DIRECT HOUSEHOLD ENERGY CONSUMPTION, 1973-74, 1975-76: THE IMPACT‘OF FAMILY MICRODECISIONS UPON LEVELS OF CONSUMPTION Dissertation for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY JOANNE GOODMAN KEITH 1977 L [B R A R Y Michig: State UDIV'C Ly This is to certify that the thesis entitled DIRECT HOUSEHOLD ENERGY CONSUMPTION, 1973-74, 1975-76: THE IMPACT OF FAMILY MICRODECISIONS UPON LEVELS OF CONSUMPTION presented by Joanne Goodman Keith has been accepted towards fulfillment of the requirements for _Rh_.J.L__..degree in Family ECO 1°8Y 0-7639 ,, ,7 . ...-L. w... A... L... .__,__._.__ 7 7 EC: am. ”3} .." ~"I -‘ JUL'31 57F-" ‘5 “,1 . 4"! . ‘ ‘ :C'fi v ‘1‘ 2.;v4- .“__ ABSTRACT DIRECT HOUSEHOLD ENERGY CONSUMPTION, 1973-74, 1975-76: THE IMPACT OF FAMILY MICRODECISIONS UPON LEVELS OF CONSUMPTION BY Joanne Goodman Keith The development of this research was based upon the assumption that decision making is the main adaptive feature of human systems and that the family is one critical societal unit where this process occurs. The focus of the study was upon one area about which families make decisions: the consumption of direct household energy. Two major questions were evaluated. Did house- holds reduce their consumption of direct energy since the year of the Arab Oil Embargo, 1973-74? Did the conserva- tion measures (i.e., household microdecisions) reported by heads of household contribute significantly to reduced levels of consumption? Energy consumption data from utility and oil com- panies and conservation measures reported by household members were the basis for the evaluation. The household was the unit of analysis for the sample of 130 families. Joanne Goodman Keith Dependent t-tests and multiple regression analyses were the statistical procedures employed. An overall reduction of 6.3 percent in direct household energy consumption was found between the years 1973-74 and 1975-76 (t = -5.62, p = .000). The decreases occurred in fuel oil and natural gas, the major sources for space heating. Electrical consumption increased 2.2 percent. The second major objective was to evaluate the impact of household conservation measures upon levels of consumption. Insulation in the walls or ceiling, instal- lation of storm windows and lowering the thermostat setting on the hot water heater were each reported by 15 percent or less of the sample. Questions were asked concerning nine daily or seasonal household conservation behaviors. Moderate to high levels of increased adoption were reported for most of the sampled practices. The collective impact of these behaviors on levels of consumption was hypothesized as a major variable in this research. A family scale was developed to reflect the extent to which the behaviors were practiced and the number of adult heads of house- holds who reported their adOption. Through stepwise regression the behavioral and structural conservation measures that took place within households between June, 1974, and June, 1976, were Joanne Goodman Keith analyzed for their impact upon the levels of consumption in 1976. Consumption 1973-74 was used as the baseline from which change was evaluated. Alternate explanations for variation in consumption were included as variables: changes in family size, addition of appliances, and replacement of the furnace. Previous level of consumption, 1973-74, was the best predictor of consumption during 1975-76 (t = 29.12, p = .000). Three change variables met the minimum cri- teria for inclusion in the regression equation: the installation of a new furnace (t = -3.16, p = .002), increased intensity of conservation behaviors (t = -2.99, p = .003), and installation of insulation in the ceiling (t = -1.l9, p = .235).v The importance of energy- efficient technology in the reduction of household energy was demonstrated. The role of the behavior of household members was equally significant, i.e., the accumulation of many microdecisions was important in the overall reduction of consumption. DIRECT HOUSEHOLD ENERGY CONSUMPTION, 1973-74, 1975-76: THE IMPACT OF FAMILY MICRODECISIONS UPON LEVELS OF CONSUMPTION BY Joanne Goodman Keith A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Family Ecology 1977 Dedicated to Robbie and Julie and their cousins-- whose energy future is important to me. ii ACKNOWLEDGMENTS The completion of this graduate degree represents the time and commitment of many people other than the graduate. The acknowledgments written here in no way adequately express the appreciation for the differing contributions of each person. I received academic support throughout the gradu- ate training from many faculty members and fellow stu- dents. My committee members were especially helpful in developing my educational program and research problem. The members included Dr. Vera Borosage, committee chair- man, Department of Family and Child Sciences; run Beatrice Paolucci, research chairman, Department of Family Ecology; Dr. W. D. Collings, Department of Physiology; Dr. Jane Oyer, Department of Family and Child Sciences; and Dr. Bonnie Morrison, interdisciplinary research team representative, Department of Human Environment and Design. The data for the research were from the Family Energy Project, an interdisciplinary study supported by the Michigan Agricultural Experiment Station. I learned many research skills while working with the research team of faculty and students. Dr. Bonnie Morrison, a iii member of my academic committee, provided very necessary and generous understanding, support, and advice through- out the research process. Dr. Peter Gladhart and Dr. James Zuiches gave invaluable assistance. Special recog- nition is due the fellow students who over the years worked willingly as we learned the research process together. Connie Whitaker, Michael Lawrence, Mark Roosa, Jeanne Alesi, and Karen Cottledge helped especially dur- ing the later stages of data preparation, analysis, and reporting. An essential contribution to the completion of this degree was the support from my family and neighbors, not only at the time of the dissertation writing, but also throughout the entire graduate program. My husband challenged me to start the degree; he and the children, Robbie and Julie, supported me in every way possible. During this process the children were young and were gen- erously loved and cared for by a host of relatives and neighbors. The list is long and of necessity incomplete, but each was very important: my aunt, Beula Goodman; brothers' and sisters' families, the Robert Wings, the Jack Goodmans, the Robert Goodmans, and the Brian Sheens; and friends, Dorothy Beasley, Marjorie Higgins and Sue Walter. I wish to especially acknowledge my parents and parents-in-law who each contributed generously with time iv and encouragement throughout this part of my educational career. My mother, Clara Goodman, warrants special recognition for carefully proofreading the manuscript and bibliography. TABLE OF CONTENTS LIST OF TABLES . . . . . . LIST OF FIGURES . . . . . . Chapter I. INTRODUCTION . . . . Conservation: A Macrodecision Conservation: Microdecisions the Household . . Conceptual Framework . Research Questions . The Research Model . II. REVIEW OF THE LITERATURE Energy Conservation: A Priority . Conservation Within the Consumption . . . Surveys Related to Household Energy Consumption . . Within National Household . Empirical Studies Related to Household Experimental Studies in Household Energy Consumption Household Conservation Measures . . III. METHODOLOGY . . . . Data Collection Procedures The Sampled Community The Research Subsample Measurement Procedures Consumption Variables Annual direct household energy Weather-adjusted direct household energy consumption Percent change in annual consumption . . vi Page ix xi 10 12 16 17 22 23 31 33 4O 41 42 45 52 S4 54 55 S7 Chapter Page Structural and Behavioral Change Variables . . . . . . . . . S7 Furnace change . . . . . . 57 Installation of air conditioning . 58 Index of additional appliances . . 58 Changes in family size . . . . . 59 Change in employment status . . . 60 Installation of insulation . . . 60 Lowered thermostat setting on hot water heater . . . . . . 61 Intensity of repetitive household conservation practices . . . . 61 Research Hypotheses . . . . . . . 67 Hypothesis 1 . . . . . . . . . 67 Hypothesis 2 . . . . . . . . . 68 Statistical Analysis . . . . . . . 69 Assumptions . . . . . . . . . . 70 Limitations . . . . . . . . . . 70 IV. FINDINGS AND DISCUSSION . . . . . . 72 Changes in Household Energy Consumption . . . . . . 72 Hypotheses 1.1 and 1. 2 . . . . . 73 Findings . . . . . . . . . 73 Discussion . . . . . . . . . 74 Hypotheses 1.3, 1.4, and 1.5 . . . 76 Findings . . . . . . . . . 76 Discussion . . . . . . . . . 76 Impact of Household Microdecisions Upon Levels of Energy Consumption . . 79 Hypotheses 2.1-2.9 . . . . . . . 80 Findings . . . . . . . . . 81 Discussion . . . . . . . . . 85 V. SUMMARY, CONCLUSIONS, AND IMPLICATIONS . 90 Overview of the Study . . . . . . 90 Conclusions . . . . . . . . . . 92 Levels of Energy Consumption, 1973-74, 1975-76 . . . . . . . 93 Impact of Conservation Measures . . 94 Implications . . . . . . . 98 Energy Policy and Educational Implications . . . . . . 98 Implications for Family Theory . . . 102 Implications for Further Research . . 103 vii BIBLIOGRAPHY . APPENDICES . A. B. C. CONSUMPTION VARIABLES STRUCTURAL AND BEHAVIORAL VARIABLES CORRELATION MATRICES AND REGRESSION TABLES 0 viii Page 105 118 119 124 131 Table l. 10. LIST OF TABLES Summary of Research Articles Reporting Household Conservation Behaviors Related to Space Heating . . . . Summary of Research Articles Reporting Household Conservation Behaviors Related to Uses Other Than Space Heating . . . . . . . . . Family Energy Project Sample Comparisons, 1974' 1976 o o o o o o o o o 0 Selected Characteristics of Households: Comparison of the Family Energy Project Sample, 1976, and Research Subsample . Selected Characteristics of Respondents: A Comparison of the Family Energy Project Sample, 1976, and the Research Subsample . . . . . . . Selected Characteristics of Dwelling Units: A Comparison of the Family Energy Project Sample, 1976, and Research Subsample . . . . . . . Sample Households With Major Appliances, 1976; Sample Households Adding Major Appliances, 1974-76 . . . . . . . Sample Households With Energy Conserving Features, 1976; Households Adding Energy Conserving Features, 1974-76 . Reported Adoption and Increase of Household Energy Conserving Practices, 1976 . . T-test of Difference Between Annual Means of Millions of Btu's for Total Direct Household Energy, 1973-74, 1975-76 . . ix Page 37 38 46 48 49 SO 53 54 63 75 Table Page 11. T-test for Difference Between Annual Means of Millions of Btu's for Natural Gas, Fuel Oil and Electricity, 1973-74, 1973—74, 1975-76 . . . . . . . . . 77 12. Regression Analysis--Forward Inclusion Method: Regression Coefficients, Standard Errors, T-Ratios, Probability of Sampling Error, and Multiple Cor- relation of Independent Variables on Millions of Btu's Consumed, 1975-76 . . 82 13. Stepwise Regression Analysis: Regression Coefficients, Standard Errors, T-Ratios, Probability of Sampling Error, and Multiple Correlations of Independent Variables on Btu's Consumed, 1975-76 . . 83 LIST OF FIGURES Figure 1. Number of Adult Heads of Household Within One Family Who Reported an Increase in Conservation Practices to the Extent of All or Most of the Time by the Mean of Millions of Btu's Decrease from 1973-74 to 1975-76 0 O O O I O O O O 0 xi Page 66 CHAPTER I INTRODUCTION Actually the world's present problems are by no means unmanageable in terms of present biologi- cal and technological knowledge. The real cri- sis confronting us is, therefore, not an energy crisis, but a cultural crisis. During the last two centuries, we have evolved what amounts to an exponential growth culture, with institu- tions based on the premise of an indefinite continuation of exponential growth. One of the principal consequences of the cessation of exponential growth will be the inevitable revi— sion of some of the tenets of that culture (Hubbert, 1973, p. 37). A major concern of the 19703 was the rapidly increasing interdependence of the American lifestyle and fossil fuel energy, contrasted with the emerging recog- nition of the finiteness of this energy source. This strong interrelationship between energy1 and lifestyle has led many to the observation that the energy crisis is really a cultural crisis (Hubbert, 1973; Energy Policy of the Ford Foundation, 1974; Fritsch, 1974; Mazur, 1974; Odum, 1974; Hannon, 1975; Boulding, 1976; Downs, 1976; Lapp, 1976; Caldwell, 1976). 1Energy is used as a generic term meaning fuels, such as natural gas and petroleum and fuel equivalents such as electricity. The finite fuels of petroleum and natural gas have become the major sources of energy worldwide. The United States received more than 75 percent of its total energy in the early 19705 from these liquid fuels. Although the information reported has varied as to the extent of the fossil fuel reserves, there is agreement that fuel oil and natural gas supplies are finite and demand may exceed supply as early as the mid 19805 (Hubbert, 1973; Odum, 1973; Koenig, 1976; F. Murray, 1976; Lapp, 1976). The projected capital costs required to develop remaining fossil fuel resources and alternate sources are exorbitant and in some cases may be prohibi- tive. Net energy gains are also questioned (Odum, 1974; Lovins, 1974; Commoner, 1976; Koenig and Edens, 1976). Even if the development of alternate sources of energy were feasible and the capital to develop them were available, the length of time required for them to become operational and make substantial impact on energy supplies is not projected to be within this century. It is probable that liquid fossil fuels may become very short in supply prior to the year 2000 (Hubbert, 1973; Fritsch, 1974; Koenig, 1976; Lapp, 1976). Time is needed to seek alternate energy solutions while making necessary social, economic and structural changes. Through conservation, the period when the dwin- dling resources are available can be extended. Therefore, conservation of fossil fuels (i.e., policies and practices that reduce consumption) has been advocated for all levels of society (Odum, 1974; Hannon, 1975; Downs, 1976). Conservation: A Macrodecision The need for conservation of fossil fuels, although of worldwide importance, is particularly relevant within the United States. Estimates during the early 19703 were that the United States with six percent of the world's population consumed thirty percent of the total amount of energy used worldwide. It also had the highest national per capita use of energy (Rocks & Runyon, 1972; Fritsch, 1974). In 1960 the United States imported 19 percent of its oil; this increased to 41 percent in 1976 (U.S. Bureau of Mines, 1976). This growing dependence on for- eign oil has made conservation a matter of national political importance. Substantial conservation could be realized with- out difficulty for it has been estimated that Americans waste half of the energy they use (Hayes, 1976; Ross & Williams, 1976). Koenig hypothesized that the United States could reduce consumption one-third without lowering the standard of living (Downs, 1976). Some European coun- tries have standards of living comparable to the United States but have lower energy consumption (Schipper & Lichtenbert, 1976). Two diverse approaches to the energy problem have been advocated: increasing collectivism and cen- tralized governmental control or decreasing centralized control and increased dependence upon free enterprise and individual responsibility. A strategy which would incorporate both approaches has been outlined by Harman of Stanford Research Institute: If the basic problem is the unsatisfactory macrodecisions arising from microdecisions based on self-interest, then the obvious thing to do is reverse the situation, that is to identify the appropriate macrodecisions--per- haps by selecting appropriate national and planetary goals that are most in accord with the best available knowledge concerning human fulfillment--and then see what patterns of microdecisions would be necessary to achieve those macrodecisions (Harman, 1977, p. 10). Based upon the available evidence, energy conser- vation warrants selection as one major national goal.1 The implementation of this macrodecision involves micro- decisions from all sectors of society--business, industry, local and national government, and households. These deci- sions entail both the amount of direct energy consumed and indirect energy embodied in goods. Monitoring and under- standing actual consumption as well as decisions and behaviors leading to conservation are imperative and complex (Hannon, 1974; National Research Council, 1977). 1At the time of this writing, President Carter had sent to Congress a proposed National Energy Plan with emphasis upon conservation and conversion to alternate energy sources. Conservation: Microdecisions Within the Household The household is one of the most important sec- tors in society in relation to energy conservation. Many of the microdecisions related to energy consumption are made within the context of this unit primarily for the welfare of its members. Energy is consumed directly in households for uses such as space and water heating, cooling, transportation, or indirectly in the purchase of goods and services. It has been estimated that American households are responsible for over 30 percent of the consumption of direct energy and an estimated additional 40 percent in the form of indirect or embodied energy. Both the direct and indirect dimensions of energy consumption pre- sent major opportunities for evaluation of potential conservation of energy (Hannon, 1975; National Research Council, 1977). Indirect or embodied energy reflects the amount of fossil fuels or other energy sources used in production and distribution of goods and services. This research examined only the direct energy consumed within the household, i.e., mechanical energy derived or transformed from fossil fuels, namely, fuel oil, natural gas, liquid propane and electricity. Conservation in existing household structures is a staggering problem. Harrje (1977) reported there are 60 million homes already built that will be occu- pied for many years. Choices open to households in these structures are in many cases circumscribed; many basic housing features such as furnaces and major appli- ances (hot water heater, range, refrigerator, freezer) represent sizable economic investments; often they are already present within the house when purchased or rented (Newman and Day, 1975; Williams, Kruvant and Newman, 1976; Morrison and Gladhart, 1976). However,Grot and Socolow carefully controlled for structural and technological housing variables and found considerable unexplained variance in residentialI energy consumption. Their conclusion was that lifestyle decisions accounted for the additional variation. In our study, where many of the technological factors are standardized, we were prepared to discover that nearly all of the "lifestyle" effect had vanished, in which case we would have been in a position to emphasize the role of technology and to deemphasize the role of individual behavior. It is already clear that the truth lies somewhere in the middle. As each technological variable is separated out, the observed variation in gas and electric con- sumption is reduced, but when many of the tech- nological variables of which we are currently aware are separated out, considerable variation remains (Grot and Socolow, 1974, p. 489). Conservation measures are a part of those life- style decisions affecting variation in energy consump- tion. In general, these measures can be classified as structural or behavioral. Structural conservation measures are relatively permanent once they have been implemented, but they may require considerable economic investment. Installation of insulation and storm windows are examples. In contrast, behavioral conservation mea- sures are repetitive requiring seasonal or daily imple- mentation by one or more members of the household with lit- tle or no economic investment. Turning out lights and lowering the thermostat setting are two examples. It has been documented that households report both structural and behavioral conservation measures (Tables 1, 2, and 8, pages 37, 38, and 54). Some of these represent potential conservation of considerable impact; however, most contribute only a very small amount individually to the total household energy con- sumed. The impact of these reported measures on actual reduction of direct household energy consumption, indi- vidually or collectively, has not been reported. Spe- cifically, are families' reported perceptions of how they have conserved reflected in lower levels of energy consumption? It was hypothesized that examination of the relationships between longitudinal patterns of energy consumption and reported microdecisions related to energy conservation could provide insight into the feasibility of energy conservation within the household as one means of meeting the cultural crisis. Conceptual Framework In the development of this research, the family was viewed from an ecological systems perspective. Major concepts deemed particularly relevant included: the family perceived as a living system within the hierarchi- cal organization of society and decision making as the family's main adaptive feature. In an ecological systems approach, the family is conceptualized as a living system interacting with its environment. Emphasis is not only upon the family as a social, psychological system processing information, but also as a system dependent upon and impacting on the natural and man-built aspects of the environment (Hook and Paolucci, 1970; Bubolz and Paolucci, 1976; Morrison, 1975; Andrews, 1977; Bubolz, Eicher and Sontag, 1977). As with all living systems, the family must have inputs of both information and matter energy to remain viable (Miller, 1971). Hierarchal theory views the universe as orga- nized in levels from elementary particles to global and supra-global systems. Each level demonstrates increasing complexity and is composed of independent stable sub- systems each of which is actually a system encompassing lower levels of organization. This concept of whole- ness and interdependence, yet uniqueness of parts, has been derived from General System Theory (Bertalanffy, 1968; Koestler, 1969; Miller, 1971; Laszlo, 1972; Odum, 1977). Within this framework the hierarchy of living systems includes the following levels of organization: cells, tissues, organs, organisms, groups, societies, nations and supranational systems (Miller, 1971). Within the hierarchy including human beings, the family is perceived as a critical system at the group level; the family is also inextricably linked with higher level systems and also with its subsystems. Decision making is viewed as the main adaptive feature of human systems; it is the process of selecting and implementing responses under conditions where the stimulus situation, the valences, and/or possible responses are complex. Decisions are selected by com- plex interactions of preferences and perceptions (Kuhn, 1974, 1975). Decisions are affected by the availability of resources--human resources (e.g., knowledge, skills, values) and nonhuman resources (e.g., economic and natural resources). From a managerial perspective, the family is viewed as a basic decision-making unit where members, interrelated with each other, are involved in crucial decisions, affected by the environment surrounding the family, but also making an impact on the environment through their decisions or nondecisions. 10 Household energy is considered one resource to be managed. The household has become increasingly dependent upon inputs of liquid fossil fuels and their transformations to provide heating, cooling, lighting and work. With increasing prices and potential short- ages, energy conservation is viewed as a necessary man- agement strategy for American households. The decisions within the household related to energy consumption, although microdecisions, are assumed to be a crucial part of the overall macrodecision of energy conservation. The testing of this theoretical assumption was the major objective of this research. Research Questions The purpose of this study was to provide empiri- cal evidence for answers to the following research ques- tions: 1. Have households reduced their consumption of direct energy since 1973-74, the year of the Arab Oil Embargo? The time periods for making the comparisons were the two years July 1, 1973, through June 30, 1974, and July 1, 1975 through June 30, 1976.1 1Throughout the research report these specific dates were the reference points for the years 1973-74 (or year ending 1974) and 1975-76 (or year ending 1976). 11 One national studylmusdocumented a 1.8 percent decrease in annual household consumption between the 'years 1972-73 and 1974-75; natural gas was reduced by 3.9 percent,but electricity showed an increase of 1.2 percent (Grier, 1976; Williams, Kruvant and Newman, 1976). Replication of these results for these years or succeeding years has not been reported. On the basis of this information it could be hypothesized that the level of consumption of direct household energy consumed within the household changed from July l973-June 1974 to July 1975-June 1976 in the direction indicated: A decrease in total Btu'slof energy used within the household with electricity adjusted or unadjusted for conversion and transmission loss. A decrease in Btu's of energy derived from natural gas. A decrease in Btu's of energy derived from fuel oil. An increase in Btu's of energy used in the form of electricity with or without the adjustment for electrical conversion and transmission loss. 2. What conservation measures (microdecisions) did households report having taken within the two years prior to June, 1976? Did the reported conservation measures singly or collectively contribute significantly to reduced levels of consumption? b‘ 1Definition of Btu, p. 54. 12 The development of the specific research model and hypotheses to answer these research questions was an iterative process encompassing a review of the litera- ture, data availability and research methodology. These topics have been reported sequentially in Chapters II and III. The model that was developed as a result of this process has been presented here to give the reader an overview of the research problem. The testing of the hypotheses and empirical results have been presented in Chapter IV. The Research Model The theoretical approach advanced in this dis- sertation and review of the household energy literature suggested that total direct household energy consumption is a function of several major groups of variables: Natural environmental variables such as temperature, wind and humidity variations, natural resource availability. Human regulatory environmental variables such as the social, economic and political structure. Man-built environmental variables, such as land use patterns and unique structural dwelling unit variables (housing type and size, energy using equipment type and size). Family structural variables and resources such as size of household, human and non- human resources. Family behavioral variables such as energy consuming/conserving practicescnrdecisions. 13 This research focused only upon a part of these variables: the reported implementation of decisions related to energy conservation within households and the effect of these decisions upon levels of consumption. Both behavioral and structural conservation measures were evaluated. Although variables such as price increases and changes in income may be related to changes , in energy consumption, they were perceived as antecedent variables which may have prompted behavioral or struc- tural changes and were not examined. Careful attention was given to evaluate or con- trol for confounding variables which could have affected direct energy consumption. Because of the overpowering / effect of the dwelling structure on consumption, only families who resided in the same dwelling units from July, 1973, through June, 1976, were evaluated. Weather variations are critical to assessing total household energy consumption. However, this study was conducted in a geographic region in lower Michigan covering a radius of less than 100 miles; therefore, the weather e variations were considered constant across the sample but not over time; i.e., the years l973-74 and 1975-76 were not assumed equal in severity of cold weather. Both models were evaluated-~changes in absolute consump- tion and changes when adjusted for weather effects. 14 Since energy is consumed by the household as a whole, it was concluded that the appropriate unit for analysis was the household. The specific model tested in this research was derived from the following equations: For any household at a given point in time (t), BTUtl BTUt2 where BTU NATENV HOUSE FAMSTRUC FAMBEH f(NATENth; HOUSEtl, FAMSTRUCtl, FAMBEHtl) + e f(BTUtl, ANATENVt1,t2; AHOUSEtltZ, AFAMSTRUCtl,t2, AFAMBEHt1,t2) + e Total annual direct Btu's of household energy Natural environmental variables such as wind, temperature Individual structural dwelling unit variables Structural characteristics of occupants in household such as number of peOple, time spent at home Behavior of occupants in household. Specifically, these groups of variables were broken down into individual variables based upon previous research findings and available data, and the model to be tested became: BTU76 where BTU74, BTU76 WEATHER STRUC AIRCON APPLNDX FAMSIZE JOBLOSS INSULC INSULW HOTWATER CONBEH 15 f(BTU74, AWEATHER ; 74—76 ASTRUC74_76 AAIRCON74_76 AAPPLNDX74_76 AFAMSIZE74_76 AJOBLOSS74_76 AINSULC74_76 AINSULW74_76 AHOTWATER74_76 ACONBEH74_76) + e Millions of Btu's consumed annually by households in form of natural gas, fuel oil, liquid prOpane and elec- tricity Heating degree daysg'l973-74, 1975-76, a constant across the sample Installation of new furnace, 1974-1976 Addition of air conditioning, 1974- 1976 Index of appliances added, 1974-1976 Change in number in household, 1974- 1976 Household member retired or lost job, 1974-76. Installation of insulation in ceiling, 1974-1976 Installation of insulation in walls, 1974-1976 Lowered thermostat setting on hot water heater, 1974-76 Increased intensity of energy conser- vation behaviors, 1974-76. 1Definition of heating degree days, p. 56. CHAPTER II REVIEW OF THE LITERATURE With the growing awareness of the interrelated- ness of people and the natural environment, energy and its social impacts have become a subject of concern to social scientists as well as geologists and engi- neers. Cottrell (1955) identified the interrelationship between social values and the energy resource base of a population. Energy conservation has been defined as a set of ; policies and practices that reduce consumption. The N rationale for conservation as a national priority was briefly outlined in Chapter I. An overview of the information and sources that led to the statement of that position has been included in this chapter. Two compre- hensive bibliographies of energy/society literature have been prepared by D. Morrison et a1. (1975, 1976). Energy consumption often has been divided by economic sectors and end use. The three major economic sectors include household, industrial and commercial/ service. This research has dealt only with the household sector and the end uses of direct energy consumed within 16 17 the dwelling unit. In the literature review, emphasis was placed upon those studies concerned with direct household consumption and its structural and family behavioral correlates. Transportation, although a sig- nificant part of household energy consumption, was not included. Virtually all the empirical studies related to household energy consumption were developed immediately prior to, or since the time of the Arab Oil Embargo, 1973. Many of the studies remain unpublished. Those that have been published have often been broadly based surveys exploratory and descriptive in nature. Several annotated bibliographies have focused upon the soci— ological and psychological dimensions of the energy research (Frankena, Buttel & D. Morrison, 1976; Lopraeto & Meriwether, 1976; Olsen & Goodnight, 1977; Frankena, 1977). Schwartz (1977) has reviewed the literature per- taining specifically to the social consequences of the changing energy supply. Energy Conservation: A National Priority In the United States consumption of energy resources (coal, oil, natural gas, falling water, and uranium) doubled between 1950 and 1970 with an annual growth rate of 3.5 percent (U.S. Bureau of Mines, 1972). -By the 19703, the two fuels--petroleum and natural gas-- 18 were supplying over 75 percent of the total energy needs. These two liquid fossil fuels have varying pro- jected time lines for proved reserves. However, there is general consensus that the supply is finite and that the peak production either has been reached, or will be reached in the 0.8. within the next decade. World pro- duction has a somewhat longer projection period, peaking perhaps around 2000 (Berg, 1973; Hubbert, 1973; Udall, 1973; Cook, 1975). Alternative sources include conversion to coal, nuclear fission and fusion, solar, geothermal and wind energy. The estimated supply of coal in the United States is abundant and there is available technology for conversion to synthetic gas. But coal has limitations; it is a solid fuel, requiring energy and capital invest- ment to get it from the ground and into usable form. The environmental and pollution effects are also serious considerations (Hubbert, 1973; Udall, 1973; Walsh, 1974; Commoner, 1976). Nuclear energy is derived from uranium, a non- renewable resource with limited proved reserves. High levels of capital and energy investment are needed before the plants become Operational. The controversy 50%, <25%, etc. Milstein thinks figures should be reported to nearest 10% (1977) . CAll the research reported took place after the Arab Oil Embargo, 1973-74, but prior to the winter of 1977. 38 TABLE 2.--Sumary of Research Articles Reporting Household Conservation Behaviors Related to Uses Other Than Space Heating. Conservation Behaviors Respondents Related to Uses Other Potential Impact Reporting Researcher and Date Reportedc Than Space Heatinga Adoption Limit amount of hot water for 97% own hot water heaters,d 2-52\ Perlman & Warren 1974 washing dishes, clothes, 15% total directe Gottlieb 5 Hatre 1976, Thompson a bathing household energy MacTavish, 1976 Doner 1975 Hogan 1976, Morrison et a1. 1976 Turn down thermostat on hot 26 Hogan 1976 water heater Dishwasher full before 25\ own dishwashers.d 45—89 Perlman 5 Warren 1975 running =4 Btu‘lOb/year Milstein 1976 Air conditioning: use less 48% own air conditioning,d Perlman & Warren 1975 1—194 BtquOP/year efficiency varies. 2% total direct house- 46-69 Kilkeary 1975; Gottlieb 5 Hatrel976 hold energy Burdge 1976 Warkov 1976 Cooking: cook several dishes 97% own stoves, at one time =1o-13 Bturio‘v/year 3-48 Doner 1975 55 total direct house- Kilkeary 1975 hold energy Morrison et al. 1976 Lighting: turn off lights Lights included in "other" 22-98 Bartell 1974 not in use; smaller bulbs category which equaled Perlman 5 Warren 1974 53 total direct household Sears et a1. 1974 energy Gottlieb G Matte 1976 Doner 1975 Hogan 1975 Kilkeary 1975 Milstein 1976 Warkov 1976, Morrison et a1. 1976 Use electrical appliances Indeterminantn SlBtuxloe/ 2—90 Murray et al. 1974 less year was used for electric Bartell 1974; Sears et a1 1974; War- and gas appliances ren 1974: Thompson S MacTavishl976 Doner 1975 Clothes drying: cutting use 53% own dryersd, 2-77 Murray et a1. 1974 of dryer-—full load, ”7~ll BtuKIOG/Year Perlman 8 Warren 1974 clothesline 2\ total direct household Doner 1975 energye Milstein 1976 Morrison et a1. 1976 aBehaviors have been grouped conceptually rather than by exact wording. Wide variation in responses; therefore, only range of responses is reported. Milstein (1977) thinks figures should be reported to the nearest 10%. All the research took place after Oil Embargo 1973-74, but prior to winter of 1977. Newman 5 Day (1975). eDepartment of Interior (1975). 39 total energy use by turning the thermostat from 72° to 68° during the daytime and 55° at night. However, the reversibility and required repetitive implementation of these behaviors by household members are obstacles not to be ignored (Pilati, 1974; Hirst, 1976). In many cases the studies reviewed equated con- servation with reported conservation behaviors. The pervasiveness and complexity of energy use in the house- hold as well as the unreliability of the self-report make this position untenable. The major outcome criterion is energy consumption. Evaluation is needed as to what extent perceived conservation behaviors reduce that measure. No field study research was reviewed which assessed the extent to which these reported conservation behaviors were contributing towards energy conservation, i.e., reduced consumption. CHAPTER III METHODOLOGY The description of the following steps in the research process tum; been included in this chapter: data collection, sampling procedures and resulting sam- ple; description of subsample used in this analysis, measurement procedures, data reduction and analysis strategies; hypotheses to be tested; assumptions and limitations of the study. The data were gathered as part of an interdis- ciplinary study entitled "Functioning of the Family Ecosystem in a World of Changing Energy Availability," also designated the Family Energy Project (FEP) and funded by the Michigan Agricultural Experiment Station.1 This project was a field survey carried out in the greater metropolitan area of Lansing, Michigan, May and June, 1974, and repeated in May and June, 1976. The surveyed unit was the "family" defined as two or more related individuals living together, one of whom was 18 years of age or older. lMichigan Agricultural Experiment Station Project No. 3152. 40 41 Data Collection Procedures Comparable data collection procedures were car- 1 . . . A tra1ned 1nterv1ewer ried out in both 1974 and 1976. contacted the male or female head of the household, screened the household for eligibility, and obtained agreement for participation in the study. Participation was defined as (1) individual responses to a self- administered questionnaire by all qualifying members present within the household; those qualifying were male head of household, female head of household and oldest child if between 12 and 20 years of age; (2) either male or female head of household responses to interviewer- administered questions about the demographic and housing structural characteristics of the household. Personal interviews of about two half-hours were completed with a household head from each sampled resi- dence, one upon initial contact and one when picking up the self-administered sections. The interview data included a large number of characteristics about the housing unit, as well as details of household composi- ton and an array of socioeconomic and demographic char- acteristics. In 1976, permission was requested for 1A more detailed account of 1974 procedures can be found in Morrison (1976) and Zuiches, Morrison and Gladhart (1976). 42 release of household consumption data from the utility and oil companies. The self-administered questionnaire focused on beliefs, attitudes and behaviors related to energy and families. A ten dollar honorarium was given each house- hold upon completion of the self-administered and interviewer-administered questionnaires. The data were checked by the research team for completeness and individuality of responses. The raw data were coded, keypunched and verified by trained per- sonnel. Ninety-eight percent of the sampled families granted permission to obtain energy consumption informa- tion. These data were obtained from Consumers' Power, Board of Water and Light and numerous oil companies and were also checked for completeness, coded, keypunched and verified by trained personnel. The Sampled Community The 1974 sample was selected from the greater metropolitan area of Lansing, Michigan. The Lansing S.M.S.A. is considered to be a well-defined social, eco- nomic, and political metropolitan area characterized by a diversity of functions. It is the seat of state government, the site of a major university, and the location of light and heavy industry related to the automotive industry. The Lansing S.M.S.A. is an area 43 of commercial enterprise surrounded by an agricultural sector. The Lansing S.M.S.A. had a total population of 378,000 persons and 89,610 families (1970 Census). A multi-stage probability sample of urban, suburban, and rural families was drawn from the tri-county area of the S.M.S.A. Some portions of Clinton, Eaton, and Ingham counties fall within the S.M.S.A. which is considered to be a viable geographic area with a heterogeneous popula- tion. Ten census tracts were randomly selected, each tract having a probability prOportionate to the number of households therein. From the 34 blocks contained within the ten census tracts, over 600 houses were selected through the use of the 1973 Polk City Directory. In the ruralareas, townships with no incorpora- ted places and specific sections within townships were selected from the counties in the S.M.S.A. The house- holds sampled were from randomly selected addresses within the sections. Sampling procedures assured attain- ment of at least 150 urban and 50 rural families. The final sample contained 216 families, 160 urban and 56 rural. In the 1976 survey the interviewers were instructed to place a priority on obtaining interviews from households surveyed in 1974. To achieve approxi- mately equivalent-size samples for 1974 and 1976, 44 taking into account panel attrition, additional house- holds were chosen to be contacted for screening and interviewing. These new families were selected from the same tracts, blocks and township sections as the original families. Rather than draw a completely new sample, the additional addresses consisted of the 228 households not contacted, not at home, or listed as vacant addresses in 1974. The new rural sample consisted of all remaining households in the originally designated 12 sections. Upon completing the second wave of interviews, the sample total was 263 families, 135 new families and 128 families contacted in 1974 and reinterviewed in 1976. This was a follow-up rate of 59 percent. A sum- mary of the samples for 1974 and 1976 is found in Table 3 (page 46). A comparison was made between the census data of 1970 for the Lansing S.M.S.A. and the surveyed households at both points in time to assess the repre- sentativeness of the sample.3 In general, the area proba- bility samples were determined to be representative of the Lansing S.M.S.A. with single member households excluded (Zuiches et a1. , 1975; U.S. Bureau of Census, 1975). Lansing and Morgan (1971) observed: When repeated interviews use the same basic sam- ple down to the rather small areas, added preci- sion is provided for estimates of change even without re-interviews, because the correlation between the first interview and second in 45 demographic characteristics, etc. is higher than chance . . . and the sampling error is smaller than with two completely different random sam- ples (p. 348). The Research Subsample The primary objective of this present research was to assess the relationships between household micro- decisions related to energy consumption and changes in levels of consumption from 1973-74 to 1975-76. The household was selected as the appropriate unit for analy- sis. Three basic criteria were judged necessary for determining inclusion of a household in the sample. To control for the effect of the change in dwell- ing structure upon changes in level of consumption, only those families living in the same dwelling unit 1973 through 1976 were included. A second requirement was completeness of house- hold energy consumption data for July 1973-June 1974, and July 1975-June 1976. Energy data were judged com— plete for each household if not more than four months of electricity and natural gas were missing, and both winter heating seasons (November-March) were complete for the fuel used to supply space heating. Two percent of the data needed estimation. (See Appendix A for computation procedures for calculating and estimating household energy.) 46 Avhmv ooa Ammmv ooa Renew ooa Aoamv ooa . . . . . unoccommon Amav m N Amav m e Avav m m Aeav m m ucoumm mamcam . . . . mucoocommon Udano Ammo m m Away H o Away m m Any N m can ucoumm mamcwm . . . . mucmccommou caaco Amway 0 am Aaov m mm Aenav m om Ammv m mm one onEmm\onz m cm common Amemv m.om Amway m.mo Aenmv m.sm Ahmav «.mm oamwomoocm can: 2 w z w z w z w m on a>a c o m on a>a a mod HEM mucmocommmm H o. .c H m HHHEmm a c. .o H .H. m caonomsom .mnma .ehma .chmHHomEoo onEmm poonoum mmnocm maflammnu.m mqmo .cofluosomn wm.o Hamuo>o .sum Nam.oa Houoom aoflmum>coo Hecauuooam .sum Nae.m Houomm cowmuo>coo HOOHHuOOHmM c o mm.m mm.n me.h HOHHO cumocmum em.~m em.mm nm.em :oHuma>mo tumocmum ooo. mma oooét com.aai om.mom he.mnm coo: DUOHmS fl U4 muwowuuomaml.mmousom Had .Hmuoa oaocOmsom mm.m mn.m Ho.m Hound oumccmbm me.om mm.mm mm.mm cofluow>oc pudendum ooo. mma Omw.m| oao.mau mm.ema mm.>o~ memos moonsom Had .Hmuoe caoemmsom WWWMMMGRMMV manmama «atmsma a mo Onam>lu mm. anomoumu moa x Sum A.oma u zv .mhnmhma .enumhma .mmuocm caonomoom uomuflo Hmuoe now m.:um mo mcoaaaaz no means amazed cmmsumm moamnwmmao mo ammuneuu.oa mamas 76 those found by the Washington Center for Metropolitan Studies for 1974-75 (Grier, 1976, Williams et al., 1976). Hypotheses 1.3, 1.4, and 1.5 H01.3 There is no difference between the means of the Btu's of natural gas consumed within dwelling units during 1973-74 and the Btu's consumed within the same dwelling units during 1975-76. H 1.4 There is no difference between the means of the Btu's of fuel oil consumed within dwelling units during 1973-74 and the Btu's consumed within the same dwelling units during 1975-76. H 1.5 There is no difference between the means of the Btu's of electricity consumed within dwelling units during 1973-74 and the Btu's consumed within the same dwelling units during 1975-76. Findings.--Two of the null hypotheses were rejected with probability 1evels<fif<.002. The means for natural gas and fuel oil were significantly lower for 1975-76 than for 1973-74. An overall percentage reduc- tion of 6.6 percent was noted for natural gas and 11.1 percent for fuel oil. The null hypothesis related to electricity was not rejected. There was an overall increase from 1973- 74 to 1975-76, but this was not statistically signifi- cant. The results of these analyses have been summarized in Table 11. Discussion.-—The reductions in natural gas and fuel oil can be interpreted as reduction of fuel used in space heating since that was the primary end use 77 w~.N mmcmno w Hamuw>o .wH.HHI mmcmbo w Hamum>ow .wm.o I omcmco w Haouo>o .maamoanuooao mcwumo: moHocomsoc 03p moooHoch Ho.m Ne.e ma.e uouuo oumocmum mm.mm me.om He.hv cofluma>op pudendum mm. mma mo.a cma.~ vm.ooa av.mm com: LomH u 21 emumsflca muaoauuomam mm. mm.H om.H uouuo oumocoum 5H.n mn.mH mm.va coHumH>oo cumocmum mm. mma mo.a Owe. mv.am >5.om cmoz Roma u zv asuaoanuomam mm.m mo.m v>.m Hound ouncesew mm.mm mo.mm mo.oo coduma>mo oumocmum moo. hm mm.mu Omm.wai mo.mea Ho.mma com: Amm u zv Hao Hose mm.m hn.o mo.> Hound ouoocmum en.mm mm.mo me.mo cowuma>oo camocoum 000. mm mo.m| bem.HH| om.mma vm.oma coo: Lem u zv mmo Hmnsumz a mo Osam>lu Aenmanmnmav mocOHOMMHQ mhumhma enlmnma mousom monocm .maumsma .asumsma .suaoanuomam use Hao Hess .mmo amusumz How w.5um MO MCOHHHHE m0 mammz Hwficnfi Gmmzfiwm OOCOHOMMflQ How umOHIBII.HH mqmfifi a 78 for both fuels. Evidence for this speculation was apparent when compared with electricity where no overall decrease occurred. In this sample only two houses were heated electrically, providing an empirical basis for comparison between space heating end use, and end uses other than space heating. These findings were also con- sistent with the national survey reported by Grier (1976) and Williams et a1. (1976). The percentage reduction was greater for fuel 011 than for natural gas. When interpreting this result, two factors need to be considered. First, the measure- ment of the annual fuel oil usage was not as precise as the natural gas. Fuel oil was delivered at irregular intervals; this required the implementation of some estimation procedures which are presented in Appendix A. Second, fuel oil was used to provide space heating only, except in two households. Natural gas totals Often con- tained energy used by the hot water heater, stove, and dryer. Before concluding that fuel oil users conserved more, households which use different fuels for space heating should be compared on the total amounts of energy consumed within the household and not with these subtotals only. 79 Impact of Household Microdecisions Upon Levels of Energy Consumption The second major objective of this research was to examine the impact of conservation measures or prac- tices upon levels of consumption. Stepwise multiple regression was judged to be an appropriate analysis mode to test the hypothesis. The significance tests associ- ated with multiple regression were based on the follow— ing assumptions: sample selection was random; each array of the dependent variable for a given combination of independent variables followed the normal distribu- tion; there was a linear relationship between indepen- dent and dependent variables; there was homogeneity of variance of the arrays of the dependent variable (Nie et al., 1970; Blalock, 1972; Kerlinger, 1973). The assumptions of random selection and normality of the distributions can be violated without serious consequences. In addition, this sample was randomly selected and the size was large enough not to be con- cerned with the assumption of normality. However, violation of the assumption of homo- geneity of variances is important and was tested through . . . 1 the exam1nat1on of re31duals. The cases were ordered lResiduals are the difference between the actual and estimated value of the dependent variable for each case. -- -— Dune—1‘ 1” I L I, ilzmwa1HQ: n .9... ..~ v 80 by level of consumption 1974 and a scattergram of the residuals for the regression equation was plotted. NO pattern emerged. This suggested independence of errors and no violation Of homogeneity of variance. A violation of the assumption of linearity would mean some relevant information might have been obscured if a nonlinear relationship existed between an indepen- dent variable and the dependent variable. Tests for nonlinearity were made. No significant nonlinear rela- tionships were elicited. F. 1:77-“7‘7Jvi7: 3' I: ~ ~ «.1 Hypptheses 2.1-2.9 H02: Given the level of Btu's consumed within the household 1973-74, there is no linear relation- ship between the Btu's consumed within the household l975-76,and the selected independent variables related to structural and behavioral changes 1974-76. The alternate hypothesis or the hypothesis of interest has been stated below with the expected direc- tion of the relationship indicated following the variables. Hypothesis 2: Given the total Btu's of direct energy consumed within households 1973-74, there is a linear relationship between the level of Btu's 1975-76 and the following independent variables: 2.1 Installation of new furnace: There is a nega- tive relationship between installation of a new furnace 1974-76 and the level of consumption 1976. 2.2 Installation Of air conditioning: The greater the number of rooms air conditioned 1974-76, the greater the level of consumption 1976. 2.6 2.9 81 Addition of energy consuming appliances: The greater the level of Btu's consumed by appli- ances added 1974-76, the greater the level of consumption 1976. Number of persons: There is a positive relationship between change in family size 1974-76 and level of consumption 1976. Loss of job or retirement: There is a positive rela- tionship between change in employment status which increased the likelihood of time spent at home 1974- ,' 76 and the level of energy consumption 1976. Installation of insulation in ceiling: There is a negative relationship between installation Of insu- lation in the ceiling 1974-76 and level of consump- tion 1976. Aggro aa' . v I . Amps—H Installation of insulation in walls: There is a negative relationship between installation Of insulation in the walls 1974-76 and the level of consumption 1976. '31-:- Changed thermostat setting on hot water heater: There is a negative relationship between having lowered the thermostat setting for the hot water heater 1974-76 and level of energy consumption 1976. Intensity of conservation behavior: The greater the level of reported increase in household energy conservation behaviors 1974-76, the lower the level of consumption 1976. Findings.--The overall F-test was computed and found to be 218.68 which has a probability of .000. With this level of probability the null hypothesis was rejected and it was concluded there was a significant linear relationship between dependent and independent variables. These results have been presented in two steps. First, the forward regression procedure forcing all 82 variables into the equation has been reported to give an understanding of the relationships of all the variables in the proposed model, but pg: to test the hypotheses (Table 12). By comparing the results in Table 13 with the overall model, the stepwise regression process was clarified. Previous level of consumption 1973-74 was forced ‘ft'r-glj into the regression on step one. This was included as a baseline measure from which change could occur and was ! TABLE 12.--Regression Analysis-—Forward Inclusion Method: :_, Standard Errors, Regression Coefficient, T-Values, Probability of Sampling Error, and Multiple Correlation of Independent Variables on Millions of Btu's Consumed, 1975-76. Independent Std. Std. T-Value Probability of Variables Error Beta Sampling Error BTU 74 .32 .94 28.4 .00 CONBEH .51 -.11 -3.16 .002 STRUC 9.03 -.09 -2.85 .005 INSULC 7.49 -.04 -l.l .270 JOBLOSS 6.50 .03 1.00 .315 HOTWATER 5.98 .03 .95 .341 APPLNDX .74 .01 .33 .738 INSULW 7.98 .01 .28 .780 FAMSIZE 2.87 .007 .21 .832 AIRCON 5.87 .005 .16 .875 Overall F = 85.02 df regression = 10 Multiple R = .94 df residual = 119 R square = .88 83 .o.H codumsow m>moH on m “we mmH.H .Nm.H coauoovo House on m museum>flcs EDEacHEO .mmfidzeom .Zoude .3ADmZH .xozqmmfi .mNHmzdm .mmOAmOh "GOHDODJO OSH CH HOG mmaflmwhm> mNH "Hmscflmmn me e "conmonoou mo om. "ouozom m em. "m OHQHpHDE ooo. mm.mam "m Hamno>o I mca amo mMNo aim-HI. VOol No.0 mHoNul mm. VFGH .4... ca coflbmasmcHIIUQszH molenma coHum>Hmmcoo moo. oo.mu oo.: oo. oo.au mo suamcmucHuummmzoo moo. oa.m- oH.- Na.o Hm.s~- moowmmwaszmwwmmem ooo. ma.o~ so. Hm. om. omesmcoo m.spmmnwmomem “seems E E. moabmflnc> ucoocmmmccH m mhlmhma .Umfismcoo m.dum .mhlmnma .omEsmcoo n.5um co moabmaum> useccomoocH mo mcowumamnnou mamfluaoz can .uowum gawamamm mo upwaflbmnoum .mosao>le .muouum pudendum .mucOAOHmmmoo scammoumom "mflmwamcd :oflmmoumom Oma3moumnu.ma mamas 84 not the primary observation of interest. The strong correlation (r = .92) between consumption 1973-74 and 1975-76 was apparent with the very large t-value of 29.1 which was significant at the .000 level. In terms of prediction, the one best indicator of consumption was the previous level of consumption. L wfl‘ The primary interest of this research was not to ; predict but to assess the relative contribution of other 5 selected variables on the level of consumption 1975-76. These independent variables were permitted to enter the ’thCS .‘E In equation only if they met certain statistical criteria. The order Of inclusion was determined by the respective contribution of each variable to the explained variance. The minimum criterion for a variable to enter the equation was set at an F of 1.32 which is the F-value at the 75th percentile point of the F—distribution with 1,128 degrees of freedom. By comparing the overall model (Table 12) with the stepwise regression (Table 13), the relative strength of the independent variables was apparent through observation oftfimaunivariate levels of significance.1 Three variables in addition to lThe t-value for the univariate tests is often reported since the degrees of freedom are l,N-2; t = /f and the direction of the result is clear from the t-value. 85 consumption level were found to meet the minimum statis- tical requirements for inclusion in the model: new furnace added (t = -3.l6; p = .002); increased intensity of conservation behaviors (t -2.99; p = .003); and .235). No insulation in the ceiling (t = -1.19; p other changes in household structure or behavior, given the four previously entered variables, contributed sig- nificantly to the overall model. Discussion.—-There were only eight households limoq—‘ITVT '_ V. . which addedrunvfurnaces, but this was very significant in the model. Two speculations have been suggested: the new furnace was no doubt highly efficient; probably those households getting new furnaces had very old and inefficient ones, making a pronounced comparison between the two. The significance of this variable clearly indicated the importance of energy efficient technology. The second variable which was included in the model was not limited to a few households, but was dis- tributed across households. This variable, increased intensity of conservation, was not correlated with pre- vious level of consumption (r= -.08) suggesting that conservation behaviors were reported or not reported across all levels of previous consumption. This behavior change variable was developed to represent an overall household conservation behavior pattern; 86 ’ although most of the items were heat related, others not heat related were included. Elements included in the scale building seemed relevant for interpreting the results. The greater the number of people and the greater the intensity of the practice, the higher the level on the conservation scale. From these findings it can now be added--the greater the _~.. ._._.....— T1 I increased intensity in conservation behaviors, the greater the reduction Of energy. ”3‘ J‘K‘Ifl .1- -A ‘ Overall reduction was 6.3 percent; therefore, .15” I u it cannot be assumed from these findings that the behav- 1. ior change produced phenomenal results. It can be con- cluded that what change did occur was significantly impacted upon by increased intensity of conservation behavior. It was also noted that this finding reflected behavior change which was not tied to technology. At least for this sample, microdecisions as a whole did make some significant impact on consumption. The third variable that entered the model was less significant. The effect of insulating the house has been well documented; it was therefore surprising that the insulation variable did not demonstrate a stronger levelcflfsignificance. Several factors may have reduced its effectiveness here; it is only one measure Of a reported behavior and the effect of error in reporting could be great. NO information was elicited 87 as to the amount of the insulation added or the date of installation. Two other conservation variables were not sig- nificant enough to be included in the regression equa- tion--installation of insulation in the walls or reduc- tion of the temperature setting for the hot water heater. The reasons for this may be similar to those cited for insulation in the ceiling. The hot water heater has been evaluated as 15 percent of the total household energy, the second largest single end use within the household; this figure includes the energy used by electric water heaters with adjustment for conversion and transmission of electricity included. When evaluating the contribution of the hot water heater to the total energy consumed within a dwelling unit only, the percentage is reduced to within the range of five per- cent. Viewed in this manner, reduction of the tempera- ture of the hot water becomes one small microdecision within the context of total household consumption and would need more detailed analysis to demonstrate its effect. In the process of analyzing these data it was thought that the structural conservation measures would be powerful enough by themselves to make a significant difference. This was not supported by this analysis. 88 The development of a summary index of permanent conser- vation measures might demonstrate more impact. Although not a part of the overall plan for this research, two additional statistical procedures Were carried out. The results helped clarify the meaning of the data analyses. The first analyzed the same regression pro- cedure as described previously, using as the dependent variable the Btu's with adjustment for transmission and conversion of electricity. The second compared the effect of the behavioral and structural change variables . on changes in Btu's, omitting the previous level of con- sumption. Stepwise regression was performed including the adjustment factor for electrical conversion and trans- mission in the consumption variables. The analyses were compared; the results were nearly the same for the model using electricity adjusted as those using electricity unadjusted. The regression table is included in Appen- dix C. For both models the same three variables were entered first--Btu's 1973-74, new furnace, and increased conservation. In the unadjusted model where electricity was unweighted and heat appeared more important, the con- servation behavior of insulating the ceiling met the minimum criteria for inclusion in the model. In con- trast, in the model where electricity was more heavily 89 weighted, the change in the number of people in the household met the minimum requirements for inclusion in the model. This was consistent with the idea that the number of people affects electrical use more than heat use. In both analyses the increased intensity of con- servation behavior was significant. This observation helped confirm the speculation that the measure was an overall assessment of conservation rather than solely a reduction in space heating energy. When using the change in Btu's from 1973-74 to 1975-76 as the dependent variable and excluding the pre- vious level of consumption as an independent variable, the same three household changes were significant and had comparable levels of probability. However, the entire set of structural and behavioral changes explained 18 percent of the variance when the previous level of consumption was not included and only three percent when it was included. This demonstrated the moderate nega- tive correlation (r = -.30) between change in Btu's and previous level of consumption; the more energy that a household had used, the more it had reduced consumption. It was also noted that the more energy a household had consumed, the more latitude it had to reduce. CHAPTER V SUMMARY, CONCLUSIONS, AND IMPLICATIONS Overview Of the Study The development of this research was based upon the assumption that decision making is the main adaptive feature of human systems and that the family is one critical societal unit where this process occurs. This study examined one area about which families make deci- sions: the direct consumption of fossil fuels within the household. It has become increasingly apparent that fossil fuels are finite and may become short in supply and more expensive within this century. The family has become dependent upon fossil fuels to provide energy for heat- ing, cooling, lighting and work. It is within the context of the household that many decisions are made to consume or conserve energy. These observations give sup- port to the perspective that energy conservation is an important and urgent management strategy for households. Decision making is a complex process of perceiv- ing, selecting and implementing a variety of responses under widely ranging circumstances. This research 90 91 examined one aspect of the decision making process in the context of energy conservation: the reported imple- mentation of family conservation measures and their impact upon direct household energy consumption. It was hypothesized that the relationships Observed between longitudinal patterns of energy consumption and reported household conservation measures could provide insights into the potential contribution of voluntary conservation as one means of reducing overall energy consumption. Energy consumption data from utility and oil companies and conservation measures reported by house- hold members were the basis for the evaluation. The household was the unit of analysis. The data used were a part of an ongoing interdisciplinary study, The Family Energy Project,1 conducted within the College Of Human Ecology, Michigan State University. From the total area probability sample of 263 households, 130 were selected for this research. The self-report measures were taken from the 1976 survey; the energy consumption data were for the years 1973-74 and 1975-76. The major objectives were to determine if house- holds had reduced levels of energy consumption and if reported conservation measures had contributed signifi- cantly to levels of energy use. Careful attempts were lMichigan Agricultural Experiment Station Project 3152. 92 made to control for what might be alternate explanations that could have affected household energy patterns. The effects of the dwelling structure on total household energy were controlled for by including only those fami- lies living in the same residence for the years studied. Weather variations, although considered constant across the sample, were examined for their impact over time. Family change in size, 1033 of job, addition of major appliances, and replacement of furnace were used as vari- ables to control for their effect on changes in consumption. The conservation measures which were evaluated for their impact included installation of insulation in ceiling or walls, lowered thermostat setting on hot water heater and an overall scale measuring increased intensity of household conservation behaviors. Conclusions Did households reduce their consumption of direct energy since the year of the Arab Oil Embargo? Did the conservation measures reported by households contribute significantly to reduced levels of consumption? Several conclusions can be drawn from the analyses. 93 Levels of Energy Consumption, 1973-74, 1975-76 An overall reduction of 6.3 percent in direct household energy consumption was found between the years 1973-74 and 1975-76. When adjusting for electrical con- version and transmission, the reduction was 4.2 percent. These results were statistically significant at the .000 probability level. When examining consumption by spe- cific energy sources, it was apparent the decreases had occurred in fuel oil and natural gas. A 2.2 percent increase in electricity was noted but this was not sta- tistically significant. The reductions in natural gas and fuel oil can be interpreted for this sample as reduc- tions in space heating since this was the largest single end use for these sources. Statistical significance is a minimum criterion in social science research, but another question must also be addressed: were the results meaningful? The average decrease was 13 million Btu's. In order to achieve this average, 31.5 percent of the households had lowered their consumption 10 percent or more; one-tenth of the sample had reduced energy use by at least 20 per- cent. These results seemed very meaningful. The direction of lowering consumption in a slightly colder winter strengthened the suggestion of a movement towards conservation. This was a higher income 94 sample which consumed greater than average amounts of energy. It can be concluded that some household conser- vation has occurred at these levels. Impact of Conservation Measures The second major Objective of this research was to evaluate the impact of conservation measures reported by households since the time of the Arab Oil Embargo, 1973-74. Fifteen percent of the households reported installation(Ifinsulation or storm windows. Fifteen percent reported lowering the thermostat on the hot water heater. Questions were asked concerning nine conserva- tion behaviors, Moderate levels of increased adoption were reported for these seasonal or daily repetitive behaviors. When considering the impact of a repetitive behavior, two factors were thought to be of significance: the extent to which the behavior was prac- ticed and the number of people within the household who reported its adoption. For each item a family score was developed which reflected the number Of adult heads of household practicing the behavior all or most of the time. When considering the report of both adults for all the practices, two items were reported by more than 80 percent of the households: caulking, sealing windows with plastic or storm windows, and turning out 95 lights when not in use. For all other practices the range Of adoption reported by both adults was from 25 percent to 63 percent. This suggested that there were areas where substantial behavior change could be realized. Two behaviors which represented the greatest potential reductions were related to space heating: 59 percent reported keeping the daytime temperatures at 68 degrees or less, and 39 percent, the nighttime temperature at 60 degrees or less. The collective impact of these behaviors on levels of consumption was hypothesized as a major vari- able in this research. A family scale was developed to reflect the composite effects of increased intensity Of household behaviors, using the conservation items from the self-administered questionnaire. Through stepwise regression, the reported behavioral and structural changes that occurred within the households between June, 1974, and June, 1976, were analyzed for their impact on the level of consumption in 1976. The specific conservation measures that were included were installation of insulation in the walls“ or ceiling, lowering the thermostat setting on the hot water heater, and increased intensity of conser- vation behaviors. Other household changes which were not specifically conservation measures but thought to affect consumption were included as variables:changes 96 in family size, addition of appliances and replacement of furnace. Consumption during 1973-74 was used as the baseline from which change was evaluated. Two variables were clearly statistically significant: the installa- tion Of a new furnace (p = .002) and the increased intensity of conservation behaviors (p = .003). Instal- lation of insulation in the ceiling was the next variable to enter the equation, but the probability level was .235. All three change variables--installation of new furnace, increased intensity of conservation behaviors and installation of insulation in the ceiling--were negatively related to levels of consumption, which can be interpreted as changes which were significant in the reduction of consumption. When using the consumption variables with elec- tricity adjusted for conversion and transmission, the results were very similar except for a variable which was marginally significant. In the model in which elec- tricity was weighted more heavily, the change in family size was included; in the analysis in which heat was the predominant end use, insulation in ceiling was more important. Although these variables were not highly significant, they did suggest that the impact of family size is greater on the level of electrical consumption than upon space heating. 97 Previous level of consumption was the best pre- dictor of consumption at the later point in time (r = .92). However, energy use during 1973-74 and change in Btu's for the two years under study had a moderate nega- tive correlation (r = -.30). This provided empirical evidence that conservation occurred somewhat more at higher levels of consumption than at lower levels. Intensity of conservation behavior was not correlated with previous consumption (r =-u08). This indicated that reported behavioral change was not related to pre- vious energy use. The only structural change that contributed sig- nificantly towards lowered consumption was the installa- tion of a new furnace. In the designing of the research problem this was not included as a conservation measure, but was entered into the model as an alternate explana— tion of reduction in consumption. Its significant rela- tionship clearly demonstrated the importance of energy- efficient technology. The permanent structural conser- vation measures of insulation and reduced thermostat settings of water heaters were entered individually into the regression model assuming that each could have sig- nificant impact. The data analysis did not support this and it was concluded that a summary measure of structural change should be considered in further research. 98 Implications It can be said of many problems, however com- plicated, that when they are looked at intensively they become increasingly more complicated. This is espe- cially true of the energy problem which is inextricably tied to economic, environmental and social issues. From the myriad of possible implications for the family and energy policy, a few basic ideas were selected for discussion. Energy Policy and Educa- tional Implications The voluntary nature of conservation has been discussed extensively. People in general have reported that they do not want forced conservation (Olsen, 1976; Zuiches, 1976). Seventy-seven percent of this sample was against conservation through government imposed controls. run: will voluntary conservation work? We often hear that Americans need to be compelled to change their habits, that patterns of waste are too deeply ingrained to expect much volun- tary change; if this is so, then we may expect that pricing, taxes, and other forms of disincen- tive are necessary to implement conservation. I believe the potential for voluntary change is largely unexplored and may be underestimated, e.g., water conservation in California (Unseld, 1977, p. 4). Voluntary conservation is a relative concept and not an either-or situation. It seems reasonable to 99 assume that with the increasing shortage of fossil fuels, prices will rise. The relevant question then becomes, given the context of rising prices, will people conserve voluntarily? The analysis of these data has suggested that a substantial percentage of people have already begun to do so. In view of the urgency of the energy problem it must be assessed whether the reduction that occurred was enough. Population increases, formation of new house- holds and rising expectations of lower level energy users can be expected to increase levels of consumption. When considering such factors, based on a national sample for the years 1972-73, 1974-75, what appeared to be a decrease was actually an aggregate increase in household consumption (Grier, 1976). Given this information, the reductions demonstrated by this research sample and the national sample would not be enough. If estimates are accurate that reductions of one- third to one-half can be made without seriously affect- ing lifestyles, much more reduction can be realized by households UM»nus,l976; Hayes, 1976). Harrje (1976) has set a goal of 50 percent reduction in household energy. If reductions of this magnitude are necessary, what insights could be gained from this research for their implementation? 100 The largest reductions were in the area Of space heating. This can be interpreted several ways. In addition to the impact of increased prices, it can be argued that households changed what was easiest and this had minimal effect on lifestyle (Milstein, 1976). Also, a major emphasis after the Oil Embargo via the media had been to reduce fuel for space heating and lower the thermostats to 68 degrees. Therefore, it can also be maintained that households responded to what had been emphasized. If one accepts this interpretation, clarifi- cation Of further dimensions to the public looks hOpeful. Space heating does offer the greatest potential savings and additional ways of conserving space heating energy need to be identifed and disseminated (Harrje, 1976; Pilati, 1976). However, the emphasis upon space heating only is too narrow. Electricity represents another major target area for conservation, especially when understanding the amount of primary fuel used to generate the electricity. This study has dealt with direct energy used within the house, but limiting the emphasis to direct energy only may be insufficient and even counterproductive. The generalization to be understood by the public is basic: we are dealing with a finite supply Of fossil fuels; the direct and indirect uses of these fuels must become apparent to households as well as to other users if energy conservation is to 101 enter into decision making at all levels (Bullard & Herendeen, 1975; Hannon, 1973, 1975). In various experimental studies information has been found to be ineffective in reducing consumption. It is not assumed that information alone would bring about the behavior change, but it is maintained that broadening the information base could help reduce the alienation that rising prices mayknflxmn Economic values force people to make certain choices; ecological under- standing would give people a basis for making and accepting those choices. A broadened understanding of energy issues would help clarify that households'limited choices are inex- tricably tied to institutional decision making. This information is basic if households are to impact on public policy. Although the role of the individual household has been emphasized throughout this research, the con- tribution of the household is clearly limited, but can be a significant part of the overall Situation. IRISweden the success of the lower energy-intensive lifestyle without a reduced quality of life has been attributed not only to the cooperation of the individual households (persons),but also to government and institutional planning (Schippper and Lichtenbert, 1976). The task of additional behavior change is not to be understated. 102 Increasing prices, information, exhortation, feedback and incentives, as well as legislative and institutional support, have all been identified to varying degrees as elements to bring about reduced consumption. The most difficult and least researched dimension of the problem may have been identified by Leik & Kolman (1977): get- ting people to do more conserving now. Implications for Family Theory The increased intensity of conservation behavior within households was a significant variable in the regression equation (p = .003). The amount of change that this variable represented was not of great magni- tude, but it was one of two change variables that were significant in explaining reduced levels of consumption. The importance of energy-efficient technology was demonstrated clearly. But the role of the behavior of household members was equally significant, i.e., the accumulation of many microdecisions was important in overall reduction of consumption. While this seems like a truism needing no sup- port, the role of the individual family in society has been increasingly de-emphasized; nearly one-half Of the research sample reported that the amount of energy all American families could save is unimportant compared to the amount of energy that government and industry 103 could save. Although this research focused upon deci- sions related to energy consumption, the theoretical position was that decision making is the main adaptive feature of human systems. This study supported the per- Spective that a family's microdecisions may have a small, but significant impact upon a complex process like energy consumption. Implications for Further Research Time series analysis was recommended as an opti- mal method for analyzing these data. This was deemed too costly and extensive for an individual research project, but appropriate for a research team analysis. This pro- cedure would be more sensitive to seasonal and monthly variations such as weather, holidays and vacations. This analysis could clarify the effect of conservation mea- sures such as insulation and other retrofitting. The environmental influences such as price increases and political/historical events could be assessed with greater precision. From this research it has been established that L/ some households have reduced consumption through struc- tural as well as behavioral changes, but are there demo- graphic correlates that "explain" these behavior changes? Discriminant analysis would be one appropriate strategy. 104 Households in this sample have demonstrated reduction in consumption. Continued monitoring of total direct household energy over time as well as for more broadly based samples seems imperative. Other dimensions of household energy use such as transportation need to be examined as well. Microdecisions were found to contribute signifi— cantly to change in consumption. Technological analyses about additional microdecisions that can be implemented . by families to reduce consumption would assist households. 3 Evaluation of the methods for disseminating the informa- tion is equally important. Observational studies or detailed self-reported documentation of energy use and behavior patterns within the household could be insightful. These studies could assist in helping families recognize areas for behavioral change and bring about immediate as well as future reduc- tion in consumption. One hypothesized long-term impact would be the socialization of children and youth towards a more conserving lifestyle. This study has focused upon direct household energy consumption. Less is known about the impact of microdecisions upon dimensions such as transportation and indirect household uses of energy. Models similar to the one tested here could be applied to these areas. BIBLIOGRAPHY 105 BIBLIOGRAPHY Andrews, Mary. "An Ecological Approach to Problems in Home and Family Life: An Education Perspective." Paper presented at Home Economics Education Conference, College of Human Ecology, Cornell University, May, 1977. Bartell, Ted. "The Effects of the Energy Crisis on Atti- tudes and Lifestyles of Los Angeles Residents," University of California, Los Angeles. Paper presented at the 69th Annual Meeting of the American Sociological Association, Montreal, August, 1974. Battalio, Raymond C.; Kagel, John H.; Winkler, Robin C.; and Winett, Richard A. "Residential Electricity Demand: An Experimental Study." Unpublished manuscript, 1976. Berg, Charles A. "Energy Conservation through Effective Utilization." Science, July 13, 1973, pp. 128-38. Bertalanffy, Ludwig V. General System Theory. New York: Braziller, 1968. Bethe, H. A. "The Necessity of Fission Power." Scien- tific American 234 (January 1976): 21-31. Blalock, Hubert M., Jr. Social Statistics. New York: McGraw Hill, 1972. Boulding, Kenneth E. Review of The Poverty of Power: Energy and the Economic Crisis by Barry Commoner. Society, November/December 1976, pp. 87-90. Bubolz, M.; Eicher, J.; and Sontag, S. "The Human Eco- system: Conceptual Clarification." Manuscript, Michigan State University, March,l977. Bubolz, M., and Paolucci, B. "An Ecological Systems Approach to the Family: Preliminary Concep- tualization." Paper presented at the annual meeting of National Council on Family Relations, October, 1976. 106 107 Bullard, C. "Energy Conservation Through Taxation." Document No. 95, Center for Advanced Computation, University of Illinois, Urbana, February 1974. Bullard, Clark W., III. The Illinois Consumer's Role in Energy Conservation. Document No. 73, Center for Advanced Computation, University of Illinois, Urbana, 1974. Bullard, Clark W., II. and Herendeen, Robert A. "Energy Impact of Consumption Decisions." Proceedings of the IEEE 63 (March 1975). Bultena, Goudon L. "Public Response to the Energy Cri- sis: A Study Of Citizens' Attitudes and Adap- tive Behaviors." Sociology Report No. 130, Iowa State University, 1976. Burdge, Robel J.; Warner, Paul; and Hoffman, Susan. "Public Opinion on Energy." Unpublished paper, Department of Sociology, University of Kentucky, 1976. Caldwell, Lynton K. "Energy and the Structure of Social Institutions." Human Ecology 4, No. l (1976): 31-44. Carter, Lewis F. "Interactive Monitoring System for Evaluating Energy Policy Effects on Private Non- industrial Consumption." Project #0024, Social Research Center, Washington State University, Pullman, Wa. Ongoing. Cohen, Reuben. "Setting Equitable National Goals for Household Energy Conservation;" Paper presented at the Annual Meeting of the American Sociologi- cal Association, New York, August, 1976. (Available from Response Analyses Corporation, Princeton, New Jersey.) Commoner, Barry. The Poverty of Power: Energy and the Economic Crisis. New York: Alfred A. Knopf, 1976. Commoner, Barry; Boksenbaum, Howard; and Corr, Michael. Energy and Human Welfare: A Critical Analysis. New York: Macmillan, 1975. Connecticut Power and Light Co., Hartford, Conn. Experi- mental study reported by Associated Press in the New York Times, August 21, 1976. 108 Cook, Earl. "Forecasting Depletion," in Committee on Ways and Means, U.S. House of Representatives. Background Readings on Energy Policy. Washing- ton, D.C.: Government Printing Office, March 1, 1975, pp. 86-177. Cook, Stuart W.; McClelland, Lou; and Belsten, Laura. "Encouraging Energy Conservation in Master- Metered Buildings." Ongoing study at University of Colorado, September 1976-June 1977. Cottrell, Fred. Energy-Society. New York: McGraw-Hill, 1955. Doner, W. B. "Consumer Study: Energy Crisis Attitudes and Awareness." Detroit: Market Opinion Research, March 1975. Downs,Charles R., ed. "The Harmoniously Confused World." Center for Environmental Quality, Michi- gan State University, 1976. Eichenberger, Mary Ann. "A Comparison of Ownership of Selected Household Appliances and Residential Energy Use by Employed and Nonemployed Homemakers in the Lansing, Michigan Area." M.A. thesis, Michigan State University, 1975. Energy Policy Project of the Ford Foundation.‘ A Time to Choose: America's Energy Future. Cambridge, Mass.: Ballinger Publishing Co., 1974. Frankena, Frederick. "Behavioral Experiments in Energy Conservation, An Annotated Bibliography." Department of Sociology, Michigan State Univer- sity, 1977. Frankena, Frederick; Buttel, Frederick H.; and Morrison, Denton. "Energy/Society Annotations." Depart- ment of Sociology, Michigan State University, November 1976. Fritsch, Albert J. Lifestyle Index. Washington, D.C.: Center for Science In Public Interest, 1974. Gladhart, Peter; Zuiches, James J.; and Morrison, Bonnie M. "Impacts of Rising Prices Upon Residential Energy Consumption, Attitudes, and Conservation Policy Acceptance." Paper presented at symposium, "Social and Behavioral Implications of the Energy Crisis," University of Houston, Energy Institute, 1977. 109 Glass, Gene V., and Stanley, Julian C. Statistical Methods in Education and Psychology. Englewood Cliffs, N.J.: Prentice Hall, 1970. Gottlieb, David, and Matre, Marc. "Conceptions of Energy Shortages and Energy Conserving Behavior." Social Science Quarterly 57 (September 1976). Grier, Eunice. "Changing Patterns of Energy Consumption and Costs in U.S. Households." Paper presented at Allied Social Science Association Meeting, Atlantic City, N.J., 1976. Gross, Gordon, Project Engineer. Study of household electrical consumption, Midwest Research Insti- tute, Kansas City. Ongoing. Grot, Richard A., and Socolow, Robert H. "Energy Utiliza- tion in a Residential Community." In Energy, pp. 483-98. Edited by Michael S. Macrakes. Cambridge, Mass.: MIT Press, 1974. Hannon, Bruce M. "An Energy Standard of Value." The Annals of the American Academy of Political and Social Science 410 (November 1973): 139-53. . "Energy Conservation and the Consumer." Science, July 11, 1975, pp. 95-102. Harman, Willis. "The Coming Transformation." Futurist 11 (February 1977): 5—12. Harrje, David T. "Retrofitting: Plan, Action and Early Results Using the Townhouses at Twin Rivers." Center for Environmental Studies Report No. 29, Princeton University, Princeton, N.J., June, 1976. Hayes, Denis. "Energy: The Case for Conservation." Worldwatch Institute Paper 4, Washington, D.C., 1976. Hayes, Steven C., and Cone, John D. "Reducing Residential Electrical Energy Use: Payments, Information, and Feedback." Journal of Applied Behavior Analy- sis, in press (as of May, 1977). Hays, William L. Statistics. Chicago: Holt, Rinehart and Winston, 1963. 110 Heberlein, ThomasZL "Conservation Information: The Energy Crisis and Electricity Consumption in an Apartment Complex." Energy Systems and Policy 1, No. 2 (1975): 105-18. Herendeen, R., and Sebald, Anthony. "Energy, Employment, and Dollar Impacts of Certain Consumer Options." Ford Foundation Energy Policy Project Report: The Energy ConservatiOn Papers. Washington, D.C.: Ford Foundation, 1975. Hirst, Eric. "Residential Energy Use Alternatives: 1976 Q to 2000." Science, December 17, 1976. Hirst, Eric, and Moyers, John C. "Efficiency of Energy Use in the United States." Science, March 30, 1973. Ffmfic 4.1m template pr .cn-rr. v- a Hogan, Janice M. "Energy Conservation: Family Values, Household Practices and Contextual Variables." Ph.D. dissertation, Michigan State University, 1976. Hook, N. C., and Paolucci, Beatrice. "The Family As an Ecosystem." Journal of Home Economics 62 (May 1970): 315-18. Hubbert, M. King. "Survey of World Energy Resources." Mining and Metallurgical Bulletin 66 (July 1973): 37-53. Hyland, Stanley R., et al. The East Urbana Energy Study, 1972-1974: Instrument Development, Methodologi- cal Assessment and Base Data. Champaign-Urbana: University of Illinois, College of Engineering, 1975. Kerlinger, F. N., and Pedhazur, E. J. Multiple Regres- sion in Behavioral Research. New York: Holt, Rinehart and Winston, 1973. Kilkeary, Rowena, and Thompson, Patricia J. "The Energy Crisis and Decision Making in the Family." Herbert K. Lehman College, City College of New York, New York, 1975. Klausner, Samuel E. "Energy Shortages and the Poor." Paper presented at the Energy Institute, Uni- versity of Houston, "Social and Behavioral Impli- cations of the Energy Crisis: A Symposium," 1977. 111 Koenig, Herman B., and Edens, Thomas C., et a1. Resource Management in a Changing Environment: With Applications to the Rural Sector. East Lansing: Michigan State University, 1976. Koestler, Arthur, and Smythies, J. R. Beyond Reduc- tionism: New Perspectives in the Life Sciences. New York, 1969. Kohlenberg, Robert; Phillips, Thomas; and Proctor, Wil- 7 liam. "A Behavioral Analysis of Peaking in Resi- , dential Electrical-Energy Consumers." Journal of Applied Behavior Analysis (Spring 1976): 13- 18. Kuhn, Alfred. The Logic of Social Systems. San Fran- cisco: Jossey-Bass, 1974. . Unified Social Science: A System Based Introduction. Homewood, 111.: Dorsey Press, 1975. Lansing, John B., and Morgan, James N. Economic Survey Methods. Ann Arbor: University of Michigan, 1971. Lapp, Ralph E. America's Energy. Greenwich, Conn.: Reddy Communications, 1976. Laszlo, Ervin. Introduction to Systems Philosophy. New York: Harper & Row, 1972. Leik, Robert K., and Kolman, Anita Sue. "Isn't It More Rational to Be Wasteful?" Unpublished paper presented at the Social and Behavioral Implica— tions of the Energy Crisis: A Symposium, Uni- versity of Houston, June 27-28, 1977. Lieberman, M. A. "United States Uranium Resources: An Analysis of Historical Data." Science, April 30, 1976, pp. 431-36. Lopraeto, S. C., and Meriwether, M. W. "Annotated Bibli- ography on Energy Attitude Surveys." Unpublished Manuscript Center for Energy Studies, University of Texas at Austin, 1976. Lovins, Ammuur B. "World Energy Strategies." Bulletin of the Atomic Scientists (May 1974): 14-32. "The Case for Long-Term Planning." Bulletin of the Atomic Scientists (June 1974): 38-50. 112 Mazur, Allan, and Rosa, Eugene. "Energy and Life Style." Science, November 15, 1974, pp. 607-10. Miller, James G. "Living Systems: The Group." Behavioral Science 16 (July 1971): 302-98. Milstein, Jeffery. "Attitudes, Knowledge and Behaviors of American Consumers Regarding Energy Conserva- tion with Some Implications for Governmental Action." Washington, D.C.: Office of Energy Conservation and Environment, Federal Energy Administration, October 1976. . "How Consumers Feel About Energy: Attitudes and Behavior During the Winter and Spring of 1976 and 19766." Paper presented at the Energy Insti- tute, University of Houston, "Social and Behav- ioral Implications of the Energy Crisis: A Symposium," 1977. Morrison, Bonnie Maas. "Socio-Physical Factors Affecting Energy Consumption in Single Family Dwellings: An Empirical Test of a Human Ecosystem Model." Ph.D. dissertation, Michigan State University, 1975. Morrison, Bonnie M., and Gladhart, Peter. "Energy and Families: The Crisis and Response." Journal of Home Economics (January 1976): 15-18. Morrison, Bonnie M.; Keith, L. Joanne; and Zuiches, James J. "Energy Impacts on Michigan Families." Report to the Sociopolitical Risk/Impact Resource Group of CONAES, National Research Council, National Academy of Science, Michigan State Uni- versity, 1976. Morrison, Denton E. "Equity Impacts of Some Energy Scenarios." Paper presented to Sociopolitical Impacts Resource Group, Risk/Impact Panel, CONAES, National Research Council, National Academy of Sciences, November 1976. Morrison, D. E.; Bemis, V.; Frankena, F.; Buttel, F.; Galin, J.; Mejorado, 0.; and Cardinal, J. "Energy/Society Reference Update: A Bibliogra- phy of Recent Social Science and Related Litera- ture." Department of Sociology, Michigan State University, 1976. . Energy: A Bibliography of Social Science and Related Literature. New York: Garland, 1975. 113 Murray, Francis x. Energy: A National Issue. Washing- ton, D.C.: Center for Strategic andilnternational Studies, Georgetown University, 1976. Murray, James R., et a1. "Evolution of Public Response to the Energy Crisis." Science, April 19, 1974, pp. 257-630 National Research Council Committee on Measurement of Energy Consumption. Energy Consumption Measure- ment: Data Needs for Public PoIicy. Washington, D.C.: National Academy of Sciences, 1977. Newman, Dorothy K., and Day, Dawn. The American Energy Consumer. Cambridge, Mass.: BaIlinger Publish- 1ng Co., 1975. Nie, Norman; Hull, C. Hadlai; Jenkins, Jean C.; Stein- brenner, B., and Bent, D. H. Statistical Package for the Social Sciences. New York: McGraw Hill, 1970. Nunnally, J. C. Psychometric Theory. New York: McGraw- Hill, 1967. Odum, Eugene P. "The Emergence of Ecology As a New Integrative Discipline." Science, March 25, 1977, pp. 1289-93. Odum, Howard. "Energy, Ecology and Economics." Man Environment Systems 4 (1974): 227—34. Odum, Howard T., and Odum, Elizabeth. Basis for Man and Nature. New York: McGraw-Hill, 1976. Olsen, Marvin B., and Goodnight, Jill A. Social Aspects of Energy Conservation. Seattle, Wash.: Battelle Human Affairs Research Centers, 1977. Ophuls, William. Ecology and the Politics of Scarcity. San Francisco: Wm. H. Freeman & Co., 1977. Palmer, Michael H.; Lloyd, Margaret E.; and Lloyd, Ken- neth E. "An Experimental Analysis of Electricity Conservation Procedures." Journal of Applied Behavior Analysis, in press (as of May, 1977). Paolucci, Beatrice, and Hogan, M. Janice. "The Energy Crisis and the Family." Journal of Home Economics (December 1973): 12-15. 114 Perlman, Robert, and Warren, Roland. "Energy Saving by HouseholdscfifDifferent IncomesiJIThree Metropoli- tan Areas." Reports 1 and 2. Florence Heller Graduate School, Brandeis University, May, 1975. Pilati, D. A. The Energy Conservation Potential of Winter Thermostat ReductIOns and Night Setback. ORNL-NSF Environmental Program Project supported by National Science Foundation Interagency Agree- ment NO. AEC 40-237-70 and NSFAG298. ORNL-NSF- EP-8. Oak Ridge, Tenn.: Oak Ridge National Laboratory, 1974. Pilati, David A. "Energy Savings Via Behavioral Changes._ Industrial Forum 7, 2-3 (1976): 103-6. Rocks, L., and Runyon, R. P. The Energy Crisis. New York: Crown Publishers, 1972. Rodgers, Willard. "Analysis of Survey Data." Mimeo- graphed class notes, Psychology—Sociology 616, University of Michigan, 1975. Ross, M. H., and Williams, R. H. "Energy Efficiency: Our Most Underrated Energy Resource." Bulletin of Atomic Scientists (November 1976): 30-38. Schipper, Lee, and Lichtenbert, A. J. "Efficient Energy Use and Well Being: The Swedish Example." Report 4430, Lawrence Berkeley Laboratory, Uni- versity of California, Berkeley, 1976. Schumacher, E. F. Small Is Beautiful: Economics As If People Mattered. New York: Harper and Row, 1973. Schwartz, T. P. "Short End of the Shortage: luiAppraisal of the Social Consequences of the 1973-74 Energy Crisis and of Related Research." Paper presented at symposium "Social and Behavioral Implications of the Energy Crisis," University of Houston, Energy Institute, 1977. Sears, David 0., et a1. "Political System Support and Public Response to the 1974 Energy Crisis." Paper presented at Conference on Political Alienation and Political Support, Stanford, California, May, 1976. Seaver, W. Burleigh, and Patterson, Arthur H. "Decreas- ing Fuel Oil Consumption Through Feedback and Social Commendation." Journal of Applied Behavior Analysis 9 (Spring 1976): 147-52. 115 Seligman, Cleve, and Darley, J. M. "Feedback As a Means of Reducing Energy Consumption." Paper pre- sented at the 1976 meeting of the American Soci- ety of Heating, Refrigerating and Air Condition- ing Engineers, 1976. Seligman, Cleve; Darley, John M.; and Becker, Lawrence J. "Psychological Strategies to Reduce Energy Con- sumption: First Annual Progress Report." Center for Environmental Studies, Princeton University, November, 1976. Socolow, Robert H. "Energy Conservation in Housing: Concepts and Optionsfl' In Future Land Use, Energy, Environmental, and Legal Constraints. Edited by Robert W. Burchell and David Listokin. New Brunswick, N.J.: Rutgers, 1975, pp. 311-23. Stanford Research Institute. Patterns of Energy Con- sumption in the United States. Washington, D.C.: U.S. Government Printing Office, 1972. Teller, Edward. Energy: A Plan for Action. New York: Commission on Critical Choices for Americans, 1975. Thompson, Phyllis T., and MacTavish, John. "Energy Problems: Public Beliefs, Attitudes and Behaviors." Unpublished manuscript, Urban and Environmental Studies Institute, Grand Valley State College, Allendale, Michigan, 1976. Tuso, Margaret A., and Geller, E. Scott. "Behavior Analysis Applied to Environmental/Ecological Problems: A Review." Unpublished manuscript summarized in Journal of Applied Behavior Analy- sis 9 (Fall 1976): 526. Udall, Morris K. Committee on Interior and Insular Affairs. America's Energy Potential: A Summary and Explanation. Washington, D.C.: Government Printifig Office, 1973, in U.S. House Committee on Ways and Means. Unseld, Charles T. "Social Impact Assessment of Energy Policy: Behavior, Lifestyle, Values and Pri- orities." Unpublished paper presented at "The Social and Behavioral Implications of the Energy Crisis: A Symposium," University of Houston, June 27-28, 1977. 116 Bureau of the Census. Census of Population and Housing: 1970, Census Tracts, FinaI Report PHC(1)-106 Lansing, Michigan SMSA. Washington, D.C.: Government Printing Office, 1972. . Population Estimates and Projections: 1973 (Revised) and 1975 Population Estimates and 1972 (Revised) and 1974 Per Capita Income Estimates for Counties, Incorporated Places, and Selected Minor Civil Divisions in MiChigan. Washington, D.C.: Government Printing Office, Series P-25, Number 670, 1977. Bureau of Mines. "U.S. Energy Use at New High in ° 1971." News release, March 31, 1972. . Minerals and Materials. Washington, D.C.: Government Printing Office, September 1976. Community Services Administration. Small Farm Energy Project, Kansas, ongoing. Department of Interior. Energy Perspectives, pre- pared by H. Enzer, W. Dupree and S. Miller. Washington, D.C.: U.S. Government Printing Office, 1975. Walsh, John. "Problems of Expanding Coal Production." In Energy: Use Conservation and Supply, pp. 118-22. Edited by Philip Abelson. Washington, D.C.: American Assn. for Advancement of Science, 1974. Warkov, Seymour. "Energy Conservation in the Houston- Galveston Area Complex: 1976." The University of Houston Energy Institute, October, 1976. Warren, Donald I. "Individual and Community Effects on Response to the Energy Crisis of Winter 1974: An Analysis of Survey Findings from Eight Detroit Area Communities." Ann Arbor: Program in Com- munity Effectiveness, Institute of Labor and Industrial Relations, University of Michigan, 1974. Warren, Donald 1., and Clifford, David L. Local Neigh- borhood Social Structure and Response to the Energy Crisis of 1973-74. Ann Arbor: Program in Community Effectiveness, Institute of Labor and Industrial Relations, University of Michigan, 1974. 117 Williams, John; Kruvant, William; and Newman, Dorothy. Winett, Metropolitan Impacts of Alternative Energy Futures. Washington, D.C.: Metrostudy Corp., 1976. Richard, and Nietzel, Michael T. "Behavioral Ecology: Contingency Management of Consumer Energy Use." American Journal of Community Psychology (1975). Zuiches, James J. "Coercion and Public Acceptance: Thex/f Case of Energy Policies." Paper presented at the annual meeting of the Society for the Study of Social Problems, New York, August 1976. (Depart- ment of Sociology, Michigan State University, East Lansing, Mich.) . "Changing Family Energy Behavior through Infra-Red Heat Loss Evaluation: An Experimental Approach." Interim Report on Project EA-77-X- 01-2118, submitted to ERDA by Institute for Family and Child Study, Michigan State University, 1977. Zuiches, James J.; Morrison, Bonnie M.; and Gladhart, Peter M. "Interviewing Families: Methodology and Evaluation of 'Energy and the Family' Sur- vey." Research Report 311, Michigan State Uni- versity Agricultural Experiment Station, East Lansing, Mich., September, 1976. APPENDICES 118 APPENDIX A CONSUMPTION VARIABLES 119 APPENDIX A CONSUMPTION VARIABLES Fuel Oil Calculations Fuel oil was delivered at irregular intervals; it was necessary to determine the amounts of fuel oil that were used during the years under study and not to include amounts from the preceding or succeeding heating seasons. The heating season was determined to be from September 1 through May 31; that period included 96 per- cent of the heating degree days for this section of Michigan. To determine what amounts were used during 1973-74 and 1975-76, the following decision rules were implemented: If the first oil fill-up in the fall of 1973 was preceded by a fill-up in the summer, the fall fill-up was the usage from September 1 until the date of the fill-up. If the first fill-up of the fall occurred with- out a preceding summer fill-up, the amount used between September 1 and the first recorded fill-up was calcula- ted on a unit-per-degree-day measure derived in the following manner: 120 _...1_-. -. _. 121 F1 = Gallons of fuel oil for first fill-up, Fall 1973 F2 = Gallons of fuel oil for second fill-up, Fall 1973 DD = Number of degree days from 9/1/73 to date of first fill-up (Fl) DD = Number of degree days between first fill-up and second fill-up, Fall 1973 PDD = Fuel used per degree day for second period PDD = FZ/DD2 F = PDD x DD Calculated fuel usage for Septembei 1 until date of first fill-up F1 was then considered fill l and put into the date of the first fill-up. For each succeeding period from one fill to the next, the fuel oil was distributed on a fuel-per-degree- day basis by the number of degree days between the fill-up dates, excluding the months of June, July and August. Heating degree days were recorded for those months but including these heating degree days made the estimate per degree day unstable; it seemed probable that peOple did not heat their houses even though heating degrees were recorded occasionally. If the first fill-up in the Spring of 1976 was followed by a fill-up in the summer, that fill was con- sidered the usage from the last spring fill-up to May 31, 1976. 122 If there was not a summer fill-up recorded, a fuel estimate was calculated on a degree day usage in the following manner: Ff = Gallons of fuel oil for final fill-up, spring 1976 Ef_1 = Gallons of fuel oil for next to the last fill-up, spring 1976 DD = Number of degree days from date of f-l F to F f-l f DDf = Number of degree days from Ff to 5/31/76 PDD = Fuel oil per degree day calculated for f—l . next to last period F5/31/76 = PDDf-l x DDf This was considered the final fill-up and put into the date of May 31, 1976. The one household using liquid propane was treated in the same manner. The fuel oil and liquid prOpane were converted to Btu's. Natural Gas and Electricity Calculations Consumers' Power of Lansing, Michigan, and the Board of Water and Light, Lansing, Michigan, were the utility companies which provided the monthly natural gas and electricity data for the respondent households, from July 1973 through June 1976. These data were converted to Btu's. 123 The data were collected at different points in time and inadvertently the month of July 1973 or June 1974 was missing for some respondents; also, some fami- lies moved into the houses between July 1, 1973, and November 1, 1973; this missing data equaled less than 2 percent of the monthly data. The missing data were estimated by using the following proportion: Mlt Mzt + M3t + M3t+ M1 M2 t+1 t+1 l where t = year data were missing t+l = year following t M1 = the missing data month the two months immediately following missing data. M2, M3 APPENDIX B STRUCTURAL AND BEHAVIORAL VARIABLES 124 lMfil o c APPENDIX B STRUCTURAL AND BEHAVIORAL VARIABLES Questionnaire Items STRUC: Have you replacedymnn:furnace in the past two years? If respondent answered YES, STRUC = 1. AIRCON: Have you added room or central air conditioning in the past two years? If respondent answered YES, AIRCON = number of rooms air conditioned. APPLNDX: Have you added any of the following appliances in the last two years? If respondent answered YES, the household was assigned the annual estimated usage of Btu's for each appliance added between June 1974 and June 1976. The values were summed to form an appliance index. , Btu's x 10° Appllance ___1Year___ Electric stove Gas stove l Dishwasher Television--black and white Television--color Electric clothes dryer Gas clothes dryer Refrigerator--self defrost Refrigerator--no defrost Freezer--self defrost Freezer-~no defrost bmmflqul—‘I—‘Ob 1Not adjusted for electrical conversion and transmission. .To estimate for conversion and transmis- sion, multiply values for electrical appliances by 3.2. 125 126 FAMSIZE: (A) For households resurveyed in 1976 (n = 79): FAMSIZE = the number of persons in the household during 1976 minus the number in the household 1974. (B) For those households newly surveyed in 1976 (n = 51): If a child was born between June 30, 1974, and June 30, 1976, FAMSIZE = 1. If a divorce or death of spouse occurred between July 1, 1974, and December 31, 1975, or if the household respondent answered YES to any of the following questions: Has your family experienced the following events during the past year? (1) Marriage of son or daughter; (2) Son or daughter leaving home (other than marriage). FAMSIZE = -1. NO household had both a gain and a loss of members. All others: FAMSIZE = 0. JOBLOSS: If an adult head Of household retired or lost a job between June 30, 1974, and December 31, 1975, JOBLOSS = 1. INSULC: Have you installed insulation in the ceiling in the past 2 years? If respondent answered YES, INSULC = 1. If NO, INSULC = 0. INSULW: Have you installed insulation in the walls in the past 2 years? If the respondent answered YES, INSULW = 1. If NO, INSULW = 0. HOTWATER: Have you ever lowered the setting on the dial of the hot water heater? If the respondent answered YES, when did you do this? If within the past 2 years, HOTWATER = 1. All others, HOTWATER = O. CONBEH: Cover or seal windows and doors with storm windows or plastic Turn down thermostat while sleeping to 60° or less in the winter Maintain daytime tem- perature at 68° or less in the winter Have heating equip- ment cleaned and serviced Turn off lights not in use Dry clothes on clothesline rather than in dryer Do not heat some rooms in winter Limit amount of hot water for bathing, dishwashing and washing clothes Cook several dishes in oven at one time 127 The following questions were asked of both male and female heads Of household. For each head of household for each question: If the respondent checked YES, INCREASED, and checked ALL or MOST OF THE TIME, the conservation practice = 1. All other combinations = 0. For each question a family score (FINC) was formed: FINC = value for the male head plus value for female head. HAVE YOU INCREASED THIS PRACTICE IN THE 116T1mw>rmum? Never All Most Once but the of the in a would time time while try YES NO 128 FREQUENCIES STRUC NEW FURNACE 74-76 CATEGORY LAtEL CODE 1. TOTAL AIRCON ADDFD AIR CONDITIONING CATEGORY LAFEL CODE 1. TOTAL APPLNDX INDEX APPLIANCES ADDED CODE O 1. 2. 3. A. 5. 6. 7. 8. 9. 10. TOTAL CATEGORY LALEL FAMSIZE CHANGE CODE '3. '2. -1. CATEGORY LABEL 1. 2. TOTAL opmtwpwutmooc p (N O - a ,0 m m V < in u m m N w P up DA)» 0 o o o o o o e ()m-I P c (D 9 m V" 0" U1 HDH 1.5 74-76 NUHBE? PEOPLE IN FAMILY 129 ABSOLUTE FREQ 11k 16 130 ABSOLLTI FREQ 111 TI FREQUENCIES JOBLOSS DID H OR W LOSE J08 IN 79-75 CATEGORY LABEL CODE N0 0 YES 1. TOTAL INSULC INSULATION CEILING 7h-76 CATEGORY LABEL CODE 0 1. TOTAL INSULW INSULATION HALLS 7Q-76 CATEGORY LAEEL CODE 1. TOTAL HOTWATER CODE 0 1. TOTAL CATEGORY LAzEL CONEEH CATEGORY LAFEL 10. 11. 12. 13. 1“. 15. 16. 17. TOTAL AFSOLUTE FRE 4 11A ”Ntd‘mood‘ p 130 ICUOO ILCWEASED INTENSITY CONSERVATION 74-7F V< '7. t‘ I aomwm-qgm H H mum-(fliro 00.... tQJN .8 100.0 , o o o o I o u o o R S C A L E 1 C 0 N 8 E H F O A N A L Y S I S o . O U U i ‘ ‘ ' R E L I A 8 I L I T Y m z m B H v o m o as [\ H m 9 fl 2 I S m u :2 E: a. E m B 0 “TH L) o no >c mm B P-N 23' ‘H H ||g 22 s H éz A Gig 9.9 é. @a—mnmflnoa mooooomoc HH'HV‘V‘v-JHHH UQUUUUUUU ZZZZZZZZZ HhWflfiHb+flHH uuuuuuuuu 00.0.0.0. «NMmekrO STD DEV HE ANS DODDBDOOB O O C O O O O O . OGDDODQOQ mnnmmnnnn ddfiflflfldv‘d \DMKNO‘U‘GNM NOWW¢N$¢® WOQTUHJDm ADO? J Of0¢¢ mfifhhmfikh 0.0.0.0.. lflhthrnvuMn m.u«=cnmwnd mocmmo‘mmo fiJJNmCOHJ nscmwmwoo 000...... v1 d 0CwHUMJUMDO momoooooo HfidAHHv-drdd UUUUUUUUU 222222222 hflHHPflHwaHH uuuuuuuuu coo-ooo.. CORRELATION FATRIX FINC159 FINF1F1 FINC162 FINC163 FINClEb EINC165 FINC166 FINC169 FIN0160 1.00010 FINClEq FINC160 FIN0161 1.00000 .3087? .25715 .16366 .h3397 .2799“ .k2626 .3273? 1.00000 05“553 039930 .1 1130 1.00000 .27173 .27920 .05107 .15265 .22015 .2699“ FINC162 FINC163 1.00000 .31057 .15072 .20730 .37927 .h2972 1.00000 .20698 .33h03 .3hhh3 .05756 .0800h .21h07 .262“? .30058 FINC16b 1.00000 .19159 .31hhh .09023 FINC1E5 1.00000 027503 033025 FINC166 1.00000 .h5113 p. F) FINC159 150.0 S w y. N OF 0A STATISTICS F09 q N OE VARIADLES H‘AN 6.h0769 SCALE STATISTICS er mowm .Jlbb d to m ww 42h» CCU-(u; (Mu A mrum HO ITEM-TOTAL okndNGOml U‘C‘v-‘NOU‘UU‘Q Nnmmmu00(c mrco®0¢mm Bhkhhhhhb O 00...... Ooosmkwhm «cuwwsneoa cn13NOWmON0 owmmmmJN¢ :Mmmmamrm 0.0.0.... ovnommmao oOuoaJMwO H‘HNDO‘HIUMD rowaammuo m33mm~4mm 0 .00 O I 0.. «004:0nno ONKU‘MO‘O‘OU‘ K J 3'»HO‘~DNU‘ waaomsmna CFQO‘CNONO‘.’ ooooooooo MJJJJuUJJ udduddddu «63Nmo(u: nmuor3nmm Naa:nom~« Ommrmmwow oomouacn~ 000.0000. mmmmmomwm OOdNMJmOO momomococ «dddwduda QUUUUUUUU 221224224! HHHHHHHHH umuuuuuuu D IF A CCFEFICIENT CANNOT RE COAPUTED . L IS POINT A VALUE DE 99.0 F.m~a'$uw'lu -.».. ‘ . . l A APPENDIX C CORRELATION MATRICES AND REGRESSION TABLES 131 STANDARD 05V CASES MEAN VARIAQLF OOOOOGOCDC‘Dr‘DC? MMMMF‘MMMMMMMMM fiflfiwfifiwddfiwdwfi NHmCFNC‘xDI‘NQ33N" mNJDdHJO‘NU‘COM's :tr‘ Ohvamu- Almond-Drama mmp omevar‘MNMm: 00000000000000 W"! N GIN-1&0 9.0.0 cl‘N U'fl‘fl‘U‘U‘U‘OJHt‘o-tkows... marmarwow‘mMNGJm HMMBMDMJNIAR'BGVJVI Dl‘fiwaofld—IdJO-TQ 0000000000000. 3'1. . 0-4 wrath” 0‘9 080-1 HN NNI 0. Lu)‘ LIN/T \DJ NC‘ ZU’FV‘IF P WJHZUCAA IIIIIIOOIOI ZOquDcu th 2H #02 mu4¢ IIIIIIUIII bun A.AA >AH4HuaHuc> Lo Lumwu “www¢.om 2oHpaH>uu ppm . ma.-.~mc an¢~¢.aoaoMA .mNA JanuHmma 04wuo. uoqzcm a curmumu< ego. moo-.¢o« Arumc.QU¢.Qd sno-.m~:mu~ .: onmmuaeuq ~¢mw¢. mucocu a uLzaLHuuchm L uoqacm zqmr muuqzcm uc 23m Lo mc‘chmq> uo mHm»4qzq ~4owo. u mdonpdnx >amzqu 2H maaowc numz:z wuvan mcquu a 0 0 0 0 0 I 0 0 I I I n 0 I I 0 I 0 0 I I 0 2 D H m w w c .zomhchnaroo auxhcau nun wag. , cemaacm.n o:~ao.u new. o::c~:o. n~4m¢s~.u mmumo.l rwu. momswro.u msocncn.~0 swaoo.u own. «maocso.o mu«u~¢m.~o Gamaw. a nchdmw. ouaoao.n~ «o >hwunhw<4w wuzmatw0240wuo~ snmwwo.u« mseuwc.m~ manwwos.n nwmomns.a mac::mno. whoaaw:.«1 wwooa~.«d suscwc.o~- -wmsowm:wm. uc353wcv. m coaum gym m co 40>wavu .h‘chm‘uuw wNnmzau Iuwzvo usuhm cracker WJo 0-0-0000'0000000000000 ZOHFCDUW WIP ZH mwddefluq, 0-0-0-0000000000'0--- wNHurdu ..: thmafioq FUCJM I» mbbw Jake» uwszz QMpw zo Dauphzm .w-waaqmua> wscqwo» ..w4mquoq> hzuczmuuc can UJQMHADXtIIOIIlnoflnlniitlIItono 1234 .2unhcacw w:» ‘H my: qur<~uc> 44¢ Non. «NNNNOBMII wONNNINIa GhrccomoaI .hzqhmzou. momoo.n moo. . Nu6h4eco «oiwnmawwmwao MN¢M03~or ndckhn¢n0 SJDWZH (mama. who. Ndooasmol ssmcowcomI r0h0ucm0o «:Mo¢m.mml UDuhw «Lawn. Now. . wowmroon ananuomI «NncmeU0 dmwhmawodl rwazcu NMNNd. «0H0 wwmwmmuol MM4m0w30«I ndnnmwhoh mummococuI UJDMZH I Q WWWmmwc0 wWHmm4mu0 deflctaom mcnmachI ZOOQH‘ .. .& mauOJMGI mmn«mda:o meowwdwso MONKmHuMI xczdnuu na.oo.- No“. . , whmwmmno wwmm««m.d ouwhanom Ncuaechoc wmuanOfi (a I .- wwHWWhoo MMucncnwo ecbnnowoo Mmooamcom awhqzboz c ‘ 0 -¢0 hmwkcauo NONBOBcso BJCmmowom menu0460d uNumtdu uLZQUHu—ZUHM >PHCthddw wozququch u w02 drwm h r UOQam urw E wa boa ;.o@« >pH4HacHu<> Lo uuwuu Hoo.u.s~ . zcaquouo o»m «odom.ado opfimo.c-nh .9.“ Snaohmma “mow". uoqzcm a oupmzsoq mug. .NAao.~ Loack.efiym Luawn.no~o« .o onuwwuoua naqu. naczcm a uUZILHLHonm L uuqzcm zen: mwuq:cm Lo zzm no mu‘aHuc> Lo mHm>4qzq «.mwa. o WJUHpua; ¢~-:~ m44¢x onAqJszH :JawzH okuan uuqzazu :wz uaapm o~-:~ onpq>uLmzou >~Hmzusz owmdmuqu :umrou mh-a~ ozHIHuc Iguhqgamzw odoqu o~-45 cszoHAHonu uaq smog. zouwuq as-.. swoon mwuzqwgoaq xuczH xczgoaq uk-au 2“ was awe; x no : ewe wwwgmws 0 3 I >4quu z“ waaowc wwwzaz mncas ucquo m-mraL ..A «meta: awpw zo omamhzu .m.wamauuq> mosas.o~ .>wu .o»m mamdu.p_- umzcnmwa 24»: a». gape» 2H wuzczu oo:»m ..maa .zmc‘wowo I I I I I I I I I I I I I I I I I I I I I I I 2 o H m w u a o u n w 4 a H h A D x I I I I I I I I I I I I I I I I I I I I I I I