight of the nine contextual vari-
ables and adoption of energy conservation practices was investigated
simultaneously using a step-wise multiple regression model. Differ-
ences in practices among the stages of the family life cycle were
tested using analysis of variance.

Hg_§; Adoption of energy conservation practices vary
with contextual variables--occupation, educa-
tion, and age of wife; family size, income,
stage of family life cycle, and urban-rural
reSidence.

The eight contextual variables were entered into the
regression equation to predict adoption of energy conservation
practices (Table 19). Wife's education was the only variable that
met the F-test of significance criterion and since it predicted
less than 1 percent of the variance in energy conservation prac-
tices, wife's education was not a meaningful predictor. The
contextual variables were not meaningful predictors.

The stage of the family life cycle was entered in a one-way
analysis of variance model as a factor and adoption of energy con-
servation practices as the criterion variable. No support was noted

for differences among the stages of the cycle, in adoption of energy

conservation practices. Thus, Ho 5 was rejected.

75

TABLE 19.--Multiple Regression Analysis Predicting Adoption of
Energy Conservation Practices From Contextual Variables.

 

StepwisegRegression

 

Correlation

 

 

C°"tEXt“a‘ variab‘e Coefficienta F-Statistic Additive
(Probability) R2

Wife's education .09 4.905 (.03)b .008
Husband's occupation -.10 2.562 (.11) .026
Wife's employment status -.08 1.150 (.28) .010
Husband's education -.03 .458 (.50) .004
Family income -.07 .701 (.40) .004
Family Size .01 .449 (.50) .003
Urban-rural residence .04 .198 (.66) .001
Wife's age .02 .525 (.81) .001
Husband's age .03 .244 (.88) .000
Multiple R - .238 R2 _ .157 F-Statistic - .978

df - 9 and 147 Probability - .460
a.159 required for .05 significance level.

bSignificant level: p < .05.

TABLE 20.--One-Way ANOVA Test for Differences Among the Stages of
Family Life Cycle Relationship in Adoption of Energy
Conservation Practices.

 

 

. Degrees Mean F-Statistic
Source Of Var1ation Freedom Squares (Probability)
Between groups 5 .0568 .671 (.646)
Within groups 151 .0846

TOTAL 156

 

76

Family Values and Contextual Variables

 

A difference between intrafamilial value patterns and
contextual variables was posed in a general hypothesis.

Ho 6: There is a difference among the nine intrafamilial
value patterns of self-esteem, familism, social
responsiveness, and eco-consciousness with respect
to contextual variables--occupation, education, and
age of husband; employment status, education, and
age of wife; family size, income, and stage of
family life cycle.

TABLE 21.--Summary of ANOVA and MCA Tests for Differences Among
Intrafamilial Patterns of Social Responsiveness Value
in Respect to Contextual Variables.

 

 

 

One-Way ANOVA Testsa MCA Testsb

Contextual Variable F-Statistic _——;;—_—-'
(Probability) a
Husband's education 4.476 (.001)c .44
Wife's education 3.429 (.001)c .40
Family income '2.320 (.023)c .33
Husband's occupation 1.789 (.083) .30
Family size 1.501 (.161) .27
Wife's age 1.244 (.277) .25
Husband's age 1.210 (.296) .25
Wife's employment .628 (.999) .18

 

aDegrees of freedom 8 and 148 in Analysis of Variance Test.

. bEta statistic equivalent to multiple R in Multiple Classifi-
cation Analysis Test.

cSignificant level: p < .05.

77

A series of one-way analysis of variance tests was performed with
all possible combinations of contextual variables and each family
value. The contextual variables were entered into the model as
criterion variables and the intrafamilial value patterns were factor
level cells. Relationship coefficients, eta, were calculated by
entering each of the 32 combinations of variables in multiple clas-
sification analysis procedure.

Social responsiveness family value patterns with respect to
husband's education, wife's education, and family income had signifi-
cant F-ratios (Table 21). The strength of the relationship was
greatest for education of the husband and the wife with social
responsiveness family value, .44 and .40 eta coefficients respec-
tively (Table 21). Family income had a significant but less mean-
ingful variance with the value.

Means and standard deviations of husband's education, wife's
education, and family income among the family patterns of social
responsiveness were examined for a priori contrasts (Table 22). Post
hoc comparison tests were conducted in respect to the three contex-
tual variables (Table 23). Couples with congruent high commitment
to social responsiveness were more highly educated and had higher
incomes than couples with congruent low commitment to the value
(contrast: family type 1 and 9). There was additional support that
level of education is related to high value commitment in contrast
to families with low commitment and incongruency (wife's education

contrast--family type 1 and 8/9; husband's education contrasts--

78

.quoe Fmowumwuepm mucmwce> mo mmmapmcm cm m_o>m_ Louume ppmup

 

 

 

 

 

Am.m_v Am.m.v A~.N_v Amp Aeaoz neocov o<eoe
A~.mv .mnflfli Am.NV.imJHMt Aa.ev.iwufl~i .mmi eons zo_oa ace; .eeoaazz .m
Aa.mv m.o_ AN.ev e.F_ Am.Nv ~.__ A. code zopoa ace: .eeaoca oceans: .w
Ap.ev a.e_ A~.mv m.m_ Am.Nv m.mp NF code zopoa oceans; .eeaaca ace: .5
Am.~v N.m_ Ae.Nv m.m_ Ac._v F.m_ F_ cede zopoa ace; .o>oaa oceans: .a
Am.av m.op AN._V a.m_ Aa.~v m.mp a code zopoa oceans; .o>oae out: .m
Am.ev m.mp Am.NV m.mF Ao.NV a.NF m_ eaoe eeaoco oer: .eeeamax .e
Ae.mv a.ep Am.Ne e.mp Am.mv c.N_ op cape eesoco eeaaaae .o>oaa dew: .m
Ae.ev N.m_ A_.~V N.mp Am.Nv ~.~_ N. eaoe eeaoca ace; .o>oae oceans: .N
A~.av “.mp Am.mv a.mp Ae.~v a.mp mm eooe o>aaa ace: .eeaamsz ._
.a.m eaoz .o.m coo: .o.m eaoz
xmo on» » Fae
Mammmw ”wwwmmmmm cowwwwmmm z mzfim> mmmcmmwmcwammw wauom

 

.mccmupma mspm> PemmemCeLucH mmocm>wmcoammm Pmmuom zn commm msoucH apwsmm
eee .eowoaosem m.oeoam== .eoeoeoaom m.oea3 co aeoeoa_>oo oceeeeom pea memos--.mm mom<e

79

TABLE 23.—-A Priori Contrasts of Wife's Education, Husband's Educa-
tion, and Family Income Based on Family Value of Social

 

 

 

Responsiveness.
Pooled Variance . Separate Variance
Contextual Variable
and Contrastsa t-Statistic df t-Statistic df
(Probability) (Probability)

 

Wife's education

Family type 1 and 9 4.178 E.
Family type 1 and 8/9 4. 856 000 b 148 4.568
Family type 1 and 5/6 1.071 (.286) 148 1.037

Husband's education b
Family type 1 and 9 5. 234 (. 000)b 148 5.145 (.OOO)b 66.8
Family type 1 and 5/6 2.049 (.O42)B 148 2.325 (.028): 27.7
Family type 1 and 5/7 2.361 (. 020)b 148 2.555 (.017)b 27.0

Family income b
Family type 1 and 9 2.033 (.044) 148 1.814 (.074) 66.1
Family type 1 and 5/6 .457 (.649) 148 .474 (.614) 21.0

.000

.000 b 148 4.300 é.000)§ 63.3
(.314) 16.9

 

aFamiiy types classified in Table 22.
bSignificant level: p < .05.

family type 1 and 5/6, family type 1 and 5/7). There was insuffi-
cient support for linkage of family income to value congruency
(contrast: family type 1 and 5/6).

Eco-consciousness family value patterns with respect to
wife's education and husband's education and occupation had signifi-
cant F-ratios (Table 24). Husband's education and occupation had a
significant but less meaningful variance with eco-consciousness than
wife's education.

The differences among family patterns of eco-consciousness

and the three significant contextual variables were thnther examined

80

TABLE 24.--Summary of ANOVA and MCA Tests for Differences Among
Intrafamilial Patterns of Eco-consciousneSs Value in
Respect to Contextual Variables.

 

One-Way ANOVA Testsa MCA restsb

 

 

Contextual Variable F-Statistic

 

(Probability) Eta
Wife's education 5.343 (.001)C .47
Husband's education 3.342 (.002)c .39
Husband's occupation 2.868 (.006)C .37
Husband's age 1.560 (.141) .28
Wife's employment 1.361 (.218) .26
Wife's age 1.344 (.226) .26
Family income .906 (.999) .22
Family size .275 (.999) .12

 

a8 and 148 df in Analysis of Variance Test.

bEta equivalent to multiple R statistic in Multiple
Classification Analysis Test.

CSignificant level: p < .05.

by comparing the distribution of means (Table 25). Post hoc com-
parison tests were conducted in respect to wife's education and
husband's education and occupation (Table 26). The tests were
supportive of both direction and congruency contrasts for the three
contextual variables. Higher levels of education for both husbands
and wives and occupational prestige for the husbands were associated
with high commitment to eco-consciousness (contrast: family type

1 and 9). And there was support for the contrast between highly

congruent and committed husband-wife eco-consciousness and

81

.Pmoos Pmuwumwampm mucmwcm> mo mwmapmcm up m—o>mp Loaume ppmom

 

Am.mpv Am.m_v A¢.pev “mp “cam: vccswv 4<FOP
Am._v .wnflfli Ao.mv .wdflflt_ Ao.~Fv .mnwmn .mm: cams zopoc amp: .vcmnmax .m
Am.mv ¢.p_ A~.¢V m.mp Ao.mpv m.o¢ u come zopma mew: .uczosp panama: .w
Am.FV m.- Aw.~v n.—F Ao.m V ~.em m come zopmn ucoaman .ucsoem mmmz..u
Am.~v o.~p AF.~V ~.mp Am.~pv <.oe op cums sopoa mew: .o>onm panama: .m
Ao.~v ~.NP Ao.Pv o.mp A~.npv m.mm mp cums sopmn vcwnmsg .m>oam owe: .m
A~.pv F.NF Am.~v ¢.NP Am.~ V m.mm m_ cams vcsocm mew: .vcmnmax .e
Am.PV o.~p A~.ev o.ep Am.m_v ~.me op some ucsogm ucmamas .m>onm mew: .m
Am.FV ~.mp Am.mv «.mp AF.NPV m.oe my game ucaoem mew: .m>onm ucmnmzz .N
Am.mv m.v~ Am.mv ~.m— Aw.m_v 0.0m on come m>onm mew: .vcmnmaz .—

 

A.a.mv com: A.o.mv cam: A.o.mv cum:

 

 

 

mxmopoaxp aFwEmu
cowucuzum cowumusum comumaaouo w:_m> mmmcmzoromcooioum
m.m$m3 m.ucmam== m.v:mam:: .

 

.mmmcmsomumcoUioom we mspm> xpwsmm co vmmmm comuoozum m.m$mz ucm
.cowumuzcm m.u:mnm:: .cowuoaauuo m.u:mnm=: mo mcowumw>mo cceccmum can mcmmzii.m~ m4m<H

82

TABLE 26.--P0st Hoc Comparisons of Differences Among Family Patterns
of Eco-consciousness Value in Respect to Wife's Education
and Husband's Education and Occupation.

 

b

Pooled Variance Separate Variance

 

 

Contextual Variables
and Contrastsa t-Statistic t-Statistic

(Probability) (Probability) df

 

Wife's Education
Family type 1 and 9 5.815

(. 5.671 é.000) 62.6
Family type 1 and 6/8 4.552 (.000) 3.170 .010) 10.4
Family type 1 and 5/6 4.046 (.000) 3.705 (.001) 35.1
Husband's Education
Family type 1 and 9 4.730 (.000) 4.643 (.000) 66.0
Family type 1 and 4 2.765 (.006) 3.321 (.002) 31.7
Family type 1 and 5/6 2.267 (.025) 2.676 (.010) 51.4
Husband's Occupation
Family type 1 and 9 3.890 .000; 3.821 (.000) 60.2
Family type 1 and 5/7 3.547 .001 3.610 .000) 34.7
Family type 1 and 5/6 3.109 .002) 2.809 .008) 40.2

 

aFamily types are classified in Table 25.
b148 df. p < .05-

incongruent value commitment (contrast: family type 1 and 5/6) with
the three contextual variables.

There was insufficient support to accept a systematic rela-
tionship among the intrafamilial value patterns of self-esteem and
familism with respect to contextual variables based on one-way
analysis of variance tests (Appendix C).

Lambda and uncertainty coefficients were calculated to
determine the relationship between the intrafamilial value patterns
and the six stages of the family life cycle. Knowing the stage of

the family life cycle would improve prediction of familism and

83

TABLE 27.--Summary of Lambda and Uncertainty Coefficient Tests of
Relationship between Stages of Family Life Cycle and
Intrafamilial Patterns of Values.

 

 

. Uncertainty
Fam1ly Value Lambda Coefficient
Familism .09 .10
Social Responsiveness .09 .08
Self-esteem .03 .07
Eco-consciousness .03 .06

 

social responsiveness value patterns by about 9 percent (Table 27).
The sample size did not permit significance testing to determine
probability of error; the predictive nonparametric statistics are
reported only tentatively.

In review of Ho 6, there was variance in family income and
husband and wife's education among the intrafamilial differences in
social responsiveness. In addition, husband and wife's education
and husband's occupation were related to differences in eco-

consciousness. Thus, Ho 6 is partially supported.

Summary
The results of the analyses are summarized in Table 28. A
difference was found among the nine intrafamilial value patterns of
eco-consciousness in adoption of energy conservation practices.
There was a significant difference among the nine intrafamilial value

patterns of eco-consciousness irt wife's education, husband's educa-

tion and his occupation; and among the patterns of social

84

TABLE 28.-—Summary of Findings by Hypothesis with Significance Level
and Correlation Coefficient.

 

 

Focus of Hypothesisa HOb pc 5;? Lambda Supported
Energy conservation related to:

self-esteem l .16 .27

familism 2 .75 .18

social responsivensss 3 .80 .17

eco-consciousness 4 .02 .34 X
Energy conservation related to:

husband's education 5 .50 (.22)

wife's education 5 .03 (.09)

husband's occupation 5 .ll (.18)

wife's employment 5 .28 (.21)

family income 5 .40 (.23)

family size 5 .50 (.23)

husband's age 5 .88 (.24)

wife's age 5 .81 (.24)

urban-rural residence 5 .66 (.24)

family life cycle 5 .65 .18
Eco-consciousness value related to:

wife's education 6 .OO .47 x

husband's education 6 ,OO .39 x

husband's occupation 6 .01 .37 x

family income 6 .99 .22

family size 6 .99 .12

husband's age 6 .14 .28

wife's age 6 .23 .26

wife's employment 6 .22 .26

stage of family life cycle 6 -- .03

85

TABLE 28.--Continued.

 

 

Focus of Hypothesisa HOb pc E3? Lambda Supported
Social responsiveness value to:

husband's education 6 .OO .44 X

wife's education 6 .OO .40 x

family income 6 .02 .33 x

husband's occupation 6 .08 .30

family size 6 .16 .27

husband's age 6 .30 .25

wife's age 6 .28 .25

_wife's employment 6 .99 .18

stage of family life cycle 6 -- -09
Self-esteem value related to:

husband's education 5 .05 .30

wife's education 5 .07 .30

husband's occupation 5 .25 .25

wife's employment 5 .99 .20

husband's age 5 .99 .22

wife's age 6 .99 .20

.family size 5 .13 .28

family income 6 .24 .26

stage of family life cycle 5 -- .03
Familism value related to: ,

husband's occupation 5 .99 .19

wife's employment status 5 .99 .21

family size 6 .99 .20

husband's age 5 .35 .24

wife's age 6 .38 .23

family income 5 .33 .24

wife's education 5 .99 .14

husband's education 6 .38 .23

stage of family life cycle 5 -- .09

 

o c 0
aHypothesis stated on page 63-64. bHypothes1s number 519. level

86

responsiveness iri wife's education, husband's education, and
family income.

There was no difference among the family value patterns of
self-esteem 0r familisni'hi energy conservation practices or con-
textual variables. And, there was no meaningful relationship between

adoption of practices and contextual variables.

CHAPTER V

SUMMARY, CONCLUSIONS, AND IMPLICATIONS

The study will be summarized and findings discussed in rela-
tion to related research studies and theory. Implications of the
findings will be related to research and educational programs.
Headings in the chapter include: overview of the study, discussion
of the findings, limitations of the study, and implications for

research and educational programs.

Overview of the Study

Energy conservation was viewed as desirable family managerial
behavior in an era where finite resources are assessed to be incap-
able of meeting high consumption levels on a sustained basis. The
family, a principal social unit in which values are internalized
and patterns of energy consumption are learned, was the unit of
analysis. Theinfinery focus of the research was to determine if
there were differences in the adoption of household energy conserva-
tion practices among families with varying husband-wife patterns of
congruency and commitment to values. This exploration included an
investigation of the interaction effects of family member's value
congruency, i.e., synergetic behavior of families.

The values included in the study were: self-esteem,
familism, social responsiveness, and eco-consciousness. Contextual

87

88

variables were also studied in respect to practices and intrafamilial
values. The sets of relationships among the variables with the great-
est magnitude were used in building a conceptual model of family
management using an ecosystem perspective.

Data collected in the Michigan Agricultural Experiment Sta-
tion Project 3152, "Functioning of a Family Ecosystem in a World of
Changing Energy Availability," were used to study the differences
in husband-wife value patterns with respect to the adoption of
energy conservation practices and contextual variables. A subsample
of 157 families was selected from the larger multistage probability
sample of the Lansing Standard Metropolitan Statistical Area.

Behavioral scales from the larger study were adapted to
measure the values of eco-consciousness, social responsiveness, and
familism. The value of self-esteem was measured using the Rosen-
berg (1965) scale. Nine types of value configurations were concep-
tualized based on the husband-wife degree of congruency and commit-
ment. The typology of husband-wife patterns was developed to
measure the synergetic properties of family interaction. Each of
the values was studied separately rather than combined into a com-
posite scale of values because of sample size restrictions.

A scale of 14 practices was used to measure the adoption
rate of household energy conservation. The family score for prac-
tices was basically a percentage of practices adopted based on
husband and wife responses. The contextual variables included
occupation, education, age, family income, size of family, urban-

rural residence, and stage in family life cycle.

89

Reliability was eStablished for each of the value scales
and the energy conservation scale. The coefficients, which ranged
from .74 to .89, were a measure of internal consistency of the
scales. Both parametric and nonparametric statistical models were
used to test the hypothesized relationships and differences among

variables.

Discussion of Findings
Discussion of the results of data analysis is organized
around three research questions.

Research Question 1: Is there a difference in adoption of
household energy practices among the nine intrafamilial
patterns of values--eco-consciousness, self-esteem,
familism, and social responsiveness?

(The results of the analyses indicated that the value of eco-

 

consciousness is a meaningful predictor of energy conservation
behavior. In families where the husband and wife had a high commit-
ment to eco-consciousness, they were high adopters of energy con-
servation practices; low husband-wife commitment resulted in low
adoption of conservation practices.)

While values have been postulated to be a directive element
in family management (Gross, Crandall, and Knoll, 1973; Deacon and
Firebaugh, 1975), there is little empirical support for this gener-
alization. The Kimball (1960) study offered some support in terms
of congruency of husband-wife hierarchy of 12 values. He reported
that very low managerial practice adoption was associated with

dissimilar value hierarchies. As similarity of values increased,

90

up to a point, the total adoption of practices increased. While the
present research used a monopolar value scale, i.e., ranged from

no commitment to high commitment on each valueg in general the
results empirically support a relationship between values and

1

practices.}

(The finding of no differences in adoption of conservation
practices among the husband-wife value patterns for self-esteem,
familism, and social responsiveness is difficult to discuss rela-
tive to prior researchti The content of values measured has varied
from study to study and furthermore, few studies have tested the
relationship of values to practices. The terminal/nature of the
values of self-esteem, familism, and social responsiveness versus
the instrumental properties of eco-consciousness may have contributed
to the differential findings in this study. Rokeach (1974) defined
ideal states of existence as terminal and ideal modes of behavior
as instrumental properties of values. Hence, instrumental-properties
of eco-consciousness may lead to energy conservation behavior’ hile
terminal values give more generalized preference to behavior.

The terminal-instrumental dichotomy may be a helpful dis-
tinction for classifying values significantly related to managerial
situations. For example, while familism may have terminal properties
in household energy conservation decisions, familism may have
instrumental properties in marriage, divorce, childbearing, and
other life style decision situations. Only the question of rela-

tionship between values and energy conservation behavior is reported

in this research.

91

Rescher (1969b) and Williams (1970) theorized that value
change is linked to the finiteness of resources. While the dynamics

of value change was not measured in the present study, the strength

/”
of the relationship between valuing eco-consciousness and adopting 0/

energy conservation practices may be dependent on the explicit
"scarcity of energy resources" component in the measure of eco-
consciousness. Hence, change in managerial practices may be more
closely related to instrumental values than terminal values and to
those values which are being rescaled because of competing demands
for scarce resources.

Research Question 2: Is there a relationship between adoption

of household energy conservation practices and contextual
variables?

 

Finding no significant relationship between adoption of
household energy conservation practices and contextual variables is
contrary to some research findings. It should be noted that wife's
education was weakly linked to practice adoption. It predicted
less than 1 percent of the variance in practices and thus, was not
a meaningful contextual variable.

Warren (1975) and Kilkeary and Thompson (1975) reported
family income to be positively related to adoption of energy conserva-
tion practices. Therefore,fit was expected that the higher the
income, the greater would be the adoption of household energy prac-

\

tices in this study. This was not the case.)

I

V

One problem in comparing the findings is the difference in
measure of conservation practices. While the Warren and Kilkeary-

Thompson studies included practices related to the use of the car,

92

this study excluded conservation of gasoline practices. Warren sug-
gested government intervention into private decisions regarding
gasoline availability during the winter of 1973-74 distinguised
conservation of gasoline from conservation of energy for maintenance
of the household. Furthermore, Newman and Wachtel (1974) found that
gasoline had a stronger positive relationship with income than
natural gas. Thus, studies which included conservation of gasoline in
the measure of practices for the time period under study might
postulate a stronger relationship between conservation practices and
income than studies which excluded gasoline.

Rogers and Shoemaker (1971) reported research which supported
the relationship between education, social status, and adoption of
practices. Education and husband's occupational prestige, indices
of socio-economic status, were not supported in respect to energy
conservation practice adoption. Age was not an important predictor
of practice adoption in the Rogers-Shoemaker research review. This
concurs with findings in this study.! Rate of household energy
conservation practice adoption was not found to be related to educa-

tion, occupation, or age of the married couple, family size, income,

urban-rural residence, or stage of the family life cycle.)

Research Question 3: Is there a difference among the intra-
familial value patterns of familism, social responsive-
ness, eco-consciousness, and self-esteem with respect to
contextual variables?

 

The focus on differences among the husband-wife patterns of
values in relation to contextual variables was established to explore

the values explicit in family management from an ecosystem

93

perspective (Steidl, 1969; Hook and Paolucci, 1970). Familism and
social responsiveness values were conceptualized to measure the
interdependence of human beings in their decisions involving
resources. Eco-consciousness value represented the interrelation-
ship of man and the physical environment. Self-esteem value had
been suggested as an antecedent and/or parallel value to social
integrative and adaptive values (Gross, Crandall, and Knoll,
1973;118:119).
(;The social responsiveness value was found to be related to

family income, husband's education and wife's education. Hence,
the magnitude of the family income and education levels was pre-
dictive of high commitment to the social responsiveness valueti From
the nature of the scale used to measure this value, low levels of
social responsiveness were equated with social alienation (Srole,
1956). Is it possible that money and formal education are prerequi- ”
sites for realization of a social responsiveness value commitment?
Or with higher levels of education is there opportunity to increase
one's social map and hence learn to become more socially responsive?

t/TThere was a positive relationship between education of hus-
band and wife with eco-consciousness. In addition, husband's occupa-
tional prestige was significantly related to eco-consciousness./
Hence, the higher the wife's education level, the husband's educa-
tion and his occupational level, the greater the commitment to eco-
consciousness value. It may be that the access to primary data on

the finite stock of resources is greater for families with higher

94

education and in families with husbands in professional and mana-
gerial occupations. Thus, there may be more opportunity for
families with higher quality of information to rescale their value
commitment and adopt new practices of energy conservation.

The values of self-esteem and familism were not signifi-
cantly related to contextual variables. Families with highly con-
gruent and incongruent value patterns occurred across stages of the
family life cycle, age ranges, levels of income, education, occupa-
tional prestige, and employment status of the wife. Thus, we have
no recognizable pattern of information about the variance to predict
the effects of congruency‘levels in families regarding self-esteem

and familism.

1

In summary, there was support for research questions 1 and [pm a
’ ' L4‘4

3. \It was found that husband-wife commitment to eco-consciousness

x..- "

value was the strongest predictor of household energy conservation“
behavior. There was no systematic relationship between adoption ofi
practices and contextual variables.? The values of social respon-
siveness and eco-consciousness werefirelated to selected contextual
variables.: Husband's education, wife's education, and family income
have a positive relationship with social responsiveness; husband's
education, his occupational prestige level, and wife's education

are associated with high commitment to an eco-consciousness value.
ifIh the broadest sense, a partial family management model

has been empirically supported, i.e., an interdependence of value

and practice (Figure 2). The model suggests that eco-consciousness

.mmpnmwem> Pmauxmucou ucm wm=Fo> »_wsmu .mmuPuumca compo>cmmcou xdemcm
"mawnmcopumpmm umucoaaam xp—muwgpasu mo povozti.~ mgamwu

epocmmaoz ecu mm=_m> zpmsmm

 

mopnmpco> Fmapxmucou

 

onh<a=uuo
m.oz<mm=:

 

 

zo-<u=om zomh<u=om
m.oz<mm=: m.uuH3

 

 

mzoqu
>4~z<u

 

 

95

 

mmuahu<ma onh<>mumzou

 

 

>wmmzu ogoxmmso:

 

 

 

mmmzmmmaomzcu , mmmzm>Hmzoamu¢
-ouu S<Huom .

 

 

mwspm> apwsmm

 

 

96

and energy conservation practices are organized in relation to each
other and that this relationship accounts for some of the family's
managerial behavior. In addition, there is a reported interdepend-
ence between commitment haeco-consciousness and wife's education,
husband's education, and his occupation. It should be noted that
(there was a slight improvement in the prediction of practice adop-
tion when there was knowledge of wife's education and husband's
occupation in addition to commitment to an eco-consciousness value.
This was not the case for other contextual variables. A
While the outcome variable in the model was not related to
social responsiveness, there was found to be strength in the linkages
between the value and contextual variables--family income, wife's
education and husband's education. There was no significant rela-
tionship found between contextual variables and practices.)
Although the model is incomplete, it is postulated that a
holistic model can be built with empirical support found in partial
models. This study can make a contribution to a family management

model.

Limitations of the Study
1. The subsample used in the study underrepresented families
in the latter stage of the family life cycle. Therefore, generaliza-
tions pertaining to older couples' values commitment and adoption
of practices are limited. Single parent families were excluded.
2. The subsample size did not permit significance testing

for the differences between the six stages of the family life cycle

97

and the nine types of husband-wife value patterns. Therefore,

support was not postulated between these variables.

Implications of the Study
The suggestions for further research and implications for

educational programs will be discussed in this section.

Research

Further development of the measure of conservation practices
regarding direct energy use would assist in building a more com-
prehensive model of family management. The scaling dimension of
the present measure might be converted into a degree of intensity
measure to improve the discrimination among adoption levels. For
example, is the thermostat turned down to 68° in the daytime always,
sometimes, or never? Also, subscales on conservation of gasoline
in the family automobile(s), electricity in the use of major appli-
ances, and energy for household heating and cooling could be devel-
oped to test the relationships among the socio-economic status
variables and each of the types of energy intensive activities.
Other researchers have reported differences among income groups in
the magnitude of discretionary energy consumed. The degree to which
families conserve energy may be related to energy choices other
than those measured in this study.

The exploration of changes in indirect energy consumption
patterns in relation to an eco-consciousness value commitment would

be another area for research. 00 families perceive the relationship

98

between their consumption choices and energy used for production,
transportation, and distribution of food, appliances, and other
goods? How are the values of thrift and eco-consciousness related
to adoption of indirect energy consumption patterns?

Further research into family value congruencies could include
one-parent families and husband-wife-children families as respond-
ents. The size of the sample to examine these intrafamilial value
patterns in survey research would need to be much larger than the
current study and/or the pattern typology would need to be parti-
tioned into fewer congruency-commitment categories. Congruency of
value systems and outcome behavior among different types of families
and with additional family respondents would expand the model of
family management to be more representative of the population.

The effect of congruency in values needs to be further
studied in terms of outcome behavior. Because there was limited
support in this research to link levels of congruency with educa-
tional levels, and no support that congruency levels were related
to rate of practice adoption, this area of family interaction
remains to be explored further. Would it-be helpful to know which
family members controlled the thermostat and which family members
made decisions about the temperature of the water for the laundry?
Would composite values give greater insight into consensus-dissensus
patterns in the family than monopolar value scales? The size of
the sample required for combining monopolar values into composite
value patterns as developed in this research would be sufficiently

large that alternative measures should be considered.

99

Educational Programs

 

Based on the finding that eco-consciousness value has a
positive relationship to the adoption of energy conservation prac-
tices and that eco-consciousness is linked to husband and wife's
education, programs aimed at conservation of energy resources will
be more effective if the transactional properties of eco-
consciousness are examined. The finite supply of renewable
resources and the capacity of the earth to assimilate wastes can
be discussed in relation to the demand brought about by alternative
consumption patterns. The costs and benefits of high and low com-
mitment to eco-consciousness could be estimated and technical
energy information included which would bring values to the level
of consciousness and motivate changes in behavior. Education '
is recommended to bring about increased energy conservation
behavior and minimize the need for external regulations mandating

life style changes in families.

APPENDICES

100

APPENDIX A

ACCEPTED AND REJECTED SUBSAMPLE FAMILIES

101

OPERATIONAL DEFINITIONS

Accepted Subsample: Meets the sample criteria for the study:

 

presence of a husband and wife respondent in the family, complete-
ness of data on value orientation and managerial practice variables
and independence of respondent response on value orientation varia-
bles. Included are 157 families.

Rejected Subsample: Does not meet the criteria for complete-

 

ness of data and/or independence of respondent response on value

orientation variables. Included are 38 families.

102

103

TABLE A-l.--C1assification of Accepted and Rejected Subsample
Families by Stage of Family Life Cycle.

 

 

. . Accepted Rejected

Stage of Fam1ly Life Cycle Families Families
Couple, wife under 40 years,

no children 12.7 7.9
Couple, oldest child six

years or less 22.3 5.3
Couple, oldest child seven

to 12 years 19.7 5.3
Couple, oldest child 13 to

19 years 22.3 13.2
Couple, oldest child 20 years

and over 10.2 26.3
Couple, wife over 40 years,

no children 12.7 42.1

TOTAL 100.0% 100.0%

Number 157 38

 

TABLE A-2.--Classification of Accepted and Rejected Subsample
Families by Size of Household.

 

 

Number in Household figgggggg §:%$gt:g
2 24.2 50.0
3 21.0 15.8
4 28.7 15.8
5 13.4 7.9
6 6.4 2.6
7 or more __1151 __Z;2.
TOTAL 100.0% 100.0%

Number 157 38

 

104

TABLE A-3.--Classification of Accepted and Rejected Subsample
Families by Role and Age.

 

 

 

 

Accepted Families Rejected Families
Age Categories
Husbands Wives Husbands Wives
Under 18 O O O 0
18-29 years 35.0 40.8 13.2 13.2
30-44 years 37.6 35.7 13.2 21.1
45-64 years 24.8 21.7 52.6 50.0
65 years and over __g;§_ __l;2 _21;l_ _l§;§
TOTAL 100.0% 100.0% 100.0% 100.0%
Number 157 157 38 38

 

TABLE A-4.--Classification of Accepted and Rejected Subsample
Families by Educational Attainment.

 

 

 

 

Years of School Accepted Families Rejected Families
Completed Husbands Wives Husbands Wives
0—11 years, less than

high school 17.4 16.6 44.7 34.2,
12 years, high school

completed 33.8 47.1 23.7 31.6
1-3 years college or

vocational school 21.3 18.5 10.5 15.8
4 years or more college '

professional training 27.1 17.8 18.4 15.8
No response 0 O 2.6 2.6

TOTAL 100.0% 100.0% 100.0% 100.0%

Number 157 157 38 38

 

105

TABLE A-5.--Classification of Accepted and Rejected Subsample
Families 1973 Income.

 

Gross Family Income
Categories

Accepted Families

Rejected Families

 

Less than $4,999
$5,000 to $9,999
$10,000 to $14,999
$15,000 to $24,999
$25,000 or more
No response

TOTAL

Median Income

Number

8.3
15.9
32.8
28.7
14.6

__9;Q
100.0%
$13,400

5.3
21.1
31.6
26.3
10.5

__§;§.
100.0%
$12,600

1

57

38

 

TABLE A-6.--Occupational Classification of Accepted and Rejected
Subsample Families as Defined by Husbands and Wives.

 

Occupation Category

Accepted Families

 

Rejected Families

 

 

 

Husbands Wives Husbands Wives

Professional 22.3 11.5 7.9 15.8
Managerial 7.6 4.5 21.1 5.3
Clerical-sales 12.1 33.8 10.5 26.3
Blue collar 51.6 8.9 55.3 7.9
Service worker, A

private household

worker 43 _1_1_.5 .5. _5._3

Subtotal 97.4 70.2 100.0 60.
Homemaker O 20.4 0 28.
No response __2;§_ __2;4_ O _1Q;§_

TOTAL 100.0% 100.0% 100.0% 100.0%

Number 157 157 38 38

 

106

TABLE A-7.--Employment Status of Wives in Accepted and Rejected
Subsample Families.

 

 

Employment Category Accepted Families Rejected Families
Fulltime, 35 hours minimum 36.3 34.2
Parttime, 34 hours or less 14.0 10.5
Not employed _4_9_._7_ __5_2._6

100.0% 100.0%

 

TABLE A-8.--Classification of Accepted and Rejected Subsample
Families by Geographic Location.

 

 

Geographic Location AccepéegeEEmil1es Rejegzegegim1l1es
Urban 74. 5 60. 5
Rural 25. 5 39. 5

TOTAL 100.0% 100.0%

Number 157 38

 

APPENDIX B

SCALES AND CODING RULES

107

108

 

 

 

.cwsmuw>cmm umwmwpmzc a an

zocmwuweem no» Low umumswco

ecu .umcompo .umxomgu
ucmsawzam acmpmw; o>m= .o

.Azpcos a

maze pmmmp umv zucmwummwm

:mopcwes op appcmzcmce
cmupwm ewe womccam «mango .m

 

 

 

 

 

 

.oso;
Co women woman: on new;
moo cozy new mcoou mmopu .¢

.cowpeuo> co co ucmxmmz
on» com aozm cos: “mum
tosses“ Lao» czou cczh .m

. awe mg» mcweav mean

me one em>m Cw xppmwumnmmv

mmm— Lo mmmcmmu me an moss»
incense“ mewpamu cwmpcmmz .N

.oemee ea geomoELoeo
Lao» czou con» .H.

 

 

 

 

 

 

 

 

 

mac; Lao cw mpnwmmoa comemm ummF mm meow commmm ummp cos»
xppmowmxca uoz ecu commmm mesh woos commmm mesh

.cowpmauwmcmpzuwucma Lao» use woepa mcwppmzu Lao» op apnea one» mmcwcp we“ zpco nxv xomsu ommmpav
.xmcmcm o>mecou use mppwn Pose muzume op on agave mmwpwsmw mazes» Co pmw_ ucoomm m we zopmm

.mmowpumcm coppo>cmmcou amemcm x—_Emm "mpmum

109

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

use; Lao cw opnwmmoa
appeommxcn uoz

 

 

comwmm Hmmp mm 05mm

on» comamm mesh

 

common “map can»

egos comemm mesh
.Avozcmpcouv mupuumca cowum>cmmcou amcmcm a~_5ou

.mcmgmm: moguopu Low mmpuzu
upou Lo Eco: xpco om:

.meom; Laue: no; mo Pocucou
mcsumcmaemu axon sea»

.mmmuumz Lozop mo mnpaa
gov: mucmwp meow oompnmm

.mm: cw appmapum

ac: mpnmwp ppm ewe acne
.meo;

um mcwcpopu cmm>mmc com:

.mxmu xccsm co
mmameu ecu mmuwgm cmno

.pgmw: pm moumgm
szn Lo\u:m menace mmopu

.mw: cw no:
can; Loosen mumpnmgwe omopu

.ep

.m—

.m—

.FF

”mFoum

110

.uoom o: Em H xcqu H .mmEHH p< .oH

.mmEHu pm mmmHmm: Home HHchuLmu H .m
.HHome

Low pumammc egos m>on uHaoo H cmHz H .m
.HHmmHE

ngz nowmmwumm an H .mHocz we» so .5
. .mesz

ecezoo oeaeHooa o>HoHaoa a deep H .a

.mo

yucca on on cuss m>oz no: oc H Home H .m
.mHaoma

Hmoe mm HHmz we mmcwnu on cu mHnm an H .e

.wgaHHmH m Em H .HHm cw PH< .m

.mmHuHHmac
coon mo Logan: m m>az H was» Home H .N

.mcmzpo op Hence HmpmH
Hm .cucoz Ho comcma m s_H Hosp Home H .H

mmcmemwo

mmcm<
AHmcoLum

mecoepm

H.mpcoemumpm mcmonHoH mcp Ho mcwH some co.wmm.H\H xumcu ommmHav
.mucmEmHmpm mcHonHoe mew cow; mmemmmHe co mmemm so» aHmcocum zoc mpmuHucH on so» mxHH 3o: uHaoz m3

mmgmmmwo mmcm<

Emmummthmm ”mHmum

111

 

.zum>ng m.Lm:po some Huoammc w: .HH

.AHHEoc oeococcHe
HHmcHHcm cm on quoo o: cmHz mwerosom m3 .oH

.mmcqu acme «zone accoz op camp «3 .m

.Hoz
muecmamm :30 mH; mmom m: me some AHHmama .m

.aoz HeauH>Hucw :30 Lo; Lo
mH; cH gon>mv op Losuo some mmmcaoucm m: .m

.meuo sumo pcweHHanu Lo memen cmumo m: .m

.ou cu
ups: Logao one HHmu op mace: m: Ho zoom .m

.HHHEMH gzo :H muuHHmcou Acme use mcmsk .e

.pHauHHmHu an op
msmwm oc on use: m: was: mchmHHasouu< .m

.xHHEmH pcmuwasou mcocpm m use as .N

 

 

 

 

 

 

 

 

.m:

com mmco mHn msoumn :mpmo meHnocg meuHH .H
«memomHo
spaeaeom

cowcHao
oz

.mcoHpmmza.wH mcHonHoH ecu mo some Low mmcHHmmH Lao» mpHm ammo zuHsz
xcoHn we» Hxv xumcu mmomHa .xHHEmH Lao» op mHmHmL mono mm mpcmewaepm mcHonHow esp cmzmcm mmmmHa

smHHHEeH ”mHmum

mmcm<

magmammo aHmcome

mmcm<

112

oogaomHo
aHmcoLHm

 

 

 

 

 

 

 

 

oogmomwo

:oHcHoo
oz

.Loouomop mazes» Home on o: .mH
.aHHEom uoNHcomLomHo o woo o3 .mH
.moHuH>Hpoo oonpoo mcoguo comm ucomog o3 .HH

.mo op HoopcooEH
xgo> coon maozHo mo; xHHEoH och .mH

.mocoHcH m.Locpo sumo oxHH poo oo o: .mH

.mooHo> one mocoocopm
goo :H moococomHHo moomcom moo moose .oH

.mLoguo an m: co ooocom ago
one .czo Loo woo woo mconHooo Loo .mH

.meuo comm ucoumcmcc: no: cu oz .NH

. [EU]
E

oocm<

oocm< szcome

.Aomchucouv EmHHHsom ”oHoom

113

 

 

. .mz HHH3 zzozooo
mzo Hmo empomz mzo mHHH Ho am: Loo mmzozo
ozo mHzo oo o: mooe m3 Lmzoom mzo .zHooom

ozozm cw mzm mmozoommz Homoom: onoz zoo .m

.mozmzmmwpo o msz aHHomL oHooz
pH .xmzmzm m>zmmzoo op omHzo xoozzgm>m CH .5

 

mmzmmmwo
szzoLum

onzHoo
oz

mmzm<

mmzmomHa szzozom

mmzm<

A.oszmooom zoom Lo» mzHH zoom zo.mmm
oxv zomzo mmomHo .HXHHoHLHomHm .mmm Hozoooz .mcHHommmv Amzmzm poozo monummoo mo HmHH m mp onmm

 

 

onoosomzooiLm>o .m

mzwzono com .m

mmHHooom xmzmzm moooomoo mo onocmom .o
m

N

H

 

 

 

 

 

 

 

onpmHooooiLm>o .
mommz .
mmozoommz xmemzm Ho zuwozmom .

 

 

 

 

 

HH< o< ozmoxm ozmoxm ozmoxm ozmoxu
ozmoxw oz ozmHHm mEom oomzo Hmmzu zgm>

H

iii. .HmzHH zoom co zmzmzm
mzo Axv xomzov wemHzozo zwzmzm mzo oo mooszocoo mzwonHoH mzo mo zoom mmoo ozmuxm omzz oh

mmmzmoowomzooioou "mHmom

114

 

.zo ozooo zoo mz 2oz:
zozz zHHomz ooz mmoo zomzmo o mxoo mmmzh .m

.mzooom mzo
Low zooH mmszo zoz mzo zHHz oHLoz mzu
oozH zmonHzo mzwzz op zHoH xHozoz mH HH .o

. .Lmoomz
ooz mmzoz mzwoomm mH cos mmozm>o mzo mo
ooH mzo .zom mHoomo mEom oozz Ho moHom zH .m

.HHmmoH
Ho mLoo mzoo 3oLLosoo omH ozo zoooo Low
zoos zoomzo m>HH oo moz zomzmo o ozoooZoz .N

.zoe mmozm>o

mzo mo mEmHzoco mzo zH omommzmozH zHHomz

ooz mLo zmomo amzo mmooomz mHoHoHHmo
oHHzoo on mzonzz zH mm: mHooHH m.memz... .H

 

 

 

mmzmomHa.

mmzm<
szzozom

mmzmomHo zHooLmozz mmzm< mezozom

H.mmzHHmmH Loo» oo mzHozoooo 3onz mzowommoo m mzo Ho zoom Low mooom.wmm oxv zomzo mmomHov
HmszoHHom mzo mo zoom on3 mmzmo.mmx oo zoos 2oz .moommH oHHzoo mzHoLommz moszmooom meow moo onmm

mmmzm>Hmzoommz HoHoom "mHoom

115

.ooHeLo

zH oo oonoHoHHoo mmszo mzo zm>o Homozoo zoos m>oz oo mHoomo Loo oHoonHHo mH oH.tit .z
.onooozzoo HooHoHHoo poo mon zoo m3 oLoHHm zmoozm on3.tii..o .m
.mozm>m oHLo3.111
Hozozoo zoo mHoomo mzo .mLHoHHo HoHoom ozo HooHoHHoo zH ozoo m>Hooo zo mzwzoo zm .z
.Hozozoo Lo: .ozoomzmoz: zmonmz.tii
zoo m3 mmozom Ho msHooH> mzo mzo m: Ho omos .omzzmozoo mzo mLHoHoo oHLoz mo Loo m< .o .H
.oH ooozo oo zoo.111
zoo mHooHH mzo zoos no: mH mLmzo ozo .Lmzoo zH mHoomo 3mm mzo an zoo mH oHLoz mth .z
.monmHomo osz:Lm>om zH mozmoHHzH co m>oz zoo szHoHo mmozm>o mzh.ttt .o .o

.Hmmm so» 3oz op mzwocoooo mm Lo.m zmonm Axv zomzo noon
.emzmzo ozmHz oz mH mzmzp .mozmemooom mo mzHom mszoHHom mzo Ho zoom Loo mzoe zoo; mmcmo so»
oozo Lmzmzo.mmm.A\H zomzo mmomHo .mzcmozoo moo>HLo ozo oHHzoo ooozo moszmooom msom mzo onmm

.Aomoszzoov mmmzm>Hmzoommz HoHoom "mHoom

116

RULES FOR MISSING DATA ON
VALUE SCALES

Self-Esteem and Eco Consciousness Scales: If more than one

 

item was omitted on either the husband's or wife's scale, family
was rejected from the subsample; if one item was blank on either
the husband's or wife's scale, the respondent's mean score for the
scale was coded as the response.

Social Responsiveness Scale: If any item was omitted on

 

either the husband's or wife's scale, the family was rejected from
the subsample.

Familism Scale: If more than two items were omitted on

 

either the husband's or wife's scale, the family was rejected from
the subsample; if one or two items were blank on either the husband's
or wife's scale, the respondent's mean score for the scale was coded
as the response.

Alternative Rule for Scales: If a respondent recorded two

 

responses for one scale item, did not record a response for an
adjacent item and answered a similar item(s) in another part of the
instrument: response consistency was assumed and missing data
recoded according to the researcher's judgment of the respondent's
intended response. There were ten cases included in the subsample

using this rule.

117

Coding: Value Scales

A. Self-esteem scale

4-strongly agree items: 1, 2, 4, 6, and 7
3-agree

2-disagree

l-strongly disagree

4-strongly disagree items: 3, 5, 8, 9, and 10
3-disagree

2-agree

l-strongly agree

8. Familism

5-strongly agree items: 2, 6, 7, ll, 17, and 20
4-agree

3-no opinion ‘

2-disagree

1-strongly disagree

5-strongly disagree items: 1, 3-5, 8-10, 12, 13, 15,
4-disagree l6, l8, and 19

3-no opinion

2-agree

l-strongly agree

C. Eco-consciousness

5-very great extent items: 1-6
4-great extent

3-some extent

2-slight extent

l-no extent at all

5-strongly agree items: 7 and 8
4-agree

3-no opinion

2-disagree

l-strongly disagree

0. Social responsiveness

5-strongly disagree items: 1-5
4-disagree

3-uncertain

2-agree

l-strongly agree

4-(a) items: 6 and 8
2-(b)

4-(b) item: 7

2-(a)

118

*
TABLE B-l.--Use of Rule 2 in Family Energy Conservation Practice
Scores.

 

Energy Conservation Practice Use Of RUIE 2

 

Number
1. turn down thermo . . . night 6
2. maintain 68° daytime . . . 5
3. turn thermo . . . when away
4. close doors . . . unused areas 10
5. cge furnace air-filter . . . 7
6. heating equip checked . . . 9
7. close fireplace damper . . . O
8. close drapes . . . shades . . . 4
9. open drapes . . . shades . . . l
10. wear heavier clothing . . . O
11. turn off lights not using . . . O
12. replace lower wattage light . . . 2
13. turn down temp water heater . . . 17
14. use only warm or cold cycle . . . _5,
TOTAL 66
Number of Families 1 44

 

*If one spouse indicated "not physically possible" and
the other spouse recorded "this season the same as last season,"
additional information from the AESP data bank was used to validate

the answer.

119

Coding: Contextual Variables

 

Age (year of birth) Size of Household/Family
Education Actual number used
OO-kindergarden 01" none Wife's employment

Ol-ll grades completed
12-high school graduate 0'"°t employed
l3-special schooling: barber, 1-employed
business school etc. Resid

l4-A.A., R.N. ence
16-B.S., B.A. 1-rural
18-M.A., M.A., R.N. with 8.5. Z-urban
21-Ph.D., M.O., etc.

Occupation

The 3-digit classification of
1970 Census of Population was
recoded into 2-digit prestige
scores adopted by National Opinion
Research Center (Siegel, 1975)

Income (gross family income)

OO-no response
Ol-under $2,000
02-$2,000-$2,999
03-$3,000-$3,999
04-$4,000-$4,999
05-$5,000-$5.999
06-$6,000-$6.999
07-$7,000-$7,999
08-$8,000-$8,999
09-$9,000-$9,999
lO-$l0,000-$10,999
11-$11,000-$11,999
12-$12,000-$12,999
13-$13,000-$13,999
14-$14,000-$14,999
15-$15,000-$15,999
16-$16,000-$17,999
17-$17,000-$20,999
18-$21,000=$24,999
19-$25,000-$29,999
20-$30,000-$49,999
21-$50,000 and above

Stage of family life cycle

Earliest to latest stage
coded 1-6; refer to Tablel
for classification

APPENDIX C

SUPPLEMENTARY FINDINGS

120

121

TABLE C-l.--Summary of ANOVA and MCA Tests for Differences Among
Intrafamilial Patterns of Self-Esteem Value in Respect
to Contextual Variables.

 

 

 

 

One-Way ANOVA Testsa MCA Testsb

Contextual Variable F-Statistic Et

(Probability) a
Husband's education 1.890 (.066) .30
Wife's education 1.846 (.072) .30
Family size 1.582 (.134) .28
Family income 1.321 (.236) .26
Husband's occupation 1.267 (.264) .25
Husband's age .961 (.999) .22
Wife's employment .768 (.999) .20
Wife's age .736 (.999) .20

 

a8 and 148 df. bEta equivalent to multiple R statistic.

TABLE C-2.--Summary of ANOVA and MCA Tests for Differences Among
Intrafamilial Patterns of Familism Value in Respect to
Contextual Variables.

 

 

 

 

 

One-Way ANOVA Testsa MCA Testsb
Contextual Variable F-Statistic Eta
(Probability)
Family income 1.162 (.326) .24
Husband's age 1.117 (.355) .24
Wife's age 1.081 (.379) .23
Husband's education 1.073 (.385) .23
Wife's employment .853 (.999) .21
Family size .780 (.999) .20
Husband's occupation .670 (.999) .19
Wife's education .374 (.999) .14
b

a8 and 148 df. Eta equivalent to multiple R statistic.

BIBLIOGRAPHY

122

BIBLIOGRAPHY

Andrews, F. M.; Morgan, James N.; Sonquist, J. A.; and Klem, L.
Multiple Classification Analysis. Ann Arbor: Institute
for Social Research, UniVersity of Michigan, 1973.

 

Babbie, E. R. Survey Research Methods. Belmont, California:
Wadsworth Publishing, 1973.

Bates, Frederick L., and Harvey, Clyde C. The Structure of Social
Systems. New York: Gardner Press, Inc., 1975.

Boneau, C. Alan. "The Effects of Violations of Assumptions Under-
‘ lying the t Test." In Contemporary Problems in Statistics,
pp. 357-370. Edited by Bernhardt Lieberman. New York:
Oxford University Press, 1971.

Boulding, Kenneth E. "The Emerging Superculture." In Values and
the Future, pp. 336-350. Edited by Kurt Baier and Nicholas
Rescher. New York: Free Press, 1969.

 

Bredemeier, Harry G. "On the Complementarity of Partisan and
Objective Research." American Behavioral Scientist l7
(September-October 1973): 5-25.

Brim, Orville 6., Jr.; Glass, David C.; Lanvin, David E., and
Goodman, Norman. Personality and Decision Processes:
Studies in the Social Psychology of Thinking, Stanford,
California: Stanford University Press, 196 .

Cottrell, Fred. Energy and Society. Westport, Conn.: Greenwood
Press, 1955.

Cutler, Virginia F. Personal and Family Values ip the Choice of
a Home. Cornell University Agricultural Experiment Station
Bulletin 840. Ithaca, New York, November, 1947.

 

Deacon, Ruth, and Firebaugh, Francille M. Home Management: Context
and Concepts. Boston: Houghton-Mifflin, Inc., 1975.

 

Dyer, Doris M. "Students' Wives Values as Reflected in Personal
and Family Activities." Master's thesis, Michigan State
University, 1962.

123

124

Eichenberger, Mary Ann. "A Comparison of Ownership of Selected
Household Appliances and Residential Energy Use by Employed
and Nonemployed Homemakers in the Lansing, Michigan, Area."
Master's thesis, Michigan State University, 1975.

Eigsti, Marilyn,li."Interrelationships of Value Orientation,
Decision-Making Mode and Decision-Implementing Style of
Selected Low Socio-Economic Status Negro Homemakers." Ph.D.
dissertation, Michigan State University, 1973.

Energy Policy Project of the Ford Foundation. A Time to Choose:
America's Energy Future. Cambridge, Mass.: Ballinger
Publishing Co., 1974.

 

Engebretson, Carol L. "Analysis of a Constructed Typology of Wives'
Values Evident in Managerial Decision Situations.” Ph.D.
dissertation, Michigan State University, 1965.

Field, Anne E. "A Study of Water Consumption Practices in House—
holds." Ph.D. dissertation, Michigan State University,
1973.

Fortenberry, Frances E. "Measurement of Values Relating to Kitchen
Design." Master's thesis, Kansas State Universtiy, 1963.

Fritsch, Albert J., and Castlemen, Barry I. Lifestyle Index.
Washington, O.C.: Center for Science in the PubTic Interest
(1779 Church Street, N.W.), 1974.

Gladhart, Peter M. "The Energy Problem Continues: Impact and
Implications for Urban and Industrial Centers." Remarks
prepared for conference quoted by Michigan Farm Economics,
390 (July 1975): 3.

Glass, Gene V., and Stanley, Julian C. Statistical Methods in
Education and Psychology. Englewood Cliffs, N. J.:
Prentice Hall, Inc., 1970. -

 

Granbois, Donald, and Willett, Ronald. "Equivalence of Family
Role Measures Based on Husband and Wife Data." Journal of
Marriage and Family 32 (February 1970): 68-72.

Gross, Irma. Proceedings of Conference on Values and Decision-making
in Home Management. Michigan State University, East Lansing,
Michigan, 1955.

 

Gross, Irma; Crandall, Elizabeth; and Knoll, Marjorie. Management
for Modern Families. New York: Appleton-Century Crofts,
1973.

 

125

Grot, Richard A., and Socolow, Robert H. "Energy Utilization in a
Residential Community." In Energy. PP. 483-498. Edited
by Michael S. Macrakis. Cambridge, Mass.: MIT Press,

974.

Handy, Rollo. The Measurement of Values. St. Louis, Mo.: Warren
H. Green, Inc., 1970.

 

Hannon, Bruce. "Energy Conservation and the Consumer." Science 189
(July 11, 1975): 95-102.

Heberlein, Thomas A. "Conservation Information, The Energy Crisis,
and Electricity Consumption in An Apartment Complex."
Department of Rural Sociology. UniverSity of Wisconsin, 1974.
(Mimeographed).

Hook, Nancy, and Paolucci, B. "Family as Ecosystem." Journal of
Home Economics 62 (May 1970): 315-318.

 

Imig, David R. "Masculinity-Femininity as Related to Family-Marital
Adjustment." Ph.D. dissertation, Michigan State University,
1971.

Jacob, Phillip E., and Flink, James J. with collaboration of Hedvah
L. Shuchman. "Values and Their Function in Decision-Making."
Supplement to American Behavioral Scientist 9 (May 1962):
7-34.

Kilkeary, Rovena, and Thompson, Patricia J. "The Energy Crisis and
Decision-Making in the Family." New York: Herbert K. Lehman
College, City University of New York, 1975. (Mimeographed.)

Kimball, William J. "The Relationship Between Personal Values and
the Adoption of Recommended Farm and Home Practices." Ph.D.
dissertation, University of Chicago, 1960.

Kluckhohn, Clyde, et al. "Values and Value-Orientations in the
Theory of Action." In Toward a General Theory of Action.
Edited by Talcott Parsons and Edward'A. Shils. Cambridge:
Harvard University Press, 1954.

Kluckhohn, Florence, and Stodtbeck, Fred L. Variations in Value
Orientations. Evanston, Illinois: Row, Peterson and Co.,
1961.

 

Koenig, Herman E., and Edens, Thomas C. "Energy, Ecology and Econ-
omics: The Realities of Thermo-dynamically Based Economies."
Departments of Electrical Engineering and Systems Science
and Agricultural Economics, Michigan State University,
September 22, 1975. (Mimeographed.)

126

Lindquist, E. F. Design and Analysis of Experiments in Psychology
and Education. Boston: Houghton Mifflin, 1953.

 

"The Norton Study of the Effects of Non-normality

and Heterogeneity of Variance." In Contemporary Problems
in Statistics, pp. 348-356. Edited by Bernhardt Leiberman.
New York: Oxford University Press, 1971.

 

Martin, Esther A. "Analysis by a Constructed Typology of Family
Members' Values Evident in Managerial Decision Situations."
Ph.D. dissertation, Michigan State University, 1965.

Meeks, Carol 8., and Deacon, Ruth E. "Values and Planning in the
Selection of a Family Living Environment." Journal of
Home Economics 64 (January 1972): 11-16.

 

Mehrens, William A., and Lehmann, Irvin J. Measurement and Evalua-
tion in Education and Psychology. New York: Holt, Rinehart,
and Winston, 1973.

 

Morrison, Bonnie M. "Socio-Physical Factors Affecting Energy Con-
sumption in Single Family Dwellings: An Empirical Test of
a Human Ecosystems Model." Ph.D. dissertation, Michigan
State University, 1975.

National Institute of Child Development and Human Development. Ipe_
Acquisition and Development of Values. Report of a confer-
ence May 15-17, 1968. Washington, O.C.: U. S. Department
of Health, Education, and Welfare, 1969.

Newman, Dorothy, and Wachtel, Dawn 0. "The American Energy Consumer:
Rich, Poor, and In-Between." In A Time to Choose. Pp. 113-
130. Cambridge, Mass.: Ballinger Publishihg Co., 1974.

Nie, Norman H., et a1. Statistical Package for the Social Sciences.
New York: McGraw-Hill Book Company, 1975.

Nye, Ivan F. "Values, Family, and Changing Societies." Journal of
Marriage and Family 27 (August 1967): 241-248.

 

Odum, Howard T. Environment, Power and Society. New York: John
Wiley and Sons, Inc., 1971.

Paolucci, Beatrice. "Contributions of a Framework of Home Manage-
ment to the Teaching of Family Relationships." Journal of
Marriege and Family 28 (August 1966): 338-342.

 

127

Paolucci, Beatrice, and Hogan, M. Janice. "The Ener y Crisis and
the Family." Journal of Home Economics 65 (December 1973):
12-15.

 

Rescher, Nicholas. Introduction to Value Theory. Englewood
Cliffs, N. J.: Prentice-Hall, Inc., 1969a.

. "What is a Value Change? A Framework for Research.“ In
Values and the Future. Edited by Kurt Baier and N. Rescher.
New York: Free Press, 1969b.

 

Robinson, J., and Shaver, P. Measures of Social Psychological
Attitudes. Ann Arbor: University of Michigan, Institute
for Social Research, 1973.

Rogers, Everett M., and Showmaker, F. Floyd. Communication of
Innovations: A Cross-Cultural Approach. New York: Free
Press, 1971.

Rokeach, Milton. "Change and Stability in American Value Systems,
1968-71." Public Opinion Quarterly 38 (Summer 1974):
222-238.

Rosenberg, M. Society and the Adolescent Self-Image. Princeton,
N. J.: Princeton University Press, 1965.

Rotter, J. B. "Generalized Expectancies for Internal Versus
External Control of Reinforcement." Psychological Mono-
graphs 80 (1966): 1-28.

Rozeboom, William ML, ''The Fallacy of the Null-Hypothesis Signifi-
cance Test." In Contemporary Problems in Statistics, pp.
116-126. Edited by Bernhardt Lieberman. New York: Oxford
University Press, 1971.

Safilios, R. C. "Family Sociology or Wives' Family Sociology? A
Cross-Cultural Examination of Decision-Making." Journal
of Marriage and Family 31 (May 1969): 290-301.

Schlater, Jean 0. "The Management Process and Its Core Concepts."
Journal of Home Economics 59 (February 1967): 93-98.

Siegel, Paul M. "Prestige and the American Occupational Structure."
Ph.D. dissertation, University of Michigan, 1975.

Srole, L. "Social Integration and Certain Corollaries." American
Sociological Review 21 (1956): 709-716.

128

Steidl, Rose E. "An Ecological Approach to the Study of Family
Managerial Behavior." Proceedings of Conference on The
Family: Focus on Management. Pennsylvania State Univer-
sity, June 30, 1969.

 

Strodtbeck, Fred L. "Family Interaction, Values, and Achievement."
In Talent and Society. Edited by David C. McClelland,
et al. New York: 0. Van Nostrand Co., 1958.

 

Turk, James L., and Bell, N. W. "Measuring Power in Families."
Journal of Marriage and Family_34 (May 1972): 215-223.

van der Veen, Ferdinand; Huebner, 8., Jorgens, 8., and Neja, P., Jr.
"Relationships Between the Parents' Concept of the Family
and Family Adjustment." American Journal of Orthopsychia-
try 34 (1964): 45-55.

van der Veen, Ferdinand, and Novak, Arthur L. "The Family Concept
and the Disturbed Child: A Replication Study." American
Journal of Orthopsychiatry 44 (October 1974): 763-772.

Warren, Donald I. Individual and Community_Effects on Responses to
the Energy Crisis of Winter 1974: An Analysis of Survey
Findings from Eight Detroit Area Communities. Study con-
ducted by the Program in Community Effectiveness, Institute
of Labor and Industrial Relations, University of Michigan,
December 1974.

Williams, Robin M., Jr. "Values and the Family: Consensus, Con-
flict, and Change." Paper presented at Clara Brown Army
Symposium on Family Values, University of Minnesota,
March 15-17, 1970. (Mimeographed.)

Winett, Richard A., and Nietzel, Michael T. "Behavioral Ecology:
Contingency Management of Consumer Energy Use." American
Journal of Community Psyphology. 1975. In Press.

Zuiches, James J. "Coercion and Policy Acceptance: The Case of
Energy Policies." Unpublished Manuscript from Agricultural ‘
Experiment Station Project 3152: Functioning of the Family
Ecosystem in a World of Changing Energy Availability. East
Lansing, Mi.: Institute for Family and Child Study, Michigan
State UniverSity, 1975.

Zuiches, James J.; Morrison, Bonnie M.; and Gladhart, Peter M.
"Interviewing Families: Methodology and Evaluation of
'Energy and the Family' Survey." Occasional Paper No. 2,
Agricultural Experiment Section, Project 3152: Functioning
of the Family Ecosystem in a World of Changing Energy
Availability. East Lansing, Mi.: Institute for Family
and Child Study, Michigan State University, 1975.

 

"111 11111411141111.4111“