" “3&1 .. ‘ ;‘;t:.L .4 .J", 1 . yaw W“ «‘3‘ :‘j‘n “ ”a? '. 1 I ‘ ’ 2% “-4 i1 - j» ' ‘ . 1"”"J fir. . ifs. .3 .31.; 'L ,. a. - I“- I ‘;.i I \t_ 4731.: 1a": -, . :‘ H95 E. If”: 9” '. )L IVA? '3' «\“.""éi'flr v “. II 5",” ‘P' .f , ’ .‘.'..‘;|‘" ‘99 ‘IJI{ v, ,‘3‘ h ' I it. i! I'- ;.i ‘ _*II I“- I?” IM‘” IIII , ,I ‘ I 1&‘31‘; t. I. n.1,] “i'I‘ .. .. ‘ {I HI”'?"N - ' (1," “' “:7"; ‘ A (:5? ‘. .r . . I‘ ‘: J m. V "k“ (I: H ‘. ‘I .“ ’11'"L:.‘.."”_I;""'I‘h 5"" .II‘.‘I .‘jy _ 1II _.-'I “J‘H'H' , H ' ,I .. " F" | '. .. I ‘v'ijggrfg‘1i-I‘Ij; ”Fri? "I “I; by...“ ” L '4" IAIN E: "' I!YI:.1(¢IVP; [ .I 0 jkfi .' I .".'II'I.\‘_'I ” l I' M I 1: :: u 4"“. W"; “ II. .“~"I~‘1"I'~. IIIII I III" "III"! I?" ‘ I ,. 1‘!“ ' II"; II II‘.I ~"' "It”; I” Lita.” ”firm (113" ' ' 5.3%? V .IWIII 4" .s- M.“ ‘17 LIBRARY "5 Michigan Stan: University lllllllllllllllllllllllllllllllllllllllUllllllllllllllllll 3 1293 10429 0568 This is to certify that the thesis entitled THE EFFECT OF PAST EXPERIENCE ON CURRENT ENERGY CONSUMPTION AND CONSERVATION PATTERNS: THE INTERACTION OF HISTORICAL TIME, SOCIAL TIME AND LIFE TIME presented by Susan Lee Merkley has been accepted towards fulfillment of the requirements for Ph.D. Fam'l E degree in 1 y co logy Major professor Date ,2//f/{/ 0-7639 OVERDUE FINES: 25¢ per day per item RETURNING LIBRARY MATERIALS: Place in book return to name charge from circulation records THE EFFECT OF PAST EXPERIENCE ON CURRENT ENERGY CONSUMPTION AND CONSERVATION PATTERNS: THE INTERACTION OF HISTORICAL TIME, SOCIAL TIME AND LIFE TIME by Susan Lee Merkley A DISSERTATION Submitted to i ‘ ‘ Michigan State University ‘5. in partial fulfillment of the requirements ' n for the degree of DOCTOR OF PHILOSOPHY n3;.‘-r Department of Family Ecology 1981 ABSTRACT THE EFFECT OF PAST EXPERIENCE ON CURRENT ENERGY CONSUMPTION AND CONSERVATION PATTERNS: THE INTERACTION OF HISTORICAL TIME, SOCIAL TIME AND LIFE TIME by Susan Lee Merkley This research was designed to examine the influence on current energy consumption and conservation behavior exhib- ited at the micro—level of past experiences with macro-level crises characterized by shortages or deprivation of natural resources. More specifically, the research was designed to determine if households headed by individuals exposed to varying levels of natural resource shortage or deprivation in the past were currently exhibiting differing patterns of energy consumption and, in response to the energy crisis, change in consumption behavior. Two research questions guided the study. (1) Did exposure to shortages and deprivation of natural resources in the past influence current energy consumption patterns? (2) Did exposure to shortages and deprivation in the past influence change in energy consumption patterns over time ; . | Wis—— Susan Lee Merkley (i.e., conservation behavior), especially in response to the urgency placed on conservation since the Arab Oil Embargo in 1973-74. Three macro—level events in twentieth century American history were identified as showing basic similarities to the current energy shortage crisis. Exposure to these three events by household heads, as determined through identifi- cation of the age of the household head, was usedtxaanswer the research questions. The effects of World Mars I and II were considered because during both periods rationing of basic foods and fuels was imposed by the Federal government in order to extend supplies for the armed forces fighting overseas. Americans thus faced shortages in both types of products. The impact of the Great Depression was considered because the general economic decline witnessed during this period forced drastic wageanuiworking hour limitations on a large number of income earners. Thus, household purchasing power was reduced and ability to obtain goods was hampered. Energy consumption data from utility and oil companies as well as sociodemographic and attitudinal measures reported by household members, along with age of the house— hold head, were the basis of the analysis. Multiple regression procedures were employed to test the research questions. Results of the first analysis indicated that, net of the effects of aging—related factors which could influence energy consumption patterns, level of past experience with shortages and deprivation was a statistically significant 4—~_ -flm l i l Susan Lee Merkley predictor of current energy use behavior (Beta = .114, p = .02). Households headed by individuals with higher levels of deprivational experience currently appeared to be consuming larger amounts of energy than comparable house— holds headed by individuals with less experience. The second research question studied proportional change in consumption behavior between 1976—77 and 1978-79. This analysis revealed that, net of the influence of aging- related factors or change in the aging-related factors between 1978 and 1979, level of past experience with short- ages and deprivation was a statistically significant predic— tor of percentage change in energy consumption behavior (Beta = —.ll4, p = .055). Households headed by primary income earners with higher levels of exposure to deprivation and shortages in the past appeared to be responding to the energy crisis by reducing proportional energy consumption to a greater degree than equivalent households with less exposure to hardship in the past. The findings suggested that households headed by older individuals, those having faced hardship in the past, were responding to this experience by currently consuming more energy than households headed by individuals with less trau- matic backgrounds. In addition, it appeared that households headed by younger individuals, those reared in the affluence of the 19505 and 19605 and lacking a backlog of experiential knowledge to draw upon to help them implement conservation practices, were not adapting as readily to the current r . .I ’ 3.‘ ‘. . l‘ 0 . -.§:‘I ‘ Susan Lee Merkley «a ' ;rgy crisis. On the other hand, households headed by J. ‘ ‘awgfder individuals, those having faced deprivational expe- .jg§%iences in the past, seemed to respond more readily to the fil-fiéed to adopt conservation practices. For my Mother and Father, and my aunts, Elaine and Jean ACKNOWLEDGMENTS Completing this research project brought to a close 3 years of study and learning with individuals who have willingly shared their time and their specialized knowledge. Being exposed to their interests and insights has expanded my own, and to each of them,l want to express my appreciation. Dr. Bonnie Maas Morrison's interest in the history of energy use at the household level awakened my own and led to the development of this project. Dr. Craig Harris' knowledge of research methodology and procedures expanded mine and led to my having a fruitful learning experience carrying out the analysis of this research. Dr. Anne Meyering's interest in the history of the family helped broaden mine and helped me realize the impor— tance of historical influences on the everyday lives of individuals and families. Dr. Beatrice Paolucci's interest in the study of the family from an ecological perspective awakened mine and helped me in conceptualization of this research problem. Others have helped in numerous ways to make working on this dissertation both an enjoyable and profitable learning experience. iv Dr. Barbara Stowe, through the 3 years, was able to insure that financial assistance was available so study could continue. In addition, her involvement on my Ph.D. committee in the early stages and her ability to view the problem very practically were appreciated. Dr. Joanne Keith helped me with conceptualization of the research problem. She then very graciously consented to become involved in the project once again when a committee member was unable to attend the oral defense. Her ability to analyze the problem and visualize alternative hypotheses added dimensions to my thinking. Mari Wilhelm and Paul Winder, members of the research team responsible for completing the evaluation phase of "Pilot Project Conserve" in Michigan, spent many hours preparing the data for analysis. Their efforts, along with those of Dr. Harris and Dr. Keith, made my task of analyzing the data used in this study much easier. The College of Human Ecology, the Michigan Energy Administration and the Michigan Agricultural Experiment Station assisted by providing funds which were used to cover expenses involved in analysis, typing and preparation of illustrations. Their generous support was a great help. And finally, but perhaps most importantly, my family deserves a special thank you. My parents, Robert and Betty Merkley; my aunts, Elaine and Jean Merkley; and my brother and his wife, David and Donna Merkley, have been a constant source of support whenever I have undertaken a new endeavor. V ing a Ph.D. and completing this dissertation were only . latest of many challenges which have been met becausé {their love, concern and help have always been available and r " 'il‘l'i'iillingly shared. Chapter I. III. TABLE OF CONTENTS LIST OF TABLES . . . . . . . . LIST OF FIGURES. . . . . . . . INTRODUCTION . . . . . . . . . Energy Shortages Past Experiences with Shortages and Deprivation. . . . . . Problem Statement. . . . Research Objective and Research Questions. Conceptual Framework . . . . Summary. . . . . . . . . . . REVIEW OF THE LITERATURE . . . Dimensions of the Energy Problem as They Relate to Age. . . . Zero- -Order Levels of Analysis. Belief in the reality of the energy problem. . . . . Consumption of energy. . Conservation of energy . Higher—Order Levels of Analysis. the Belief in the reality of problem. . . . . . . Consumption of energy. . Conservation of energy . Summary. . . . . . . . . . . CONCEPTUAL FRAMEWORK . . . . . Temporal Dimensions of the Aging Life Time. . . . . . Psychological aging. . . . . Biological aging . . . . Summary. . . . . . . . . Social Time. . . Social aspects of. aging in States . . . . . . . . Summary. . . . . . . . . energy a o o o o o o Chapter Historical Time. . . . . . . . . . . . . . . Summary. . . . . . . . . . . . . . . . . Historical time and this research. . . . . Structured Research Models Conceptual Research Models . . . . . . . . . Model A. . . . . . . . Model B. . . . Specific Research Models Model A. . . . . . . . Model B. . . . . . . . IV. METHODOLOGY. . . . . . . Sampling and Data Collection Procedures. . . . The Research Subsample . . Measurement Procedures . Independent Variables. . Experience with deprivation and shortages. . . Biological, psychological and social aspects of aging . . Health influences. . Educational influences . . . . . . . . Attitude and belief influences . . . . . Income factors . . Household size and number of rooms . . . Technical modifications. . . . . . . . . Installation of a new furnace. . . . . . Dependent Variables. . Household energy consumption . . . . . . . Weather— —adjusted energy consumption. . . Total change in annual Percentage change in levels . . . . . . Research Hypotheses. . . . Hypothesis One . . . . . Hypothesis Two . . . . . Assumptions. . . . . . . . Analysis . . . . . . . . . V. FINDINGS AND DISCUSSION. Consumption of Energy. . . Hypothesis One . . . consumption levels. annual consumption Bivariate Regression Analysis. . . . . . . . Multiple Regression Analysis . . . . . . . Discussion . . . . . Conservation of Energy . . Hypothesis Two . . . . . Bivariate Regression Analysis. . . . . . . . Multiple Regression Analysis . . . . . . . . Discussion . . . . . . viii o o a o o o o o o a 110 112 112 116 119 120 121 122 124 126 ‘JJHL'; Chapter VI. SUMMARY AND CONCLUSIONS. . . . . . . . . . . . . Overview . . . . . . . . . . . . . . . . . . . Temporal Orientation . . . . . . . . . Need for Age-Energy Use Research . . . . . . Analytical Procedures. . . . . . . . . . . . Conclusions. . . . . . . . . . . . . . . . . . Limitations. . . . . . . . . . . . . . . . . . Implications . . . . . . . . . . . . . . Future Research Efforts. . . . . . . . . . . Educational Implications . . . . . . . . . . BIBLIOGRAPHY. . . . . . . . . . . . . . . . . . . . . . APPENDICES APPENDIX A. Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Subsample. . . . . . . . . . . . . . . . . . . B. Description of Age Stratification of Principal or Oldest Income Earners: Year of Birth, Age in 1977, Years of Experience with Deprivation and Shortages. . . . . . . . . . . . . . . . . C. Tables Describing the Cross-Distributional Characteristics of the Research Sample: Biological, Psychological and Social Variables by the Age Stratification Variable . . . . . . D. Comparison of Total, Direct Btu's Consumed per Heating Degree Day, 1978- 79: Bivariate Regression Analysis and Multiple Regression Analysis . . . . . . . . . . . . . . . . . . . ix Page 128 128 129 133 134 134 137 139 139 142 145 159 170 171 178 LIST OF TABLES l 5 Table Page i 1. Age, Year of Birth and Sample Weighting in . Grand Rapids, Michigan Study, 1976. . . . . . . l7 2. Income Distribution: Comparison of Michigan Households, 1976, and Research Sample, 1978 . . 83 3. Age Characteristics: Comparison of Age of . Household Heads in Michigan, 1976, and Age I of Principal or Oldest Income Earner in the Research Sample, 1978 . . . . . . . . . . . . . 84 4. Educational Attainment: Comparison of Michigan Population, 1976, and Research Sample, 1978 . . 85 5. Form of Tenure: Comparison of Michigan House- holds, 1976, and Research Sample, 1978. . . . . 86 l 6. Number of Rooms in Dwelling Unit: Comparison x of Michigan Households, 1970, and Research E Sample, 1978. . . . . . . . . . . . . . . . . . 87 l l 7. Age Stratification System: Year of Birth, Age ‘ in 1977, Years of Experience with Shortages 2 and/or Deprivation and Sample Size. . . . . . . 92 l 8. Simple Bivariate Regression Analysis of Total, Direct Btu's Consumed per Heating Degree Day, 1976-77, on Independent Variables: Unstan- dardized Regression Coefficients, Standard Errors, Standardized Regression Coefficients, F-Ratios and Probability of Sampling Error. . . 114 9. Multiple Regression Analysis of Total, Direct Btu's Consumed per Heating Degree Day, 1976- 77, on Independent Variables: Unstandardized Regression Coefficients, Standard Errors, Standardized Regression Coefficients, F-Ratios and Probability of Sampling Error . . . . . . . 117 Table Page 10. Simple Bivariate Regression Analysis of Per- centage Change in Btu's Consumed per Heating Degree Day, 1976—77 to 1978—79, on Independent Variables: Unstandardized Regression Coef— ficients, Standard Errors, Standardized Regression Coefficients, F-Ratios and Proba— bility of Sampling Error. . . . . . . . . . . . 123 11. Multiple Regression Analysis of Percentage Change in Btu's Consumed per Heating Degree Day, 1976—77 to 1978—79, on Independent . Variables: Unstandardized Regression Coef- ficients, Standard Errors, Standardized Regression Coefficients, F-Ratios and Proba- bility of Sampling Error. . . . . . . . . . . . 125 A-l. Household by Age of Principal orOldest Income l Earner: Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Sub- Sample and Research Subsample, 1978 . . . . . . 159 A-2. Household by Total Family Income: Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Sub— sample, 1977. . . . . . . . . . . . . . . . . . 160 A—3. Household by Educational Attainment of Principal or Oldest Income Earner: Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Subsample, 1978 . . 161 ”f .5 Household by Number of Members: Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Subsample, 1978 . . . . . . . . . . . . . . . . 162 A-5. Housing by Dwelling Unit Type: Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Sub— sample, 1978. . . . . . . . . . . . . . . . . 163 -.. __.____.__.__. .— .— A-6. Housing by Form of Tenure: Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Subsample, 1978 . . 164 A-7. Housing by Number of Rooms: Comparison of Total 1 Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Subsample, 1978 . . 164 Table Page A—8. Housing by Type of Heating Fuel: Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Sub- sample, 1978 . . . . . . . . . . . . . . . . . 165 A-9 Household by Total Btu's Consumed per Heating Degree Day, 1976—77: Comparison of Total Pilot Conserve Sample, Total Energy Con- sumption Data Subsample and Research Sub— i sample . . . . . . . . . . . . . . . . . . . . 166 i A—lO. Household by Total Btu's Consumed per Heating Degree Day, 1978-79: Comparison of Total Pilot Conserve Sample, Total Energy Con— sumption Data Subsample and Research Sub— sample . . . . . . . . . . . . . . . . . . . . 167 A-ll. Household by Total Change in Btu's Consumed per Heating Degree Day, 1976—77 to 1978—79: Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Subsample . . . . . . . . . . . . 168 A-12. Household by Percent Change in Btu's Consumed per Heating Degree Day, 1976-77 to 1978-79: Comparison of Total Pilot Conserve Sample, 1 Total Energy Consumption Data Subsample and Research Subsample . . . . . . . . . . . . . 169 B-l. Description of Principal or Oldest Income Earners: Year of Birth, Age in 1977, Years of Experience with Deprivation and Shortages and Sample Size. . . . . . . . . . . . . . . . 170 Setting by Age of Principal or Oldest Income Earner . . . . . . . . ... . . . . . . . . 171 I l f ' C-l. Health Reasons for High Winter Temperature l i C—2. Educational Attainment of the Principal or , Oldest Income Earner by Age of Principal { or Oldest Income Earner. . . . . . . . . . . . 172 C-3. Energy Attitudes by Age of Principal or Oldest Income Earner, 1978.. . . . . . . . . . . . 173 C-4. Belief in the Energy Problem by Age of Principal or Oldest Income Earner, 1978.. . . . . . 174 C-5. Income Level by Age of Principal or Oldest Income Earner, 1977. . . . . . . . . . . . . . 175 Table Page C-6. Number of Occupants in the Household by Age of Principal or Oldest Income Earner, 1978. . . 176 C-7. Number of Rooms in the Dwelling Unit by Age of Principal or Oldest Income Earner, 1978. . . 177 D-l. Simple Bivariate Regression Analysis of Total, Direct Btu's Consumed per Heating Degree Day, 1978-79, on Independent Variables: Unstan— dardized Regression Coefficients, Standard Errors, Standardized Regression Coefficients, F-Ratios and Probability of Sampling Error. . . 178 D—2. Multiple Regression Analysis of Total, Direct Btu's Consumed per Heating Degree Day, 1978— 79, on Independent Variables: Unstandardized Regression Coefficients, Standard Errors, Standardized Regression Coefficients, F-Ratios and Probability of Sampling Error. . . 179 xiii '4 L “.2- LIST OF FIGURES Figure Page 1. Illustration of the Influence of Time on Response—Behavior Variables, Organismic— Biological Variables or Stimulus— Environmental Variables as They Account for Behavior; Adapted from Baltes, 1973, p.460. . 44 i 2. Illustration of the Relationship between 1 Life-Span Development and Three Major Influence Systems: Ontogenetic (Age-Graded), Evolutionary (History—Graded) and Non— Normative (Non—Developmental); Adapted from Baltes and Willis, 1979, p. 24. . . . . . . . 45 3. Five Year Birth Cohorts, by Stage in the Life Course, as They Experienced World War I, the Great Depression and World War II . . . . 64 4. Identification of 15 Counties in the Mid- ; Michigan Area Targeted for the Computerized | Residential Energy Audit Program, "Pilot ' Project Conserve," Winter 1977-78 . . . . . . 75 i 5. Comparison of Mean Total Btu's Consumed per Heating Degree Day, 1976—77 to 1978—79, by Age Group. . . . . . . . . . . . . . . . . 99 6. Comparison of Mean Total Btu's Consumed per Heating Degree Day, 1976—77 to 1978-79, by Age. . . . . . . . . . . . . . . . . . . . 101 7. Mean Percentage Change in Btu's Consumed per . Heating Degree Day, 1976—77 to 1978-79, by Age Group. . . . . . . . . . . . . 104 8. Mean Percentage Change in Btu's Consumed per Heating Degree Day, 1976-77 to 1978-79, by Age. . . . . . . . . . . . . . . . . . . . 105 - aim-1.. CHAPTER I INTRODUCTION Energy Shortages Freezing to death in Ohio, or dying from heat pros— tration in Texas and Missouri; remaining in bed all day, or wearing overshoes inside your home in order to stay warm in New York; turning the oven on high and opening the door for heat in South Dakota and Michigan, or heating your dwelling from 60—680F as opposed to 72-780 10 years earlier; giving up hot food, using a flashlight for lighting, or having friends refuse to visit because your house is dark and cold, all in New York; waiting in line for gasoline in California, or paying $1.32 a gallon for gasoline that 10 years ago cost $.32, and thus foregoing a vacation in Iowa. A mixture of horror stories combined with some minor inconveniences, perhaps; yet these represent the personal experiences some Americans have faced since the Arab Oil Embargo precipitated a national energy crisis in the winter of 1973-74. For that first initial encounter with deficits in petroleum supplies was followed, after a short respite, by nearly annual occurrences of energy shortages. During the winter of 1976, for example, natural gas sup- plies were not available in levels sufficient to heat homes supplied with gas. A year later, in the winter of 1977, the second crisis was followed by a third characterized by still more intense shortages of natural gas. This time, however, the situation was exacerbated by shortages of coal. Finally, during the summer of 1979, Americans faced renewed deficits of petroleum and gasoline was in short supply. Thus the nation watched as nightly television reports showed Cali— fornians queuing in long lines to get their allotted ration of gasoline. In a very general sense, then, the overall impression of the "energy crisis" conveyed to the American public via mass media or through personal experiences with closed gas— oline stations or cold, dark homes has been one of shortages in the supply of commodities very much needed and demanded: shortages of natural resources that have come to be regarded as essential life—support systems and, perhaps most impor- tantly, resources that have been inexpensive and readily available for well over a quarter of a century. Past Experiences with Shortages and Deprivation Viewed from this perspective, it was the argument of this research that the current "energy crisis" was, in fact, quite similar in character to other twentieth century crises marked by shortages of essential resources or personal depri— vation and thus inability to obtain goods and materials. During World War I, for example, as employment levels increased and incomes rose under the impact of an expanding wartime economy, Americans were asked to curtail consumption of needed resources in support of the war effort. Indeed, concern at the national level was so great regarding declin- ing stocks of foodstuffs and fuels that rationing, limiting the use of sugar, wheat, meat, butter and other foods, was imposed, as were meatlessanuiwheatless days. In addition, shortages in supplies of fuel, especially coal, were acute, and Americans were asked, first of all, to curtail unneces- sary lighting and heating to extend supplies. In April 1918 more stringent regulations were implemented and general rationing to domestic users was introduced. Prior to this, fuel oil distribution had been controlled, and within months regulation was extended to natural gas and gasoline (Scheiber et a1., 1976, pp. 317—328). During World War II, with shortages in raw materials again acute and with World War I as an example, the federal government implemented rationing for a second time. Within a month after the war started distribution of tires was con- trolled due to shortages in rubber, while rationing of gaso- line and fuel oil followed shortly thereafter. Food short- ages were common within months, with sugar and coffee both in short supply due to interruptions in shipping from export- ing nations in the Far East and South America. Rationing of each started early in the war, although coffee restrictions were lifted 8 months later. Then, towards the end of 4W. .._.______- i l 1942, supplies of more basic foodstuffs began to dwindle as large quantities were shipped overseas to feed the armed forces. Canned, bottled, frozen and dried vegetables, fruits, juices and soups were thus rationed after March 1943. The list was expanded shortly to include meats and fats and then expanded still further to encompass nearly all food commod- ities. Scheiber et a1. (1976) indicated, for example, that by mid-1943, 95% of the food supply was rationed which, they pointed out, ensured fairer distribution, but by no means solved the problem (see especially, pp. 407-411; see also Walton & Miller, 1978, pp. 139-145). In many respects the character of still a third twen- tieth century event, the Depression of the 19305, was similar to yet different from that of the two World Wars. During both wars, a sense of shared responsibility and commitment to doing without in support of a national cause appeared to have been strong motives in bringing about national consensus; the urgency of the war situation thus perhaps mitigating some of the psychological impact of the forced shortages (see, for example, Clark et a1., 1977, pp. 316—317). In addition, severe hardship, in the real sense, was perhaps not realized during the war years because employment levels were high and the stability of having a job, thus, may have provided a sense of security about the future, somewhat softening the impact of shortages (see Clark et a1., 1977, pp. 117—151, especially p. 143, for comments regarding the effect of World War I on American families). ._u4‘~.—~.~—— —..__._- - The impact of the Depression, on the other hand, was different. Unemployment reached unprecedented levels with 4.34 million jobless in 1930, 8.02 million in 1931, 12.06 million in 1932 and nearly 13 million in 1933-- approximately one-fourth of the civilian labor force1 (Scheiber et a1., 1976, p. 357). Concurrently, hours of work were cut for those still employed and wages fell. Allen (1969) commented that in 1932 the total amount of money paid in wages equaled 40% of the 1929 level (p. 130), while Scheiber et a1. (1976) indicated that wage—rate cuts and hour reduction accounted for greater drops in payrolls than unemployment (p. 357). In desperation, numerous house— holds turned to public assistance for the first time. Scheiber et al. estimated that by late 1934 at least 17 mil- lion familieswerereceiving public help or welfare (p. 357). Judgment concerning overall impact of the Depression was difficult. Elder (1974), in one study of the subject, pointed out that all families were not affected directly, nor with equal severity (pp. 43—53). Manchester (1979), in describing General MacArthur's order to shoot on the bonus marchers demonstrating in Washington in 1932 for back pay from service in World War I (pp. 3-18), or in detailing the actions of a group of desperate Iowa farmers who kidnapped and beat a local county judge to force a pledge that he 1Walton and Miller (1978) indicated these estimates may be low, as precise figures on labor force participation were not available (p. 126). would sign no more mortgage foreclosures (p. 59), was more impassioned, as was Steinbeck in his description of the Depression in The Grapes of Wrath. Perhaps Allen‘s (1969) analysis somewhat captured the character of the Depression: It marked millions of people——inward1y-—for the rest of their lives. Not only because they or their friends lost jobs, saw their careers bro- ken, had to change their whole way of living, were gnawed at by a constant lurking fear of worse things yet, and in all too many cases actually went hungry; but because what was hap- pening to them seemed without rhyme or reason. Most of them had been brought up to feel that if you worked hard and well, and otherwise behaved yourself, you would be rewarded by good fortune. Here were failure and defeat and want visiting the energetic along with the feckless, the able along with the unable, the virtuous along with the irresponsible. They found their fortunes interlocked with those of great numbers of other people in a pattern complex beyond their under— standing, and apparently developing without reason or justice. (p. 131) Along with this suffering, surpluses in food, at least, abounded as farmers actually increased production by 3% between 1929 and 1932 in order to make up for a 70% decline in net income (Walton & Miller, 1978, pp. 126-127). Excesses consumers could not purchase piled up and led to the eventual destruction of surplus farm products in order to equalize supply and demand and thus raise price levels (Scheiber et a1., 1976, pp. 357-358). The Depression, then, was characterized by hardship and deprivation, just as World Wars I and II were characterized by shortages. Each event, in general, thus bore a basic resemblance to the present energy crisis situation. During each of the earlier periods, for example, resources were denied to large numbers in the general population; during the war years, because supplies of basic consumer goods such as food and fuel were insufficient; during the Depression, because incomes were reduced and thus ability to obtain resources curtailed. In addition, each past event was of major national significance; few in the society could have missed the implications of each situation, although direct experience may not necessarily have resulted in lowered resource acquisition (see, for example, Elder, 1974, pp. 18-20, 43—53). Likewise, during the current situation, as it has developed since 1973, shortages and increased prices of energy have been evident, thus limiting access to supplies by some groups (Unseld et a1., 1978). In addition, the problem has received national and local media attention. Few today remain ignorant of the general implications of the energy problem, whether they believe the situation is real or contrived. Problem Statement In light of the similarities between the current energy problem and the experiences with shortages and deprivation characterized by World Wars I and II and the Great Depression, it was the premise of this research that people exposed to these three past events would be influenced by attitudes, beliefs and perceptions gained from that set of experiences and further, that this influence would be manifest in cur- rent behavior exhibited toward the energy crisis. The primary research problem, therefore, was to discern if past experience with deprivation and shortages influenced current patterns of energy consumption and conservation. Research Objective and Research Questian The specific research objective was to determine to what extent people exposed to varying levels of past expe- rience with deprivation and shortages differed in their reactions to the current energy situation, with its impli- cations for energy shortages and thus deprivation with respect to energy supplies. Two questions, in particular, guided the study: (1) Did exposure to shortages and depri- vation in the past influence current energy consumption behavior? (2) Did exposure to shortages and deprivation in the past influence change in energy consumption behavior over time and, most particularly, reduced consumption (i.e., conservation) behavior, given the urgency placed upon conservation at the local and national levels? (Stobaugh & Yergin, 1979; Unseld et a1., 1979) Conceptual Framework Inherent within this study was consideration of the concept of age, because of the direct positive relationship existing between age and level of past experience with shortages and deprivation. In order to have experienced the three historical events considered in this research, indi- viduals necessarily must be in their sixties or older now A _ _-_._ - v.—.— ,_. .‘I‘l| 'that impact from the current energy situation was being felt. On the other hand, individuals with no prior experience involving national crises marked by shortages or deprivation would presently be in their thirties or younger. Between these extremes, level of experience varies in a unidirec- tional manner, i.e., increasingly older individuals having had more experience, younger individuals less. Age has come to be viewed as a difficult analytical tool to utilize in research, however. As Baltes and Willis (1977) and Lerner and Ryff (1978) have pointed out, in and of itself, age measured nothing and was, in fact, a noncausal variable. It became useful in empirical studies, they indicated, only when it served aS’a marker along which experience occurred (Lerner & Ryff, 1978, p. 10). Within the context of this research, experience was multifaceted, however, encompassing not only past experience with shortages and deprivation, but experiences on dimensions of the aging process linked to biological, psychological and social aspects of aging. What complicated the situation was the fact that energy consumption and conservation were also believed related to these three aspects of aging (i.e., biological, psychological and social aging,(Newman & Day, 1974, 1975; Morrison et al., 1978b, 1979; Farhar et a1., 1979)). Higher income levels, for example, were related to higher energy consumption pat- terns, just as they were related to middle-aged as opposed to younger or older age levels. Size of household was also related to consumption of energy in a positive way. Its 10 relationship to aging was curvilinear, however, with younger and older households being smaller, and size of household expanding and then contracting throughout the middle years of adulthood. In essence, then, a paradox existed. Aging defined past experience with shortages and deprivation and, in turn, was correlated with characteristics with known relationships to energy use. To the extent that these confounding linkages could have masked the relationship of interest, analysis was problematic. It was necessary, therefore, that each factor related to both aging and energy consumption and conservation be considered and controlled as past experience was studied. The issue of systematically appraising age—energy rela- tionships was most easily resolved by the development of a temporally oriented conceptual framework, capable of allowing separation of the interrelated aspects of aging, as well as specification of their characteristic linkages to energy use. Specifically, three dimensions of time were considered. Along with their aging-related components, the three temporal dimensions were: (1) life time or the measurement of aging as a psychological and biological process, (2) social time or social aspects of aging, which focused on consideration of stage in life and the process of accepting and relin- quishing roles and responsibilities defining the various stages, and finally, (3) historical time or the locational juxtaposition of life time within the span of time sur— rounding it. -._- --_ - W-“ w—.—-. _ 11 Summary This introductory statement has focused on describing similarities between the current energy situation and three twentieth century historical events: World War I, the Great Depression and World War II. The parallels identified were: (1) shortages of basic life-supporting resources, or (2) inability to obtain basic resources, and (3) major national impact. It was pointed out that during each of these four periods, resources have been denied to households within the general population. While in each past instance, there may have been households not actually deprived, each historical event was of such major significance that few experiencing it could have remained ignorant of its implications. Like- wise, dramatic increases in energy prices and shortages as well as extended media coverage have left few unaware of the present day energy situation, even though some households have not been forced to curtail consumption of energy as a result of shortages or price increases. The research problem and research questions were then identified. The study was designed to determine whether living through these three historical events characterized by deprivation or shortages had influenced current energy consumption and conservation behavior and, if so, how and to what extent. Problems particular to the study were specified as including: (1) the need to focus on the age variable as a means of determining level of experience with historical 12 events, (2) the analytical problems inherent when studying the age variable,and (3) the compounded problems of studying energy use behavior in relation to the age variable. A brief summary of the conceptual framework developed to help over— come these problems was specified last. In the following chapters, the conceptual framework is explored in detail, as are the methods used to operationalize and measure the component elements of the framework. First, however, an analysis of the current state of knowledge con- cerning age-energy research is presented. W-__—_ ____1 .__.-- CHAPTER II REVIEW OF THE LITERATURE Dimensions of the Energy Problem as They Relate to Age An exploratory review of the literature revealed that no information existed exploring the relationships between past experience with deprivation and shortages and current energy consumption and conservation behavior. An abundance of information was available, however, analyzing the rela— tionships between age and a variety of energy-related issues. For a complete discussion the reader should consult Farhar et a1. (1979), as this group has completed a comprehensive and timely analysis of 114 energy surveys conducted at the national, state and local levels since the Arab Oil Embargo in 1973—74.1 lFarhar et a1. (1979) analyzed the 114 surveys attempting to discern relationships between age and a broader range of questions than were applicable for discussion in this review. Specifically, they studied the surveys to identify relation— ships between aging and the following issues: (1) perceptions of the general situation, (2) expectations about the future, (3) perceived impact of the situation, (4) policy preferences, (5) energy and the environment, (6) knowledgeability and sources of information, (7) feelings about solar energy, (8) feelings about nuclear energy, and finally (9) energy conservation (Farhar et a1., 1979, pp. B-20 to B-4l). 13 I ~—- ._-____— _.__... 14 In conducting their analysis Farhar et al. found that numerous studies included age level analysis of survey results to provide basic sociodemographic descriptions of their sample. Yet Farhar et al. also indicated that con- fusion existed concerning age level differences with respect to behavior and attitudes towards the energy situation. In effect, no clear and consistent age—related patterns were apparent. They hypothesized that problems in discerning differences were due to the varying number and variety of age categories employed in the surveys. They also indicated that the level of analytical method employed in the surveys may have masked differences. The majority of surveys relied on zero-order levels of analysis, and to the extent that factors with identified links to energy behavior and attitudes were interrelated with age level, they could have confounded the aging patterns (Farhar et a1., 1979, pp. B—20 to B—21). To help understand the analytical problems identified by Farhar et al., the literature review was separated into two broad categories: (1) findings exploring the zero-order relationship between age and various energy issues and (2) findings exploring the relationship by using more power- ful analytical tools, which allowed separation of the impact of age from variables linked to both age and the particular issue under study. For ease in interpreting the findings, the literature was further organized into sub—groupings. Three questions in particular were investigated because of their direct 15 association with this research effort. They included: (1) belief in the reality of the energy problem, (2) actual and reported energy consumption behavior, and (3) measured or self-reported energy conservation behavior. Zero—Order Levels of Analysis As Farhar et a1. (1979) pointed out, the majority of information exploring relationships between age and energy use or attitudes had been analyzed by cross-tabulation pro- cedures. The findings were somewhat contradictory. They were useful, however, because they indicated general patterns. Belief in the reality of the energy problem Morrison et a1. (1979) studied the relationship between age and belief in the reality of the energy problem. Their findings indicated that over time older respondents, born before or during the earlier years of the Great Depression, appeared to have become less willing to acknowledge that an energy problem actually existed. In 1974, for example, when asked if they believed the energy problem was real, 50% (n = 169) of older respondents (40 and over or born duringtnrbefore 1933) indicated yes, while 51% (n = 236) of younger respon- dents (less than 40 or born after 1933) indicated yes. By 1976, however, only 39% (n = 243) of older respondents (40 and over or born during or before 1935) indicated willingness to believe the energy problem was real, while 50% (n = 248) of younger respondents indicated belief in the reality of the problem (p. 15). 16 Morrison et a1. (1979) found that high percentages in both age groups expected an energy problem to emerge in the near future, however. Fifty—eight percent of the older group in the 1976 survey expected an energy problem to exist by 1981, while 64% of the younger group concurred. Percentages were still higher for both groups concerning a belief that a problem would exist by 1985, 70% and 69% respectively (p. 15). Thompson and MacTavish (n.d.) found similar results regarding belief in the reality of energy problems when they surveyed a group of Grand Rapids, Michigan residents. Their survey was conducted in February 1976, 2 months prior to the second phase of the Morrison et a1. (1979) study. The Thompson and MacTavish study differed from the Morrison et a1. study in two important ways, however. A broader range of age-belief questions were studied,and more age categories were employed to analyze sample responses. Thompson and MacTavish employed four age categories, for example, in com- parison to Morrison et al., who used only two. The cate- gories and the approximate years of birth of sample members in the Thompson and MacTavish study are shown in Table 1. With respect to age patterns and belief in the energy crisis, Thompson and MacTavish found that, in general, older respondents were more reluctant to believe a problem existed, were more cynical about its causes and were more pessimistic about the future. Overall, for example, respondents believed an energy problem currently existed (63% - yes; 28% - no). Yet the 45 to 64 year age group was less willing to 17 Table l.--Age, Year of Birth and Sample Weighting in Grand Rapids, Michigan Study, 1976a Age Year of Birtha N = 515 19 thru 24 1951 to 1956 14% 25 thru 44 1931 to 1950 42% 45 thru 64 1911 to 1930 27% 65 and over 1910 or before 17% Source: Phyllis T. Thompson and John MacTavish, "Energy Problems: Public Beliefs, Attitudes and Behaviors" (Allen— dale, Mi: Urban and Environmental Studies Institute, Grand Valley Colleges, 1976, mimeographed), p. iii. aYears of birth are approximate, as they were estab— lished by the author for purposes of comparison. The base year used to determine year of birth was 1975, as the survey was conducted early in 1976. acknowledge existence of the problem (55% — yes; 34% —no (p. 6)). Concerning belief in future energy problems emerging, the Thompson and MacTavish findings differed from those of Morrison et a1. (1979) because the percentage believing that there would be future problems decreased with age. A cor— responding increase in the "don't know" category emerged as age increased (p. 10). Linked to belief in future energy problems, Thompson and MacTavish hypothesized, was the belief that oil and gas supplies in the United States would be exhausted and that shortages of these products would be prevalent in the future. Young people were more likely to expect oil and gas supply depletion in the future. This 18 expectation consistently decreased with age (pp. 19-20). Likewise, the expectation of future shortages, in general, decreased with advancing age, i.e., over 65 (p. 47). Expectations regarding increases in energy price levels revealed similar relationships. Anticipation of major price elevations decreased with increases in age (pp. 31-32). Linked to this was the finding that the 25 to 44 age group was most likely to expect utility price increases in 10 years, while the elderly least expected utility costs to increase (p. 34). Thompson and MacTavish hypothesized that these cost findings were possibly a "sign of wishful thinking on the part of older respondents," as incomes were likely to be fixed for this age group and, thus, unable to keep up with price increases (pp. 31—32). The elderly, Thompson and MacTavish seemed to feel, were simply blocking reality due to their economic situation. In citing reasons that utility bills might increase, the age groups differed as well. The under 25 group empha- sized shortages and increased demand for energy, while they de-emphasized greed and production costs. The 25 to 44 age group emphasized increased demand and production costs; they de-emphasized greed. The 45 to 64 age group showed a slight tendency to emphasize greed, while de—emphasizing increased demand and shortages of supplies. The elderly, however, emphasized greed as a major cause of utility price increases; they de-emphasized increased demand (p. 34). 19 With respect to the possibility of solutions to the problem, fewer older individuals than expected felt answers ‘ would be found. This corresponded with the finding that ‘ older people, in general, were more pessimistic about the g future. As age increased, expectations that solutions would be found decreased, while expectations that solutions would not be found increased (p. 13).2 One final interesting finding concerned the fact that all age groups indicated they did not believe a shortage "caused" the gasoline crisis of 1973-74 (i.e., they appar— ently felt it was a fraud). The elderly, in particular, were likely to indicate that shortages had not caused the crisis, however (p. 25). In general, it appeared that as age level increased belief in the seriousness of energy supply problems decreased. This same pattern also seemed evident regarding the prospect of future shortages. Farhar et a1. (1979), for example, when summarizing findings from a broader variety of surveys, I I indicated that younger age groups appeared more likely to i report greater belief in the possibility of future shortages (see pp. B-21 to B—24, B-40). Consumption of energy Newman and Day (1974) indicated that analysis of actual energy use on a per household basis suggested that consumption 2Thompson and MacTavish offered no explanation of the inconsistency between this finding and the finding indicating that older individuals did not know if energy shortages would be prevalent in the future. A 20 increased if children were present, decreased if they were not. They concluded that younger and elderly households thus consumed less energy (p. 6). Support for their findings came from an energy data bank tracing a 1 year period of energy consumption behavior for a national sample of households.3 Surveying consumption of natural gas and electricity usage for the year ending May to July 1973, Newman and Day found that average annual con- sumption equaled 234.9 million British Thermal Units (Btu's). Households headed by persons under 45 (approximate birth date‘ after 1927) used an average of 7% more than this overall mean amount; if they included children, they consumed 16% more than the national average. On the other hand, if no children were present, this same age group used 36% less than average. If the head was over 45 years of age (birth date prior to 1927), the household consumed 4% less than the national mean. Finally, elderly households, headed by people over 65 (birth date 1907 or before), used 18% less than the national average (Newman & Day, 1974, Table 1). The findings of Morrison et a1. (1979), based upon a sample in a geographically distinct area as opposed to a national sample, confirmed Newman and Day's conclusions. Studying total, direct energy consumption patterns for the 3A description of the procedures used to select the national income-stratified sample of 1,455 households and the consumption data subsample were discussed in Newman and Day, 1975 (see especially Appendix A-2, pp. 237—265). 21 1 year period from 1973-74, Morrison et al. found that a curvilinear relationship seemed to exist with age level of the household head. The data represented a sample of households in the Lansing, Michigan Standard Metropolitan Statistical Area. Overall, average annual consumption was 207.83 million Btu's (p. 22). Households headed by persons less than 30 (n = 8, approximate birth date after 1943) consumed 150.83 million Btu's on the average. Households headed by people between 30 and 45 (n = 53, approximate birth date between 1928 and 1943) used 228.82 million Btu's, while households headed by persons over 45 (n = 69, approximate birth date prior to 1928) consumed 198.32 million Btu's. In general, it seemed that a curvilinear relationship existed between age and energy consumption. It was apparent, however, that in both of these studies age acted as a surro- gate for stage in the family life cycle (i.e., the temporal pattern of family life emerging as the size and composition of the unit changes,(Zimmerman, n.d.)). And to the extent that these patterns were captured, understanding of age and the characteristic it measures were masked. Conservation of energy The literature varied concerning the relationship>between age and energy conservation. Some studies, for example, indi— cated no relationship existed between age and self-reported con- servation behavior (Kilkeary & Thompson, 1975),while others indicated a relationship was evident. If a relationship was 22 identified, the direction of the association varied. Some studies, for example, found individuals over 65 conserving little if any energy (Smith, 1976), while others, based on aCtual consumption data, indicated major conservation was occurring in households headed by individuals over 65 (Mor— rison et a1., 1979; Williams et al.,1979). In addition, some studies documented that middle—aged groups were con- serving, while others indicated they were not. Finally, methods used to obtain results varied. Both methodology and results will be discussed in this section, therefore. During the summer of 1974, Kilkeary and Thompson (1975) surveyed middle and working class households in two New York City communities to determine self—reported conservation behavior. The two communities were selected purposefully; one because it had experienced an extended power failure in the previous summer; while the second, because it had not and could therefore serve as a control sample. To assure com- parability, the second community was chosen based upon similarity to the first in socioeconomic and ethnic char- acter (pp. 4, 6). In addition, an attempt was made to assure that samples within communities were randomly selected by using U.S. Census Bureau Block Statistics to determine households to be interviewed by personal visit (p. 6). How- ever, due to circumstances particular to the study (i.e., urban distrust and'concern regarding allowing entry of strangers), much of the control was lost as interviewers 23 were forced to begin arbitrarily interviewing willing respon- dents from groups on the streets (p. 7). The final sample consisted of 602 households, uniformly represented in each of six age strata (e.g., 18-25, n = 117; 26-35, n = 136; 36-45, n = 87; 46-55, n = 85; 56-65, n = 76; 66 or older, n = 90),4 while age was one of the major independent variables hypothesized related to self-reported conservation behavior. Chi square analysis revealed, how- ever, that no relationship existed between the two (p. 48). This finding was interesting and somewhat problematic, as it contradicted results in other studies that indicated a relationship was evident between age and conservation behavior (see below). The Kilkeary and Thompson finding could, perhaps, simply have been a matter of sampling error. On the other hand, problems existed in their study which may have accounted for their inability to identify a relation— ship. Respondent selection, for example, did not allow for random sampling, as people "out-of-doors," (p. 7) willing to be interviewed, were interviewed and then winnowed if answers seemed to indicate a lack of interest in the inter- view (p. 4). Response on the conservation scale may have been uniformly high across age groups as a result and, thus, analysis incapable of detecting a relationship that, in fact, could have been identified in a more representative sample. 4Eleven respondents refused to give their age. 24 A second problem identified in the Kilkeary and Thomp— son study concerned measurement reliability. The study used self-reported practices to index conservation behavior. Yet Milstein (1978) indicated that reported actions were a poor measure at best, as respondents, aware of the social desir- ability of conservation, tended to overestimate their activ- ity. Thus Kilkeary and Thompson may have missed identifying a relationship because nonconserving age groups may have been reticent to report their behavior. If this was the case, it was interesting to note that Smith (1976) found evidence of a curvilinear relationship between age and reported conservation practices. His data base seemed more representative as the 1,400 households studied were selected via a national probability sampling technique, and thus the findings would appear to have more validity. One additional difference between the Smith and Kilkeary and Thompson studies which may have influenced results concerned interviewing technique. Smith's inter- views were apparently conducted by telephone rather than personally, and this more remote contact may have induced respondents to feel somewhat freer to admit nonconservation behavior.5 In any event, Smith reported a curvilinear relationship between self-reported conservation behavior and age. He 5Milstein (1978), however, argued against self-reported behavior in general as a measure of conservation activity, making no allowance for interview technique. 25 indicated that 27% of the age group under 25 (approximate birth date during or after 1950) reported making no change in energy use behavior between 1974 and 1975, while 49% of the 65 and over age category (birth date during or prior to 1910) reported no change in behavior. Fewer in the middle age ranges indicated no change in energy consumption pat- terns: 17% in the 25 to 34 group (approximate birth date between 1941 and 1949), 21% in the 35 to 49 group (approx- imate birth date 1926 to 1940), and 22% in the 50 to 64 group (approximate birth date 1911 to 1925 (p. 5)). Smith indicated one additional finding which had impli— cations for this research. Respondents in the elderly age group who had not made adjustments in living patterns were questioned about the future. And Smith reported that "a significant number" indicated that a further 25% increase in» energy prices would not influence them to conserve (p. 3). The findings of Williams et al. (1979) concerning con— servation somewhat contradicted Smith's findings. Like Smith, Williams et a1. documented a relationship between age and conservation behavior. The discrepancy came in direction of the relationship. Williams et al. proposed that households headed by individuals 45 to 64 have conserved minimal amounts, while those headed by people 65 and over have shown major, rather than minor, changes in energy use patterns. No evidence was presented regarding conservation behavior of age groups under 45. .. ...\~LA'.. ‘ 26 Findings in the Williams et a1. study were based on energy consumption levels of a national sample of households in 1972-73, compared to consumption level in 1974—75.6 Although change between the two periods was discussed, two independent sets of cross-sectional data, rather than longi- tudinal data, were used for comparison in the analysis. Thus, the same households were not studied over time. In addition, the sample was urban in character, while only con- sumption patterns of households utilizing natural gas and electricity were analyzed. Williams et al. indicated that, overall, there was a reduction of 1.8% in total, annual Btu's consumed between 1972-73 and 1974-75, with conservation of natural gas being the major contributor. Natural gas consumption was reduced 3.9%, for example, while consumption of electricity actually increased by 1.2%. The findings related to age revealed divergent patterns. Those over 65 (approximate birth date 1909 or before) reduced overall consumption by an average of 7.1%, while those between 45 and 64 (approximate birth date between 1910 and 1929) increased consumption by 1.4%. Increased use of electricity accounted for most of the over- all increase within the 45 to 64 age group. They increased electrical consumption by 3.0%, for example, while those over 65 reduced electrical consumption 7.4%. Use of natural 6Base year data were the same as those analyzed by New- man and Day (1974, 1975), discussed earlier in the section on energy consumption patterns. 27 gas showed similar patterns; those 45 to 64 increased con- sumption by an average of .6%, while those over 65 decreased consumption by 6.9% (p. 45). The final research studies to be discussed linking age to energy conservation were reported by Morrison et al. (1978b, 1979).7 And the finding once again contradicted those presented above. Morrison et al. found a distinct linear relationship between age and conservation, ranging across three age levels from less than 30 to over 45. In addition, they found that the relationship was positive, suggesting that as age level increased, conservation increased; rather than increasing only in the older years as reported by Williams et a1. (1979); or only in the mid— dle years as Smith (1976) indicated. Methodologically, the Morrison et a1. (1979) study was similar to the Williams et a1. (1979) study. The dependent variable, for example, was change in actual consumption behavior over a 3 year period. The time periods differed, however, as Morrison et al. studied conservation patterns between 1973-74 and 1975-76, while Williams et a1. studied change between 1972-73 and 1974-75. In addition, geograph- ical focus differed, as Morrison et a1. concentrated on the more limited Lansing, Michigan Standard Metropolitan Statis- tical Area. Finally, methods of selection and design differed. 7The Morrison et a1. (1979) study was discussed earlier in the section on energy consumption. 28 Morrison et a1. studied households using a wider range of fuel sources than Williams et a1., and Morrison et al. utilized a modified panel design. The same households, therefore, were studied over the 3 year period (p. 12). In general, Morrison et al. found a reduction of 6.3% in total, direct annual energy consumption between 1973—74 and 1975—76. More specifically with respect to age, they found that households headed by persons less than 30 (n = 8, approximate birth date after 1945) decreased consumption by 1.9% on the average, while households with heads between 30 and 45 (n = 53, approximate birth dates being 1930 to 1945) decreased consumption by 4.9% on the average. Finally, households over 45 (n = 69, approximate birth date prior to 1930) conserved an average of 6.35% (p. 22). Thus, their evidence indicated that conservation was occurring across all age levels, although it seemed more pronounced as age level increased. Morrison et al. also attempted to determine household members' perceptions of the difficulty involved in future conservation efforts. The entire 1976 sample of male and female respondents was utilized to answer these questions. In general, the prospect of additional conservation was not alarming, as most sample members felt it could be accom- plished with little disruption to lifestyles. With regard to age, however, individuals 40 or over (n = 243, approximate birth date during or prior to 1935) viewed anticipated con— servation with somewhat more reluctance. Thirty-three 29 percent, for example, felt it would be very difficult to reduce miles currently driven, while 24% of those under 40 (n = 248, approximate birth date during or after 1936) felt reduced driving would cause hardships. In addition, 26% of those 40 or over felt it would be burdensome to reduce the amount of electricity they currently used. In comparison, 11% of those under 40 felt this would be a difficult change. The only prospect that concerned those under 40 more than those over 40 was reducing consumption of material goods. Twelve percent under 40 felt this would be a difficult alternative, while 6% over 40 were concerned about the impact of reduced consumption levels. Thus, it seemed that even though future conservation did not appear to concern sample members, older respondents, in general, were somewhat less likely to feel it could be accomplished easily. In carrying analysis of the Lansing, Michigan data base still further, Morrison et al. (1978b) utilized a discriminant analysis technique to profile factors characterizing conserver as opposed to non-conserver households. In general, they found conservers had higher incomes ($15,000 annual gross income), while non-conservers were characterized by lower incomes($lZ,600 gross annual income). In addition, they found that conservers had slightly more education (13.4 years), while non—conservers had slightly less (13 years).‘ With respect to age, they found that conservers were slightly younger than non-conservers. The measure used to compare households was the average age of the two spouses. 30 In aggregate, the conserver group of households averaged 48.6 years of age, while the non-conserver group averaged 53 years. Difference in age was not a statistically sig- nificant characteristic between the two groups, however. Detailed discussion of these studies was included to illustrate the diversity of information available on the age-conservation question. Not only did findings differ, but measures of conservation and age differed, as did methods employed to gather information. Farhar et al. (1979), having surveyed and summarized a much wider selection of material, indicated that, in general, they found the same variation. They concluded, as a result, that there appeared to be no obvious age-related patterns in energy conservation behavior (pp. B-34 to B-38, B-4l). It was somewhat difficult to concur with Farhar et a1., however, because they reached this conclusion by relying mainly on studies focusing on self-reported behavioral change to document conservation behavior, rather than studies using actual consumption data. When actual data were used, as they were in the Williams et a1. (1979) and Morrison et a1. (1979) studies, rather evident age-related differences existed. The older age groups in each study appeared to be conserving more energy. Patterns of differences in actual conservation behavior of middle and younger age groups were less apparent. A basic problem existed in these two studies, however, as it did with others reviewed in this section which attempted to make generalizations regarding relationships 31 between age level and beliefiJithe realitycflfthe energypmoblem or energy consumption or conservation patterns. Analysis in each of the studies was conducted at the zero-order level, and to the extent that age was intercorrelated with other sociodemographic characteristics with identified relation- ships to various energy questions, its effect was confounded in each study. Morrison et al. (1978b), for example, indi- cated that higher income level and higher educational level were both significant predictors of energy conservation behav- ior. These two variables are related to age as well, and this multicollinearity makes the net contribution of each variable difficult to assess. Higher-Order Levels of Analysis Few studies were identified which explored the relation- ships between age and various energy issues by utilizing higher-order levels of analysis. When used, these procedures have enabled the impact of various independent variables to be analyzed simultaneously, along with allowing the net effect of each to be studied. Thus, they have permitted greater understanding of the separate contribution of each variable. Belief in the reality of the energy problem No studies were identified utilizing higher-order levels of analysis to measure the impact of age on belief in the reality of the energy problem. 32 Consumption of energy Gladhart (1977a) studied a number of factors to estimate both annual demand for electricity and total residential energy consumption.8 Included in the analysis were two age- related variables: (1) head of household 35 or under (born in 1940 or after) and (2) head of household 60 or over (born in 1915 or before). Both age factors were included as dummy variables, with the characteristic being given a value of 1 if it was present, a value of 0 if absent. Gladhart found a negative relationship between head of household under 36 years and electrical energy consumption (B = -ll.14, p = .045). A negative relationship was also identified between head of household over 59 and electrical energy consumption. At the .12 probability level, however, the finding was not statistically significant (B = -9.93 (Gladhart, 1977a, Table 1)). The final question concerned demand for total residential energy. With respect to age, Gladhart found a negative rela- tionship between head of household under 36 and total energy use. The relationship was not statistically significant, however (B = ~12.45, p = .124). Head of household over 59 was not included in this analysis (Gladhart, 1977a, Table 3). 8Gladhart‘s analysis was based on the Lansing, Michigan data utilized by Morrison et al. (1978b, 1979). To predict consumption levels, Gladhart analyzed sociodemographic characteristics and energy consumption patterns collected in the 1976 wave of interviews. 33 Any generalizations concerning a relationship between actual use of energy and age based upon one study would be tenuous. Yet Gladhart's findings seemed to suggest that younger households consumed less energy. He indicated that family life cycle factors may account for this finding, with younger households using less energy because children are younger and thus consuming less independently of parents (p. 10). Conservation of energy Hogan (1976), utilizing the same sample as Morrison et al. (1978b, 1979) and Gladhart (1977a),9 analyzed a number of sociodemographic variables in order to determine their influence on reported energy conservation behavior. Included as predictor variables in the regression analysis were: (1) husband's age, (2) wife's age, (3) wife's edu- cation, (4) husband's education, (5) husband's occupation, (6) wife's employment status, (7) family income, (8) family size, and (9) urban or rural residence. Only wife's edu- cation was identified as having a statistically significant relationship with conservation behavior (p = .03). Wife's 10 age and husband's age did not appear to be related (B = .02, p = .81; B =.03, p = .88, respectively (Hogan, 1976, p. 75)). 9Hogan utilized data collected in the first wave of interviews with the Lansing, Michigan sample. Thus, her analysis was based upon data collected in 1974. 10Hogan's sample included age levels ranging from 18 (birth date prior to or during 1955) to over 65 (birth date prior to or during 1908). 34 Bailey (1980), on the other hand, indicated that age was negatively related to reported conservation practices. She studied the effect of a more select set of sociodemo- graphic factors, including size and income level of house- holds as well as age and educational attainment of the house— hold head. The age relationship accounted for 6.2% of the variance in the conservation practice scale, while the stan- dardized regression coefficient equalled -.18 (p = <.01). The larger sample size used in the Bailey analysis (n = 1,875), compared to the Hogan (1976) study (n = 156), may have accounted for this difference, as large sample sizes, in general, have a tendency to reveal significant patterns which are sometimes not apparent in smaller samples. In addition, use of fewer independent variables may have made some difference, although Bailey entered the age variable first, allowing it to account for the maximum amount of variance possible before entering the three additional variables. Hogan used stepwise procedures as well; however, age of husband and wife were entered last (Hogan, 1976, p. 75). Finally, it should be noted that both Bailey and Hogan used self-reported activities to determine conservation behavior. As indicated previously, Milstein (1978) found that this was a somewhat unreliable measure, because indi- viduals seemed to have a tendency to overestimate conservation activity when responding to formal inquiries about energy use behavior (p. 81). Thus, to the extent that this was occurring 35 with Bailey's respondents,generalizations about a rela- tionship between age and energy conservation would be dif- ficult. As indicated at the beginning of this section, few studies have been conducted using higher-order levels of analysis to explore relationships between age and questions concerning belief in the reality of the energy problem or energy consumption and conservation behavior. And of those existing, some were based on unreliable measurement and analytical procedures. Summary The literature review revealed that three inherent weaknesses existed in research studies conducted thus far exploring the relationship of age to belief in the reality of the energy problem and energy consumption and conservation behavior. They included: (1) inappropriate analytical pro- cedures (i.e., use of zero-order levels of analysis when higher-order levels would have been more appropriate), (2) unreliable measurement procedures (i.e., reliance on self-reported behavioral response rather than more precise energy consumption data measures), and finally (3) inade- quate conceptualization of the meaning of age and thus its possible influence on attitudes about the energy situation or consumption of energy and change in consumption resulting from the energy crisis. This research effort was designed to overcome the weak- nesses of other studies, by first, providing a meaningful 36 rationale for use of age as an independent variable in energy research (i.e., as a measure of past experience with depri- vation and shortages) and secondly, by exploring the rela- tionship of past experiences to actual energy consumption and conservation behavior. Finally, higher order levels of analysis (i.e., multiple regression procedures), which E allowed separation of the influence of past experience from other aging related sociodemographic and attitudinal measures, were employed to test the research questions and establish the net contribution of the age/experience variable towards b explaining energy consumption and conservation patterns. CHAPTER III CONCEPTUAL FRAMEWORK Temporal Dimensions of the Aging Process Since the late 19605, as emphasis has been placed on viewing phenomena from both a holistic and an ecological perspective, scholars focusing on similar problem areas have attempted to establish links or bonds between their various disciplines, the links serving as basic unifying themes and enabling discourse and the integration of knowledge. As integrative thinking has progressed, the concept of time has emerged as one such link for scholars of human development and family life. Historians of the family, for example, as well as sociologists and psychologists concerned with study of family and individual change have all acknowledged the importance of time as a causal agent, helping to explain growth, development and change in both individual and family behavior (Kantor & Lehr, 1977; Elder, 1978a, 1978b; Foner, 1978; Baltes & Willis, 1977, 1979; Lerner & Ryff, 1978). In each of these disciplines, as advances in model building and theory or prototheoretical thinking have occurred over the past 10 years, the meanings of time have been clarified and expanded. Time has come to be viewed as 37 38 both an element and a process as a result, and its multi— dimensional character has been recognized. At the conceptual level, cognizance of time as an element and as a process has illuminated how time and movement through time, especially movement at different periods,have influenced human and family behavior (Riley et a1., 1972; Riley, 1976a, 1976b; Foner, 1978). Concurrently, thinking regarding the multi- dimensional character of time has led to conceptualizing its component parts of life time, social time and historical time. The systematic and simultaneous interaction of these three dimensions, scholars have come to believe,luusinf1uenced behavioral change and developmental change (Elder, 1978a; Neugarten & Datan, 1973; Baltes & Willes, 1977, 1979). Visualization of the impact of time and movement through time can be best illustrated by a concrete example, indicat- ing not only the influence of time but the linkages between aging and time. The process of aging occurs on two temporal dimensions; first, it takes place over time, and secondly, within time (Elder, 1978a, p. 823). Thus for a particular individual or group of people born in one year, aging occurs over a specific span of time, such as 1900 to 1980. Numerous other age groups experience the same period, obviously; yet realization of the rather simple but important fact that the years from 1900 to 1980 are unique to the particular cohort born in 1900 is important. No other age group will experience those years in the exact same way (Riley, 1976a, pp. 193-194; Elder, 1978a, pp. 523-524; Ryder, 1965). 39 Consider a second set of cohorts, for example, born in 1920 and living until 2000. Sixty years in the lives of the two groups coincide, covering the period from 1920 to 1980. During that shared time, however, the ages of the groups dif- fer by 20 years, thus makingtflmaimpact and perhaps the perceptions of events and occurrences within those years dif- ferent for the two groups. Riley refers to this phenomenon as cohort-centrism (1976b, pp. 24-25) and indicates that successive cohorts have a distinct character because they age in different ways as a result of their interaction with historical events and the unique sequence of roles they fill in various historical periods (Riley, 1976a, pp. 191-192). The impact of the Great Depression can be used as one example of the unique influence that time has on various cohorts. The 1900 to 1909 cohort, for example, would have reached adulthood by late 1929 when the stock market crashed, marking the beginning of the Great Depression. Thus they would have been employed or shortly seeking employment at a time when jobs were becoming increasingly scarce. The 1920 to 1929 cohort, on the other hand, experienced the Depression during childhood or adolescence, which was undoubtedly trau- matic, but phenomenologically distinct from the experience of the other cohorts. In essence, as viewed from the perspective of this research, life time, social time and historical time capture these differences by defining cohort-centrism, just as they, in turn, define: (a) the process of developing or aging 40 over time, (b) the influence of role sequencing through time and within time,and finally (c) the impact of aging through historical events or processes which occur within time. In the remainder of this chapter the nature of these three temporal dimensions is explored, as are their structure and process orientations and the systematic way in which they affect human development and behavior over time--which is the aging process. Life Time Life time is associated with chronological aging or the number of years lived since birth (Foner, 1978, p. S341), while aging is the process of accumulating years over time. The process is universal, inevitable and irreversible (Riley, 1976b, pp. 29-30; Riley, 1976a, p. 195). In addition, it is lifelong. Riley (1979b), for example, refers to aging as a process of growing up and growing old, commencing with birth (or conception) and ending with death (p. 4). The process is not unidimensional, however, but rather, is defined by a set of processes. Change, growth and some- times decline in psychological, social and biological func- tioning each play a role in carrying individuals through their life time, and the three are closely interrelated. Riley (1979b), for example, indicates the three are "system- ically interactive." As they move people forward through their life time, each process influences the others, as change in one stimulates changes in the other two (p. 4; see 41 also Foner, 1978, p. S341). Consider the transition from childhood to adolescence, for example. The biological tran— sition between these stages calls forth new psychological needs as well as a new set of social needs. Among these three processes, the two linked most closely to life time are psychological and biological aging. Theyanxa discussed in this section. Social aspects of aging, on the other hand, are somewhat distinct conceptually from life time because of their relationship to cultural definitions of the aging process. They are best studied as a separate dimension, therefore, and are explored in a later section on social time. Psychological aging Emphasis by psychologists on time and historical time placement as causal agents affecting behavior and development is relatively new. They have only recently been explored, for example, after initial concern surfaced and died in the past 2 centuries (Baltes & Willis, 1979, p. 17). Tempo- ral thinking is influencing perspectives, however, and help- ing shift focus from traditional age-developmental or bracketed specialitiesl towards a life-span developmental perspective which recognizes the impact of external, 1Examples of age-bracketed specialities include fields such as infant development, child development or gerontology which emphasize: (ljznormative, universal patterns of development influenced by genetic, maturational factors, rather than focusing on individual differences in develop- ment, and (2) intraorganismic sources of developmental change influenced by simple age functions, rather than considering external, contextual factors (Baltes & Willis, 1979, pp. 17, 18-21). ' 42 contextual forces on prospective developmental change (Baltes & Willis, 1979, pp. 17, 18-21). In explaining this new focus, Baltes and Willis (1979) indicate that individuals continue to develop and change throughout their lives, and that aging, therefore, should be viewed as a continuous, lifelong process. They maintain, in fact, that aging can only be understood if experiences throughout the life time are considered (p. 15), along with factors relating to bio-cultural changes occurring around the aging individual. More specifically, they View life-span development as occurring within the context of both micro- level individual life events and macro-level ecological or social events. Thus, they feel that aging is not simply a process of inherent developmental change over time, but rather a dynamic process occurring within the context of societal level events and change (Baltes & Willis, 1979, p. 15). With regard to the field of behaviorism as it relates to aging processes, Baltes and Willis (1977) have a somewhat new perspective as well. They indicate, for example, that behaviorism traditionally focuses on three classes of ante- cedent or causal variables as determinants of behavioral response: (1) response or behavior variables (R), (2) stim- ulus or environmental variables (5), and (3) organismic or biological variables (0). Baltes and Willis argue, however, that this perspective is simplistic because it neglects consideration of time as it 43 influences the set of three causal factors. They maintain that the relationships between the (R), (O) and (8) variables can themselves be either concurrent or historical (see Figure l), with concurrent relationships focusing on deter- minants of behavior which are close or proximal in time to the response being explained. Historical determinants, on the other hand, focus on influences or chains of influences from the past and are thus more distant in time (Baltes & Willis, 1977, p. 140). They indicate, for example, that loss in intellectual functioning in advanced age may be the result of either concurrent or historical influences. Concurrent aging-specific conditions, such as a reduction in cerebral blood flow, could cause the decline, or the loss in mental ability could be the result of an historical life event, such as lack of aging-relevant education in childhood (Baltes & Willis, 1977, pp. 140-141). In schematic form, temporal influences (i.e., concurrent or historical) on behavioral and/or developmental change can be Viewed as a synthesis of three major sets of influence systems: (1) ontogenetic age-graded, (2) evolutionary history-graded,and (3) non-normative (see Figure 2). Again, however, it is the interaction of the three influence systems which induces developmental or behavioral change, just as it is the interaction of (R), (O), (S) factors which determines behavioral response (Baltes & Willis, 1979, p. 23). Because the field of life-span analysis is new, Baltes and Willis indicate that definitions of the age-graded, 44 owe .m .mhma .mouammfl EOHM poummpa “Hofl>mcom How ucooood >o£e mm moanoaum> Hmucoacouw>cmnmoaoeflum no moanmflum> HMOfimOHOMmIofiEmflcmmuo .moanoflum> Hoa>mcom noncommom :0 mafia mo mocooawcH on» no cofiumnumsHHHlu.H ousmflh 8.8:; ewe .on .2“: 95:5 aEchEoE eo EcmEEeoeoo 32.6%.... 2.: meant; 6V .on .Ev ocoEm aEmcoszm do mEmEEeBoo «cot—6:00 A9 339.; .mEmEcozém 8 2.38.5 6V meant; 605205 .0 o_Ew_cm90 6V mo_nm_ew> 83935 to $533. EV mmcoamom 3.25:3 E 6:33 a AI:c< one How coflumHOOmma sooflumad wn mbma ADV unmflu>mou .woawm ouwnz mpflflumz >n Ub»flpo .mo>fiuoommumm humcflamwomfipuoucH "gamma 0» cuuwm Eoum mcwmm Eoum mmoum 3ow>um03 mo :oflmmfishmm >3 poumopda Hem .m .mhma .mfiaafiz can mouamm Eoum poumm0< “AamucosmoHo>ooicozv o>ADMEHOZ|soz can “poncho |>Houmflmv >umCOADSHo>m .Apopmuuloom. cepocomouco umEoumam monogamcH Home: mouse pom ucoEmoHo>oQ cmmmlomaq coo3uon mangOADMHom on» no coaumuumsHHHii.m gunmen m2: m>F2“ amo<¢0->¢Ohw_1 L2. 552m: #229535 2 -ZO¢_>zw.O_ u ..w mhzr- -2.5°/o '— -3.0°/o - - 3.5% '- -4.0% )- -4.5% '— - 5.0% - .9 -4L5%:- -6.0°/o — -6.5% L— -7.0% (— 4.5% .. -8.0°/o '— -8.5°/o - -9.0°/o — Btu ll 1 l J l l 23 thru 37 thru 41 thru 45 thru 60 thru 64 thru 36 40 44 59 63 87 (n=140) (n=44) (n=26) (n=115) (n=36) (n=72) Age Figure 7.-—Mean Percentage Change in Btu's Consumed per Heating Degree Day, 1976—77 to 1978—79, by Age Group 105 2.2 S .233, 2 2-22 .60 $660 9.28: too cmEamcoo m.2m 5 35:0 32:023. cams—ix 9:9”. 00¢ cs mm co mm cm at on an 8 mm B we 8 ms 41.34::_::_:_J::_:::1:::_::_::_::_::_:: 4.3. u fine 353 II. .33. u 1 "no 258 .__ lllllllllllllllll o\om.0 I n 1 ”an E53 .33.. u a "3 Eda... $3. u 1 5c 2434" «and. u m "8 358 o\oNN I o\oON . o\oc_. . $0.. I 0\oQP I 0\oNv I $6.. I o\oQ I o\oO I o\o' I o\oN I o\oN {ow $6 $0 ++++ o\oo—.+ 106 supporting teen-age and college-age children, could be more V pressured to conserve energy dollars. Research Hypotheses To test the research questions posed for this study in the introductory statement, two hypotheses The research questions are repeated below, stated in the null form, follow. Question 1. Did exposure to shortages and past influence current energy behavior? Question 2. Did exposure to shortages and were developed. and the hypotheses, deprivation in the consumption deprivation in the past influence change in energy consumption behavior over time and, most particularly, reduced consumption (i.e., conservation) behav— ior, given the urgency placed upon conservation at the local and national levels? Hypothesis One Years of past experience with deprivation and shortages has no linear effect on energy consumption level during 1976- 77, when controlling for the effects of the following variables: 1. Health status 2. Educational attainment 3. Attitude regarding responsibility solve the energy crisis for helping to 4. Belief that the energy situation is or will be a problem 5. Income level 6. Household size 7 . Dwelling unit size, measured in number of rooms 107 Hypothesis Two Years of past experience with deprivation and shortages has no linear effect on percentage change in energy con- sumption (i.e., conservation) between the heating years July through June 1976-77 and 1978-79, when controlling for the effects of the following variables: 1. Health status 2. Educational attainment 3 Attitude regarding responsibility for helping to solve the energy crisis 4. Change in attitude regarding responsibility for helping to solve the energy crisis, 1978 to 1979 5. Belief that the energy situation is or will be a problem 6. Change in belief regarding the energy situation being or becoming a problem, 1978 to 1979 Income level 8 Change in income level, 1977 to 1978 . Household size 10. Change in household size, 1978 to 1979 ll. Dwelling unit size, measured in number of rooms 12. Change in dwelling unit size, measured in number of rooms, 1978 to 1979 13. Installation of a new furnace 14. Energy consumption level in 1976-77 \DCDQ Assumptions Four assumptions underlay this research effort: 1. The primary focus was consideration of the impact of past experience with deprivation and shortages on household energy use. To determine experience levels, it was appro- priate to use the age of the principal or oldest income earner (i.e., household head), as this individual's decision making power would allow exertion of subtle pressure to insure conformity to his/her desires. 2. Categorizing primary income earners on experience with events in the past characterized by physical or material 8Data were collected in 1978 and 1979 and reflected income levels in the previous year. 108 resource shortages or inability to obtain resources due to contextual constraints was appropriate for exploratory research concerning the effects of past experience on current behavior. 3. Survey research methods were appropriate for gather- ing sociodemographic measures of household characteristics and objective measures of technical and behavioral changes undertaken to conserve energy. 4. For purposes of determining energy consumption and conservation behavior, data gathered from utility, propane and fuel oil companies, indicating actual energy use pat- terns, were the most precise and reliable measures available. Analysis Both research hypotheses were analyzed using regression procedures because the focus of the research was determin- ation of the effect of one variable while controlling the influence of an array of interrelated variables believed cor- related with the criterion variable.9 For each hypothesis, in order to determine if intercorrelation was present and thus masking the influence of the major predictor variable, simple bivariate regression coefficients were determined 9Multicollinearity (i.e., high levels of intercorre- lation, in the range of .8 to 1.0) was apparent between some of the variables, such as the 1978 and 1979 measures of household size and number of rooms in the dwelling unit, as well as measures of consumption in each of the years. These variables were not used simultaneously in regression equations, however, and thus did not cause estimation prob- lems (see Nie et al., 1975, pp. 340-341). 109 between each independent variable and the dependent variable under study. The dependent variables were then regressed simultaneously on the mix of independent variables. The multiple regression procedures answered the researohquestions, indicating the direction and extent to which the major inde- pendent variable seemed to influence consumption and con— servation behavior when the influence of the confounding variables was removed. In the following chapter, where findings are discussed, the results of both the simple bivariate and the multiple regression analyses have been presented. Although not stan- dard procedure, it seemed appropriate in this situation because of the particular focus of the study and the emphasis placed upon confounding factors masking the effect of the primary independent variable under consideration. CHAPTER V FINDINGS AND DISCUSSION Findings related to the two research questions guiding this study are reported in this chapter. They are discussed in two sections: Consumption of Energy and Conservation of Energy. Consumption of Energy The first objective of this research was determination of the net effects of an array of variables measuring aspects of the aging process on energy consumption levels. The ultimate purpose, however, was to control or hold the effect of each variable constant in order to determine the magni- tude, direction and statistical significance of the variable measuring level of experience with deprivation and shortages during the life course. Because the relationships between the independent variables were believed intercorrelated and thus capable of confounding past experience, multiple regression procedures were deemed the most appropriate and powerful statistical tool available for analysis. The following assumptions underlying the statistical significance tests associated 110 111 with multiple regression were considered: (1) the sample being random, (2) each array of the dependent variable for a given combination of the independent variables following a normal distribution, with the error terms being randomly distributed about the composite independent measure, (3) the relationship between the independent and dependent variables being linear,and (4) the dependent variable displaying homo- geneity of variance for each array of the independent variables, with the error terms being independent and having a mean of zero at each level (Nie et a1., 1975; Kerlinger & Pedhazur, 1973; Blalock, 1979). As pointed out in the previous chapter, within each of the "Pilot Project Conserve" participating and non- participating groups, sample members were drawn at random, thus making the total sample and the research subsample random. The second assumption regarding the distributional characteristics of the dependent variable can be violated without serious consequences, especially if sample size is large, as it was within this study (Nie et a1., 1973, p. 341). The final two assumptions concerning linearity of the relationship between the independent and dependent variables and homogeneity of variance across the independent variables were explored by a scattergram plotting the residuals against the consumption variable. The relationship was distinctly linear. In addition, no patterns with respect to the error terms seemed evident, indicating that the final assumption 112 concerning homogeneity of variance around the array of independent variables had not been violated. The hypothesis tested to determine the influence of aspects of the aging process on energy consumption is stated below in the null form. Hypothesis One Years of past experience with deprivation and shortages has no linear effect on energy consumption level during 1976- 77, when controlling for the effects of the following variables: 1. Health status 2. Educational attainment 3. Attitude regarding responsibility for helping to solve the energy crisis 4. Belief that the energy situation is or will be a problem 5. Income level 6. Household size 7. Dwelling unit size, measured in number of rooms The alternate hypothesis, or the one of interest, was that level of experience with deprivation and shortages in the past would have a statistically significant effect, net of the influence of the additional variables. Because this was an exploratory study, no directional estimates were hypothesized. Bivariate Regression Analysis To assess the general impact of the predictor variables, the criterion variable of total, direct energy consumed within the household per heating degree day in 1976-77 was regressed separately on each. By using bivariate regression procedures in this manner, estimation of the zero-order 113 influence of the variables could be determined and thus pro- vide a base line for comparative appraisal when the multi- variate regression was done. The results of the bivariate analysis are shown in Table 8. At the zero—order level, stage in the family life cycle and lifestyle factors (i.e., social processes of aging) appeared to be the major influences determining level of consumption. Number of rooms, number of occupants and income level were the first, second and third strongest predictors, each was statistically significant (p < .000, < .000, and .000 respectively), and each was related to consumption in a positive way. Interestingly, however, in terms of magnitude of influence, unit changes in income level produced much smaller incremental increases in consumption, while house size and household size were more dramatic. Consideration of the standardized regression coefficients somewhat reduced these differences, however, as standard deviation unit changes in household size and income both produced nearly the same level of change in consumption (.29 and .21 respectively). Next in order of importance were psychological aspects of aging, with educational attainment and belief in the energy problem both appearing to influence consumptionlevels to a significant degree (significant at p < .05). The most interesting aspect regarding their impact, however, was the directional effect. Both were positive, indicating that the higher the level of education and awareness and the greater the belief in the reality of the problem, the higher was 114 mmv "Hmzpflmmu mp H ”coflmmoumoh mo coflum>nmmcoo monoco mmw. mmm. mmo.| ma.omm om.m~mt mpumzou mooufiuufl mom. mom. ovo.l mm.mwm ha.mamu muouoom spammm mommunonm cam :ofium>flumoo ova. mHH.N Oko.- mm.om Ha.~mr uooooauooxo ammo Ewanoum xmuocm mmo. mm¢.v Hoa. mm.vmw mm.mmma may :a mofiaom useEcflmuum moo. mmw.m ova. mm.VVH vm.mmv Hmcoflumospm ooo. www.0m mam. vo. om. Ho>mH mEoocH ooo.v moa.mm mmm. mh.ovm mm.momH mucmmnooo mo umnEsz ooo.v oma.mma Now. wv.me mo.momm mEoou mo Honfisz nouum m ucofloflmmoou Houum ucofloflmmoou mcflHQEom mo coflmmoumom pumpcmum coflmmoumom >uflaflnmnoum UoNHoumccmum powwoumpcmumca moHQMHHo> ucoocomoch hhlwhma .xmo oonmoo mcflumom pom UTESmcoo n.5um I. . . Ill- I..I‘ ‘ I). II...‘ 1 Houum mcflHmEmm mo muflawnmnoum pom moflummlm .mucofloflwwoou :oflmmonmom pouflpumpcmum.muonum pumpcmum .mucofloflwmooo coflmmmnmom pomfloumpcmumCD ”moanmflum> ucopcomoch co .hnlmbma .>mo woumoo mcwummm nod coasmcou m.:um uomuflo .Hmpos mo mflm>amc4 coflmmoumom mamaum>flm maafiflmll.m magma 115 consumption. These findings were opposite to intuitive expec- tations regarding their influence and suggested a "get mine now while it lasts" feeling on the part of households. Some strength for this interpretation was given in that attitude regarding responsibility for helping to solve the problem (the final psychological measure) was the weakest predictor of consumption level, coming in last as an insignificant predictor, even though it was related to consumption in the expected negative direction. Finally, cognizance must be taken of the fact that both educational attainment and belief could have been highly related to income level, which, to the extent that it influenced higher consumption, could have "washed out" the expected effect of these two variables. Pearson product moment correlations between income and belief and education were .09 and .41 respectively, which somewhat supported this argument with regard to educational attain— ment. It did little to explain the belief finding, however. The two final aspects of aging, level of past experience with deprivation and shortages and biological functioning, both appeared poor predictors of consumption (p = .146 and .342 respectively). In addition, even though statistically insignificant, both appeared to have a negative effect on consumption level, indicating that as experience increased, consumption was apparently depressed (i.e., past experience with shortages limited profligate use of energy resources). The negative direction of the health factor was somewhat more problematic, as intuition would have seemed to indicate 116 that if health was poor, energy would have been consumed in higher quantities to compensate. The influence of income, however, could have intervened, thus moderating use of energy. Two factors related to income level came to mind. To the extent that income was needed to maintain health, less was available for energy consumption. In addition, if lower income households, in general, appeared to have health prob- lems, the budgetary squeeze could have influenced the amount spent for energy. Indeed, this appeared to be the case, as the relationship between health and income was negative, although quite low (r = —.22). Multiple Regression Analysis Results of the multivariate analysis depicting the interactive influence of the variables are found in Table 9, indicating that some rather significant shifts in the order- ing of influences as well as in the direction and magnitude of the variables' influences had occurred. The two measures of social aging related to stage in the family life cycle identified as major influences on con- sumption level in the bivariate analysis retained their first and second order rankings. Number of rooms and house- hold size, thus, both continued to appear as major contrib— utors to consumption level. In addition, their influence remained positive, although their effect had been somewhat ‘ lessened (Beta = .443 and .229 respectively in comparison to .492 and .288 in the bivariate analysis). The effect of 117 .moanmflum> ucoocomopcfl may mo nouomumco o>fluomuoucfi may pump on buonmEo mm; ponuoE coflwmoummu oum3nom oumocmumm mmv "Hmsoflmou mo mam. “mumsqm m m "cofimmmumon mo mvm. "m mamfluasz ooo. www.mm m Hamuo>o mmm. mom. hmo. mm.wva mm.am ucoEcflouflmHmcowumooom mom. mom. mmo.| mm.~mm no.vamu muouomm Spammm mmm. mmm. woo. me. we. Ho>ma oEoocH Emanonm amm. oaN.H omo. mm.mam NH.mmo sonooo on» on moaaom coflum>uomcoo mmo. omm.v Hao.- H~.mam vo.maNH- smuooo mouozoo moooaooa mommuhonm pom cofium>flumw© omo. Hma.m «Ha. 4H.mm aa.mma -ooooauooxo ammo ooo. mom.mm mum. mm.om~ mm.mmHH mucmmsooo mo HwQEsz ooo. bmm.m0H mew. mb.vom mm.amcm mEoou mo Honfisz nounm m ucofloflmmoou nonum ucofloflmmoou mcflamemm mo coflmmoumom pumccmum coflmmoummm xuwaflnmnoum woufipuwocmuw powflpnmocmumco moanmfium> pamccomwwcH hhlmbma .wmo ooumoo mcfiumom Mom ooasmcou m.:#m .II n‘.(l|l1. lultl ..I Iva-Ill (I‘l.)’|'llllv)(.lll['.ll.l)) .I. all; I .7! muonum mcwHQEmm mo xuflafinmnoummfl:~moflummlm .mpcmfloflmwoou coflmmoumom woufioumocmum .muouum pumpcmum .mucofloflmmoou coflmmoumom pomfionmpcmumco "moanmflum> ucoocomoocH co .hhlmhma .>MQ moummo mcflumom Mom poEsmcoo n.5um poouflo .Hmuoe mo mflmwamcd coflmwmumom mfimfluHDZIl.m magma 118 income level, on the other hand, appeared to have been diminished. The variable shifted from third to sixth place, and its Beta weight was reduced to .044 from .213 in bivariate regression form. The statistical significance of income was lowered as well, thus indicating that probability was low regarding its ability to predict consumption level (p = .355).1 Concerning psychological aspects of aging, the effects of the various measures were transposed. Educational attain- ment and belief in the problem both became insignificant predictors of consumption level (p = .582 and .257 respec- tively), while retaining their positive directional influ- ences. Attitude regarding conservation, on the other hand, was elevated to fourth place and appeared, in its negative directional orientation, to have a significant influence on consumption (p = .038), indicating that households identi- fied as feeling greater responsibility for solving the energy crisis seemed to be consuming less energy. 1The effect was not as dramatic in the analysis con— ducted on consumption in 1978—79 (see Appendix D, Table 2). Although the magnitude of the income effect was reduced (Beta = .117 versus .264 in the bivariate analysish income level retained third rank in the ordering of variable influ- ence. In addition, it still appeared as a significant pre- dictor of consumption (p = .022). Part of the difference in rank ordering of the variables between the 1976-77 and 1978- 79 multiple regression runs could possibly be explained by changes in the level of income measurement between the two interviews used to collect data. In the 1978 interview, for example, six income categories were employed, while in the 1979 interview seven categories were used. The range in 1979 thus captured more variability in income, and the wider spread could account for the differences. 119 Biological aspects of aging still appeared as insig- v nificant predictors of consumption. Any effect from the health factor, in fact, was reduced further in the inter- active analysis (Beta : —.026, p = .546; versus Beta = -.046, p = .342 in bivariate analysis), indicating other factors were apparently more meaningful predictors of consumption. Finally, rather important shifts took place with regard to level of experience with shortages and deprivation in the past. In the multivariate analysis its effect was heightened, as past experience did appear as a statistically significant predictor of consumption level (p = .020). Thus, the mul— tiple regression analysis supported rejection of the null hypothesis and acceptance of the alternate hypothesis that net of influences relating to dimensions of the aging process, level of experience with deprivation and shortages appeared to affect current consumption behavior in a positive direction. Thus it seemed that the greater the level of exposure to deprivation and shortages in the past the greater the propensity to consume energy. Discussion The findings of this hypothesis suggested that with regard to temporal dimensions, two mutually interactive influences were important considerations in determining levels of energy consumption. The first influence concerned social aspects of aging, while the second pertained to exo- genous factors influencing the life-span development of 120 household members and, in turn, contributing to consumption behavior. The magnitude of the relationships between dwelling unit size, household size and energy consumption were indi- cations that changes in size and composition of the household over time (i.e., the family life cycle) appeared to play a role in consumption behavior.2 Beyond this influence, how- ever, the findings suggested that consumption behavior was influenced by the locational juxtaposition of the household head with respect to the historical events taking place during his or her life-span. In essence the findings implied that factors associated with perceptions and memories of events in the past when access to resources was blocked or basic resources were in short supply tended to influence current behavior. Conservation of Energy The second objective of this research was similar to the first, and thus procedures were conducted in exactly the same manner.3 In the second instance, however, the focus 2Note that in the analysis concerning energy consumption in 1978-79, income, the third factor defining social pro- cesses of aging, was found related to consumption as well. 3The four assumptions underlying regression analysis discussed on pp. 110—112 with respect to the first hypothesis tested were also considered for this hypothesis. The first assumption concerning random selection of the sample was met, as pointed out previously, because the sample was drawn at random. A scattergram plotting residuals from the multiple regression analysis against the conservation variable indi- cated that assumptions 3 and 4 had been met. The 121 changed in order to determine the effect of past experience on change in energy consumption levels (i.e., energy con- servation) while controlling the effects of the aging pro- cess as well as any effects of changes in the processes that had occurred between the two measurement periods. The hypothesis tested to determine the influence of various dimensions of aging on conservation of energy is stated below in the null form. Hypothesis Two Years of past experience with deprivation and shortages has no linear effect on percentage change in energy con- sumption (i.e., conservation) between the heating years July through June 1976-77 and 1978-79, when controlling for the effects of the following variables: 1. Health status 2. Educational attainment 3. Attitude regarding responsibility for helping to solve the energy crisis 4. Change in attitude regarding responsibility for helping to solve the energy crisis, 1978 to 1979 5. Belief that the energy situation is or will be a problem 6. Change in belief regarding the energy situation being or becoming a problem, 1978 to 1979 7. Income level 4 8. Change in income level, 1977 to 1978 9. Household size 10. Change in household size, 1978 to 1979 relationship was distinctly linear and no patterns were evi— dent in the error terms (i.e., they were randomly scattered about the regression line). Finally, sample size was large enough that concern was not warranted regarding assumption 2 and the distributional characteristics of the error term about the sets of combined independent variables (i.e., the multivariate independent variable values). 4Data were collected in 1978 and 1979 and reflected income levels in 1977 and 1978 respectively. 122 ll. Dwelling unit size, measured in number of rooms 12. Change in dwelling unit size, measured in number of rooms, 1978 to 1979 13. Installation of a new furnace 14. Energy consumption level in 1976-77 The alternate hypothesis was that level of experience with deprivation and shortages in the past would have a statistically significant effect, net of the influence of the other variables. Because the study was exploratory in nature, no directional estimates were hypothesized. Bivariate Regression Analysis To determine the general influence of the predictor variables, the dependent variable of percentage change in direct energy consumed within the household per heating degree day between 1976-77 and 1978-79 was regressed sepa— rately on each. The results, shown in Table 10,suggested that no distinct patterns were apparent with regard to the relationships between conservation and social, psychological, biological or historical aspects of aging. Only two variables had coefficients statistically dif— ferent from zero at the .05 level or less. The most influ- ential predictor was consumption level in 1976-77, which had a negative relationship with percentage change in consumption (Beta = —.202, p = .000). This finding was expectable due to the statistical phenomenon of regression to the mean: yet, it was also interesting because it suggested that those consuming higher levels in 1976-77 were conserving greater amounts in 1977-78 and 1978-79. Likewise, change in 123 .Aomq. n o .aoo. u a .amo.- n moon "woumom.t u my mmocop0wudp ucmoflmmcowm C: pw3ocw mu~3m0h u-o3 mm venomous mo: hbou Cw ~m>oH 0500:“ we quQEHm mmv "accommwu up "cowmmoummu up hem. hmc. ace. monumvm. mcnmmnv. ucoECflmuuc Hmcoflumospm mHa. ema. afic.n monunmfi. wormmmc.u chad 0» whoa .Ho>o~ oeoucM :« umcmcu who“ can Hmm. 5mm. owo.n «No. «Ho.u mood cowzuwn momcuom 36: mo newumfiawumcH owe. new. vmc.- woo. moo.u whoa .uwcs ocflafio3c osu c“ mEOOu no umbezz mhmfl cu mnma .c0au6>uwmcoo has. amt. mmo.- one. aco.- sonata acacuoocoo oesuauuo ca oocozo «we. cam. one. fine. aoo. whoa .Emanouo sonata on» ca sonata mood Du mnma oam. mar. Heo.n NHo. oHo.u .uwc: mCMHHo3c c“ meoou mo uonEs: cw omcocu mom. one. nvo.- oc-mmmo. oo-MeHo.- mmamfl .Hm>oH weoucH vow. wc~.H omc. wHo. hao. muouunu guano: mom. vcw.~ ch.| wcc. Hco.t mmomuuocm pom cofium>fiuawpnmocoauwmxo ammo «Ha. ocm.m who. “He. sac. mhmfi .co«u6>ummcou amumco mouozou upsuauu< Naa. ~4m.~ oao.- «cc. aco.- whoa .mucmosuuo mo amass: ofiu. omm.~ rho. ooc. mfic. area ou whoa .mucoosooo no amass: ad omcmcu shad 0» whoa .anoouo souoco on» mvo. mmo.m moo.n HHc. «No.1 cfi uwfiaon mcficuoocoo cowunwouwm c“ mucoso occ. wmm.ma Ncm.a ccammmn. moumhmm.u bhuonma .Hm>w~ COquESmcou neuum ocmdecm m acoaoauuooc neuum ucowuacoooo Cc >uwafi£m£CLm ccmmmonqom puoccmum commmwuoom pouwcuopcouw pwumpuocccuwca thmhoH .1. ill .I h. Cu outcrop .>mo ocuooo ocmumm: poo coEsmcou m.sum cfi occmcb ooaacoouoo I I. I. . v rt . 1.1! .||l.-IuIF I 1. .5. ELK! IE)FRI r . A. .IVP“...-val. r I (Phwhphl I'llrn’u" 6......(1! III-g I conflcuocccum penum mcLHQEmm uo >uflafinonoum can moduomum .wucmMUMMGQOU :oflmmwumom .muouum cumccoum .mucofioduwooo ccwmmoumcm ownwpumpcoumca mwfinomum> ucmpcmnoch h..||I.. 1H .n.'llafl1..vln .i) .Iv.l|l 'ii'lJ l . .h‘ll 1.“..1 "u...” 9 .l "moanmwum> ucmpcwmmpcw co ambumhoa Cu nonwbofi .>cc wouooo mcwumo: you cwEsmccu w.:um cm occocu wqmycooumo no mwm>awc< cowmmoumwm wuomuo>wm mHQEmmII.oH manna 124 perception regarding belief in the energy problem had a nega— tive relationship with change in consumption (Beta = -.095, p = .048), indicating that as belief level increased between the 2 years, energy conservation resulted. No other variables had relationships that could be con- sidered influential, with the exception of three that were close to the .10 level of significance. They included social aspects of aging related to household size and change in household size and a psychological measure concerning atti- tude towards responsibility for helping to solve the energy crisis. At the zero-order level of analysis, past experience with deprivation and shortages had a significance level of over .20, indicating no apparent relationship to change in conservation behavior. Multiple Regression Analysis Results of the multivariate analysis revealing the interactive influence of the variables are shown in Table 11. The findings indicated that original consumption level and change in perception concerning belief in the energy problem retained their first and second rank orders in predicting change in consumption level. Both, in fact, were strength- ened with the effects of the other variables held constant (Beta = -.217 and -.125 respectively versus -.202 and -.095 in the bivariate analysis). Third in rank order contributing to change in con- sumption was past experience with deprivation and shortages 125 .mofinmflum> acupccaoccw on» we Louoouoco o>wuocpoucw may umou Ou poxcaaEm moB ccxyos :Cwmmoumou pumzu0u pumccmumo one “Haztwman or moo. nonoscm a ma ”commmchmon up mom. “m w~awafisz moo. omm.m L HHouo>o onoa Cu whoa .c0aum>uomcoo moo. Monmomv. Hoo.u Nfio. mcnmflmm.n >ouoco ocflcuoocco woouwuum c“ ovcmzu moo. moo. ~oo.n moo. monmmHH.| ucchwmuuo Hmcowumusom whoa 0» whoa mob. moo. «Ho.u «no. moc.l .uwco oc«-w3p CH meoou uo umnESC Cw wocmzu oao. am". mmo.n oonmxoo. coummcm.- whoa .H0>oH oecoc_ How. mow. omo. mcumHvH. ooImFMh. whoa Ou who“ .Ho>oH 0506:“ c“ oococu whoa can omm. mom. mmo.| «no. vHo.n whoa cmozumn momcusu so: no :OwuofiaoumcH mmm. mmm. ovo.- mac. ooo.- asafi .eoanouo smuoco or» an Condom ofim. oHv. Nmo. hao. Hao. mucuuou Ludoo: mom. ooh. emo.u moo. moo.n whoa .mucoaouuo mo nunssz mom. syn. omc. v—o. mHo. whoH .cowum>uwmcoo >ouwcm puo30u opoufiuud amm. woo. Nmo. cfic. opo. onoH Ou whoa .mucmaoooo uo nonezc aw wococo omm. nov.H woo. woo. moo. whoa .uflco azaHHo3p may no mEOOu mo umbssz mmo. coo.m «HH.: Hoo. ~oc.s mooouuonm one cofium>wumupuwucowuomxo umom mama 0» whoa .soaoouo souoco on» mvo. moH.v mm~.n vac. omo.| cm wowyon ucflcumucoo newumoouom Cw mucozu coo. mmv.¢H BHN.I oonmemo. moumMom.n shuohoa .Hm>mH cowuoesmcoo uCuLm camanmm L ocowoooucou Leuum ucmwowowmou no >um~wnonoua ccwmw0uccm cumpcoum cofimmouqoz commoumccmuw touwpumccmumc: ohlmhma Cu or: l.n .I. 1%...“ .I‘A)I coca .>cc oouqcc ocmumoz you caszw:Ob n.2um :_ mocogu oqoucwuuom x h I rut...» h .I ».h....vl... (.I. ......h..n. cclmuosvcm pcnwcumpccim .mLCLLu Unaccoom .onnmno_ Cu hblmhow ... n... ..I.E|Nvl LILIIIhI. 1| hl) (firtu .I. . . ; Errtlh mucuum ccq~1Eom we .. ul ._ U mmfinmfiuo> acmpcoachH 91! E). il...l...EIHlu.(l(I..ll.lu.flth1l)(.(.'...x. (n .. .91.) l..t 13h.” I Lhfl («1M . l 1. I u r >uw~w£oo0um cco mcflummth .mocoaofluumou .mucowummocCC :Cmuoonccm TLNMCLocccumc: "mofinowuc> acoCCmQQCCH cc .>c£ ochre; Uc_+c:: Loo ccs:mCCC m.:+m cm cccozp occlcouuca Cc mmm>~oc< commmoucom ofiawu~szur.afi wanmb 126 (Beta = -.114, p = .055). Thus in the second multivariate analysis, the effect of level of past experience was again heightened by controlling the effects of factors related to the aging process. This finding led to rejection of the null hypothesis and acceptance of the alternate hypothesis that when controlling the influence of age-related confounding variables, level of experience with deprivation and short- ages in the past seemed to influence change in consumption levels in a negative direction. Thus it appeared that the greater the level of exposure to deprivation and shortages in the past, the greater the propensity to conserve energy. Discussion In general, it appeared that households at most age levels were conserving energy (see Appendix A, Tables A-11 and A-12), and that underlying this behavior could have been psychological factors motivating increased concern for energy depletion as time has passed, and thus conservation ofemergy. The relationship between high consumption and conservation led to an additional consideration: that increased prices of energy could have motivated high energy users especially to conserve. Likewise the impact of higher prices could have influenced the various age levels to conserve due to the strain increased energy costs have placed on family incomes, which must be stretched to cover the various expenses households encounter across the family life cycle. For younger families, the expenses involved in rearing and 127 launching children would be prime examples. While in old age, the pressure of making fixed, limited incomes cover costly energy expenses could have forced conservation. In addition, however, the findings suggested factors beyond simple family life cycle considerations. It appeared that households with heads having more experience with dif— ficult times in the past were conserving a larger proportion of their initial consumption level than households headed by individuals with lesser experience. Two influences could have accounted for this behavior; the first, a psychological affect, while the second, a practical consideration. Older people, having experienced national crises and knowing the impact of shortages and deprivation, could have willingly been conserving energy with the hope that early individual sacrifice would prevent a larger, more dramatic and forced confrontation with future shortages. On the other hand, the explanation could be that, given increased costs of energy, older individuals were conserving more simply because they knew how, having experienced or witnessed general or forced conservation efforts in the past. Younger individ- uals, on the other hand, lacking this backlog of skills, may not have been as aware of ways to implement conservation and thus conserved less proportionately. CHAPTER VI SUMMARY AND CONCLUSIONS Overview Since the Arab Oil Embargo of 1973-74, sporadic short- ages of energy have affected some Americans, while the dramatically increased price of energy has affected all. As a consequence of these two factors, this research has viewed the current energy situation facing the nation as a social problem (Schwartz, 1978; Schwartz & Schwartz-Barcott, 1974; Smelser, 1979; Morrison, D., 1977; Newman & Day, 1974: Unseld, 1978). And to the extent that social problems are often accompanied by social change, this research has attempted to determine in what direction and to what degree micro—level change in energy use patterns may be indicative of broader, more extensive change at the societal level. More specifically, it has attempted to determine to what degree micro-level conservation of energy may be indicative of a shift in American attitudes away from profligate con- sumption of natural resources towards more restrained use. In order to focus more directly on this process, the research investigated longitudinal patterns of energy use at the household level. Patterns of energy consumption and 128 'l 129 conservation evidenced by households stratified by the age of the principal or oldest income earner were studied, in particular. Temporal Orientation Selection of the age variable seemed intuitively logical, the reasoning being that age, if viewed as a marker defining a distinct set of life experiences, would necessarily influ— ence current behavior, thus influencing adaptational patterns and change in energy use. From a macro—level perspective, age—related behavioral change in energy use seemed especially important with respect to the future. Which age levels were conserving energy, for example, and how much? And were these patterns a function of aging-specific or historical experience influences? And if historical experience affected current energy conservation, what were the social implications of additional shortages or still more costly energy in the future when cohorts with differing historical experiences would be forced to deal with these problems? Older cohorts' lives have included experiences with resource shortages or with resources being highly priced in relation to disposable income and thus ability to obtain them. Their lives have included training, in other words, in how to deal with situations such as shortages of energy, just as they have included experience with hardship and the suffering resource shortages could bring. Younger cohorts, 130 on the other hand, have not had this experience, in essence, having been prepared for a life of abundance. Did this make a difference? And if so, how? In effect, in terms of the future and social problems occurring in the future from heightened price levels in relation to declining supplies of fossil fuel sources, focus- ing on age seemed appropriate in order to determine micro— level change. For if age levels have differed in their adjustments to the energy crisis, social conflict between generations could occur, just as it has seemed to be occurring with respect to inflation (Quinn, 1980). Study of household level change in energy use patterns from a historical experience perspective posed analytical problems. In order to examine past experience, cohort groups studied within this research were stratified by vary— ing levels of exposure to shortages of resources and inability to obtain resources in the distant past; age level of the cohort acted as the operationalized stratification variable. This resulted in interactional problems because age level defined not only levels of past experience, but energy depen- dence of the cohorts due to factors related to processes of aging. To overcome these problems, it was necessary to study each of the various aspects of aging believed related to energy use, along with past experience, which was hypoth- esized to be accounting for a substantial proportion of current energy use behavior. This was accomplished by study 131 of the aging process within a temporal framework, which allowed consideration of the dynamic and constant change occurring as aging progresses over and within time. Specif- ically, three temporal dimensions of aging defined the frame- work: (1) life time, (2) social time,and (3) historical time. More personal aspects of aging were viewed as life time changes, and the processes of change were identified as psychological and biological aging. In turn, psychological aging or life-span growth and development was viewed as occurring within time, while biological aging was perceived as taking place over time. Thus, in terms of psychological aging, stress was placed upon the importance of context and interaction between the aging individual and the physical and social environment encountered within the individual's life time. The influence that this interaction had on cur- rent behavioral response was the special consideration. Specifically, the psychological impact of aging through three recent events in American history was studied. World Wars I and II were considered because during both of these periods Americans were forced to limit consumption of food and fuels, as these resources were rationed by the federal government. The third event considered was the Great Depression, which was phenomenologically distinct from the other two, yet related. During the Depression, people were not faced with macro-level shortages of material resources as much as they were simply faced with inability to purchase goods and resources due to lowered consuming power. Massive 132 unemployment, affecting approximately 25% of the work force, and drastic wage and working hour cuts reduced purchasing power by at least 60% in the early years of the Depression. Suffering and hardship obviously resulted. In this research, aging through the World War I, the Depression and World War II experiences were viewed as highly intense situations, capable of creating strong psychological impressions and influencing behavior throughout the life time. Thus experiences with these events were perceived as temporal antecedents, affecting behavioral adaptation in response to the current problems of energy shortages and increased energy prices. Biological aging was viewed as a process of gradual deterioration taking place over the life course, but mani- fest most apparently in the later years. This factor was considered because, as physiological functioning declines with age, people could be forced to rely more intensely on environmental support to help them maintain health and independence. Fossil fuel energy, thus, could be an impor- tant life—sustaining factor in the lives of older individuals. Along with biological aspects of aging, social aging was perceived as occurring over time. Special attention was focused on temporal dimensions of moving through age— related stages, changing social roles and responsibilities as transitions between life stages occurred. Of special concern were micro-level turning points related to family 133 life or the process of expansion and contraction occurring over the family life cycle. Even social roles outside the family orbit were viewed in relation to dimensions of house- hold change over time. Income level, for example, was per- ceived as an index reflecting family lifestyle changes over time as income increased and decreased with the aging of income earners. Historical time or the process of aging through twen- tieth century American history provided the general frame- work for the study. To the extent that age levels varied with respect to past experience with historical events characterized by deprivation or shortages, response to the energy crises occurring since 1973—74 was expected to differ. The study attempted to determine how and to what degree. Need for Age—Energy Use Research A review of the age-energy literature revealed that numerous studies had been conducted attempting to ascertain whether relationships existed between age and energy con- sumption and conservation (Farhar et al., 1979). The majority of these research efforts focused on the age variable as a general determinant, however, neglecting to consider the various processes of aging which could have accounted for energy use patterns. This study attempted to overcome those deficiencies by decomposing the aging variable and considering its component parts in relation to actual energy consumption and conservation patterns. 134 Analytical Procedures Multiple regression procedures were employed to deter- mine the strength and magnitude of the relationships between age and energy consumption and conservation. Two structured linear models were developed depicting energy use behavior and change in behavior over time as functions of aging. These models were then further developed to illustrate the linkages between energy consumption and conservation pat— terns and the various processes of aging defined as psycho- logical, biological, social and historical aging. This final operationalized model was used to test the hypotheses. The two specific questions addressed concerned the net influence of historical aging on current energy consumption and con— servation patterns. Conclusions Results of the first multiple regression analysis revealed that historical aging was positively related to energy consumption (Beta = .114, p = .02). Additional aging- related factors identified as influencing consumption level included: (1) size of dwelling unit in rooms (Beta = .443, p = .000), household size (Beta = .229, p = .000)1 and attitude concerning responsibility for helping to solve the 1In the analysis of consumption in 1976-77, income level was not identified as a significant predictor of energy consumption (Beta = .04, p = .36). However, in the analysis of consumption in 1978-79, income level was a significant predictor (Beta = .12, p = .02). 135 energy problem (Beta = -.091, p = .038). The amount of variance in consumption accounted for by the overall, age- related linear model was 30% (R2 = .298). These findings seemed to indicate that two temporal dimensions related to aging could be influencing energy con- sumption patterns. The first concerned family life cycle changes occurring over the life-span, with indications that as the family expanded during the early stages of develop- ment, energy consumption increased, while it decreased as the family contracted in the later stages of development. The second temporal dimension affecting consumption appeared to be related to the influence of external, contextual forces from the past. The findings suggested that house- holds headed by individuals having higher levels of past experience with deprivation and shortages were currently consuming more energy than equivalent households headed by individuals with less experience. The implication seemed to be that if life course patterns of household heads included resource deprivation or exposure to resource deprivation in the past, energy consumption was somewhat more extravagant in the present. The findings of the second multiple regression analysis, testing the influence of aging related factors on change in energy consumption patterns (i.e., percentage change in consumption between 1976-77 and 1978-79), revealed that level of past experience with deprivation and shortages was a statistically significant predictor of proportional change 136 in energy use behavior. The relationship, in this instance, was negative (Beta = “.114, p = .055). The majority of other factors related to aging did not appear to influence change in consumption levels (all probability levels were greater than.20), with one exception. Change in belief regarding the reality of the energy problem showed a negative rela- tionship to percentage change in consumption (Beta = -.l25, p = .042). The final factor influencing change was original consumption level, which was negatively related (Beta ='u217, p = .000). The amount of variance in the change variable accounted for by the overall, aging-related linear model was minimal at 7.5% (R2 = .075). These findings suggested that factors related to aging, per se, did not appear to have an important influence on energy conservation patterns. There was one exception, how- ever, related to the household head's level of experience with deprivation and shortages in the past. It appeared that households composed of primary or oldest income earners with higher levels of past experience were presently con- serving a larger percentage of energy than equivalent house- holds headed by individuals with less experience. Two fac— tors could have accounted for this finding. The most probable explanation could be that households headed by individuals forced to reduce consumption of resources in the past knew how to live more frugally and utilized this know- ledge to help them adjust to the current situation. 137 Alternatively, households headed by individuals who had experienced forced shortages in the past could simply have been determined to forestall a major crisis and, thus, more willing to conserve now to extend current supplies until effective new sources have been found. Limitations Limitations inherent within this study were related to both the state of knowledge in household energy research, in general, and to the research design utilized. More specif- ically, the limitations concerned questions about the unit of analysis as well as measurement and precision factors involved with secondary analysis of the "Pilot Project Con- serve" survey data. Because study of household energy use has only recently received attention, having been actively studied in only the last 7 or 8 years, little has been determined about the nature of decision making processes underlying consumption and conservation behavior at the micro-level. The limited amount of knowledge available had implications for this exploratory study because of the decision to focus on the age of one household member in order to stratify the house- hold into a level measuring its exposure to past events characterized by deprivation and shortages. In conducting the analysis, the assumption was made that the age of the principal or oldest income earner (i.e., household head) would be the logical choice because this person, due to a -"W,q 138 decision making power, would exert direct or indirect influ- ence, and, in effect, control many energy use decisions. Thus, while the unit of analysis was the household, the energy consumption and conservation decisions of the household were attributed to the household head. This assumption could greatly oversimplify the process involved, as some research efforts have questioned whether various household members have exerted differing control over energy use decisions (i.e., males controlling heating decisions and females regulating household equipment decisions (Hogan, 1976)). Hopefully, future research efforts exploring energy-use decision making at the household level will clarify this question. The second limitation concerned the fact that precise information measuring the direct impact of each historical event upon the head of the sample households was not available in the data bank. The study, therefore, through using the age of the household head as a measure of exposure to past crises characterized by deprivation and shortages, relied on indirect rather than direct measures of the influ- ence of each crisis. In essence, the variability of the household head's personal experience with each crisis was not considered (Elder, 1974). This was a weakness and hope- fully future research efforts can be designed which will allow measurement of experience with past deprivational or resource shortage situationsrmnxadirectly and explore the relationship of this more precise measure to current energy consumption and conservation behavior. 139 One of the most logical ways to attempt such an analysis would be through recall techniques. In structured inter- views, household heads of varying age levels could be asked their perceptions of the effects of the war and the Depression years on them personally and their impressions of the general impact of the events upon their families, neighborhoods and communities. Alternatively, the development of reliable scales measuring perceptions of the severity of past personal experience with deprivation of resources could be considered. The results, in either case, could be used to analyze whether a relationship between perceived severity of past personal hardship and current energy use behavior can be detected. Recall, although often criticized as a highly subjective measurement tool, could, ifluxxialong with the type of scale developed in this study, prove a valuable research tool. Implications Two implications are suggested by the findings of this research. They concern future research efforts in the area of aging as a factor affecting energy use patterns and edu- cational programming with respect to energy conservation. Future Research Efforts The findings of this study suggest that households headed by older individuals are conserving proportionately more energy than households headed by younger age levels. And to the extent that differences in behavioral adaptation are occurring between age groups, social conflict may be a 140 future problem. InflatiOn tux; already caused tension between generations; Quinn (1980) recently indicated that under current economic conditions, our society (meaning employed, income—tax-and—social—security-paying younger generations) can no longer afford to pay the growing bill for the retired. This emerging strain could be exacerbated as increased energy costs or shortages of supply take place in the future, particularly as younger generations are asked to pay the higher energy bills of older generations. To help understand this tension and monitor its growth, future research seems imperative to clarify further the effects of the various aspects of aging upon energy use pat- terns. Most particularly, however, the influence of his— torical experiences on conservation behavior should be understood, as the particular aging pattern of older cohorts seems to have influenced them to conserve proportionately more energy than comparable younger cohorts. In addition, trend analysis of aging—related consumption and conservation behavior should continue to determine stability or changes in patterns in the future. Such studies, although needed, would not be easy. One inherent difficulty concerns the fact that long—term longi- tudinal analysis of household energy consumption data is needed from the past and continuing into the futureu Research designs based upon historical information of this type would be especially beneficial in helping to determine the mag- nitude of aging-specific, as opposed to historicalexperience, 141 influences on energy consumption and conservation behavior. Studies such as this would be costly, however, involving the collection of data difficult to access, if indeed they exist. The extent to which utility companies keep detailed house— hold consumption records dating into the more distant past is questionable, while attempts to gather sociodemographic data on households serviced by the utilities would be prob- lematic, due to the mobility characterizing American house- holds in the twentieth century. In addition, further research of a more current nature is needed regarding the relationship between age and energy conservation behavior. If, as this research suggests, older age levels are conserving proportionately more energy, initial factors needing clarification are differences in con— servation actions undertaken by various age groups. Are the lower-income elderly saving energy by behavioral adaptation or are they relying on technical modifications? And what specifically are younger age groups doing differently from older cohorts that reduces the impact of their actions? Further research is also needed to determine causal relationships accounting for energy conservation behavior. To what extent does income level modify the influence of historical experience? Or perhaps more importantly, how does the temporal patterning of the family life cycle influ- ence or interact with historical placement of cohort members? Analysis of these causal linkages was beyond the scope of 142 this exploratory study. It certainly should be considered in future research, however. Questions concerning the current energy situation are a rich field of study for understanding processes of behav— ioral change at the individual or household level (i.e., micro—level change) and their affect on social change at higher levels (i.e., macro-level change). This brief dis- cussion concerning the relationship of age to social change has touched on only three issues relating to the impact of micro-level change on macro-level processes. Educational Implications The 1970s marked a turning point for Americans as con- cern became manifest that diminishing supplies of natural resources, in general, could not continue to support growing world—wide and United States demand in the future (Smelser, 1979). Americans especially have been chastised for their contribution to the problem in an effort to encourage con- servation of energy, in particular.2 While this study suggests that American households across all age levels have conserved energy since the winter of 1973-74, when the initial energy crisis was felt, it sug- gests, more specifically, that households composed of older individuals, knowledgeable of ways to live less 2B. Morrison (1975), for example, citing information published in the Scientific American, indicates Americans constitute 6% of the world's population, while they consume over 30% of the world's energy resources. 143 resource-intensive lifestyles, have conserved proportionately more. Thus they appear to be using this knowledge in order to adapt to the new social and economic circumstances. Younger households, on the other hand, appear to have conserved less in comparison to their original consumption levels. While this suggests that these households may be less willing to conserve, it also suggests that younger households, having experienced less hardship in the past due to life course placement, could be somewhat ignorant of "how" to lead less resource-intensive lifestyles. The implications for educational programming are direct, and efforts to reorient thinking regarding priorities in educational outreach have already begun. Renewed interest in skill building, especially among young families, is apparent in state Cooperative Extension Services across the country as well as in adult education classes offered through community colleges or high schools. Examples include in- depth educational workshops on home food production and preservation as well as clothing construction, simple home repairs and maintenance and repair of home furnishings. In a broader sense, emphasis on skill building is renewed recognition of the household as a viable economic unit, one capable of providing for many of its own goods and services through home production, rather than through con- sumption in the traditional market-place (Burns, 1980). Skill building need not focus strictly on production 144 activities, however. Emphasis can just as easily be directed towards conservation activities, as this research suggestsit should. __._._ ufll’e‘ VII-"mm :- ‘- r BIBLIOGRAPHY BIBLIOGRAPHY Adler, Jerry; Carey, John; and Rogal, Kim. "A Slight Sigh of Relief." Newsweek, 4 August 1980, pp. 28-29. Allen, Frederick Lewis. The Big Change: America Transforms Itself, 1900-1950. New York: Harper & Row, 1952; Perennial Library, 1969. Allen, Frederick Lewis. Only Yesterday: An Informal History of the Nineteen—Twenties. New York: Harper & Brothers, 1931; Perennial Library, 1964. Allen, Frederick Lewis. Since Yesterday: The 1930's in America, September 3, 1929-September 3, 1939. Originally published, 1939; New York: Perennial Library, 1972. Andrews, Mary; Bubolz, Margaret; and Paolucci, Beatrice. "An Ecological Approach to Study of the Family." East Lansing, Mi., College of Human Ecology, Michigan State University, n.d. (mimeographed) Aries, Phillippe. Centuries of Childhood: A Social History of Family Life. Translated from the French by Robert Baldick. New York: Vintage Books, 1962. Atchley, Robert C. The Social Forces in Later Life: An Introduction to Social Gerontology. Belmont, Calif.: Wadsworth Publishing Company, 1972. Bailey, Ann Wiseman. "Socioeconomic Variables and Attitudes of Consumers Related to Energy Conservation Behavior." Oxford, Ohio, Miami University, 1980. (mimeographed) Baltes, Paul B. "Prototypical Paradigms and Questions in Life-Span Research on Development and Aging." The Gerontologist 13 (Winter l973):458-467. Baltes, Paul B., and Willis, Sherry L. "Life-Span Develop- mental Psychology, Cognitive Functioning and Social Policy." In Aging from Birth to Death: Interdis- ciplinary Perspectives, pp. 15-46. Edited by Matilda White Riley. Boulder, Colorado: Westview Press, for the American Association for the Advancement of Science, 1979. 145 146 Baltes, Paul B., and Willis, Sherry L. "Towards Psycho- logical Theories of Aging and Development." In Hand- book of the Psychology of Aging, pp. 128—154. Edited by James E. Birren and K. Warner Schaie. New York: Van Nostrand Reinhold Company, 1977. Blalock, Hubert M., Jr. Social Statistics. Revised 2nd ed. New York: McGraw—Hill Book Company, 1979. Boocock, Sarane Spence. "Historical and Sociological Research on the Family and the Life Cycle: Methodo— logical Alternatives." In Turning Points: Historical and Sociological Essays on the Family, pp. 5366-8394. Edited by John Demos and Sarane Spence Boocock. Amer- ican Journal of Sociolggy, Volume 84, Supplement. Chicago: University of Chicago Press, 1978. Boorstin, Daniel J. The Americans: The Democratic Experience. New York: Random House, 1973; Vintage Books, 1974. Borgatta, Edgar F. "My Student, the Purist: A Lament." The Sociological Quaterly 9 (Winter 1968):29-34. Brocklehurst, John. "Ageing and Health." In The Social Challenge onggeing, pp. 149—171. Edited by David Hobman. New York: St. Martin's Press, 1978. Burns, Scott. "Your Castle, Inc." Next, July/August 1980, pp. 34-39. Clark, John G.; Katzman, David M.: McKinzie, Richard D.; and Wilson, Theodore A. Three Generations in Twentieth Century America: Family, Community, and Nation. Home- wood, Illinois: The Dorsey Press, 1977. Clausen, John. "The Life Course of Individuals." In Th3 Sociology of Aging: Selected Readings, pp. 38-50. Edited by Robert C. Atchley and Mildred M. Seltzer. Belmont, Calif.: Wadsworth Publishing Company, 1976. Curtin, Richard T. "Consumer Adaptation to Energy Shortages." Paper prepared for presentation at the 83rd Annual Con- vention of the American Psychological Association, Division 23, Consumer Psychology, Chicago, Illinois, 31 August 1975. (mimeographed) Daly, Herman E. "On Thinking About Future Energy Require— ments." In Sociopolitical Effects of Energy Use and Policy. Study of Nuclear and Alternative Energy Sys- tems, Supporting Paper 5, pp. 229-241. Edited by Charles T. Unseld, Denton E. Morrison, David L. Sills, and C. P. Wolf. Washington, D.C.: National Academy of Sciences, 1979. 147 Demos, John. "Demography and Psychology in the Historical Study of Family—Life: A Personal Report." In Household and Family in Past Time, pp. 561-569. Edited by Peter Laslett and Richard Wall. London, England: Cambridge University Press, 1974. Demos, John. A Little Commonwealth: Family Life in Plymouth Colony. New York: Oxford University Press, 1970; Oxford University Press paperback, 1971; reprint, 1977. Dorf, Richard. Energy, Resources, and Policy. Reading, Mass.: Addison—Wesley Publishing Co., 1978. Duvall, Evelyn Millis. Marriage and Family Development. 5th ed. Philadelphia: J. B. Lippincott Company, 1977. Edelstein, Michael. "Energy, Conservation and Lifestyle." In The Social Dimensions of Energy Options Conference Proceedings, pp. 279—300. Edited by Joel Kameron. Mahwah, N.J.: Ramapo College of New Jersey, 1979. 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. Elder, Glen H., Jr. "Approaches to Social Change and the Family." In TurningyPoints: Historical and Socio- logical Essays on the Family, pp. 81-838. Edited by John Demos and Sarane Spence Boocock. American Journal of Sociology, Volume 84, Supplement. Chicago: University of Chicago Press, 1978a. Elder, Glen H., Jr. Children of the Great Depression: Social Change in Life Experience. ChiCago: The University of Chicago Press, 1974. Elder, Glen H., Jr. "Family History and the Life Course. lleransitions:The Family and the Life Course in Historical Perspective, pp. 17-64. Edited by Tamara K. Hareven. New York: Academic Press, 1978b. Energy Policy Project of the Ford Foundation. A Time to Choose: America's Emergy Future. Cambridge,.Mass.: Ballinger Publishing Co., 1974. Farhar, Barbara C.; Weis, Patricia; Unseld, Charles T.; and Burns, Barbara A. Public Opinion About Energy: A Literature Review. Solar Energy Research Institute, SERI/TR—53—155. Springfield, Va.: National Technical Information Service, U.S. Department of Commerce, 1979. 148 Foner, Anne. "Age in Society: Structure and Change." American Behavioral Scientist 19 (November/December 1975):144-l65. Foner, Anne. "Age Stratification and the Changing Family Group." In Turning Points: Historical and Sociological Essays on the Family, pp. 8340-8365. Edited by John Demos and Sarane Spence Boocock. American Journal of Sociology, Volume 84, Supplement. Chicago: University of Chicago Press, 1978. Foner, Anne, and Kertzer, David I. "Intrinsic and Extrinsic Sources of Change in Life-Course Transitions." In Aging from Birth to Death: Interdisciplinary Perspecthmxh pp. 121-136. Edited by Matilda White Riley. Boulder, Colorado: Westview Press, for the American Association for the Advancement of Science, 1979. Fritsch, Albert J. The Contrasumers: A Citizen's Guide to Resource Conservation. New York: Praeger Publishers, 1974. Gelman, David; Weathers, Diane; Whitman, Lisa; Abramson, Pamela; Maitland, Terrence; and Copeland, Jeff B. "The Games Teen-Agers Play." Newsweek, 1 September 1980. PP. 48-53. Gladhart, Peter Michael. "Determinant of the Demand for Residential Energy: A Family Level Analysis." Revised. Michigan Agricultural Experiment Station Journal Article No. 8317. East Lansing, Mi., Department of Family Ecology, Michigan State University, 28 July 1977a. (mimeographed) Gladhart, Peter Michael. "Energy Conservation and Life- styles: An Integrative Approach to Family Decision Making." Michigan Agricultural Experiment Station Journal Article Number 7915. Journal of Consumer Studies and Home Economics 1 (l977b):265-277. Gladhart, Peter Michael, and Roosa, Mark W. "Family Life- style and Energy Consumption: A Family-Energy Adaptation Model." Family Energy Project Occasional Paper No. 10. East Lansing, Mi., Institute for Family and Child Study, Michigan State University, November 1978. (mimeographed) Gladhart, Peter Michael; Zuiches, James J.; and Morrison, Bonnie Maas. "Impacts of Rising Prices upon Resi- dential Energy Consumption, Attitudes and Conservation Policy Acceptance." In Energy Policy in the United States: Social and Behavioral Dimensions, pp. 60-78. Edited by Seymour Warkov. New York: Praeger Publishers, 1978. 149 Gladhart, Peter Michael; Zuiches, James J.; and Morrison, Bonnie Mass. "Impacts of Rising Prices Upon Residential Energy Consumption, Attitudes, and Conservation Policy Acceptance." Occasional Paper No. 9, Project 3152. East Lansing, Mi., Michigan Agricultural Experiment Station, Michigan State University, 1977. (mimeographed) Glass, Gene V., and Stanley, Julian C. Statistical Methods in Education and Psychology. Englewood Cliffs, N.J.: Prentice-Hall, 1970. P. Goldman, Eric F. The Crucial Decade and After: America, 1945-1960. New York: Vintage Books, 1960. Gordon, Michael. The American Family: Past, Present, and Future. New York: Random House, 1978. Gordon, Robert A. "Issues in Multiple Regression." American V Journal of Sociolggy 73 (March l968):592-6l6. Gottlieb, David. "Texans' Responses to President Carter's Energy Proposals." In Energy Policy in the United States: Social and Behavioral Dimensions, pp. 33-44. Edited by Seymour Warkov. New York: Praeger Publishers, 1978. Gottlieb, David, and Matre, Marc. Sociological Dimensions of the Energy Crisis. Project E/S-S of the Governor‘s Energy Advisory Council. Houston, Texas, 1974. Greiner, Delores. "Energy Changes Throughout This Century: A Historical Look at My Own Home." East Lansing, Mi., College of Human Ecology, Michigan State University, 1980. (mimeographed) Hareven, Tamara K. "Introduction: The Historical Study of the Life Course." In Transitions: The Family and the Life Course in Historical Perspective, pp. 1-16. Edited by Tamara K. Hareven. New York: Academic Press, 1978. Hareven, Tamara K. "Preface." In Anonymous Americans: Explorations in Nineteenth-Century Social History, pp. vii-xiv. Edited by Tamara K. Hareven. Englewood Cliffs, N.J.: Prentice-Hall, 1971. Harris, Craig K.; Keith, Joanne G.; and Wilhelm, Mari S. "Final Report on the Evaluation of Pilot Project Conserve in Michigan: A Computerized Residential Energy Audit Program." East Lansing, Mi., Institute for Family and Child Study, Michigan State University, 19 August 1980a. (mimeographed) 150 Harris, Craig K., and Keith, Joanne G. "Preliminary Findings on 'Project Conserve' Pilot Study—~Phase II Evaluation of a Computerized Residential Energy Audit Program." East Lansing, Mi., Institute for Family and Child Study, Michigan State University, 2 June 1980b. (mimeographed) Harris, Craig K., and Keith, Joanne G. "Preliminary Findings on the Evaluation of Pilot Project Conserve in Michigan: A computerized Residential Energy Audit Program." East Lansing, Mi., Institute for Family and Child Study, Michigan State University, 16 June 1980c. (mimeographed) Herendeen, R. A. "Affluence and Energy Demand." Paper pre- pared for presentation at the Winter Annual Meeting, American Society of Mechanical Engineers, Detroit, Mi., 11-15 November 1973. (mimeographed) Hill, Reuben. Family Development in Three Generations: A Longitudinal Study of Changing Family Patterns of Planning and Achievement. Cambridge, Mass.: Schenkman Publishing Co., 1970. Hodgson, Godfrey. America In Our Time: From World War II to Nixon, What Happened and Why. New York: Doubleday and Co., 1976; New York: Vintage Books, 1978. Hofstadter, Richard. The Age of Reform: From Bryan to F.D.R. New York: Vintage Books, 1955. Hogan, Mary Janice. "Energy Conservation: Family Values, Household Practices, and Contextual Variables." Ph.D. dissertation, Michigan State University, 1976. Hornback, Kenneth Elmer. "Orbits of Opinion: The Role of Age in the Environmental Movement's Attentive Public, 1968- 1972." Ph.D. dissertation, Michigan State University, 1974. Kantor, David, and Lehr, William. Inside the Family: Toward a Theory of Family Process. San Francisco: Jossey-Boss Publishers, 1977. Keith, Joanne Goodman. "Direct Household Energy Consumption, l973-74, 1975-76: The Impact of Family Microdecisions upon Levels of Consumption." Ph.D. dissertation, Michigan State University, 1977. Kerlinger, Fred N., and Pedhazur, Elazar J. Multiple Regression in Behavioral Research. New York: Holt, Rinehart and Winston, 1973. 151 Kilkeary, Rovena, and Thompson, Patricia J. "The Energy Crisis and Decision—Making in the Family." Bronx, N.Y., Department of Family and Consumer Studies, Herbert H. Lehman College, City University of New York, 1975. (mimeographed) King, Jill A. "The Impact of Energy Problems on the Income Distribution." Paper prepared for presentation at the Society for Government Economists Session, joint meetings of the Allied Social Science Association, Atlantic City, New Jersey, 18 September 1976. (mimeographed) Klausner, Samuel Z. "Household Organization and Use of Electricity." In Energy Policy in the United States: Social and Behavioral Dimensions, pp. 45—59. Edited by Seymour Warkov. New York: Praeger Publishers, 1978. Knutson, Bonnie J. "Energy Information Classroom Experiment: A bbasure of Student Beliefs and Attitudes." M.A. thesis, Michigan State University, 1979. Kobrin, Frances E. "The Fall in Household Size and the Rise of the Primary Individual in the United States." In The American Family in Social—Historical Perspective. 2nd ed., pp. 69—81. Edited by Michael Gordon. New York: St. Martin's Press, 1978. Kuhn, Alfred. Unified Social Science. Homewood, 111.: Dorsey Press, 1975. Labovitz, Sanford. "The Assignment of Numbers to Rank Order Categories." American Sociological Review 35 (June 1970):515—524. LaFeber, Walter, and Polenberg, Richard. The American Century: A History of the United States since the 18905. New York: John Wiley & Sons, 1975. Lerner, Richard M. and Ryff, Carol D. "Implementation of the Life-Span View of Human Development: The Sample Case of Attachment." In Life-Span Development and Behavior. Vol. 1, pp. 1-44. Edited by Paul B. Baltes. New York: Academic Press, 1978. Manchester, William. The Glory and the Dream: A Narrative Historyyof America, 1932-1972. Boston, Mass.: Little, Brown and Company, 1974; New York: Bantam Books, 6th printing, 1979. Mason, Karen Oppenheim; Mason, William M.; Winsborough, H. H.; and Poole, W. Kenneth. "Some Methodological Issues in Cohort Analysis of Archival Data." American Socio- logical Review 38 (April 1973):242-258. 152 Mayer, J. Allan, and McGuine,Stryker. "A Killer Heat Wave." Newsweek, 14 July 1980, p. 29. Milstein, Jeffrey S. "How Consumers Feel About Energy: Attitudes and Behavior during the Winter and Spring 1976-77." In Energy Policy in the United States: Social and Behavioral Dimensions, pp. 79-90. Edited by Seymour Warkov. New York: Praeger Publishers, 1978. Morrison, Bonnie Maas. "The Closing Energy Resource Environ— ment: A Multi-disciplinary Research Team Approach to the Study of a Complex Problem." Paper prepared for presentation at the 66th Annual Meeting of the American Home Economics Association, San Antonio, Texas, 24 June 1975a. (mimeographed) Morrison, Bonnie Maas. "Socio-Physical Factors Affecting Energy Consumption in Single Family Dwellings: An Empirical Test of a Human Ecosystems Model." Paper prepared for presentation at the Joint NBS/NCSBCS Conference on Research and Innovation in the Building Regulatory Process, Cranston, Rhode Island, 21 September 1976. (mimeographed) Morrison, Bonnie Maas. "Socio—Physical Factors Affecting Energy Consumption in Single Family Dwellings: An Empirical Test of a Human Ecosystems Model." Ph.D. dissertation, Michigan State University, 1975b. Morrison, Bonnie Maas, and Gladhart, Peter Michael. "Energy and Families: The Crisis and the Response." Journal of Home Economics 68 (January l976):15-18. Morrison, Bonnie Maas; Gladhart, Peter Michael; Zuiches, James J.; Keith, Joanne G.; Keefe, Dennis; and Long, Brenda R. "Energy and Families: The Crisis and the Response." Journal of Home Economics 70 (Winter 1978a): 18-21. ‘Morrison, Bonnie Maas; Keith, Joanne G.; and Roosa, Mark. "Household Energy Conservers and Non-Conservers: A Discriminant Analysis Profile." Paper prepared for presentation at the Michigan Home Economics Association Conference, Research Reporting Session, Mackinac Island, Mi., 11 August 1978b. (mimeographed) Morrison, Bonnie Maas; Keith, Joanne Goodman; and Zuiches, James J. "Impacts on Household Energy Consumption: An Empirical Study of Michigan Families." In Sociopolitical Effects of Energy Use and Policy. Study of Nuclear and Alternative Energy Systems, Supporting Paper 5, pp.7-33. Edited by Charles T. Unseld, Denton E. Morrison, David L. Sills, and C. P. Wolf. Washington, D.C.: National Academy of Sciences, 1979. 153 Morrison, Denton E. "Equity Impacts of Some Major Energy Alternatives." Paper prepared for the Sociopolitical Risk/Impact Resource Group, Risk/Impact Panel, Com- mittee on Nuclear and Alternative Energy Systems (CONEAS), National Research Council, National Academy of Sciences. East Lansing, Mi., Department of Sociology, Michigan State University, May 1977. (mimeographed) Morrison, Peter A. "Demographic Trends Impinging on Energy Use." In Sociopolitical Effects CM? Energy Use and Policy. Study of Nuclear and Alternative Energy Sys- tems, Supporting Paper 5, pp. 79—99. Edited by Charles T. Unseld, Denton E. Morrison, David L. Sills, and C. P. Wolf. Washington, D.C.: National Academy of Sciences, 1979. Mulligan, Linda W. "Energy Scarcity and Family Privacy: ' Implications for Social Policy." Paper prepared for presentation at the annual meetings of the North Central Sociological Association, Louisville, Kentucky, 6-8 May 1976. (mimeographed) Murray, James R.; Minor, Michael J.; Bradburn, Norman M.; Cotterman, Robert F.; Frankel, Martin; and Pisarski, Alan E. "Evolution of Public Response to the Energy Crisis." Science 184 (19 April 1974):257-263. Neter, John, and Wasserman, William. Applied Linear Statis- tical Models: Regression, Analysis of Variance, and Experimental Designs. Homewood, Ill.: Richard D. Irwin, 1974. Neugarten, Bernice L., and Datan, Nancy. "Sociological Perspectives on the Life Cycle." In Life-Span Develop— mental Psychology: Personality and Socialization, pp. 53-69. Edited by Paul B. Baltes and K. Warner Schaie. New York: Academic Press, 1973. Newman, Dorothy K., and Day, Dawn. The American Energy Consumer. A Report to the Energy Policy Project of the Ford Foundation. Cambridge, Mass.: Ballinger Publishing Co., 1975. Newman, Dorothy K., and Day, Dawn. "Energy, the Environment, and the Poor." Paper prepared for presentation at the Society for the Study of Social Problems, August 1974. (mimeographed) Nie, Norman; Hull, C. Hadlai; Jenkins, Jean G.; Steinbrenner, Karin; and Bent, Dale H. SPSS: Statistical Package for the Social Sciences. 2nd ed. New York: McGraw-Hill Book Company, 1975. 154 Norval, Glenn D. Cohort Analysis. Sage University Paper Series on Quantitative Applications in the Social Sciences, No. 07-005. Beverly Hills, Calif.: Sage Publications, 1977. Odum, Howard T., and Odum, Elisabeth G. Energy Basis for Man and Nature. New York: McGraw-Hill Book Company, 1976. Olsen, Marvin E. "Public Acceptance of Energy Conservation." In Energnyolicy in the United States: Social and Behavioral Dimensions, pp. 91-109. Edited by Seymour Warkov. New York: Praeger Publishers, 1978. Olsen, Marvin E., and Goodnight, Jill A. Social Aspects of Energy Conservation. Battelle Human Affairs Research Centers. Vancouver, Wa.: Pacific Northwest Regional Commission, 1977. Paolucci, Beatrice, and Hogan, M. Janice. "The Energy Crisis and the Family." Journal of Home Economics 65 (December 1973):12-15. Perlman, Robert, and Warren, Roland L. "Effects of the Energy Crisis on Households of Different Income Groups." Paper prepared for presentation at the annual meetings of the Society for the Study of Social Problems, San Francisco, Calif., August 1975. (mimeographed) Perlman, Robert, and Warren, Roland L. Families in the EnergyyCrisis: Impacts and Implications for Theory and Policy. Cambridge, Mass.: Ballinger Publishing Co., 1977. Quinn, Jane Bryant. "The Affluent Elders." Newsweek, 4 August 1980, p. 53. Ratner, Sidney; Soltow, James H.; and Sylla, Richard. The Evolution of the American Economy: Growth, Welfare, and Decision Making. New York: Basic Books, 1979. Riley, Matilda White. "Age Strata in Social Systems." In Handbook of Aging and the Social Sciences, pp. 189-217. Edited by Robert H. Binstock and Ethel Shanas. New York: Van Nostrand Reinhold Company, l976a. Riley, Matilda White. "Aging and Cohort Succession: Inter— pretations and Misinterpretations." Public Opinion Quarterly 37 (Spring 1973):35-49. 155 Riley, Matilda White. "Aging, Social Change and Social Policy." In Aging fromyBirth to Death: Interdisciplin- ary Perspectivnn, pp. 109~120. Edited By Matilda White Riley. Boulder, Colorado: Westview Press, for the American Association for the Advancement of Science, 1979a. Riley, Matilda White. ”Introduction: Life-Course Perspectives." In Aging from Birth to Death: Interdisciplinary Per- spectives, pp. 3113. Edited by Matilda White Riley. Boulder, Colorado: Westview Press, for the American Association for the Advancement of Science, 1979b. Riley, Matilda White. "Social Gerontology and the Age Stratification of Society." In The Sociology of Aging: Selected Readings, pp. 22-37. Edited by Robert C. Atchley and Mildred M. Seltzer. Belmont, Calif.: Wadsworth Publishing Company, 1976b. Riley, Matilda White; Johnson, Marilyn; and Foner, Anne. Aging and Socieny.y‘anume Three: A Sociologynof Age SnratifiEaEion. New York: Russell Sage Foundation, 1972. Ryder, Norman B. "The Cohort as a Concept in the Study of Social Change. American Sociological Review 30 (December l965):843-861. Scheiber, Harry N.; Vatter, Harold C.; and Faulkner, Harold Underwood. American Economic Histony. 9th ed. New York: Harper & Row, 1976. Schwartz, T. P. "Human Societal Energy Systems in Evolu- tionary Perspective." Paper prepared for presentation at the 70th Annual Meeting of the American Sociological Association, 25-29 August 1975. (mimeographed) Schwartz, T. P. "Short End of the Shortage? An Appraisal of the Purported Social Consequences of the l973-74 Energy Shortage." In Energy Policy in the United States: Social and Behayionnl Dimensions, pp. 110-131. Edited by Seymour Warkov. New York: Praeger Publishers, 1978. Schwartz, T. P., and Schwartz—Barcott, Donna. "The Short End of the Shortage: On the Self-Reported Impact of the Energy Shortage on the Socially Disadvantaged." Paper prepared for presentation at the annual meeting of the Society for the Study of Social Problems, 1974. (mimeographed) Sheils, Merrill; Henkoff, Ronald; Morris, Holly; and Belton, Don. "The Siege of the Sun." Newsweek, 28 July 1980, pp. 35—36. 156 Shippee, Glenn. "The Psychology of Energy Consumption and Conservation: A Review and Conceptual Analysis." East Lansing, Mi., Department of Psychology, Michigan State University, 1980. (mimeographed) Shippee, Glenn, and Leedom, Nancy. "Conservation as a Renewable Resource and the Role of Social Science." East Lansing, Mi., Department of Psychology, Michigan State University, 1980. (mimeographed) Shorter, Edward. The Making of the Modern Family. New York: Basic Books, 1975, 1977. Sills, David L. "Social Science Research and the Formation of Energy Policy." In Sociopolitical Effects of Energy Use and Policy. Study of Nuclear and Alternative Energy Systems, Supporting Paper 5, pp. 417-427. Edited by Charles T. Unseld, Denton E. Morrison, David L. Sills, and C. P. Wolf. Washington, D.C.: National Academy of Sciences, 1979. Smelser, Neil J. "Energy Restriction, Consumption, and Social Stratification." In Sociopolitical Effects of Energy Use and Poligy. Study of Nuclear and Alternative Energy Systems, Supporting Paper 5, pp. 215-228. Edited by Charles T. Unseld, Denton E. Morrison, David L. Sills, and C. P. Wolf. Washington, D.C.: National Academy of Sciences, 1979. Smelser, Neil:I.,and Halpern, Sydney. "The Historical Triangulation of Family, Economy, and Education." In Turning Points: Historical and Sociological Essays on the Family, pp. 8288-8315. Edited by John Demos and Sarane Spence Boocock. American Journal of Sociology, Volume 84, Supplement. Chicago: University of Chicago Press, 1978. Smith, Richard B. "Household Energy Adjustments." Speech prepared for presentation at the National Agricultural Outlook Conference, Washington, D.C., 17 November 1976. (mimeographed) Spitzer, Alan B. "The Historical Problem of Generations." American Histonical Review 78 (December l973):1353- 1385. Steinbeck, John. The Grapen of Wrath. New York: The Viking Press, 1932, 1967. Stobaugh, Robert, and Yergin, Daniel, eds. Energy Future: Renort of the Energy Project at the Harvafd’BuSiness School} New Yofk: Random House, 1979. 157 Taylor, John F. "The Social Implications of the Energy Crisis." Paper prepared for presentation at the sym— posium "Energy and Life," Michigan State University, 1 March 1974. (mimeographed) Thompson, Phyllis T., and MacTavish, John. "Energy Problems: Public Beliefs, Attitudes and Behaviors." Allendale, Mi., Urban and Environmental Studies Institute, Grand Valley State Colleges, 1976. (mimeographed) Tilly, Louise, and Scott, Joan W. Women, Work, and Family. New York: Holt, Rinehart and Winston, 1978. Uhlenberg, Peter. "Changing Configurations of the Life Course." In Transitions: The Family and the Life Course in Historical PerspecEIVe, pp. 65-97. Edited by Tamara K. Hareven. New York: Academic Press, 1978. Unseld, Charles T.; Clark, C.Thomas; Lieberman, Louis; Rosenblatt, A. Michelle; and Boiko, Clair. The Impact of Rising Energy Costs ontflszlderly Poor in New York State. Aibany, New York: Welfare Research, 1978. Unseld, Charles T.; Morrison, Denton E.; Sills, David L.; and Wolf, C. P. Sociopolitical Effects of Energy Use and Policy. Study of Nuclear and Alternative Energy Systems, Supporting Paper 5. Washington, D.C.: National Academy of Sciences, 1979. U.S. Department of Commerce. Bureau of the Census. Current Population Reports. Series P-20, No. 334, "Demographic, Sodial and Economic Profile of States: Spring 1976." Washington, D.C.: Government Printing Office, 1979. Verway, David I., ed. Michigan Statistical Abstract. 14th ed. East Lansing, Mi., Graduate School of Business Administration, Michigan State University, 1979. Walter, John P., and Zentner, Rene D. "Social Implications of the Energy Crisis: Are There Any?" In Energy Policy in the United States: Social and Behavioral Dimensions, pp. 132-147. Edited by Seymour Warkov. New York: Praeger Publishers, 1978. Walton, Gary M. and Miller, Roger LeRoy. Economic Issues in American History. San Francisco, Calif.: Canfield Press,ii978. Waring, Joan M. "Social Replenishment and Social Change: The Problem of Disordered Cohort Flow." American Behavioral Scientist l9 (November/December 1975): 237-256. 158 Warkov, Seymour. "Introduction: Energy and the Social Sciences." In Energy Policy in the United States: Social and Behavioral Dimensions, pp. xv—xxiii. Edited by Seymour Warkov. New York: Praeger Publishers, 1978. Whitehead, David. "Ageing and the Mind." In The Social Challenge of Ageing, pp. l97-221. Edited by David Hobman. New York: St. Martin's Press, 1978. Wicks, Malcolm. Old and Cold: Hypothermia and Social Policy. London: Heinemann Educational Books, 1978. Williams, John S., Jr.; Kruvant, William; and Newman,Dorothy. "Metropolitan Impacts of Alternative Energy Futures." In Sociopolitical Effects of Energy Use and Policy. Study of Nuclear and AIternative Energy Systems, Sup- porting Paper 5, pp. 35-77. Edited by Charles T. Unseld, Denton E. Morrison, David L. Sills, and C. P. Wolf. Washington, D.C.: National Academy of Sciences, 1979. Winsborough, Halliman H. "Changes in the Transition to Adulthood." In Aging from Birth to Death: Interdis- ciplinary Perspectives, pp. 137—152. Edited by Matilda White Riley. Bouider, Colorado: Westview Press, for the American Association for the Advancement of Science, 1979. Wolf, C. P. "Recommendations for Future Research on the Sociopolitical Impacts of Energy." In Sociopolitical Effects of Energy Use and Policy. Study of Nuclear and Alternative Energy Systems, Supporting Paper 5, pp. 377-415. Edited by Charles T. Unseld, Denton E. Morrison, David L. Sills, and C. P. Wolf. Washington, D.C.: National Academy of Sciences, 1979. Wolff, Kurt. Tne Biologigal, Sociological and Psychological Aspects of Aging. Spfingfield, Ill.: Charles C. Thomas, 1959. Woodward, Kenneth L., and Salholz, Eloise. "When Age Doesn't Matter." Newsweek, 11 August 1980, pp. 74-76. Zimmerman, Carol. "The Family Life Cycle and Energy Research)‘ n.p., n.d. (mimeographed) Zuiches, James J.; Harris, Craig; Morrison, Bonnie Maas; Keith, Joanne; and Boyd, James. "Final Report on 'Project Conserve' Phase I: Development and Operation of a Computerized Residential Energy Audit Program." East Lansing, Mi., Institute for Family and Child Study, Michigan State University, 28 April 1978. (mimeographed) APPENDICES APPENDIX A COMPARISON OF TOTAL PILOT CONSERVE SAMPLE, TOTAL ENERGY CONSUMPTION DATA SUBSAMPLE AND RESEARCH SUBSAMPLE Table A-l.--Household by Age of Principal or Oldest Income Earner: Comparison of Total Pilot Conserve Sample, Total Energya Consumption Data Subsample and Research Subsample, 1978 Total Energy Age of Principal Co::::irepéali;;le anZnZEEZZiEL 555:3: or Oldest Income Earner % N % N s N 19 thru 36 32.6 (359) 30.5 (184) 32.2 (140) 37 thru 40 9.3 (102) 10.1 (61) 10.1 (44) 41 thru 44 7.2 (79) 6.6 (40) 6.4 (28) 45 thru 59 26.5 (291) 26.0 (157) 26.4 (115) 60 thru 63 7.5 (83) 7.6 (46) 8.3 (36) 64 thru 94 15.4 (169) 17.6 (106) 16.6 (72) Missing 1.5 (17) 1.5 (9) --- (0) Total 100.0 (1100) 100.0 (603) 100.0 (435) Mean 46.6 47.6 47.1 a . . Percentages have been rounded in some instances. 159 Table A-2.-~Household by Total Family Income: 160 Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Subsample, 19775“b Total Energy . Total Pilot Consumption Research Total Family Conserve Sample Data Subsample Subsample Income 96 N % N % N Less than $5,000 6.7 (74) 6.1 (37) 6.0 (26) $5,000-$9,999 10.6 (117) 11.8 (71) 13.6 (59) $10,000-$14,999 15.5 (170) 13.9 (84) 15.4 (67) $15,000-$l9,999 19.1 (210) 19.9 (120) 22.3 (97) $20,000-$24,999 14.6 (161) 17.7 (107) 19.3 (84) $25,000 or more 19.4 (213) 20.7 (125) 23.4 (102) Missing 14.1 (155) 9.8 (59) --- (0) Total 100.0 (1100) 100.0 (603) 100.0 (435) Median $17,650 $18,300 $18,375 a . . Percentages have been rounded in some instances. bData collected in 1978 reflecting 1977 income level. 161 Table A-3.--Household by Educational Attainment of Principal or Oldest Income Earner: Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Subsample, 1978a —_ Total Energy . Total Pilot Consumption Research Education of . . Conserve Sample Data Subsample Subsample Pr1nc1pal or Oldest Income % N % N % N Earner Less than high 21.5 (237) 18.9 (114) 20.2 (88) school High school 29.8 (328) 29.7 (179) 31.0 (135) Some college 20.9 (230) 21.4 (129) 20.7 (90) Finished college 13.8 (152) 14.8 (89) 14.5 (63) Graduate school 11.9 (131) 13.8 (83) 13.6 (59) Missing 2.0 (22) 1.5 (9) --- (0) Total 100.0 (1100) 100.0 (603) 100.0 (435) a Percentages have been rounded in some instances. Table A-4.—-Household by Number of Members: Conserve Sample,Total Energy Consumption Data Subsample and Research Subsample, 162 1978a Comparison of Total Pilot Total Energy Number in Total Pilot Consumption Research Household Conserve Sample Data Subsample Subsample % N % N % N l 8.0 (88) 7.8 (47) 7.6 (33) 2 28.9 (318) 29.2 (176) 29.4 (128) 3 or 4 40.8 (449) 41.8 (252) 42.3 (184) 5 to 7 20.5 (225) 19.4 (117) 19.5 (85) 8 or more 1.2 (13) 1.3 (8) 1.1 (5) Missing .6 (7) .5 (3) --- (0) Total 100.0 (1100) 100.0 (603) 100.0 (435) Mean 3.37 3.37 3.34 a . . Percentages have been rounded in some 1nStances. Table A-5.--Housing by Dwelling Unit Type: 163 Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Subsample, 1978a Total Energy Total Pilot Consumption Research Dwelling Type Conserve Sample Data Subsample Subsample % N % N % N Single family 91.5 (1006) 93.4 (563) 94.7 (412) dwelling Single family 1.8 (20) 1.8 (11) 1.6 (7) converted to multiple family Duplex 1.3 (14) 1.2 (7) .9 (4) Two-family .8 (9) 1.0 (6) .2 (1) dwelling Four-plex .2 (2) .2 (1) .2 (l) Townhouse .5 (5) .3 (2) .2 (1) Apartment .6 (7) .2 (l) --- (0) Mobile Home 3.0 (33) 1.7 (10) 2.1 (9) Other .4 (4) .3 (2) --- (0) Total 100.0 (1100) 100.0 (603) 100.0 (435) a O I Percentages have been rounded in some instances. 164 Table A-6.--Housing by Form of Tenure: Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Subsample, 1978 — Total Energy Own or Total Pilot Consumption Research Buying Home Conserve Sample Date Subsample Subsample % N 56 N % N Yes 93.8 (1032) 96.8 (584) 97.7 (425) No 6.0 (66) 3.2 (19) 2.3 (10) Missing .2 (2) --' (O) --- (0) Total 100.0 (1100) 100.0 (603) 100.0 (435) Table A-7.——Housing by Number of Rooms: Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Subsample, 1978a Total Energy Total Pilot Consumption Research Number of Conserve Sample Data Subsample Subsample Rooms % N % N % N 2 to 5 36.0 (396) 33.7 (203) 34.0 (148) 6 or 7 42.5 (467) 46.3 (279) 46.4 (202) 8 or more 21.4 (235) 19.9 (120) 19.5 (85) Missing .2 (2) .2 (l) —-- (0) Total 100.0 (1100) 100.0 (603) 100.0 (435) Mean 6.33 6.35 6.32 a I I Percentages have been rounded in some instances. V. t thi‘f‘o'. 165 Table A-8.-—Housing by Type of Heating Fuel: Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Subsample, 1978 Total Energy Type Of Co:::::ePS::;le Daign:::::;;ie 565::26i: Heating Pnel % N % N % N Electric 5.0 (55) 4.8 (29) 4.8 (21) Natural gas 64.7 (712) 73.8 (445) 71.5 (311) Fuel oil 24.9 (274) 19.4 (117) 22.1 (96) Propane 2.9 (32) 1.3 (8) 1.6 (7) Coal .2 (2) --- (O) --- (O) Kerosene .1 (l) --- (0) --- (0) Wood 1.3 (14) --- (O) --- (0) Oil and wood .1 (1) --- (O) ’-- (0) Missing .8 (9) .7 (4) --- (0) Total 100.0 (1100) 100.0 (603) 100.0 (435) 166 Table A—9.--Household by Total Btu's Consumed per Heating Degree Day, 1976—77: Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Subsamplea Total Energy Total Btu's per Total Pilot b Consumption Research Heating Degree Conserve Sample Data Subsample Subsample Day, 1976-77 -————————- % N % N 96 N Very low --- (NA) 4.6 (28) 4.1 (18) consumption Low consumptiond --- (NA) 44.9 (271) 47.6 (207) Medium e --- (NA) 41.6 (251) 41.8 (182) consumption High consumptionf --- (NA) 8.8 (53) 6.4 (28) Total --- (NA) 100.0 (603) 100.0 (435) Mean 26,000 25,000 Minimum 3,400 3,400 Maximum 70,200 61,300 aPercentages have been rounded in some instances. bData not available for total Pilot Conserve sample. c3,000-13,000 Btu's per heating degree day. d13,000-25,000 Btu's per heating degree day. e25,000-37,000 Btu's per heating degree day. f37,000-71,000 Btu's per heating degree day. 167 Table A-10.-—Household by Total Btu's Consumed per Heating Degree Day, 1978-79: Comparison of Total Pilot Conserve Sample, Total Energy Consumption Data Subsample and Research Subsample Total Energy Total Btu's per . Total Pilot a Consumption Research Heating Degree Conserve Sample Data Subsample Subsample Day, 1978—79 111-,_111 % N % N 915 N Very low b —-— (NA) 6.1 (37) 4.6 (20) consumption Low consumptionC -—- (NA) 50.4 (304) 54.9 (239) Medium d --- (NA) 37.0 (223) 36.1 (157) consumption High consumptione --- (NA) 6.5 (39) 4.4 (19) Total --- (NA) 100.0 (603) 100.0 (435) Mean 24,000 24,000 Minimum 3,200 3,200 Maximum 69,000 58,000 aData not available for total Pilot Conserve sample. b3,000-13,000 Btu's per heating degree day. C13,000-25,000 Btu's per heating degree day. d25,000-37,000 Btu's per heating degree day. e37,000-71,000 Btu's per heating degree day. 168 Table A—11.--Household by Total Change in Btu's Consumed per Heating Degree Day, 1976-77 to 1978-79: Conserve Sample, Total Energy Consumption Data Subsample and Research Subsamplea Comparison of Total Pilot Total Change in Total Energy Btu's per Heating Total Pilot Consumption Research Degree Day, 76-77 Conserve Sample Data Subsample Subsample to 78-79 % N % N % N High con- --- (NA) 11.1 (67) 9.4 (41) servation Medium con- --- (NA) 16.4 (99) 17.7 (77) servation Low conservatione —-- (NA) 41.1 (248) 42.1 (183) No changef --- (NA) 1.3 (8) 1.4 (6) Increased -—- (NA) 30.0 (181) 29.4 (128) consumptiong Total —-— (NA) 100.0 (603) 100.0 (435) Mean -1,500 -1,400 Minimum —20,000 -20,000 Maximum 17,000 14,000 aPercentages have been rounded in some instances. bData not available for total Pilot Conserve sample. C-20,000 through -5,000 Btu's per heating degree day. d--5,000 through -2,500 Btu's per heating degree day. e-2,500 through -10 Btu's per heating degree day. f-10 through 10 Btu's per heating degree day. 910 through 17,000 Btu's per heating degree day. 169 Table A—12.--Household by Percent Change in Btu's Consumed per Heating Degree Day, Conserve Sample, Total Energy Consumption Data Subsample and Research Subsamplea Percent Change in .__1 1976-77 to 1978-79: Total Energy Comparison of Total Pilot Btu's er Heatin Total Pilot Consumption Research p g Conserve Sample Data Subsample Subsample Degree Day, 76-77 to 78-79 % N % N % N High con- --- (NA) 10.4 (63) 8.7 (38) servationC Medium con- --- (NA) 17.2 (104) 18.9 (82) servation Low conservatione —-- (NA) 37.5 (226) 37.7 (164) No changef --- (NA) 8.3 (50) 7.6 (33) Increased --- (NA) 26.5 (160) 27.1 (118) consumptiong Total —-- (NA) 100.0 (603) 100.0 (435) Mean -5.1% -4.9% Minimum -64.6% -64.6% Maximum 71.9% 71.9% aPercentages have been rounded in some instances. bData not available for total Pilot Conserve sample. C-7O.0% through -20.0% in Btu's per heating degree day. d-20.0% through -l0.0% in Btu's per heating degree day. e-10.0% through -0.5% in Btu's per heating degree day. f-0.5% through 0.5% in Btu's per heating degree day. 90.5% through 80.0% in Btu's per heating degree day. APPENDIX B DESCRIPTION OF AGE STRATIFICATION OF PRINCIPAL OR OLDEST INCOME EARNERS: YEAR OF BIRTH, AGE IN 1977, YEARS OF EXPERIENCE WITH DEPRIVATION AND SHORTAGES 170 .ucw>o vnu >3 poucozaucm on Ou conuEhouoc nacho ozu :u cocauocfl on Cu gecko Cu oucouusuoc m.u:c>o ozu vauuop puo wumo> m on cu on: doscu>uccm co mask .A>uueou uo mucouma con: uuwQEw ozu Lcscusu cououuuu ..o.a. mucouucaxo uuwu_ccu cmxu nozuou uowumc coo: wows: mm) uce>o HoULLOumu; ecu no uumaEu osu mo uceewusmoozu .oH succuzu hm mumw> n mucu gonEo>oz Ou c_o~ Huuec .u he: puuoz nod zc:c~:u m mueo> I ave“ oumu Ou oNon ucLOuuo .momoH ozu mo coummounoo one “v concuzu u muoo> n mvoa wueu ou “can ueLEcumo .HH um: capo: ”who: uocaEu Co mumw> can mount one .pouucumccu mo: ume> uncuu osu abut uucaEu mssu humo> ocucSuucou muu no one; ccouom ogu Oucm coumca uco>o Lowe echo uueoEH .pouopmmcou uo: mm: umo> umcu .Accmmmeuooo onu ..o.uo uno> e we mnucoe o uncuu ozu auzuwa coupsuuo uooQEu Heuuucm NH .umo> ouuuco ecu New pofismmm we: uco>u ecu EOuu 00:6:Hucu .AM he: puuox ..o.u. umo> a we mzucoe o umuuu ozu Cuzuma pouusuuo uco>o no no uoeafim ”cmuuCu ecu on .uocan ocuonHOu osu nu conuEquop who: oomuuonm vco couuo>uuoev uo muoo>n .muocumo oEoucw ecu Lupus mo uoo> caucusnoucp cu umo> wouuomon onu we won: on Ou boom owsou~o musk .hhma Cu posuoou mo) hocuoo oEoocM umwcuo uc Hoauucuua ecu ho» couozv ~o>oH woo uncuCOHOCGuzu ecu uozu mcoE mm: couuaizmmo onu .mbmH >6: :u pmuuuaaou one) wuov umaouomm 6 ea we Haoa o~ o co mama H oa no comm N ma mo NHmH n b no vmou 0 ma mm uomu m on we Mao” m 0 Ne mmou 0 ma mm Nomfi 0 0H no vfiom o m Av cmom 0 on we nomu mu m~ No r_o~ v~ e co hmou u oH mm vow” N Va He mama HA n on mmma 0 ea No mama Nu ma co buofl mu N on onma 0 ad Hm coma m Na om muou o H on ovo~ N mu om noon m um mm onou ca 0 on Avon N ou or moon 0 ~n hm 0N9" NN 0 mm Nvma N ofi on ooma 6 HH om HNOM m 0 en mood M ed on ooou m A" mm NNoH o o no coma m ou or moou c Mu cm mNou m o No mood n ma mo Noon m .m mm vNoH ha c an econ N o— co mood mu uH Nm mNo— «N 0 on head e on no econ v an Hm 0N0" o o mN ovou v mu Nb mcou o ufi 0m nNoH o o wN ovou 0 ea Ho woo, an Au av mNoH o 0 FN emoa m on on ncoa ed a" mo oNoH h o 6N umoa m on em mco~ n “a he o~o~ o o mN Nmma 5 0H mo oooa Q Ca 0v umoa N o VN mmon m on em oHo— 0H 0 me NmoH m 0 MN «mow mumm ..eocouuooxm beau Luuum ouum oucouquXw hood :uuum emum mucouquXm been :uuum u L o 0.; w 0.2 a egaEcm co :2 Co o~LEow 60 cu uo w_oEom uo cu no mute» oo< umo> muwo> wad umo> mueow ova new» I x .u r,.(.t L l I1). I I I III .r "In. I, p I.. I .“II L'l- n . , I. n (.31. I..|)I). . I ”M.I(III(.ILH.E§F IL .I . I J .I I. I I I .h I I I I a .. r. .I will? .II .V (II: .|)I [I .....hlrllufln....)lilnu|h I'lIhFh—hlrh' “Lu“ .l. h I I Uri-yr: oNuw ouosow poo mucouuozw can couuo>uuooo zuus occwuuoaxm No wuwo> .hhofi cu 60¢ .zuuum no uwo> "muocuom oeoocu unwouo ho unauucuuo no couuauuumooun.~nm omnofi APPENDIX C TABLES DESCRIBING THE CROSS-DISTRIBUTIONAL CHARACTERISTICS OF THE RESEARCH SAMPLE: BIOLOGICAL, PSYCHOLOGICAL AND SOCIAL VARIABLES BY THE AGE STRATIFICATION VARIABLE 171 Ammo 0.00H Ammo 0.00H AmHHV o.ooa Ammo o.ooH ”woo o.ooa onao 0.00H Hmuoe Admv H.mv Any v.0H Away o.mH Amy b.0H Amy v.HH ANNo b.mH mow AHVV m.©m Ammo 6.0m Ahoy m.vm Ammo m.mw Ammo o.mw Amado m.wm oz Z w 2 w 2 w 2 w 2 w 2 w no we m m on MN NCOmmom o 06 mm me we av .. - we- o nuamom mmsouo mod Hmcumm mfioocH (..(III III :- ul...‘ umooao uo HmouocHum mo oo< >Q wouuuom ousumuomEoB Mousez sour Mom mCOmmom nuamomII.HIU manme 172 .mmocmumcu meow cu pmccsou coon o>mn mommucooummm Ammo o.ooa Ammo o.ooH Amaao 0.00H AmNo 0.00H Aevo 0.00H onao 0.00H Hmuoe Amy o.o Amy m.m Amy m.h Amy b.0H AmHv m.mN Ammo o.wH Hoonom muMDGMHD ”no h.m “Ho m.N Away m.ma Amo m.bH Avv H.m Aomo v.HN mumspoum omoaaou Amy m.NH Amv o.mN Away N.NH Amy o.mN Amao m.mN Homo v.oN omoaaoo meow mmwa Aamv w.oo AMNV m.mo “who H.oo ANHV m.Nv Away w.Hm Ahoy o.mm no Hoonom poem 2 w 2 w z w z w 2 w 2 w no I go mo ow mm mv vv I H@ ov mm mm mm cowumonpm mmsouw mm< .I ‘I 1' III‘IIDII‘ .(lIII I) ‘()(.(II I.-.‘.II(.)‘ -II (II )4 ill I) ‘(n .I mumcumm oEoocH unopao Ho HmmuoCHum wo wo« >3 Mocumm wEoocH umopao Ho Homu02uum ecu mo ucwficumuufl Hmcofiumospmnl.NIu manme 173 .moocmumcw oEOm cfl poocaou coon o>mn mwomucmoummm Ammo o.ooH Aomv o.ooH LmHHo o.ooH Ammo o.ooH xevo o.ooH Aosuo o.ooH Houoe Amy o.m Ame m.o Ame o.m Lao o.m 1H1 m.m Lao o.m sou; sno> Ammo 6.6m Asuo m.ae Log. o.mg Loo H.mm Ammo m.om Ammo m.av 26am AMvo a.am Amuo a.ue Ammo a.me Lode H.am Asao m.Hm Ammo e.ue sou “mo m.m Luv m.m Lao H.o Ame H.a Leo H.m Loo m.g 30H Nuo> z w z w z w z w z w z N am so mo oo mm me as as o6 I am om mm menopauua mmsouw oma .||. Ii1 I!) III!) I ((1.1‘1111 11110.1(. llltllu mmhma .uocumm oEoocH umopao Ho HMQNUCAHQ mo om< >n mopsuuuufl >mumchI.mIU manna 174 .moocmumcu mEOm Cu popcsou coon m>ms mommucmouomm Ammo 0.00H Ammo o.ooH Amado 0.00H Ammo 0.00H Avvv 0.00H ondv 0.00H Hmuoe AOMV o.av AHNV m.mm Ammo m.bv Away H.hm Ahmv v.H® Ahoy m.hq swam Ammo m.mm AOHV m.hm Homo N.Nm on v.HN Away m.Hm Ammo h.o¢ Efiflwmz Awav «.ma Amy m.ma Ammo o.ON Amy v.HN Amy m.w Away ¢.HH 309 Z w 2 w Z w 2 w 2 w 2 & hm I we m@ I ow am I me we I Av 0v I mm mm I mm mmwamm masouo om< IIIIII I III .II.II.I. | IIIIIIII II. I II II I I I I I I I III,I III). II I.III I. I.I.IIIII.III (III IIIIIIIIIII I. l I .III I II II...II IIIII :IIIII I mmbma .uocumm oEoocH umooao uo Hmmuocuum wo 00¢ >Q EoHQoum >ouocm osu Cu mouHomII.vIO oHQMB .Ho>oH oEOUCN whoa mcfluooamou whoa ca pouumaaoo mumon .mmUCMUmCH 050m CH mumngOH Gmwfl 0>MS mODMUCOUHQQM 175 Ammo 0.00H Ammo 0.00H Amado 0.00H Ammo 0.00H Avvo 0.00H Aowao 0.00H Hmuoa Amy m.w Amao H.0m Aoov N.Nm Away v.00 Ammo m.mo AHov o.mv umoswwnIooo.on Amav H.mH AMHV H.0m Ahoy 0.06 Amy H.Nm Amao m.mN Ammo m.mv mmm.mamlooo.oaw www.mmlooo.mm Ammo o.Mh Aoav m.hN Amy 0.5 ANV H.h ANV m.v AOHV H.h cmnu wood 2 w z w 2 w 2 w 2 w 2 66 pm I «0 mm I 00 mm I we vv I av ov I hm em I MN TEOUCH mmsouw ova .ohhoa .uocumm oEoocH umooao uo HmmuocHum «0 mod wn Ho>mq mEoocHII.mIU magma 176 Amho 0.00H Ammo o.ooa Amaao 0.00H Ammo o.ooa “woo 0.00H AooHo 0.00H Hmuoe Aoo III Aoo III Amy o.N Aoo III AHV m.N AHV o. THOE Ho m Aoo III AHV m.N ANNV H.mH ANHV m.Nv AHNV h.ov AoNo b.0N mucmmsooo ham mucomsuoo Any >.m Amo N.NN Avmv o.bv ANHV o.Nv Away o.ov Ammo o.oo o No m “moo n.om AvNo p.66 Ammo n.mN Amo 5.0H Avo H.m AHNV o.mH mucmmsouo N ANNV 0.0m Amy m.m Amy o.N AHV o.m on III Avv m.N ucmmoooo H z w z w z w Z w z w z w no I em mo I oo om I me we I Ho 06 mm mm I MN mucomsooo mmsouo mud whoa .uwcumm mEoocH umopao uo Hmmflocflum Mo mod >n UHocomsom ocu :fl mucmmsooo mo HTQESZII.©IU magma Amhv 0.00H Ammo 0.00H Amado 0.00H Ammo o.OOH Avvo 0.00H AOVHV 0.00H Hmuoe 177 mEoou AOHV m.MH Amy m.m ANNV H.mH Awo o.wN Away v.om AoNV o.mH mHOE Ho m Ammo H.0m AONV o.mm Aaoo o.mm Aoo v.HN Ammo m.Nm Ammo H.5v mEoou hno Ammo .o.om “may H.6m ammo m.oN Avao 0.0m Amy v.HH Amvo m.vm meoou mIN Z w 2 w 2 w z w 2 w 2 w hm I we no I oo am I me we I av ov I hm om I mm mEoom mmsouw om4 whoa .uocumm mEoocH ummpao Mo Hmmuocflum mo mom >Q uHCD wowaaozo o£u cu mEoom mo Hwnfidzll.hIo magma APPENDIX D COMPARISON OF TOTAL, DIRECT BTU'S CONSUMED PER HEATING DEGREE DAY, 1978-79: BIVARIATE REGRESSION ANALYSIS AND MULTIPLE REGRESSION ANALYSIS 178 mmv "HospHmwu MU H "conmoumwu up HHo. mom. emo.I no.6Hm em.oooI muouoom nuHoom :oHum>uomcoo Hon. Ham. ovo.I om.mom mg.moou souooo ouozou oosuuuua EoHnoum omH. omm.m moo. Ho.voo HH.eHoH souoso ozu :H mmaaom moomuuonm one :oHum>Humop moo. Ham.m soo.I om.mo 6H.ooI I mucouuooxo ammo ooo. osm.mH aoH. so.smH mm.Hme ucosoemuum Honouuoooom ooo.v Nmo.om sow. om.omm sm.momH muoomsooo mo noneoz ooo.v oov.mm son. 60. um. Ho>oH osoocH ooo.v Nom.m0H oov. eo.mmH mm.mHoH meoou mo nonssz uouum mcHHmEmm m ucoHonmoou uouum ucoHonmoou mo :onmouoom pumpcmum :OHmmwummm moHQmHHm> unopcomopCH muHHHnmnoum UoNHpHmpcmum UoNHpumpcmumca . mhImhmH .>ma oomooa mcHumom Mom pofismcou m.sum uouum oCHHmEmm mo >uHHHnmooum pom moHummIm .mucoHUHmmoou COHmmwuoom poNHpumocmum .muouum pumpcmum .mucoHonumou :onmoumom poNHoumpcmumCD "moHanum> ucmpcomoocH co .mhImhmH .>mo wouowo ocHumom Hum ooEdmcou n.9um uoouHo .Hmuoe mo mHm>Hmc¢ conmonom ouMHum>Hm mHmEHmII.HIo mHnme .moHQMHHm> unoccomoch osu mo umuomumso o>HuomuouCH onu ummu ou ©o>oHQEw moB conmmHmou pumzhom wumocmumm owe "Honoummu mo omN. "mumsom m m ":OHmmmHmou up com. um mHmHuHsz ooo. mon.nH m HHmHo>O 0mm. mmo. woo. vH.mmm mm.HmH muouomm suHmmm mHm. vNo. mmo. o>.va vN.vm ucoEchuum HmcoHumoapm EmHDOHm ovm. HHo. mvo. vv.Nmo Nm.NN© >muoco onu CH mmHHom mommuuonm 6cm coHum>HHmmp moo. omm.m woo. vm.mm 05.50H I oocmHuomxw ummm Q :oHum>ummcoo WM Hmo. voo.v 00H.I mH.omm ov.HHNHI >muocm puo3ou owauHuud NNo. oom.m hHH. vo. mo. Ho>wH oEoocH Hoo. mmo.NH mbH. NH.HoN No.oom mucmmsooo mo Hmnfiaz ooo.v ova.mo ohm. hm.NoH mm.NooH mEOOH mo Honfisz uouum mcHHQEmm m ucoHonmmou uouum ucmHonwoou mo :onmoumom pumpcmum conmoummm NAHHHHQMQOHQ ©$NHMVHMMUCM¥W QmNfiCHQQCMUWCD meQMHHMNV “cwwcwmmmvCH mhlmooH .>mo moumoo mcHumom Mom ooEDmcou m.5um III I II. I I I III. I I I I.II IIIIII III IIII.I..II IIII III III I r muouum mcHHQEmm mo >uHHHanoum 6cm moHummIm .mucoHonmooo :onmoumom ooNHpumpcmum .muouum pumocmum .wucoHonmoou :onmoumom poNHoumpcmumCD ”moHanum> ucopcomwocH c0 .mnlmooH .>mo ooumoo quumo: mom poEDmcou n.5um uomHHo .Hmuoe mo mHm>HMC< :onmwumom mHmHuHDzII.NIo oHnme "71111111111144)1“