An investigation of issues surrounding the disclosure of energy use data on major home appliances


Material Information

An investigation of issues surrounding the disclosure of energy use data on major home appliances
Physical Description:
xiv, 280 leaves : ill. ; 28 cm.
McNeill, Dennis Lee, 1947-
Publication Date:


Subjects / Keywords:
Household appliances, Electric -- Labeling   ( lcsh )
Household appliances, Electric -- Energy consumption   ( lcsh )
Consumer education   ( lcsh )
Marketing thesis Ph. D   ( lcsh )
Dissertations, Academic -- Marketing -- UF   ( lcsh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis--University of Florida.
Bibliography: leaves 182-186.
General Note:
General Note:
Statement of Responsibility:
by Dennis L. McNeill.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 025843027
oclc - 03393022
System ID:

Table of Contents
    Title Page
        Page i
        Page ii
        Page iii
    Table of Contents
        Page iv
        Page v
        Page vi
        Page vii
    List of Tables
        Page viii
        Page ix
        Page x
    List of Figures
        Page xi
        Page xii
        Page xiii
        Page xiv
    Chapter 1. Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
    Chapter 2. Background to the problem
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
    Chapter 3. Hypotheses and manipulations
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
    Chapter 4. Research methodology
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
    Chapter 5. Results
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
    Chapter 6. Conclusions
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
    Appendix A. Questionnaire
        Page 187
        Page 188
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
        Page 200
        Page 201
        Page 202
        Page 203
        Page 204
        Page 205
        Page 206
        Page 207
        Page 208
        Page 209
        Page 210
        Page 211
        Page 212
        Page 213
        Page 214
    Appendix B. Model displays for the control condition
        Page 215
        Page 216
        Page 217
        Page 218
        Page 219
    Appendix C. Model displays for the annual dollars plus a comparative range condition
        Page 220
        Page 221
        Page 222
        Page 223
        Page 224
    Appendix D. Model displays for the annual dollars plus a comparative range plus annual dollars by feature condition
        Page 225
        Page 226
        Page 227
        Page 228
        Page 229
    Appendix E. Model displays for the KWH per year plus a comparative range condition
        Page 230
        Page 231
        Page 232
        Page 233
        Page 234
    Appendix F. Model displays for the monthly dollars plus a comparative range condition
        Page 235
        Page 236
        Page 237
        Page 238
        Page 239
    Appendix G. Model displays for the annual dollars condition
        Page 240
        Page 241
        Page 242
        Page 243
        Page 244
    Appendix H. Human release form
        Page 245
        Page 246
    Appendix I. Data
        Page 247
        Page 248
        Page 249
        Page 250
        Page 251
        Page 252
        Page 253
        Page 254
        Page 255
        Page 256
        Page 257
        Page 258
        Page 259
        Page 260
        Page 261
        Page 262
        Page 263
        Page 264
        Page 265
        Page 266
        Page 267
    Appendix J. Codebook
        Page 268
        Page 269
        Page 270
        Page 271
        Page 272
        Page 273
        Page 274
        Page 275
        Page 276
    Appendix K. Computer runs by hypothesis
        Page 277
        Page 278
        Page 279
    Biographical sketch
        Page 280
        Page 281
Full Text



Dennis L. McNeill





It is important to acknowledge the contributions that many have

made to this dissertation and to my career in general.

The funds provided by the National Science Foundation through a

grant to William L. Wilkie allowed this research to attain a greater

degree of sophistication than would have been possible otherwise. Also

contributions were made by the Center for Public Policy Research at the

University of Florida.

More important is the assistance which was given by more than a

few individuals along the way. I would like to thank Dr. Barry Schlenker

for his help and for enduring the rather haphazard way I communicated

with him. Dr. Olli Ahtola made major contributions to this work. Olli

has consistently provided support to me throughout my time at Florida

and I am indebted to him.

There is an individual who deserves credit for this effort who was

not even directly involved. Dr. B. Curtis Hamm was the first person who

showed faith in me academically in my early years at Oklahoma State Uni-

versity. I should have thanked him much earlier than this.

The most important person to this effort and to the direction of

my career is a person I respect totally, Dr. William L. Wilkie. Bill

has made sense out of my undisciplined pattern of work. I consider being

a Wilkie student an honor I don't deserve, and I consider our friendship

an important possession.

I would like to mention my family in this. My mom and dad, and

brother have always been my biggest fans. I wish my dad were here to

be proud of me. Knowing Jarret was coming gave me added motivation

to complete this work. He is such a happy person and we can now get

to know each other better. Finally I'd like to say something to my

wife, Jennifer, but words are not enough. I don't think you know how

important you are to me and how you have helped me. Thanks for being

here--I love you.

To all these people and many more I'm grateful for your help.

I'll never write another dissertation and I'm sure we are all a little



Chapter Page

LIST OF TABLES ..... ..................

LIST OF FIGURES . . . . . . . .

ABSTRACT ...... .. ....................

INTRODUCTION ...... .......................

The Problem ..... ... .................. 1
The Energy Policy and Conservation Act (EPCA) 2
Research Issues ...... .. ................ 3
Unit of Measurement ...... ............. 3
Time Period for Energy Use Computation 4
Comparative Disclosure ..... ........... 5
Degree of Disclosure ...... ............ 5
Impact of the Disclosure .... .......... 6
The Central Research Questions ... ....... 6
The Method ...... .. .................... 6
The Impact ..... ... ................. 7
The Task ...... .. .................. 8
Organization of the Report ..... ........... 9

II BACKGROUND TO THE PROBLEM ..... ........... 10

Introduction ............................. 10
Major Energy Programs and Research ... ....... 11
The M.I.T. Report (1972) ... .......... I.11
The Voluntary Labeling Program ... ....... 15
Energy Consumption vs. Comparative Energy
Consumption ...... .. ................ 17
Products Covered Under the Program ..... .. 18
Summary of Program Activities .... ........ 19
Response to the Program ..... ........... 21
Energy Policy and Conservation Act (Public Law
94-163) ...... ................... .... 21
Administration of the EPCA .. ......... ... 22
Covered Products .... .............. .... 22
Labeling Requirement .... ............ ... 23
General Comments on Programs ........ ... 26
Past Public Policy Research .. .......... ... 26
Public Policy Decisions ............. .... 27
Experimental Stimuli ...... ............ 31
Sample Considerations ...... ............ 33
Task Orientation .... .............. .... 35
Focus Groups ...... .................. ... 36





Experimental Manipulations ... ...........
Independent Variables .............
Unit of Measurement ... .............
Time Period for Computation ............
Comparative Information ... ...........
The Degree of Disclosure ..........
Treatment Conditions .... ..............
Control No Energy Data ..........
Energy Consumption in Annual Dollars Plus
a Comparative Range ... ............
Energy Consumption in Annual Dollars Plus
a Comparative Range Plus By Feature ...
Energy Consumption in Annual KWH Plus a
Comparative Range .................
Energy Consumption in Monthly Dollars Plus
a Comparative Range .............
Energy Consumption in Yearly Dollars .
Experimental Vehicle .... ..............
Product Class ..... ................
Attribute Choice . . . . . . .

IV RESEARCH METHODOLOGY .... ..............

Introduction ..........
Objective One ........
Objective Two ........
Objective Three .....
Objective Four .....
Experimental Design ...
Experimental Controls .
Conduct of the Experiment
Introduction ......


Initial Model Evaluation (Acquisition)
Model Evaluation ...........
Building Task ... ..............
Comments on the Experimental Procedure .
Sample ...... ..................
Sample Characteristics ........
Dependent Measures ... ............
Cognition/Recall ...........
Analysis of Energy Expectation/Recall .
Evaluation . . . . . . .



Chapter Page

IV Recall Evaluation ... .............. .100
Analysis of Recall Evaluation ... ........ 101
Non Recall Evaluation ..... ............ 102
Analysis of non recall evaluation . .. 106
Behavior ........ .................... 107
Analysis of building task .......... ...109

V RESULTS ..... ................... ..I. 111

Introduction .... .................. il
Analysis Procedures .... .............. i1
Cross Tabulation .... .............. i1
Test of a Proportion ... ............ ... 112
One Way Analysis of Variance .......... ...113
Significance Levels ..... ............. 115
Model preference ..... ............. 116
Overall impressions ............ ...119
Energy use expectations .... ......... 121
General comments ..... ............. 125
Size choice ... ............... ....133
Style choice ... ............... ... 136
Insulation choice ... ............ ... 137
Defrost choice ... .............. ...138
Icemaker choice ..... ............. 139
Power miser choice ..... ............ 140
Total Energy Consumption and Total Purchase
Price ....... ................... 141
Discussion of the Building Task Results . 143
Likelihood of Change Measures .......... 144
Relative performance of alternative format
for disclosure versus the control
condition ... ............... ....147
Relative performance of the disclosure
conditions versus the EPCA mandate . 153
Respondents' perceptions of feature/energy
use relationships ... ........... ...155
Evaluation of the Disclosure Usefulness and
Complexity ...... ................ 157
Respondent Choice for Disclosure Format . 158
Discussion ....... ................. 159
Discussion of the Findings .... ......... 165
Limitations of the Study .... .......... 170



Introduction ......
The Findings ......
The Process .......
Implications ......
Future Research .....
Conclusion .......















CONDITION .... ..............

CONDITION .... ..............

CONDITION .... ..............


CONDITION .... ..............


DATA . . . . . . .

CODEBOOK . . . . . .


SKETCH . . . . . . .




















Table Page

1 Summary of the Composition of Life Cycle Cost
for Refrigerator-Freezers and Color Televisions 13

2 Summary of Major Sources of Residential Energy Use. 15
3 Summary of the Disclosure Decisions by the
Voluntary Labeling Program ..... ............. 19

4 Demographic Profiles of Group Interview Participants 37

5 Summary of Characteristics of Refrigerator-Freezer
by Order of Discussion .... .............. ....39

6 Summary of the Comments of Groups 1 and 2 Made
About Characteristics of Refrigerators ... ...... 40

7 Attribute Values of Experimental Models ... ...... 75

8 Energy Consumption by Model by Treatment ..... 77

9 Sample Composition by Experimental Groups ....... 95

10 Demographics of the Sample .... .. ............ 96

11 Validation of the Comparative Information Scales 106

12 First Choice of the Treatment Conditions by Model 117

13 Preferences for MODEL 2 and MODEL 3 .... ........ 118

14 Overall Impressions by Model ..... ........... 119

15 One Way ANOVA on Overall Impressions of MODEL 2 121

16 Energy Use Expectations by Model .. ......... ... 124

17 Impact of the Range of Comparative Values for
Responses to MODEL 2 .... .. ............... 127


LIST OF TABLES (continued)

Table Page

18 One Way Analysis of Variance on the Impact of the
Range of Comparative Values for Judgments on
MODEL 2 ....... .. ...................... 128
19 Impacts of the Range of Comparative Values for
Models 1, 3, 4 ..... .................. ... 128

20 Evaluation of Model Performance on Energy Consumption 130

21 Dunn's Test for Significant Differences in the
Model Energy Use Evaluations ..... ........... 131

22 Summary of Feature Choice by Treatment Conditions 134

23 Analysis of Size Choice ...... .............. 135

24 Presentation of Style Choices ..... ........... 136

25 Presentation of Insulation Choices .... ........ 137

26 Presentation of Defrost Choices .... .......... 138

27 Presentation of Icemaker Choices .. ......... ... 139

28 Presentation of Power Miser Choices .......... ... 140

29 Analysis of Variance for Total Energy Use Between
Treatment Conditions .... .. ............... 142

30 Analysis of Variance for Total Purchase Price
Between Treatment Conditions ..... ........... 143

31 Likelihood of Change by Feature Choice ....... ... 145

32 Performance of the Disclosure Alternatives versus
the Control Condition of MODEL 2 Preferences and
Expectations of Energy Use ..... ............ 148

33 Performance of the Disclosure Alternatives versus
the Control Condition on Overall Impressions of
MODEL 2 ....... .. ...................... 149

LIST OF TABLES (continued)

Table Page

34 Building Task: Performance of the Disclosure
Alternatives versus the Control Condition ....... 150

35 Building Task: Performance of the Disclosure
Alternatives versus the Control Condition on
Energy Use and Purchase Price ..... ........... 152

36 Performance of the Disclosure Alternative versus
the EPCA Requirement on Judgments for MODEL 2 . 154
37 Building Task: Relative Performance of Disclosure
Alternatives versus the EPCA Requirement .. ..... 154

38 Feature/Energy Use Judgments by Condition ....... 157

39 Evaluations of the Usefulness and Complexity of
the Disclosure Alternatives ... ............ ... 158

40 Preferences for Disclosure Format ........... ....159

41 The Relationship of Sample Characteristics to the
Dependent Measures .... ... ................ 163


Figure Page

1 Model Displays for MODEL 1, 2, 4 ....... ... 80

2 Model Display for MODEL 3 .. .......... ... 81

3 Conduct of the Experiment .. .......... ... 87

Abstract of Thesis Presented to the Graduate Council of
the University of Florida in Partial Fulfillment of the Require-
ments for the Degree of Doctor of Philosophy




Dennis L. McNeill

May, 1977

Chairman: William L. Wilkie
Major Department: Marketing

The purpose of this dissertation is to examine the consumer

information processing issues within the framework of formal pro-

grams designed to disclose standardized energy consumption data on

labels attached to major home appliances. The research incorporated

both the activities and the intent of the program to provide energy

use data as the focus of the research decisions.

Essentially, the problem of interest is the impact of the

information on behaviors characteristic of consumers' evaluation of

major home appliances. The investigation was a controlled experimen-

tal setting within which subjects encountered information on refrig-

erator-freezers. The product class of refrigerator-freezers is an

interesting one from the program standpoint. Refrigerators account

for a large portion of household energy use due primarily to the fact

that the product is continually in use. However, within the product

class the energy consumption varies in response to the large number

of model features which are available.

In order that the results be significant inputs into the policy

decision the sample was actual consumers, housewives, who were paid

for their participation. The sample was drawn with the intent of

providing a reasonable cross section of the more general population

in terms of income, education, etc. The subjects were recruited

from local women's organizations in Gainesville, Florida.

The subjects encountered realistic product information on

four models of refrigerator-freezer along the attributes of size,

style, purchase price, frost free, icemaker, insulation, and energy

use. The information on energy use was presented in such a way

that certain research questions could be investigated. The first

question is whether the energy use information will assist the con-

sumer in behaviors related to major home appliances? The second

question of interest is whether the format which is currently

required by the Energy Policy and Conservation Act is the best way to

disclose energy use data? These research questions required six

treatment conditions. Five of the conditions were disclosure con-

ditions which varied the format of the energy use data on the follow-

ing dimensions; the unit of measurement by which to disclose energy

consumption, the presence of comparative energy consumption, the

time period used to compute energy consumption, and the degree of

disclosure. These five disclosure conditions represent realistic

alternatives for the format of the energy use labels. The sixth

condition is a comparison condition which represented the present

information environment without energy consumption data available.


These energy consumption treatments represented the only difference

across the conditions, and subjects were randomly assigned to one

of these conditions.

The experimental session was characterized by a complete

profile of information impact through multiple dependent responses

over a series of realistic product evaluation tasks and a product

building decision.

The findings of the research support the potential of energy

use data to impact on consumer purchase related behavior in a manner

consistent with the overriding goal of the energy labeling program,

the reduction of demand for energy.



An issue of continuing concern in the public policy arena is

the ability of the consumer to make informed purchase decisions.

Often this ability is linked to the quality of the information which

is available and used by the consumer. The assumed relationship

between consumer information use and consumer decisions has been

manifested in programs which have specified the provision of stand-

ardized product performance information. While the form and content

of the different programs has varied (e.g., nutritional labeling,

cigarette health warnings, gasoline octane ratings, etc.) the orienta-

tion to the information provision has been similar. Each of these

programs have addressed a broad problem which is possibly remedied by

the availability of information that had previously been unavailable,

or available in less than optimal form. Ultimately, by providing the

product information, the consumer decisions should be improved.

The Problem

This dissertation has come about through a concern for a broad

societal problem, and the consumer information processing issues which

characterize a program directed toward the problem. The problem is the

broad area of energy use and the attendant limitations on fuel sources.

The phrase "energy crisis" has become a household word in the

United States in recent years. The energy crisis calls to mind a

whole spectrum of issues ranging from the verity of the problem itself


and the immediacy of the response to the myriad of formal and informal

programs which stem from the problem. The consumers of energy have

been requested to slow down, urged to buy smaller energy saving cars,

encouraged to closely monitor housing temperatures and in general to

become aware of energy consumption. In many of these cases, the

behavior change has come about through a reaction to emergency energy

situations. Severe winters, oil embargoes, and the lack of necessary

contingency plans required a severe reaction in energy consumption.

This reaction is particularly strong in the U.S. due to its immense

demands for energy. With only six percent of the world's production,

the U.S. consumes one-third of the energy and mineral resources pro-

duced worldwide each year ("What's Behind the Energy Crisis,"1975).

The problem of an energy shortage is essentially one of energy

limitations. The limitations of existing fuel sources will dictate

an eventual shift to other, as yet untapped, sources of energy (e.g.,

solar energy, wind energy, nuclear energy, etc.) However, all of the

alternative power sources will not make substantial contributions to

the fuel needs for some time to come. Thus, the emergency reactive

measures will not overcome the energy shortage problem. What is needed

is a more permanent and fundamental change in energy use patterns.

The Energy Policy and Conservation Act (EPCA)

The EPCA is the most recent in a stream of energy programs. The

act, passed in December, 1975, is a very comprehensive energy program.

The various provisions of the act stipulate the powers of the President,

provide for the establishment of energy reserves, urges the further

development of existing sources of energy while attempting to reduce

the general demand for energy. Among the program activities directed

to the general goal of reducing the demand for energy is a mandated

labeling requirement for major home appliances.

The labels are to contain energy consumption in annual dollars

and a range of annual energy use for similar products. The labeling

requirement is the responsibility of the Federal Trade Commnission.

However, the overall responsibility of the act rests with the Federal

Energy Administration.

Research Issues

This dissertation has investigated two fundamental questions

about the impact of the energy information on appliance labels. The

crucial focus for this research is not whether the program will be a

success (i.e., reduce the energy demand), but to address certain issues

surrounding the character of the information. In essence the issues

relate to the ability of the information to assist consumer decisions.

The concern of this investigation was to develop a clear profile

of the impact of the disclosure issues. The disclosure of energy con-

sumption data can be broken down into different dimensions which could

relate to impact.

Unit of Measurement

The required disclosure is to be in annual dollars. The unit of

measurement question (in the case of the EPCA, the unit of measurement

is dollars) could have an impact on the disclosures. Generally, the

unit of measurement should uniformly communicate consumption. However,

this issue must be viewed in light of the program goals, reduction in

the demand for energy. If dollars are to be used as the unit of

measurement, the consumption will be computed by a standard rate per

kilowatt hour (KWH) of energy use. This will provide uniform energy

consumption on each product across the country. Thus, while the

relative energy consumption (on a product-by-product basis) would be

constant, the consumer would not be able to ascertain the exact amount

of energy use he would encounter with product purchase. It may be

that the disclosure of the actual energy used (e.g., KWH) would be

more in line with the goal of reducing the demand for energy. The

question is whether there exist some incremental benefit (in terms of

the consumer's use of the disclosure) of the dollar unit of measurement.

Time Period for Energy Use Computation

A second aspect of the disclosure is the time period chosen for

computation of energy consumption. The mandate is for the energy use

in annual terms. The time period for energy use computation is directly

related to the magnitude of the disclosure. This is often an overlooked

issue in program development but an issue which is put forth as an

after-the-fact rationale of a program evaluation (e.g., with the unit

pricing program). The issue is whether the larger time periods for

reporting, monthly vs. yearly vs. total product life, provide benefits

in impact by magnifying the relative differences between products and

providing disclosures of greater magnitude. This is an interesting

question in the area of energy use due to the fact the consumer typic-

ally encounters the energy use in monthly terms.

Comparative Disclosure

An important issue for the energy use disclosures is the impact

of a new dimension of product information, relative energy consumption.

This is the range of annual energy use for similar products. This

aspect of the disclosure format should provide the basis for compara-

tive performance judgments with regard to the energy consumption of

a particular product. This judgment should be facilitated without the

normal shopping behavior typical of the process to compare products.

A set of scales specifically designed for this study will be used to

investigate this important dimension of the disclosure format.

Degree of Disclosure

A crucial aspect of the disclosure of energy use is the ability

of the consumer to identify those attributes of the product which

account for energy use. At present, the disclosure would require a

very careful trade-off procedure by product for the consumer to dissect

the overall product's energy use into the sources (i.e., size, features,

etc.). Given the complexity of the major home appliances, this would

present a formidable task.

Of interest is the benefit of disclosing the sources of product

energy use explicitly. This would not require the mental effort and

would be expected to provide a more consistent reduction in energy

demand. This issue is actually a step beyond the required disclosure

but is consistent with the intent of the act.

Impact of the Disclosure

In order to carefully investigate the disclosure issues, it is

necessary to discern the nature of the disclosure itself. The impact

of the disclosure will provide a frame of reference through which to

judge the impact of the form of the information. Thus, a group, rep-

resenting the present information environment, will serve as a compari-

son point for this investigation.

The Central Research Questions

The issues discribed above served as a basis for the develop-

ment of the conditions used to manipulate the energy use disclosure.

There were five disclosure conditions and a comparison condition which

provided insight into the research questions which were the focal

point of the research:

1. Will the presence of energy consumption data on
labels attached to major home appliances be likely
to assist the consumer in purchase-related activities?

2. Is the reporting format required under the Energy
Policy and Conservation Act the "best" way to disclose
energy consumption?

The orientation of this research has been to develop the issues

of the program from within the program activities. Also, the assess-

ment of these issues is contingent upon the method of investigation.

The Method

This research was a controlled experiment investigating the

impact of the energy consumption data. A sample of housewives were

paid to participate in a study requiring evaluative judgments on a

major home appliance, refrigerator-freezers.

The refrigerator-freezer is an interesting product in relation

to the energy program. The product occupies a prominent role in the

home. In addition, refrigerator-freezers account for a large propor-

tion of energy consumption in the average household. This energy use

is attributed to the fact the product is continually in use, rather

than some unique product use characteristics. The variability in

energy use for refrigerator-freezers is due to various features which

make up the different models. Consequently, the refrigerators is a

useful and relevant product class for the investigation of the energy

use disclosures.

The Impact

The specifics of the data collection method reflects careful

attention to the energy program. The program is in the early stages

of development. Consequently, it is doubtful whether one level of

impact would be able to completely address the issues under question.

Further, researchers are generally urged to employ multiple measures

of impact (Heeler and Ray, 1972, Wilkie, 1975, Day, 1976). The

character of the results in total would provide better support for

the research conclusions.

The use of multiple dependent measures characterizes this inves-

tigation. The fundamental issue is to provide a profile of the impact

of information (in this case energy use information). It is felt that

the impact must be measured as directly as possible rather than infering

the information use from certain global responses. The levels of impact

are essentially a hierarchy of responses to the product/information.

These levels represent initial evaluations, information recall,

evaluations in the presence of the information, and a specifically

designed choice situation. The key goal of these measures is to

profile, as completely as possible, the impact of energy use infor-


The Task

In order to obtain the measures, the task orientation of the

subjects received considerable attention. The result was an involving

experimental session which taps the levels of impact in a realistic

fashion. The details of the experimental task will be provided later

in the text.

Thus, the research effort is characterized by careful attention

to the issues surrounding the disclosure of energy consumption data.

The treatment conditions are potential candidates for the disclosure

format, and can be compared across various levels of impact. These

levels of impact, from initial model evaluation to attribute choice,

provide a complete picture of the impact of the energy consumption

information. The measures were taken from a sample of housewives,

paid participants, responding to the product class of refrigerator-


The goal of the dissertation was to address the issues within

the program to disclose energy consumption data on major home appli-

ances. Each of the essential research decisions were made with this

problem perspective as foundation, and results with implications to the

energy program and the public policy research endeavors.

Organization of the Report

This dissertation is composed of six chapters. Following this

introduction chapter, Chapter II presents the background material to

the energy disclosures. Particular emphasis is given to three major

activities; the M.I.T. report on consumer durables, the Voluntary

Labeling Program, and the Energy Policy and Conservation Act. Past

public policy research efforts and their limitations will be discussed

along with the results of two focus group interviews, both of which

serve as foundation for certain research decisions.

Chapter III is a formal statement of the hypothesis to be tested.

Following the hypothesis is a presentation of the manipulations which

serve as experimental conditions.

Chapter IV discusses the methodology employed in collecting the

data. Within this chapter, the dependent measures, sample, experimental

procedure, and experimental stimuli are discussed.

The results of the study discussed in order of the hypotheses are

the substance of Chapter V.

Chapter VI is synthesis of the research with particular care to

the issues of future research efforts.



Within the past few years many activities have surfaced in

response to what is commonly known as the energy crisis. The concern

for dwindling resources for energy has resulted in many actions at

both the local and national level. One of the centers of activity

has dealt with the provision of energy information to the consuming

public. The object of these provisions has been the demand for

energy-related products. A recent set of activities has dealt with

the provision of energy consumption information with regard to major

home appliances.

This research effort is directed toward issues surrounding the

successful implementation of a program to provide energy consumption

information. In order to insure that research inputs will be useful

for the program, the program activities must become the focal point

of the research efforts.

This particular chapter will present the background material

which has led to the research effort. The activities to be discussed

are in three sections: 1) major energy programs or research;

2) focus group interviews; and 3) past research on public policy

information programs.

Major Energy Programs and Research

The scope of this section is limited to the energy activities

which one finds directly related to both energy consumption provision

and the products which are classified as major household appliances.

Thus, this chapter presents the major activities which led to this

research project.

The M.I.T. Report (1972)

In response to a general concern, by a broad spectrum of society,

with the costs and quality of product service, a major two-year study

of the consumer appliance industry was undertaken. The study was con-

ducted by the Massachusetts Institute of Technology's Center for

Policy Alternatives, and the Charles Stack Draper Laboratory, Inc.

under N.S.F.-RANN sponsorship. The purpose of the study was to examine

alternatives to increasing the service of a product thereby reducing

the cost of a product to the consumer.

To address this purpose, the study focused on the concept of

life cycle costs. The definition of life cycle costs (LCC) is as


the sum of all dollars spent during a
product's useful life." (MIT Report, 1972).

The concept of LCC centered on three cost components: 1) purchase

cost the price of acquisition; 2) energy cost the price of power;

and 3) service cost the price of repair. These components represent

the major portion of all dollars spent on a major appliance. The

general implication of this approach is that it represents an extended

view of the cost of a product. Given the ecological limitations on

energy, it is certainly reasonable to assume that energy consumption

be viewed as a part of the cost of a product. The major contention

of this report is that the public is not aware of cost (as it is

viewed as LCC) of owning a major appliance.

Two points should be made before continuing. The orientation

of this particular study is consistent with the thrust of the present

concerns for energy conservation. As stated previously the MIT study

focuses on the impact of LCC in terms of cost and quality of product

service. The orientation is that the energy cost is a prominent

component of LCC and that the quality inferences made about a product

should reflect the changed conceptualization of costs. To carry this

potential reasoning further, on another things equal basis, higher

energy consumption would be interpreted as lower quality.

The second point is that the MIT study does not provide data

as to either the present meaning of costs to the consumer, or to the

possible impacts of the new view of costs. The MIT study investigated

the state of the art of the appliance industry and provides credence

to the new LCC concept as a valid and needed way of disclosing product


In order to investigate the character of LCC in the appliance

industry, two representative products were chosen for study: refrig-

erator-freezers and color television sets. By choosing these products,

the project could investigate products which represented large LCC

expenditures, and also products which demonstrated a different com-

position of the LCC. Purchases of color televisions and refrigerator-

freezers accounted for $4 billion and $1.7 billion, respectively, in

1972. Further, refrigerator-freezers ranked first and color tele-

visions ranked third in terms of annual expenditures on purchase,

service, and energy when these costs were viewed collectively as

LCC. Further, the study represents two different types of goods when

viewed from a consumer good classification standpoint (brown home

electronics and white -major household appliances). Thus, when con-

sidering the product classes, refrigerator-freezers and color tele-

visions, each product is representative of the broader major home

appliance industry.

The results of the MIT investigation yielded some interesting

findings with regard to the composition of LCC for refrigerator-

freezers and color television.

Table 1
Summary of the Composition of Life Cycle Cost* for
Refrigerator-Freezers and Color Televisions

LCC Components Color Television Refrigerator

Purchase Costs 53% 36%
Service Costs 35% 6%
Energy Costs 12% 58%
100% 100%
*Source: The M.I.T. Report, "Consumer Appliances: The Real Costs."

It should be noted that the figures reported through the M.I.T.

investigation were for years prior to 1971. Since that time the major

economic problems would be reflected in changes in the composition of

the life cycle costs. The national energy shortage has resulted in

increasing energy costs coupled with strong inflationary pressures.

The costs of electricity rose thirty-six percent between 1973 and

1975, while the overall cost of living rose twenty-one percent

("What's Behind the Energy Crisis," 1975). These escalating costs

accentuate the costs of products, particularly the magnitude of the

energy component. The implication of the M.I.T. study remains, the

inclusion of energy and service cost as a component of total costs

changes the composition of the cost of an appliance dramatically.

Table 2 focuses more directly on the energy component of the

LCC. Both refrigerator-freezers and color televisions occupy prominent

roles in household energy consumption. Refrigerator-freezers alone

account for 14.1 percent of the household energy expenditures. This

finding provides some support for the attempts to provide energy con-

sumption data. If the consumer, as assumed by the MIT report, is

unaware of the LCC (particularly energy cost) it is possible that

energy-related decisions may be affected by the disclosure.

One further note, from a technological standpoint there exists

some potential for energy savings through improved design features

on refrigerator-freezers. Thus, the disclosures, as the MIT report

states, may facilitate these design changes through consumer expendi-

ture shifts.

The MIT report highlighted the extent of energy consumption by

major appliances and the apparent need for activities which lead to

lower energy consumption. Two major activities have been oriented

specifically to that purpose. The following sections will discuss

the Voluntary Labeling Program and the Energy Policy and Conservation


Table 2
Summary of Major Sources* of Residential Energy Use

Water Heaters
Clothes Dryers
Air Conditioners
Space Heating
*Source: The M.I.T. Report, "Cc

% of Total Energy Use

)nsumer Appliances: The Real Costs."

The Voluntary Labeling Program

In December, 1973, the President of the United States, in an

address on energy, directed the Department of Commerce to develop a

voluntary program for energy efficiency (McGuire, et al., 1975).

This directive became the first formal statement of the government's

role of information provision in the energy consumption problem. As

with many of the previous public policy programs, this action was in

reaction to a broad problem, which is characterized by a lack of

information available to the consumer. This program was to be carried

out in cooperation with the Council on Environmental Quality and the

Environmental Protection Agency.

The stated purpose of the program was to encourage manufacturers

of household appliances to voluntarily provide consumers with infor-

mation on the energy efficiency or energy consumption of products.

In addition to the mere provision of energy consumption data, the

program hoped to encourage consumers to utilize the consumption data.

The action with regard to this secondary purpose took the form of

educating the consumers about energy consumption. The relevance of

the voluntary program for this research project is related to the

overall goal: the provision of energy consumption data. The energy

consumption data was to be made available to the consumer through

labels affixed to the product at the point of purchase.

In the past, research efforts related to public policy programs

have been hampered by inexplicit or vague goals of the program. The

voluntary labeling program rather uncharacteristically possessed

explicit well-stated goals. The statement of goals is as follows: encourage manufacturers to provide con-
sumers at the point of sale with information on
energy consumption and energy efficiency of house-
hold appliances. Such information presented in a
manner readily understandable to consumers, would
be attached to or otherwise provided with the appli-
ance or equipment. The labels will include a system
intended to make it possible for consumers to compare
by cost or otherwise the energy consumption and
energy efficiency when purchasing household appliances
and equipment."

Thus, the program was designed to provide uniform, understandable

information on energy consumption and efficiency so that consumers

can use the energy consumption within the comparison process of a

purchase decision. Note that the program goals refer to two dimen-

sions of the disclosure: consumption and efficiency.

Energy Consumption vs. Comparative Energy Consumption

By attempting to communicate both consumption and efficiency,

the voluntary program was going a step beyond past information pro-

grams. It is worthwhile to take a moment to discuss this change.

It is difficult to understand at first glance the difference

in what is meant by consumption and efficiency. In fact, the issues

become clear only after investigating the proposed disclosures under

the program. The disclosure of energy consumption data represents the

classic information provision, the disclosure of the particular prod-

uct's performance along a specified dimension. With regard to the

energy policy, the aggregate use of energy expressed in a specified

unit per time period would be the disclosure of energy consumption

(e.g., the energy consumption of model A is $60 per year). This dis-

closure provides the performance information, but, in and of itself,

gives no clue to the relative performance of the particular product.

It is to this issue that energy efficiency is directed. By communi-

cating performance alone the consumer must make the comparative

judgements through further search activity.

However, by communicating "efficiency" (comparative energy

consumption), it provides comparative data. Thus, the term efficiency

is misleading in that the disclosure is not addressing technical

functional efficiency but rather the relative energy consumption

(i.e., does model A appear to be more efficient than model B consume

less energy). The standard for efficiency then is the relative energy

consumption of the product. The provision of comparative information

is directed toward the comparison process of the consumer. In order

to make accurate comparative judgements, the consumer would have to

engage in extensive shopping behavior. By way of example, the volun-

tary labeling program provides comparative energy consumption on air

conditioners by presenting the range of energy efficiency ratios

(energy consumption) of similar-sized air conditioners. The consumer,

by this disclosure, can make the evaluative judgements of the air

conditioners' energy consumption. This form of disclosure represents

an empirical question which is, to date, untested.

Products Covered Under the Program

Under the voluntary labeling program, a labeling specification

was to be developed for each of the following products:
room air conditioners,
water heaters,
clothes washers and dryers,
ranges and ovens, and
central heating and air conditioning. (McGuire, et al., 1975)

The focal point for the development of these specifications was the

Appliance Labeling Section of the National Bureau of Standards. The

order of the products considered was determined according to estimates

of potential energy savings. The program proceeded in a product by

product fashion in all phases of development of a labeling specifica-

tion in all phases in order to insure the information provided reflects

the energy consumption character of the product class.

Summary of Program Activities

By February, 1975, the voluntary program had determined the unit

of measurement by which to disclose the energy consumption on three

product classes: room air conditioners, refrigerators, and water

heaters. Table 3 shows the decisions with regard to the disclosure.

These decisions were made with consideration to the unique character-

istics of consumer use of the product, and/or consumer preference for

a particular form of disclosure.

Table 3
Summary of the Disclosure Decisions by the
Voluntary Labeling Program
Date of Publication
Product Class in Federal Register Unit of Measurement

Room air conditioners January 1974 energy efficiency ratio
Refrigerator September 1974 dollars per month
Water Heaters February 1975 dollars per year

The room air conditioner energy consumption was to be disclosed

by an energy efficiency ratio (EER). The EER is obtained by dividing

the cooling in British thermal units (Btu) per hour by the power

requirement in watts. This particular disclosure was chosen due to

the wide variation in product use and the critical nature of the size

of the air conditioner units. For example, the energy consumption of

a 2000 Btu room air conditioner would depend on the size of the room

to be cooled, the temperature desired, the characteristics of the

residents, etc. The unique features make the computation of an

annual, or otherwise aggregated, cost figure impractical. The compara-

tive energy consumption was disclosed by a range of EERs for similar-

sized room air conditioners. The unfamiliar nature of the EER resulted

in an education program conducted by the Department of Commerce to

explain the disclosure.

The next product class under consideration was refrigerator-

freezers. This product class disclosure decision was characterized

by a small scale survey of 100 residents in the Washington, D.C. area

(Persensky, 1975). The survey investigated preferences for disclosure

format by the sample. The results indicated a preference for the format

of dollars per month. The rationale for this type of limited primary

data collected as decision criteria rested with usage patterns of

the product class. The patterns of use, and subsequent energy con-

sumption, for refrigerator-freezers was uniform, and consequently did

not require additional format consideration as with the air condi-


The final product class undertaken by the Voluntary Program was

water heaters. The disclosure format chosen was dollars per year.

Even though this product class is faced with certain unique usage

characteristics it was felt that a standardized disclosure could

convey the energy consumption data adequately.

The disclosure formats of both refrigerator-freezers and water

heaters was characterized by energy consumption alone, and not the

new dimension of relative energy consumption.

Response to the Program

While the voluntary labeling program was relatively short-lived,

the response to the program was considered to be favorable. In the

room air conditioner product class, the 24 manufacturers who partici-

pated in the program represented 95 percent of the sales of air con-

ditioners in the United States. Of further interest, is the fact

the average EER for all 115 volt window model air conditioners

increased (improved) 6.5 percent from January, 1974 to January, 1975.

This period coincides with the implementation of the labeling program

for room air conditioners. There does not exist supportive data with

regard to consumer response to the program for the same period. The

recession, and the fact that very few labeled items were sold as

inventories were slowly worked down, is the primary reason for the

lack of documented consumer response. The improvement in the EERs

is supportive of the potential impact of the disclosures. The

activity relating to voluntary provision of energy consumption data

has subsequently become mandatory under the Energy Policy and Con-

servation Act.

Energy Policy and Conservation Act (Public Law 94-163)

On December 22, 1975 the Congress of the United States passed

into law what is commonly known as the Energy Policy and Conservation

Act (EPCA). This act is the most recent in a stream of activity

responsive to the issues related to the energy crisis. With rather

broad scope, the EPCA is directed to the following overall goals:

*" to increase the domestic energy supplies and
availability, to restrain demand, to prepare for
energy emergency, and for other purposes." (The EPCA,
p. 871).

The act has far reaching issues and goals. The realm of consid-

eration for this dissertation will focus on the aspect of the EPCA

dealing with the provision of energy consumption on labels attached

to major home appliances. The thrust of this area of the act is the

consumer demand for energy related projects. Toward this end the

EPCA specifies the following goal:

to provide for improved energy efficiencies
of motor vehicles and major appliances and certain
other consumer appliances." (The EPCA, p. 874).

The particular action related to this goal is a labeling requirement

on major home appliances.

Administration of the EPCA

The Federal Energy Administration has been designated the admini-

strator of the act. However, the Federal Trade Commission has the

most direct responsibility to determine the test procedures, while

the Commission ascertains whether such disclosures are technically

feasible and will 'assist' the consumer (The EPCA, p. 921).

Covered Products

At present the following products will come under the EPCA:
refrigerators and refrigerator-freezers,
clothes dryers and washers,
water heaters,
room air conditioners,

home heating equipment,
television sets,
kitchen ranges and ovens,
and central air conditioners. (The EPCA, p. 918).

The above list is composed of products with potential for energy

savings. The criterion used for inclusion of a product was for the

energy consumption of the product during normal usage to exceed 100

kilowatt hours per year (The EPCA, p. 918).

Label ing Requirement

At the present time the labeling requirement for the disclosure

of energy consumption data specifies the following format:

(1) ". the estimated annual operating cost of such
product," and
(2) ". . information respecting the range of estimated
operatina cost for such type or class of products"
(The EPCA, p. 922, paragraphs A & B).

Thus, the disclosure will consist of the annual energy costs and the

range of estimated annual costs for that class of products. This man-

date brings to the foreground some issues surrounding the disclosure.

First, the unit of measurement by which the energy consumption

is to be disclosed is in dollars. The question surrounding this issue

is the technical aspects of the disclosure standards and the clarity

with which the disclosure is tied to the goals of the EPCA. The goals

of the EPCA with regard to the labeling requirement are, unfortunately,

not instructive: the criterion by which the labeling effectiveness is

to be judged is that of the disclosurer's ability to "assist" the con-

sumer in making purchase decisions (The EPCA, p. 922). There exists some

criterion (i.e. assisting the consumer), but it is stated too broadly

to be directive with regards to research. As was discussed earlier,

some products are characterized by unique usage aspects, which

thwart certain kinds of disclosure (i.e. the EER for air conditioners

under the voluntary program). Thus, as a contingency measure, the

alternatives to the general disclosure requirements should be inves-

tigated. The first issue is that the present requirement must "assist"

the consumer. This requires that the disclosure requirement be

researched to attest to the superiority of annual dollars to assist

the consumer. Even without the need to justify the disclosure impact,

the technical nature of the disclosure standards suggest that contin-

gency alternatives be investigated in the case of unforeseen disclosure


The second issue stems from the disclosure of the range of com-

parative energy consumption for similar products. The act specifies

that similar products are products the functions of which are similar.

This tautology allows little guidance as to along what specific dimen-

sions this similarity may be defined (i.e. size, features, style, etc.).

As a matter of clarification this research will take the point of view

that the criterion for the range of comparative energy consumption

should consider those characteristics of a product which most closely

relate to the variance in energy consumption. This formulation is

consistent with the philosophy of the range disclosures. The range

should allow a comparative judgment with regard to a product's energy

consumption. Conceptually, this would be analagous to a consumer

sampling an entire class of "similar" products with regard to energy.

When that consumer is confronted with a product's energy consumption

that information can be evaluated against the other similar products

by using the range of comparative values. Consequently, certain deci-

sion-related judgments should be facilitated by the range (e.g., the

feasibility of shopping, evaluation of the performance, etc.).

The ability of the range to communicate this information is

subject to the manner in which the range is developed. For the con-

sumer the range should reflect the characteristics of the products

which represent energy use. In this way the range would be formulated

differently for different product classes (e.g., the air conditioners

would have products similar in size, the refrigerator-freezer would

have products similar in features, etc.).

It is important to realize that the operationalization of the

range is crucial from a program point of view. In subsequent investi-

gations the results of the comparative disclosure, either successful

or unsuccessful, must be considered in light of the particular range

under study.

With regard to the disclosure, it is encouraging to note the

implied role of research in the decisions concerning the specific for-

mat requirements. The mandatory requirement must meet the criterion

of aiding the consumer in purchase-related decisions. If this particu-

lar requirement does not meet this standard, the requirement is subject

to change.

General Comments on Programs

The discussion has centered on three major activities which have

served as primary background for this research project. These acti-

vities are not to be considered peripheral to the research, but are

the focal point of this research effort. In some ways this is a unique

thrust of academic/public policy research as the goal is to be respon-

sive to the programs, rather than viewing the programs as only a

vehicle for the research efforts. There are several ways in which

research can be approached within the public policy arena. The pro-

gram can serve as a framework from which to structure a research

situation, or as a tool to can insight into consumer response in

general. From this perspective the consumer issue becomes the focus,

and only secondary input can be claimed to the public policy program.

An opposite approach can be used. In this case the policy itself

becomes the focus and the issues under investigation are attuned to

program needs. For this purpose the research may draw upon a consumer

behavior background to investigate program issues. The results would

address the program from a consumer context. The latter is the

approach taken in this dissertation.

Past Public Policy Research

While this research is focused on the issues surrounding the

disclosure of energy consumption data, it is instructive to take a

closer look at the broader policy environment to learn from past

programs and research efforts. This type of effort is particularly

useful when the focus of a research project is the policy action

rather than a broad theoretical framework. In the same manner as

theoretically oriented efforts use the past research as foundation,

the policy oriented efforts use past public policy research as support,

and guidance.

This section presents the policy decisions and research efforts

for past programs.

Public Policy Decisions

The problem of designing a consumer information program is a

complex task. The exact flow of the policy decision has taken on

different formulations (see Wilkie, 1975, Ferguson, 1972, and Day,

1974a). The essential steps include a problem identification stage

(both product class and attribute decisions), the development of test

standards, the determination of reporting formats and the means of

disclosure, and the assessment of impact. Obviously, research efforts

can input at any stage of the decision process.

With regard to the energy consumption program, many of the

decisions must be taken as given by this dissertation. The relevant

dimensions, product classes, and testing methods and standards are

decisions which are not subject to change. However, research efforts

can be useful inputs into the other decision areas. The formal

activities in the energy consumption information programs have demon-

strated both a need and desire for research inputs into the reporting

format stage. The final format of the energy consumption disclosure

within the EPCA, is subject to the confirmation of a particular format's

ability to assist the consumer. In several academic position papers,

the lack of format-like research has been suggested as possible

rationale for program effectiveness shortcomings (e.g., Monroe

and LaPlaca, 1972 and Day, 1976).

The area of program effectiveness is an area where research

efforts can serve policy needs. The need for evaluation of programs

is crucial to future effort. Indeed, much of the research efforts

has been toward this type of problem (i.e. does the program have an

impact?). With the advent of more standardized requirements and

more complex disclosures, the issues around program effectiveness

types of research become difficult.

The issue of measurement have received attention in the liter-

ature. Day (1976), Wilkie (1975), and Heeler and Ray (1972), among

others, have urged the adoption of multiple dependent measures.

The basic issue is that a single level of impact does not allow for

a complete and accurate profile of the impact of information. This

is essentially an attempt to more carefully reflect the issues under

study. In most public policy programs the standardized performance

disclosure represents a new or changed input into the consuming

public's information environment. The objective is to establish the

impact of a particular disclosure. The essential question is what

measure to employ. Since the phenomena under study is the information

it would be instructive to tap those dimensions where information

could or should impact. This would allow for the monitoring of

impact across certain crucial dimensions; behavior, evaluations,

cognition, etc., and a more careful assessment of a program. This

dissertation will follow this guidance and address multiple levels

of information impact by incorporating multiple dependent measures.

The types of measures which are taken for program effective-

ness are usually reflected in the task orientations of the research

effort. As with much of the work in marketing a brand versus brand

decision, orientation characterizes much of the policy directed

research efforts. As Bettman (1975) states in an article on

designing consumer information environments, instead of choice

behavior, the impact of a given policy may best be reflected in

accept/reject judgments. Wilkie (1976) suggests the use of more

non-directive constructs which reflect knowledge with regard to

information disclosures. The conclusion is not that behavior

measures should not be implemented, but that behavior measures should

be considered in the context of the broader impacts of a particular

program. This impact will be measured by additional levels of effect

and the use of task orientations other than choice, particularly

brand choice, decisions.

Considering only the energy program for a moment, the program

decision framework just presented allows for identification of areas

of research potential. The energy program to date has been character-

ized by an inability to adopt specific uniform reporting formats and

then by the EPCA mandatory format without consumer impact justifica-

tion. The issues associated with the format of the disclosure repre-

sent important areas of needed research effort.

The general need to assess the effect of disclosures is faced

with some rather complex problems. The ultimate goal of the program

is to reduce energy-related demand. To assess the effectiveness of

disclosures at this time is a difficult job. Energy information is

a new informational input for consumers to use. If one were to apply

a strict behavioral criterion to the disclosures, the success of the

disclosure will be limited. It will be necessary to measure the

levels of impact across different task orientations to better profile

the success of the disclosures.

For the most part the recent research efforts have been charac-

terized by attempts to assess the success of an information program.

The early unit pricing studies, reported by Monroe and LaPlaca (1972),

showed a desire to survey the state of awareness and use of unit

pricing. However, the more recent studies by Houston (1972), Granger

and Billison(1972), Isakson and Maurizi(1973), all attempted to

measure program success. In the truth-in-lending area the study by

Day and Brandt (1974), Stokes (1972), and Walker and Sauter, (1974)

followed a program evaluation orientation.

There are few policy-oriented studies which are attempting to

be inputs to other stages of the program decision. Russo, et al.

(1975), Stokes (1973), and Chestnut (1976), in the unit pricing,

nutritional labeling, and energy labeling areas respectively were

oriented toward the reporting format stage. The crucial aspect is

not what these studies found, but their orientation to their research.

The orientation will limit to type of measures undertaken and the use

of the results.

The stage of input to the policy decision is an important con-

sideration when evaluating previous research efforts. However, a

more crucial attribute, if one is to learn from these past efforts,

is the quality of these efforts. This calls for a more critical

look at the particular research efforts with an attempt to determine


It would be outside the scope of this study to review all

aspects of each of the research efforts undertaken for public policy

purposes. However, there exist some crucial issues which reflect

shortcomings in past research, which this particular study will try

to overcome. The areas for consideration are experimental stimuli,

sample composition, and the task orientation.

Experimental Stimuli

The products chosen for use as experimental stimuli have

received very little critical attention. This decision represents

a decision with regard to product class and brand names.

The choice of product classes represents an important aspect

of the quality of the research input. In the unit pricing studies,

one sees wide variance in the experiments attempting to assess the

impact of information. Houston (1972), in a field experiment used

14 products, Gatewoodand Perloff's (1972) lab experiment used nine

products, Isakson and Maurizi (1973) used 13 products in a field

experiment, Russo, et al. (1975) used three products, and Friedman

(1966) used 20 products. The nutritional labeling program fared no

better as Asam and Bucklin (1973), used one product (canned peas)

and Stokes (1972) used no products to determine the impact of nutri-

tional information. Chestnut (1976), in the only energy labeling

study, purports to test format alternatives by using the product

class of light bulbs. With regard to the Chestnut study, the product

class is not even under consideration within the energy labeling

programs discussed in the earlier sections of the paper.

The disturbing aspect of the above studies and their choice

of experimental stimuli is the virtual lack of overt criterion used

in the selection process. The choice of experimental products should

not be left to either convenience or subjective judgments. But,

instead, the choice needs to be tied directly to the program goals.

What is suggested is the establishment of overt criterion for product

class selection. The suggested criterion are as follows:

the relevance of the product class to the program,
the variability of the product class along the
dimensions to which the program is directed,
the characteristics of product use,
the prominence of the product in the household, and
the state of knowledge of the consuming public.

When evaluated against these criterion measures, the product class

choices of previous research appear to be lacking. For example, the

impact of unit pricing would seem to vary by product class, and conse-

quently the products used will have an impact on the conclusions to

be reached. The only unit pricing study to use a criterion for

product classes was the Russo et al. (1976) study, which implemented

the criterion of variability between products or price. This criterion

resulted in the use of dishwashing liquid, facial tissue, and dog food

from which to make statements about unit pricing. In none of the

other studies on unit pricing is a criterion for product selection

given. Also, the Chestnut study, a study which at first glance value

would be relevant to the energy issue, is particularly lacking in the

choice of product class, and the subsequent conclusion with regard to

the policy should be suspect.

The second decision with the choice of experimental stimuli is

the use of real vs. hypothetical brands. Brand names have been sug-

gested and found empirically to impact on product choices (for

example, see Bettman and Jacoby, 1975). Thus, if one were interested

in the impact of information, the presence of brand may overwhelm or

cloud the effects. If one could measure the effect of a brand name,

the task of making unequivocal conclusions would be facilitated.

The researcher can opt for the use of hypothetical brand names.

However, the study would be attacked for the lack of external validity.

The situation the subjects are placed in is not "real." What can be

attempted is the use of alternatives not tied to brand choice. In the

area of energy consumption this can be accomplished by using model

rather than brand choice. The consumer (subject) is then faced with

a situation where the brand name is not known, but the task is reason-

able when compared to typical shopping behavior.

Sample Considerations

There appears to be a shift into experimental work with regard to

public policy programs. Consequently, the sample size is a function of

the study objectives, and is not the crucial consideration. The

important judgment is sample composition and the relation of the com-

position to the program. It is interesting to note that in only one

public policy research effect was the limits of sample composition

related to the results. Day and Brandt (1974) reported the deficien-

cies of the sample which may limit conclusions about the truth-in-

lending issue under study.

It is essential that the sample composition reflect the segments

of the population which will be related to the success of a disclosure.

Thus, consideration must be given to the decision making process

related to the disclosure. As the disclosures become more and more

complex, the selection of the sample must be made with care. The

energy issue is a good case in point. The energy labels will be

attached to major home appliances. The sample to be drawn must be

related to the individuals making decisions about major appliances.

It is expected that the sample composition would vary even between

product classes. For example, the refrigerator-freezer decision may

be characterized by a joint decision with the feature choice being made

by the wife and with the technical aspects being decided by the hus-

band. It is possible that the water heater decision will be made by

the husband and dishwasher by the wife. These statements are not

facts, but reflect the type of consideration that needs to be made

by the researcher in determining the sample composition. The only

energy-related study, Chestnut's (1976), is particularly lacking in

this area. The use of 36 students to test the impact of alternative

energy consumption formats render the results of the study of

questionable usefulness to the energy program.

If policy research on information programs is characterized by

deficient samples, the limitations of the sample should be stated so

as to effectively delimit the corresponding results.

Task Orientation

The task orientation is a crucial decision for the experimental

work done for public policy issues. This particular topic has been

alluded to earlier. In most of the studies designed with information

programs under consideration, the most frequent task orientation is

choice, particularly brand choice. The task orientation would result

in a limited view of impact since the only level of impact to be

measured is behavior. It is possible that choice per se does not

reflect the impact of information. As Bettman (1975) states, the

brand choice decision may not be correct task to make a research

input to a public policy program. It may be that the evaluations of

a consumer better reflects the information impact in the more typical

shopping experience. Also, this orientation would not place choice

pressures on the individual in a one-shot experimental setting.

The designs which are used to test the impact of information

must be attuned to policy needs. In order to provide a meaningful

input, the task must be structured so as to reflect the interaction

between consumers and information in a typical shopping situation.

This requires that the tasks and task orientation be determined with

regard to policy needs, and to the real life information/consumer


This section has attempted to provide a framework for public

policy research. To do this a view of policy decisions followed by

crucial issues concerned with policy research were presented. This

discussion was oriented primarily to describing past research efforts.

The highlights of this review provide the following generaliza-

tions. Most of the studies have attempted to ascertain program

effectiveness through the use of consumer surveys, focusing on the

impact in terms of consumer choices. The investigation did not provide

objective criterion for product class choices. The overall conclusion

is that this manner of investigation did not allow for meaningful

inputs to the policy program under question. This dissertation will

address the questions of program impact through the format decision.

Since the focus is on information this dissertation will attempt to

provide a complete picture of effect by measuring impact at multiple

levels. The task orientation will reflect these levels by providing

orientations consistent with the phenomena under study. With this

background the research of this study can overcome the past deficien-

cies, and provide a sound input to the energy consumption information


Focus Groups

Toward this end two group interviews were arranged through the

Central Florida Research Group, a local professional organization.

The groups were selected on the basis of education due to past policy

oriented efforts (Day and Brandt, 1974) showing variance in response

to information programs correlated with education. Group 2 was a low-

education group, operationalized as a high school education or below.

Group 1, selected on the basis of a college education, was the high-

education group. As Table 4 shows, the groups did follow these guide-

lines. In addition, as a result of the education differences, the

Table 4
Demographic Profiles of Group Interview Participants

Characteristic Group 1 Group 2

18-22 2 --
23-28 1 1
29-33 -- 3
34-38 -- 3
39-44 1 1
45 or above 4 --
11 or below -- 1
12 years -- 6
some college 1 1
college graduate 5 --
post-graduate 2
$ 0- 4,999 -- 1
5- 9,999 1 3
10-14,999 2 2
15-19,999 2 2
20,000 or above 3 --
Marital Status
married 8 6
divorced -- 2
Last Appliance Purchase
1 year 2 4
1-3 years 5 1
4-6 years 1 2
7-9 years
10 or more -- 1

groups differed on the basis of income. Each group consisted of eight

female residents of Gainesville, Florida, or nearby communities. None

of the participants were connected with the university. The partici-

pants were paid $8 for a group interview session lasting approximately

one hour. Group 1 was conducted by the author, and Group 2 was con-

ducted by R. Bruce Hutton. In order to insure some uniformity between

the groups, the group leaders determined a standard introduction,

and basic outline of the discussion. Permission was obtained from

the participants, and each session was recorded.

At the beginning of each session, the group leader explained

that this interview was conducted in conjunction with in-process

dissertations on consumer decisions relating to major home appli-

ances. The group was introduced to one another, and the group leader

requested that each participant place herself in the role of a con-

sumer. The participants were told that they were to think about the

purchase of a refrigerator-freezer. The groups were told that this

product was chosen to facilitate discussion, and because each of them

had considerable experience with refrigerator-freezers. The partici-

pants were then told to think of things they consider when they are

going to buy a refrigerator-freezer. The discussion proceeded in a

voluntary fashion with effort by the group leader to encourage partici-


Tables 5 and 6 summarize the results of the two groups. Table 5

presents the order of discussion of the characteristics of refrigerator-

freezers. Table 6 summarizes the comments made about the character-

istics of refrigerator-freezers by group.

In general, some important differences between the groups

emerged. The energy use dimension was brought up for discussion very

early by the high-education group, but had to be suggested by the

group leader in the low-education group. The high-education group

discussed the energy use by features, and felt that the information

Table 5
Summary of Characteristics of Refrigerator-Freezer
by Order of Discussion

Group 1 Group 2
High Education Low Education
1. sources of information 1. frost free
2. frost free 2. icemaker
3. energy use 3. water dispenser
4. price, costs 4. vinyl door coverings
5. icemaker 5. shelves
6. shelves 6. color
7. vegetable bins 7. service
8. meat tray 8. maintenance agreement
9. size 9. warranty
10. location of freezer 10. shopping-around process
11. expected life 11. brand name
12. rollers 12. size
13. service* 13. expected life
14. maintenance agreement 14. costs*
15. warranty* 15. energy*
16. textured steel door 16. sources of information
17. color
18. salesmen
19. shopping-around process
* characteristic suggested by group leader.

was presently available on the labels of the appliances. The low-

education group's comments indicated a general concern for energy, but

an aversion to information, since those commenting felt they would have

difficulty "understanding" the information. The participants of

Group 2 indicated that they would pay more for energy saving, but the

saving would have to be automatic--those commenting did not want "a

course on how to save energy."

The high-education group appeared to be oriented more to the

use of information. The group expressed faith in such sources as

Consumer Reports as a starting point, and said they would read the

Table 6
Summary of the Comments of Groups 1 and 2 Made
About Characteristics of Refrigerators

Group 1

It is the number one thing I
would look for.
Uses a lot of energy.
Estimate of energy cost--
1/3 more, $1 per month.

Tremendous luxury
Saved my marriage
That's an energy consumer, too.
There's a problem with water
and corrosion.

Cost--$40 plus installation

Not mentioned

Doesn't show fingerprints.
Most super-marvelous thing that
has ever happened.

Adjustable shelves are a necessity.
I'd like wide shelves on the door.
I'd look at the supports on the

I don't like to defrost.
The food doesn't spoil.
Dries food out
Didn't know they made any
that weren't frost free.

Don't spill water anymore.
My mother has one.
Ice trays don't last, and
they pinch my fingers.
Can transfer extra ice.
The arm keeps it from making
Uses more electricity, but
it's worth it.
Doesn't make ice fast enough.

I won't have the nozzle on
the door
Clogs up.
It's not worth the trouble.
Not a selling point, but the
kids love it.

Doesn't show fingerprints.
Saw a commercial that said
you could change the colors.

Don't like plastic holders
on the shelves
Adjustable shelves, in case
I want the shelves in
different places.
Wide spaces in the door for
large objects.

Group 2

Table 6 (continued)



Cost a little extra, but
not much over the life.

I would consider that
Consumer Reports--calls
per year.
$15 to walk in the door.

I don't personally think those
service contracts are that
No good the first year.
Good in the last years of life,
but expensive.
Dealers are not too reliable
in upholding them.
Lottery. Gamble.

I think they are pretty standard.
5 years on the motor.
It's different for the motor,
and other things.
It's always the part that isn't
covered that breaks.

I like colors.
Makes a gayer kitchen.
Doesn't show dirt as easy--
depends on the color.
Cost more--about $10.
It used to be more, not
it's got to be about the

So important where you buy
Would pay a little more for
my appliance to buy it
from someone with a good
service department.
Live here long enough to
find out.
If the salesman is proud of
it, he'll tell you.
Cost-too much no matter where.
Have to think of what you'll
$18 just to come out.
Like doctors, they charge
you so much then parts.

If it breaks down the first
year, it's already paid for.
It's worth it the first year.
It's like insurance.
It gets higher the older the
Usually there isn't anything
go wrong the first year.

5 years--to me that's all
you need.
Things break down as the
warranty runs out.
Not a complete one (replace-
ment) after the first year.

Table 6 (continued)

9. WARRANTY (continued)

Read in Consumer Reports, ask,
then look within price
Read label.
Go to stores that sell a lot
of brands.
Husband decides the technical
You need to be informed.

Not mentioned.

12. SIZE
Depends on the family size,
and the space in it.
14 to 15 cubic feet is big.
Mine holds 150 lbs. (freezer).
What's the largest you can get--
that's it.

9. WARRANTY (continued)
They only pay so much.
They use it in their sales.
Rather they didn't charge
me for that little
goodie, and let me buy
my maintenance agreement.
Does warranty add to price?
No, Goes with every
product. That's why it's
not worth anything.

Shop around, get pamphlets
and prices, and go to the
next store.
You have to shop.
I don't want any paper--I
have to look at the item.
See if it will hold my
Best price for what I want.

Reasonably important--don't
think I'd buy Teddy Bear
I'd have to know the brand--
something I'm familiar
Yes, there are differences
in quality--commercials
made us think this.
Pay for quality--depends on
the situation.
Wouldn't borrow money to get
a better one.

12. SIZE
I don't understand it (cubic
feet)--have to see if it
will hold my groceries.
When they say 20 cubic feet,
you don't get that much.
Just some rigamorole they
throw in.
Size of the freezer is a big

Table 6 (continued)



15 years on the average.
The more money you pay the
longer it lasts.
Look at the specifications
on the label.


$300 to $800.
The more money you pay the
longer it lasts.
What makes them last--# of
coils. Better engine.
Look at the specifications
on the label.
Smart to compare on price--
someone might be having
a wingding of a sale.

You keep it 15 years.
15 years for a Frigidaire.
Mine lasted 10 years.
Every brand varies.
Doesn't last as long anymore.

$850 to $1 ,000
Around $300
$600 with an icemaker.
Icemaker--$89. Less than
They charge you for hooking
it up.

I would be interested now in
what kind of energy savings
are available.
Supposed to be listed on a
little fact sheet.
Depends on how energy conscious
you are.
I keep track of how many KWH we
We contradict ourselves, because
I want that (frost free), but
I know for a fact that it is
one of the biggest consumers
of energy.
Depends on whether you want to
use that energy or your energy.
It's not exactly energy conscious-
more cost conscious.
No real difference across brands.

I have an energy saver; I
don't understand it at all;
it's a new little hickey;
something about when the
weather is damp you do
They give you a whole book
on it--you know I never
read the book and never
mashed the little hickey.
Too many other things.
If it is automatic--ok--but
I don't want to do anything.
Have to save a lot for me to
mash the button.
Selling point.
Got to be something you plug
in and is automatic--they
are getting paid for it.
I don't want to waste energy.
Pay more for energy saving,
but I won't read a book.

Table 6 (continued)

15. ENERGY USE (continued)

I would go to a consumer
information outfit.
Shop around.
Ask the neighbors.

I'd die without one.

Not mentioned.

Side-by-side is bad--I'd
never have another one.
Children get in.
I think I, get more in mine
now--located on top.
Top with separate door is
Side-by-side is bad for
big stuff.

Very important for cleaning.

You have to think for yourself.
A lot of these people that sell
in the stores aren't really
aware of all the specifications.
It's surprising how little they

15. ENERGY USE (continued)
Don't want a course on how
to save energy.
Couldn't understand the
books anyway.

Just being involved.
I watch commercials.
Having used the product.
Husbands handle the complex

Not mentioned.

Not mentioned.

Not mentioned.

Not mentioned.

Not mentioned.

labels. Both groups stated that considerable shopping occurred. The

expression of abstract concepts such as size and quality appeared to

be difficult for both groups, although the existence of the concepts

appeared to be universally accepted.

Each group appeared to be price conscious, and stated that they

looked for the "best deal."

Each group had a significant amount of experience with the

product, and appeared to be familiar with the features available.

However, the participants were chosen from one geographical area,

and consisted of only white females. In order to insure that the

results found are representative for other areas, additional groups

in other geographical locations should be conducted. The purpose of

the investigation was not to test hypotheses or to generate testable

data. The groups allowed for greater knowledge of the consumer and

the product class of refrigerator-freezers. Thus, the attributes

deemed important and the terminology used when interacting with the

product class were identified. This information serves a useful

purpose in stimulus development, and construction of the test instru-

ment. Further, the differences between the groups was instructive

for this research due to the fact that the sample for this study will

be taken from the population in Gainesville, Florida, and the surround-

ing area.


Given the background presented in the policy activities related

to the provision of energy use information, past policy research, and

the results of the group interviews, the discussion will now turn to

the specific research issues to which this dissertation is directed.


The hypotheses around which this research was developed stem

from the goal of identifying the relationship between certain inde-

pendent variables which characterize energy consumption disclosure and

certain levels of impact. The specific hypotheses are generated from

the general working hypothesis that the disclosure of energy consump-

tion data will have a positive impact (positive in light of the goals

of the EPCA) on the purchase decisions of consumers of major home

appliances. The following hypotheses will reflect the levels of impact

and the relationship of impacts to the different disclosures.

The first hypothesis relates the impact of energy-use data on

the evaluations of alternatives in a shopping experience.

Hypothesis One: Impacts of the Disclosure
The disclosure of energy-use data will have a
significant impact on the preferences, overall
impressions, and energy-use expectations of
alternative refrigerator-freezer models.
Specifically, subjects receiving energy-use
data will have lower preferences, lower over-
all impressions, and more accurate expectations
of energy use for models which are high users
of energy.

The first hypothesis is directed toward the impact of information in

a situation where the consumer is evaluating alternative appliance

models (i.e. models with different features). This is a very important

question. In many cases the degree of shopping behavior with regard to

consumer durables is characterized by a lack of extensive store-to-

store shopping (Olshavsky,1973, Newman and Staelin,1973). Given this

characteristic of behavior the crucial product decision appears to

revolve around model/feature combinations of particular appliances.

The question then becomes what types of impact can be desired or

expected with regard to energy consumption disclosures.

The energy information, if it is being used within a decision,

should impact on product/model evaluations. This does not mean that

the presentation of energy use data will make the energy dimension

more important, but that the consumers' model evaluations should

reflect the product's energy usage. If a model is a high energy user,

a positive impact of this disclosure of energy use should be reflected

in lower preferences and evaluations. This is not to say that the

consumer should prefer a model that does not meet other requirements,

but within the model/feature combinations under consideration the

consumer preferences and evaluations will reflect the energy consump-

tion data. For example, the energy consumption of a model of refriger-

ator-freezer that does not have a freezer section would be very low

since the freezer section accounts for a large amount of energy use.

However, a model without a freezer would most likely not be preferred,

even with exceptionally low energy use. Essentially, this is saying

that the energy use data should be handled in a realistic fashion within

a decision context.

The attempt to measure evaluative judgments centers on a dimension

that has not been represented in public policy research to date. Impact

has not been operationalized at levels of impact and, consequently, the

impact of previous standardized disclosures on the product evaluations

has typically not been tested.

The question of expectations addresses a more fundamental issue

with regard to the disclosure of energy use. Do consumers, given the

present state of knowledge, have accurate expectations of energy use

with regard to appliances? A crucial assumption behind the disclosure

is that aspects of the consumers' purchases will change when this

information becomes available; thus, there is a need for the informa-

tion. If consumers, however, accurately predict either the relative

or absolute energy consumption of alternative model/feature combina-

tions without the information, then the need for energy disclosures

becomes less apparent. This hypothesis states that consumers will be

unable to state accurately the energy consumption of a given model

without the disclosure.

This consideration becomes more important the closer the models

are in terms of features. The more substitutable the models, the more

likely the difference in energy consumption will be useful in discrim-

inating models. For example, the consumer would accurately be able to

rank two very different models (i.e. a small refrigerator with manual

defrost versus a large-size side-by-side with all the accessories).

The crucial test is that as the models become more alike does the

discrimination break down for those without the energy use disclosure?

This is more characteristic of a shopping experience. The consumer is

confronted with a set of objects differing in features, but generally

satisfying a set of needs. The decision is to then select the model

which is best from among the alternatives. It should be stated that

this hypothesis does not reflect different degrees of knowledge or

education with regard to energy consumption, but is in fact related

to the present consumer information environment as a potentially

relevant consideration in related uniform information provision areas

(i.e. unit pricing).

The second hypothesis deals with the impact of comparative

information. The shift toward the disclosure of uniform product

information is well documented. Recent actions with nutritional

information, octane rating, miles per gallon, etc. have provided uni-

form product performance information. But, these disclosures are not

really comparative. The performance information alone does not give

an indication of the product's relative performance. This does not

reflect one of the central goals of the disclosure. As a task force

report submitted to the Federal Trade Commission states, the disclosure

should be provided uniformly on all brands of a given product and in

such manner as to be readily understood and capable of being easily

compared by the average consumer (Ferguson, et al., 1972, p. 42). In

light of this goal (i.e. comparison judgments) the recent energy con-

sumption requirements specify a range of values of energy consumption

for "similar" products to accompany the performance information. The

second hypothesis tests the effect of this information.

Hypothesis Two: Testing the Provision of the Range
The disclosure of relative range of energy consumption
values for similar products will provide comparative
data which will result in significantly different
judgments with regard to the feasibility of additional
shopping, the likelihood of finding a better value,
and encouraging purchase. Specifically, for Model 2
(a model which is comparatively a high energy user),
subjects receiving the range of energy consumption

values will state that it is more likely that a better
value could be found, that it would be worth the
effort to engage in additional shopping, and would
discourage with regard to purchase when compared to
subjects who did not receive the range of values.

In developing this hypothesis it is necessary to conceptualize

the impact of comparative information. The consumer is attempting

to allocate income to the best combination of price and quality

within certain brands or models. The result of the presentation of

performance standards is to facilitate this judgment. Consequently,

if an object is comparatively poor on the performance dimension

associated with the disclosure, given that the dimension has meaning

and is important to him, the judgments with regard to the price/

quality relationship should be altered (lowered). Thus, energy use

is implicitly assumed to be directly related to quality. Theoretically,

this would result in the consumer looking to other similar objects from

which to make a choice. Thus, the comparative information should

impact on the judgments of product value, shopping behavior, and

purchase decisions of the consumer.

It should be stated that the disclosure of comparative infor-

mation can be accomplished a number of different ways, and the manner

of disclosure could affect the impact. Consequently, the operation-

alization of this dimension is a crucial consideration with regard to

this hypothesis.

The third hypothesis reflects the need for accurate perceptions

of the energy consumption of alternative models or products when con-

fronted with the information. The hypothesis is:

Hypothesis Three: Evaluation While Disclosure is Present
as an Aid to Processing
The disclosure of energy consumption data will result
in accurate perceptions of a product's energy use
when subjects are not required to rely on memory
for this judgment. Subjects will perceive the
products with high energy use as less good than
those products with low energy use.

This question addresses the disclosures from the ability of consumers

to accurately perceive a product along the dimension of energy con-

sumption. What the consumer must be able to accomplish is an accurate

evaluation of the product's energy use. This addresses a more central

impact, which in truth must be present before the more interesting

behavioral measures can be found. The dependent measures in this

hypothesis provide important diagnostic powers with regard to the

eventual impact of the energy consumption in choice situations. This

test differs from the previous two hypotheses in that the crucial com-

parison is not between treatments but between products. The consider-

ation is for the consumer to be "able" to correctly perceive the

product's energy consumption. If a model uses a large amount of energy,

the subjects' evaluative judgments should reflect this (i.e. be evalu-

ated as poor with regard to energy).

The fourth hypothesis tests the impact of energy consumption

with regard to choice behavior.

Hypothesis Four: Demand for Product Features
The presentation of energy-use data will have a
significant impact on the demand for energy-related
features. The exact impact is that the disclosures
will result in more energy conscious decisions with
regard to product/feature decisions. This will be
reflected in smaller size choices, less demand for

frost-free defrost, higher demand for poured
insulation and the power mizer, and less demand
for icemakers.

This hypothesis addresses the final level of impact, the impact the

disclosure of energy consumption has on behavior. Day (1976) regards

this level of impact to be a crucial test of a public policy informa-

tion program's success. However, the choice criterion has typically

been looked upon as object/product/brand choice. By forcing this

type of choice upon subjects, the researcher places an unreasonable

burden on the disclosures. The choice of models or products relate to

broad needs and motives regarding a purchase, and may not reflect the

more subtle impacts of the disclosures.

The behavioral impact has been directed to choice of features

each subject chooses to include on a personal model of refrigerator-

freezer. This allows for the analysis of the impact with regard to

energy-related as well as functional and convenience features. The

basic assumption is that the disclosure of energy consumption will

provide the consumer with information not only about what models use

energy, but what features associated with the model seem to account

for the energy consumption. It is possible that this disclosure is

not explicit (i.e. frost-free uses X amount of energy) so that the

consumption of a particular feature must be deduced. However, this

is crucial. For the disclosure of energy consumption to have a posi-

tive effect, the demand for energy-using features must be affected.

The subjects who receive the energy consumption data will be less likely

to choose energy-related features. This relationship must, however,

consider the broad energy use benefit decision of a feature. Also,

this method of gathering the choice dimension can also reflect a

total energy choice of subjects. By summing the feature choices and

the related energy use of these features, a respondent's model can

be viewed from the total energy consumption of the model that the

subjects choose.

The next hypothesis reflects the effectiveness of alternative

disclosures with regard to the levels of impact.

Hypothesis Five: Importance of Disclosure Format
The format of the disclosure of energy-use data
will have a significant impact on the success of
the program to disclose energy consumption. The
disclosure formats will be evaluated with regard
to preferences, evaluations, and product choice.
Specifically, the disclosure of energy consumption
in dollars will have a greater impact than the
disclosure in kilowatt hours. The disclosure of
energy consumption in yearly units will have a
greater impact than the disclosure in monthly
units. Also, the format disclosing energy use
data by feature will provide additional benefits
over a condition not providing feature-energy

The first four hypotheses deal with the issue of the success of the

program to disclose energy consumption. Thus, the crucial comparison

becomes, does a group with energy consumption data perform differently

than a group without energy consumption data available? The fifth

hypothesis addresses the issue of the impact of alternate disclosure

formats. There exists a number of different dimensions along which

the disclosures can be varied. In light of the prescriptions of the

energy act, the nature of the disclosures become crucial. Consumers

will be faced with the information on labels attached to products.

Thus, the consumer must be able to use the information for evaluative

and comparative purposes.

The levels of impact need to reflect the success of various

disclosure alternatives. The consumer must be able to use the infor-

mation for both evaluative and comparative purposes. The alternative

disclosures will be evaluated in light of the different formats'

effects on preferences, evaluation, and choice.

The impact of alternative formats has been suggested in the

literature, but rarely tested well particularly in light of policy

needs. This type of manipulation, to be meaningful, requires that the

policy become the focus of the research. The format decision has been

suggested by Day (1976) to be a crucial fundamental research decision.

This research will investigate four dimensions of disclosure formats:

the unit of measurement; the time period of the disclosure; the presence

of feature energy use; and, the disclosure of comparative information.

The unit of measurement decision is particularly interesting with

regard to energy disclosures. The disclosure can be in dollars of

energy use or in terms of the actual energy use units (i.e. KWH, gaso-

line, etc.). The disclosure in terms of actual energy use units

provides flexibility with regard to the consumer's ability to predict

the actual "energy cost" which would result from a product. With the

recent rate increase and the variations by season and geographical

area, this particular consideration can be important. However, the

dollars of energy use represent a familiar non-ambiguous disclosure

which has some appeal. The familiarity of the respondents with the

disclosure has been suggested as an important variable with regard to

information impact (Day 1976).

The time period over which to report energy consumption has been

suggested as a relevant consideration in related uniform disclosure

programs. Monroe and LaPlaca (1972) have postulated that the impact

of a disclosure will be greater for disclosures in larger units, due

to the fact that the larger units will magnify the product differences

along the relevant dimension. However, the disclosures of monthly

values may relate to the consumer budget and bill-paying behavior so

the magnitude of the disclosures becomes a relevant consideration.

An additional dimension under consideration would be the degree

of disclosure. The impact of the program may ultimately rest with the

consumer's ability to associate product energy use with the features

that most account for the energy consumption. With the complexity of

major home appliance this energy use/feature judgment may be difficult

to make. This research will incorporate a disclosure format which

explicitly discloses the feature contribution to energy use. This

additional (by feature) disclosure is expected to provide additional

benefits to the consumer.

Thus, the effects of the alternative disclosures should provide

an input for future research, while being informative with regard to

the strategical disclosure decisions.

Hypothesis Six: Who Will Benefit?
The impact of the disclosure of energy consumption
will be dependent upon the characteristics of the
individuals receiving the energy consumption data.
The disclosure of energy consumption will have a
greater effect on high-education consumers when
compared to low-education consumers.

The public policy efforts relating to information disclosures have been

faced with a success story which is potentially paradoxical in nature.

The goal of the programs has been to improve the purchasing habits of

the consuming public. In the past certain efforts have been directed

to problem areas such as nutrition and credit information. In both of

these cases the information will do the most good for those members of

the consuming public who are less informed. This informed character-

istic has been linked to the dimensions of education and racial group.

Day and Brandt (1976) found that the truth-in-lending information has

the least success for low-education consumers and minority groups.

This result suggests the success of the program may be tied to the

groups which need relatively less help in the area of credit information

practices. Although the nutritional area has not been researched and

reported to the degree which would be necessary to make any conclusive

statements, the potential for the same sort of result as in the truth-

in-lending issue is possible. There are, according to the report which

stimulated the nutritional labeling TRR, wide-spread deficiencies in

the nutritional character of the population. However, certain segments

suffer to a greater degree than others. The poor and low-educated con-

sumers are the most deficient, but will probably gain the least from

nutritional disclosures. This is not an empirical fact, but in the

CRI study on formats, the different responses to nutritional labels

along the lines of education and race was evident (see Stokes 1972).

The issue of the characteristics of the consuming public is also

relevant to energy consumption labeling. The energy consumption data

will be complex performance data. Given this, one would hypothesize

that the disclosure would be more effective for high-education con-

sumers when compared to low-education consumers. It is possible that

this could insure the success of the program. If the high-education/

high-income consumer uses the greatest amount of energy for home

appliances, the potential for energy savings would be greater for

this segment of the population. This sort of broad statement cannot

be made without additional knowledge of the consumption patterns of

various classes of consumers. This study will attempt to gain insight

into the relationship between the characteristic of education and the

disclosure of energy consumption data.

Two major limitations should be kept in mind with regard to this

hypothesis. First, the study will only use one product class as exper-

imental stimuli. Consequently, the broad generalizability to other

major home appliance disclosure will be limited. Second, the sample

of respondents is not reflective of the characteristics of the general

population. The sample is a non-random sample of residents of Gaines-

ville, Florida. The sample reflects a desire to achieve variance along

the lines of education and income. But, the ability of the sample to

mirror the ultimate impact is limited.

The hypotheses were tested in an experimental setting. The

research is designed to gain insight into the impact of alternative

disclosure formats and the impact of energy labeling in general. To

accomplish this goal a set of six experimental conditions were employed.

Each of these manipulations is discussed in the following section.

Experimental Manipulations

The focus of this research is the proposed disclosure of energy

consumption of major household appliances. At present this disclosure

is mandated by the EPCA, and is prescribed to be the annual energy use

in dollars plus a range of annual dollars for similar products. While

the act will serve as a foundation for the tests to be run within this

study, a key assumption is made with regard to the disclosure format.

The particular manner or form of the disclosure must stand the test

of consumer impact before the specific requirements or formats can be

determined. Thus, while the disclosure is recommended within the act,

the disclosure to be ultimately required is subject to empirical


Generally speaking, the manipulations to be discussed will

address two fundamental questions:

1. Does this disclosure of energy use data on a label
attached to major home appliances have an impact
on behaviors related to consumer purchase decisions?

2. Does the reporting format specified by the energy
act perform best in light of other format alterna-
tives with regard to "aiding" consumer purchase

These questions have essentially been addressed in the preceding hypo-

theses. The first question is addressed to the classical question of

impact. At what level of impact do the uniform disclosures of product

performance appear to affect consumer behaviors toward a specific

product? The second question is a less common approach with regard

to investigation of the impact of specified disclosures. This questions

the format of the disclosure and adopts the tactical orientation of

which format is most successful.

The question of format alternatives leads to the determination

of the independent variables, which could or should characterize a

particular disclosure.

Independent Variables

The independent variables selected for study are the unit of

measurement, the time period over which the consumption will be com-

puted, the presence of comparative information, and the degree of


Unit of Measurement

An issue related to the public policy decision that has con-

tinually been suggested as needing research activity is the question

of the unit of measurement by which information is disclosed. Both

Day (1976) and Wilkie (1976) assert that little systematic research

has occurred with regard to this aspect of the format decision, and

that this is a necessary area of inquiry. Research conducted by

CRI/FDA with regard to nutritional labeling tested alternative units

of measurement (% RDA, graphics, or verbal) in an attempt to determine

the "best" format for nutritional content disclosures (Stokes, 1972).

The results of this study provided meaningful inputs into the policy

decision. A recent study by Chestnut (1976) attempted to test the

alternative units in an energy consumption setting, but this issue

was clearly not the focus of the research effort; thus, conclusions

are limited.

This is a very complex area, and the particular fixed levels

tested under this independent variable represent a crucial decision.

For most consumers, energy consumption is a cost that must be paid

in more or less regular intervals. Thus, the most familiar method by

which consumers address energy consumption is with regard to dollars,

which must be exchanged for energy use. The fact that this is the

more common method leads one to assume that the dollars disclosure

would be most effective. In fact, the degree of familiarity has been

linked to positive impacts of information programs with the dimension

disclosed (Day and Brandt, 1974). However, there exists some unique

aspects of energy disclosure problems.

First, the absolute level of familiarity, with regard to energy,

may be relatively low. While the costs of energy are reflected in

bills each month, the amount of the bill gives little information about

energy consumption sources or the composition of the bill. Thus,

familiarity with dollars as a means of disclosure may not be as

strongly related to disclosure effectiveness. It is also highly

likely that the consumers do not know the rate (dollars per unit of

use) by which that bill is computed.

Second, the disclosures as mandated in the EPCA will be calcu-

lated on a standard rate per energy use (e.g., 4 per KWH). From this

calculation the consumer would be able to make relative judgments as

to energy consumption, but the absolute judgments (i.e. how much will

this cost me?) would deviate as the rates for energy use deviate from

the calculated rate. Given the particular situation which confronts

the consumer, this can be a potentially deceptive situation. Energy

rates vary greatly by season and by geographic area; thus, the presen-

tation of a standard dollar performance figure does not reflect the

energy usage correctly. By presenting the energy consumption in KWH,

the consumer with knowledge of local rates can correctly anticipate

the energy consumption that would be associated with a given product

in addition to being able to make relative judgments.

Thus, the unit of measurement decision will be tested using two

levels: dollars and KWH. The dollars disclosure is the current

mandate, while the KWH disclosure provides flexibility with regard to

energy consumption perceptions. Each of these disclosures will be

tested at all levels of impact so as to provide a more complete profile

of the comparison. While these alternatives do not exhaust the various

levels of this independent variable, the tests do reflect realistic and

reasonable policy alternatives with regard to energy provision.

Time Period Used for Computation

While the relative energy consumption will remain the same, the

time period employed to compute the energy consumption figures affects

the absolute magnitude of consumption and subsequent product differ-

ences. The issue of magnitude of disclosure has received some mention

with regard to public policy activities. In a position paper on unit

pricing, Monroe and LaPlaca (1972) state that particular weight chosen

to compute unit prices (per ounce, per pound, etc.) may have contributed

to the lack of positive findings in unit pricing studies. The crux of

their argument is that as the magnitude of differences between products

and the absolute level of the information increased, the effect of the

disclosure would increase. Thus, while the relative positions remain

unchanged, the greater the magnitude the greater the impact. With

regard to unit pricing, no research has addressed this problem.

When considering the energy provision program, the issue of

magnitude of disclosure becomes relevant. Would disclosure of the

same relative information expressed in different time levels (i.e.

yearly, monthly, over the life of the product) result in different

responses? The disclosure is specified to be computed in yearly

terms with the assumption that this represents the optional time period

over which to compute and disclose the energy consumption. However,

the consumer is faced with energy consumption in monthly increments,

and since the impact of the disclosure must be considered within the

budget of the consumer, it may be that this type of disclosure would

be best. In contrast, the longer the time period used, the greater

the magnitude of the disclosure. If the contentions of Monroe and

LaPlaca are correct, the greater magnitude would result in the greatest

impact. As an extension of this line of reasoning, the MIT study

(MIT, 1973) reports energy consumption over the expected life of the

product as a component of the product's "life-cycle costs." It

might be expected that the larger magnitude of this disclosure would

result in greater impact than either of the two previously discussed

levels (i.e. yearly or monthly).

This study will investigate the differential impact of the yearly

and monthly disclosures. Other on-going research has undertaken the

life-cycle cost framework (see Hutton, 1977). The primary question

is whether the time period over which the consumption is computed will

have differential impact.

Comparative Information

The disclosure of the range of comparative infornition provides

a new dimension for policy disclosures: relative energy consumption.

This form of disclosure has some real potential advantages for the

consumer. The consumer can make accurate good/bad judgments with regard

to product performance without the necessary shopping behavior attendant

to setting performance standards. The comparative judgments of the

consumer should be facilitated by the disclosure and the extent of the

shopping behavior necessary to make this judgment should be lessened.

Thus, the disclosure may be justified in terms of a broader "right to

know" rationale. However, the question of interest is whether or not

the disclosure facilitates the judgments. This leads to some rather

complex issues.

First,, it is very difficult to establish sound criteria for a

comparative disclosure. In the voluntary labeling program (McGuire,

et al., 1975) the range of comparative values for air conditioner

energy efficiency ratios was done by size. However, using size as the

criteria for refrigerators gives little indication of the causes of

the energy consumption or the particular feature trade-offs available.

Conversely, if the range were set up by similar features, the infer-

ences concerning size would not be reflected accurately, but the con-

sumer would have a better grasp of the effects of various feature

combinations. The crucial consideration is that the criteria for the

range reflect the dimension which results in the variance in energy

use found across products.

Second, given a correct (valid) range, the consumer must make

correct judgments with regard to the value within the range. Thus,

a model may be a high user of energy, but relatively good considering

the particular feature combination. Thus, when making statements about

the range there must be confidence that the measurements have tapped

the judgments from the comparative disclosure dimension and not some

other aspect of the disclosure.

The EPCA provides little guidance with regard to this particular

dimension. The prescription is for a "range of annual dollars for

similar products." Products can be similar along a number of dimen-

sions. The manipulation chosen for this study is that the range of

comparative values be for models with similar features and sizes. It

is felt that this reflects the source of variance of the energy con-

sumption between products.

However, the range of values is difficult to investigate and the

insights gained from this manipulation cannot be conclusive. This is

particularly true given the critical nature of the issues surrounding

the exact operationalization of the range. Thus, the attempt is to

make statements concerning the impact of comparative information given

the constraints of this particular operationalization.

The Degree of Disclosure

The final set of manipulations center on the extensiveness of

the disclosure. A major orientation of the EPCA centers around the

recognition of energy sources and reduction of the demand for those

sources. With regard to major household appliances, the source of much

of the variance in energy consumption is the features available for

inclusion on a given product.

Under the present EPCA requirement, the consumer is required to

make rather complicated trade-offs and mental calculations in order to

accurately identify the sources of energy consumption. In many cases

the consumer is evaluating few models, and thus, this sort of judgment

(feature energy use) is even more difficult. The question addressed

is whether more explicit disclosures will result in more accurate

judgments and/or ultimately less demand for energy-related features.

This type of disclosure would seem to be closely tied to the spirit of

the act.

It seems that the disclosure of energy use by feature in addition

to the total model energy use, would serve a clarifying function for

the consumer, and reduce the task which would be necessary to mentally

make such judgments. This disclosure would be standard, and each model

containing such a feature would present the same incremental energy-use

information. A disadvantage is that the feature information increases

the information available to the consumer. While detrimental effects

of information load are not expected (see Jacoby, et al., 1975 a,b),

the increased information may result in less recall overall.

Treatment Conditions

Each of the levels of each independent variable represent feasible

dimensions for a study within a given disclosure requirement, and could

reasonably be expected to affect the impact of the disclosure. The

presence of such a large number of independent variables represents a

formidable experimental design. If each of the four independent vari-

ables were to be investigated at two levels (i.e. dollars and KWH within

the unit of measurement) a completely crossed factorial design would

require 16 cells. Given both the time and monetary constraints for the

present study the number of cells had to be reduced. However, each of

the dimensions were deemed to be an important format dimension and

worthy of study. The procedure for choosing treatments was to consider

each combination of the independent variables along two lines. First,

each disclosure format represented a combination of the four independent

variables. The most urgent consideration was whether the format of the

disclosure represented a viable policy alternative (i.e., one which,

if the empirical results justified, would have a probability of being

implemented). So the first consideration was the usefulness of a dis-

closure format to the actual policy decision. The second question

centers on the degree of insight which could be gained from the dis-

closure format with regard to the independent variable levels present

in a given treatment. Because a factorial design was not used in a

completely randomized design, the results will not provide a test of

the main effects of the independent variable. But, the format alter-

natives can be expected to provide comparisons with regard to the other

alternatives under study, and with a careful comparison procedure, one

can gain insight into the relative effectiveness of a certain inde-

pendent variable. The usefulness of these results can be seen from

both a strategic and hypothesis-generation point of view. A disclo-

sure's effectiveness can be given empirical validation by the results

of the experimental setting. Also, if the effects after the comparison

procedures support the effectiveness of a format alternative, additional

experiments can be attempted to clarify and substantiate the impact of

certain independent variable considerations.

The following discussion presents the treatment conditions

chosen for study. The treatment conditions in order of discussion

are as follow:

Condition 1 a control condition which has no energy

" Condition 2 -

" Condition 3 -

" Condition 4 -

" Condition 5 -

Condition 6 -

consumption data available;
energy consumption data in annual dollars
with a range of comparative annual dollars
for similar products;
energy consumption data in annual dollars
with a range of comparative annual dollars
for similar products plus the annual energy
consumption of each product feature;
energy consumption data in annual kilowatt
hours with a range of comparative annual
kilowatt hours for similar products;
energy consumption data in monthly dollars
with a range of comparative monthly dollars
for similar products; and, 0
energy consumption data in annual dollars.

Control No Energy Data

The energy labeling program is in response to the overriding

problem of energy shortages. In order to accomplish a redirection of

energy-related activities, the policy makes an implicit assumption that

the current information environment is insufficient. And, further, the

program assumes that the disclosure of energy consumption data will

affect purchase-related behaviors. To attest to that impact, the

current information environment must in some way be related to the

changed information environment (one with energy consumption data

available). Consequently, the comparison of the control condition

with the disclosure alternatives is necessary for this measure of


It should be emphasized that the control condition represents the

current information and consuming environment. Not only is energy

consumption information not available, but there is no attempt to

sensitize any of the groups to the energy dimension via some surrogate

consumer education efforts. Thus, the comparison of control vs. treat-

ment represents the impact of energy on situations where the energy

consumption data is not present.

The comparison with the control condition will be made across

all levels of impact. Special care has been taken to insure that the

only major difference between the treatment and control conditions is

the presence of the energy consumption data. The following treatment

conditions are conditions which present the same energy consumption

information in different formats. Thus, in each condition the energy

consumption of the alternatives is in the same relative values for

the models. Some condition may transform the energy consumption into

different units or report consumption over different time periods, but

in each case the consumption of a particular object (model) remains

the same. This is a crucial aspect of the disclosure, in that for

some analysis the conditions must be combined to gain a feel for the

effects of certain independent measures. By classifying the condi-

tions as the same information transformed into a different format,

it is possible to make that particular combination of different

treatments have some conceptual meaning. What is being combined is

the same information that is communicated in different format. Thus,

when all treatment conditions are compared to the control, the treat-

ment conditions do not represent conceptually different objects and,

thus, the comparison has meaning in an information-no information sense.

Energy Consumption in Annual Dollars Plus a Comparative Range

This treatment condition represents the requirements of the EPCA

as the disclosures are presently stated. The inclusion of this condi-

tion represents a crucial test for the energy policy. It is an essen-

tial component of the assessment of the impact of the energy consump-

tion disclosures. Also, this treatment condition is essentially the

comparison group for the other disclosures, addressing the question of

which disclosure format is the most effective. Thus, the contrast of

this comparison condition with either disclosure formats will allow

for the judgment as to which format best aids the consumer in purchase-

related activities.

This condition provides a combination of the levels of three

of the independent variables. The EPCA specifies the disclosure to

be in dollar units, with consumption aggregated over a year, and pro-

vide a range of comparative values. The comparison with conditions

using different levels of the independent variables will be conducted.
Energy Consumption in Annual Dollars Plus a Comparative Range Plus

By Feature

This manipulation represents the addition of more explicit energy

information disclosure than that proposed by the EPCA. What is assumed

is that the disclosure of energy use in terms of simply annual overall

consumption of energy requires that the consumer assume the burden of

determining the incremental contribution of a feature toward the model's

or product's overall energy consumption. Thus, the feature disclosure

provides a more explicit disclosure not only of the total consumption

of energy, but also of the source of that consumption. The format

also provides the standard disclosure as specified by the EPCA, thus

facilitating a comparison as to the effect of this additional disclosure.

Energy Consumption in Annual KWH Plus a Comparative Range

The next treatment presents the energy consumption again on an

annual basis with comparative information, but uses a different unit

of measurement. Given the added flexibility of this unit of measure-

ment, it would seem that the format would offer some advantages. With

the fluctuating cost of energy between seasons and geographical areas,

the flexibility offers the consumer greater potential accuracy with

regard to absolute consumption. The comparison of this alternative to

the EPCA disclosure will provide insight into these levels of the unit

of measurement variable.

Energy Consumption in Monthly Dollars Plus a Comparative Range

The presentation of energy consumption aggregated by month repre-

sent a reasonable policy alternative to that proposed by the EPCA. The

consumer is familiar with energy consumption disclosed in this manner

due to the common billing practices. Thus, even though energy consump-

tion is typically a continuous process due to billing and budgetary

considerations, the consumer may be most familiar with this format of

disclosure. The question, however, stems not only from the consumer's

choice for disclosure, but to the empirical impact of format alterna-

tives. This remains an untested question.

Energy Consumption in Yearly Dollars

The typical disclosure of product performance information presents

only the performance of the specific product and heretofore has given

no explicit or implicit indication of the relative evaluation of that

performance. As performance standards become more and more complex, the

task of developing these standards by shopping, experience, etc. is

an unreasonable request of the consumer. Thus, the presence of com-

parative data should relieve this burden. The condition without com-

parative data facilitates this comparison.

Experimental Vehicle

Product Class

The product class chosen for the experimental stimuli is a

refrigerator-freezer. The following criteria were used in the selec-

tion of this particular product:

The relevance of the product class to the energy con-
sumption information program. The program deals with
major home appliances of which refrigerator-freezers
represent a large volume of major appliances purchase
in both per capita and dollar volume terms.

Refrigerator-freezers represent the single greatest
source of household energy consumption (see MIT report
in previous chapter).

The variability in energy consumption of refrigerator-
freezers is related to both feature and model decisions.

There exists a potential for energy savings with respect
to product technology which represents a potential
for impact of energy consumption disclosures.

The product occupies a prominent place in the American
household; consequently, the population can be expected
to possess a certain amount of product knowledge.
The product is not characterized by unique consumer
usage patterns.

With respect to the product class of refrigerators, the great

diversity of model/feature combinations allows flexibility with regard

to the specific selection of experimental stimuli. Also, the choice of

a refrigerator is characterized by model decisions. Even if a consumer

shops within only one brand, the decision becomes a model choice. Con-

sequently, this frees the research from a brand choice orientation to

a model evaluation framework.

Attribute Choice

The following attributes were available in the model evaluation


1. style,
2. price,
3. frost-free,
4. size,
5. icemaker,
6. insulation; and,
7. energy use.

The attributes were all attributes which were mentioned in the focus

group interviews conducted earlier. For the most part, each attribute

is related to the model energy consumption. This also coincides with

the type of information which is readily available on current products,

should a consumer choose to evaluate a particular model. Further, each

of the features relate to energy consumption in certain ways.

The combination of the attributes produced the particular models

which the respondents evaluated. Each consumer was faced with four

models of refrigerator-freezers to evaluate. These models were:

MODEL 1 Top Freezer Style with Manual Defrost; MODEL 2 Top Freezer

Style with Frost-Free Defrost; MODEL 3 Side-by-Side Style with

Frost-Free Defrost and Icemaker; and, MODEL 4 Top Freezer Style with

Frost-Free Defrost and an Icemaker. Within each model each attribute

took on different values. Table 7 presents the values chosen for each

of the model attributes. Thus, the models differed along the lines

of style, size, price, defrost insulation, and icemaker. The values

of the dimensions were chosen to represent model types which would be

Table 7
Attribute Values of Experimental Models

Model 1
top freezer style
with manual defrost

17 cubic feet
13 fresh food
4 freezer



Poured insulation.
Adds $20 to the
purchase price and
subtracts $7.50 each
year from energy use.

Model 2
top freezer style
with frost-free
17 cubic feet
13 fresh food
4 freezer
Frost-free adds
$100 to the purchase
price (and adds $25
each year to the
energy use).
Cut fiber glass--

Model 3
side-by-side style
with frost-free and
an icemaker
19 cubic feet
13 fresh food
6 freezer
Frost-free adds
$100 to the purchase
price (and adds $25
each year to the
energy use).
Cut fiber glass--

Icemaker adds $75
to the purchase price
and $12 each year to
the energy use.

Model 4
top freezer style
with frost-free and
an icemaker
20 cubic feet
15 fresh food
5 freezer
Frost-free adds
$100 to the purchase
price (and adds $25
each year to the energy
Cut fiber glass--

Icemaker adds $75
to the purchase price
and $12 each year to
the energy use.




found in appliance stores. This was accomplished by several visits to

local retain outlets of such brand names as Philco, G.E., Coldspot,

Whirlpool, and Frigidaire. The author consulted models on the floor

and dealer material to determine the price increase of additional sizes

and features.

The models represent realistic model alternatives, but they, in

addition, provide trade-offs on the features which account for refrig-

erator-freezer energy consumption. The effect of each feature on energy

consumption was accomplished by consulting the 1976 Directory of

Certified Refrigerator-Freezers. This source allowed for the computing

of average model energy consumption. It was determined that on the

average, frost-free increases energy consumption $25.00 each year;

icemaker increases energy consumption $12.00 each year; side-by-side

$10.00 each year; and, poured insulation decreases energy consumption

$7.50 each year. From these average figures The Directory was con-

sulted to determine the range of values for similar models. Table 8

presents the values on the energy use attribute. The values were

chosen to reflect three models (MODELS 1, 3, 4) which were in the lower-

energy use portion of the comparative range of energy consumption and

one model (MODEL 2), which was in the high-energy use portion of the

comparative range of energy consumption values. The values for each

model's energy represent models which possessed the features of the

experimental stimuli in various brands. Thus, while the models were

fictitious, there exists a model in the real world which has the same

features. The actual displays of the model information was accomplished

Table 8
Energy Consumption by Model by Treatment

Dol lars/year
plus a compara-
tive range plus
by feature
Dol I ars/year
plus a compara-
tive range

KWH/year* plus
a comparative

Dol l ars/month
plus a compara-
tive range


Model 1
$32.40 each year.
Similar models use
$32.40 to $50.80
each year.
$32.40 each year.
Similar models use
$32.40 to $50.80
each year.

648 KWH each year.
Similar models use
648 to 1016 KWH
each year.

$2.20 each month.
Similar models use
$2.20 to $4.23
each month.

$32.40 each year.

Model 2
$101.00 each year.
Similar models use
$60.00 to $108.00
each year.
$101.00 each year.
Similar models use
$60.00 to $108.00
each year.

2020 KWH each year.
Similar models use
1200 to 2160 KWH
each year.

$8.42 each month.
Similar models use
$5.00 to $9.00
each month.

$101 .00 each year.

Model 3
$96.80 each year.
Similar models use
$93.70 to $168.60
each year.
$96.80 each year.
Similar models use
$93.70 to $168.60
each year.

1956 KWH each year.
Similar models use
1874 to 3372 KWH
each year.

$8.15 each month.
Similar models use
$7.81 to $9.70
each month.

$96.80 each year.

Model 4
$76.80 each year.
Similar models use
$75.00 to $135.00
each year.
$76.80 each year.
Similar models use
$75.00 to $135.00
each year.

1572 KWH each year.
Similar models use
1500 to 2178 KWH
each year.

$6.55 each month.
Similar models use
$6.25 to $t1.34
each month.

$76.80 each year.

* computed by dividing the dollars per
Florida rate.

year by $.05 per KWH to obtain KWH/year. $.05 is the Gainesville,

by arraying the attributes and values for each model around a drawing

of a refrigerator. Figures 1 and 2 illustrate the drawing for the

side-by-side and top freezer styles. The actual stimulus material is

available in Appendix A through for each model within each treatment

condition. The subjects worked through a series of tasks using the

model displays a stimuli. Thus, the judgments for the first section

of the tasks were limited to the four models and the attendent infor-

mation. However, in the last task the respondents were asked to make

certain choices in the form of a building task.

The building task operationalized product choice in a unique


Since the subjects were asked to design the refrigerator to meet

their personal needs, the task had to be arranged so that the flow of

the task made sense. This was accomplished by having the respondents

take the features in the order that the feature would be placed or

built on a model. The features and alternatives within each feature

were: total size (16, 17, 18, 19, 20, 20+ cubic feet); style (side-by-

side, top freezer, bottom freezer); color (white, colors); insulation

(cut fiber glass, poured insulation); defrost (manual, partial auto-

matic, frost-free); icemaker (yes, no); shelves (stationary, wire

shelves, adjustable wire shelves, stationary glass shelves); power

miser (yes, no); and cold-water dispenser (yes, no).

The respondents were confronted with each alternative within a

feature and the price (purchase price) of that feature. The prices

corresponded to the prices which were given to the models evaluated


earlier, so that if a consumer designed a model with exactly the same

features as MODEL 1, the price would be $359.00. The task and attri-

bute values for the building decisions are found on pages 13-15 of

Appendix A.

This chapter has presented the hypotheses to be tested and

experimental manipulations used to test these hypotheses. The following

chapter will discuss the Research Methodology.

Figure 1

Model Display for MODEL 1, 2, 4

Model Description







*available only in the disclosure conditions

Figure 2

Model Display for MODEL 3

Model Description







*available only in the disclosure conditions




The concern with assessing the effect of information oriented

disclosures has developed into a crucial research issue. The need

is to integrate the goals and direction of a given policy activity

with the standards of research conduct in such a manner that the needs

of both are satisfied. This research has taken the posture of focusing

on the particular policy activity as the foundation of the research

endeavor. At the onset, the research goals centered upon an attempt

to assess the impact of the disclosure of energy consumption data on

the consumer. As a subset of this major goal, the research is attempt-

ing to determine where and why the impact would take a particular

shape. The following is a more explicit statement of the research

goals of this project.

Objective One

While attempting to identify alternatives to present energy

sources the government has determined that present demand for energy

must be affected (i.e. lowered). The provision of energy consumption

data on labels is directed toward this goal. While the program is in

early stages of development, there is a need to attempt to gauge the

impact of the disclosure on consumer purchase/product evaluation

behavior. The first objective is to gain insight into the impact of

the disclosure of energy consumption data.

Objective Two

The second goal of this research is to determine the effect of

the disclosure by measuring specific levels of impact. As discussed

earlier, the orientation of much policy oriented research is to deter-

mine the choice differences due to disclosures. Choice remains the

ultimate impact criterion, but it must be noted that often when this

becomes the crucial dependent measure the emphasis becomes decision

(especially brand) rather than evaluation oriented. This orientation

provides a rather incomplete picture of impact. This objective attempts

to measure the levels of information use in order to more realistically

assess the variation in response due to the disclosure.

Objective Three

One of the often suggested but infrequently tested questions

concerning standardized disclosures revolves around the issue of dis-

closure format. Does the form of the disclosure provide for, or pre-

vent the positive impacts of a given program. A large number of inde-

pendent variables come to mind, the unit of measurements, the time

frame of a disclosure, the presence of comparative information, etc.

The third objective of this study is to determine if there appears to

be significant differences due to certain dimensions which alter the

form of the disclosure.

Objective Four

The final objective is more global. This study has attempted to

represent a framework for performing policy-oriented research. The

focus of the study and subsequent design and methodology reflect the

unique considerations of a policy and, consequently, the results should

reflect partial answers to the policy oriented questions.

Experimental Design

The design contains five treatments varying certain important

dimensions of the disclosure of energy consumption data plus a control

group. The five treatment groups disclosed energy consumption as

dollars of energy consumed each year with a range of
energy consumption values for similar products, plus
each energy related feature had the energy consumed
each year in dollars for that feature,
dollars of energy consumed each year with a range of
energy consumption values for similar products,
kilowatt hours of energy consumed each year with a
range of energy consumption values for similar products,
dollars of energy consumed each month with a range of
energy consumption values for similar products, and
dollars of energy consumed each year.

Included in this design is a no energy consumption control group which

received all information about each of the products except energy con-


Experimental Controls

A complete set of controls was used to establish the source of

effects which were observed. Subjects were randomly assigned to one of

the experimental groups. The subject pool was such that the sample was

composed of members of existing organizations. This resulted in several

experimental sessions composed of the members of a particular organi-

zation. Each treatment was represented at every experimental session,

and the subjects were randomly assigned to the treatment. If the

location of the session was such that the subjects could, with a

measure of certainty, work alone the seating arrangement randomly

mixed the treatment conditions. If, however, the location was char-

acterized by close working conditions, the treatment groups were

separated and the random assignment occurred as the subjects entered

the room. In retrospect, the random assignment to physically separated

treatment areas is preferred due to the potential communication among

subjects, thereby controlling a potential source of interpersonal

communication. The random assignment of treatments once seating had

been accomplished produced the least amount of reaction by the experi-

mental subjects.

The conduct of the experimental sessions was uniform with regard

to the instructions given to the sessions. In all but one session, the

experimenter was the author. In the session with a different experi-

menter, explicit, written instructions were used, and identical visual

aids were employed. In all sessions, the interaction of the experi-

menter with subjects was standardized by use of the experimental

instrument. Due to variability between the tasks contained in each

session, the total time of each session varied. The time period ran

from 45 minutes to a little over one hour.

Further, each subject, regardless of the treatment group,

received the following material:

human subject release form,
a questionnaire,
an acquisition board,
a large envelope containing the models to be used
in the acquisition task, and
small envelope containing models for the evaluation
section of the experiment.

Conduct of the Experiment

The stages of the experimental tasks are presented in Figure 3.

The experiment consisted of five major stages; introduction, informa-

tion acquisition, model evaluation, building task, and personal data.

The following is a description of the conduct of each experimental



The experimenter introduced himself and briefly stated that this

research was a part of the author's dissertation. Subjects were then

asked to read the human subject release form, Appendix H. Upon reading

the release, each subject indicated her consent by signing the form.

The experimenter explained that to insure anonymity, the release forms

would be picked up immediately and not associated with their responses.

Upon completing this, the subjects were asked to look at the cover page

of the questionnaire and to read along with the experimenter. It was

stressed that there existed a wide range of questions and their responses

should reflect each subject's personal opinions. If a subject had a

question, either the experimenter or one of the individuals assisting

in the session would be glad to answer the question personally. The

Appendix A contains the completed questionnaire and instructions for