MSU RETURNING MATERIALS: P1ace in book drop to remove this checkout from 9? ‘ J W?L’9'J er ‘— J7; ‘/ “4-8255 ”,4 \ ( 1' ob’*‘ri§;i’6"6 LIBRARIES “ your record. FINES Ni” be charged if book is returned after the date stamped below. 4?; 51 9'1: W6 M ‘\ z 01% RESIDENTIAL ENERGY CONSERVATION: DO FOLLONUP PROGRAMS MAKE A DIFFERENCE? By John Clifford Jeppesen A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1985 ACKNOWLEDGEMENTS This research was developed and completed with the guidance and support of many individuals. First, I wish to thank the members of my doctoral committee, Dr. Charles Johnson, Dr. Robin Redner, Dr. Neal Schmitt, and Dr. Denton Morrison. My chairperson, Dr. Johnson was very helpful throughout the entire process; his direct and thoughtful review of my drafts greatly improved the final product. My thanks also to my coworkers at the Energy Administration. Marge Wilder and Ginger Macheski assisted with the substantial details of data collection and preparation. Their dedication and professionalism were an inspiration. Special thanks are due to Jan Patrick, who assisted in the early stages of negotiations with community representatives and the critical stages implementation. Her interest and support for the research were invaluable. During the three years of this research I was fortunate to have the companionship, support, and professional review of my colleagues. I am grateful for the encouragement and insightful comments offered by Marty Kushler, Craig Blakely, and Debbie Bybee. ii TABLE OF CONTENTS LIST OF TABLES.................................................. vi LIST OF FIGURES................................................. ix Chapter Page I INTRODUCTION............................................ 1 The Energy Efficiency Challenge....................... The Problem...................................... Priorities....................................... Influencing Voluntary Actions......................... Technical Expertise.............................. Compound Program Design.......................... . Factual Information.............................. 11 Persuasion Approach.............................. 12 Summary............................................... 17 Hypotheses............................................ 20 «UV‘JHHM II "ETHDDIIODOIIIIIOIIOIIIIIICIIIOOIIIOOIC IIIII .IIICIICCII 25 Setting............................................... 25 Research Context................................. 25 City Selection................................... 25 City Description................................. 26 Organization for the Residential Program.............. 27 Overall CEM Program.............................. 27 Residential Program.............................. 2 Thermograms...................................... 28 Organizational Assistance........................ 29 Volunteer Training............................... 29 Thermogram Meetings.............................. 30 A Model for Follow Up............................ 31 Participants and Sampling............................. 32 Experimental Design................................... 33 Procedures............................................ 3 Condition One: Thermogram + Norkshop............. 34 Condition Two: Thermogram + Mailed Information............................. 38 Condition Three: Thermogram Only, (No Followup)......................... ....... .. 39 Condition Four: No Thermogram, (Control, No Program Contact) ........... ....... 39 SUflmary.‘I.OIOIIICCOUIIIICIIIIIICIIIOOIIIIOIII... 39 iii Instruments........................................... 41 Thermogram Meeting Registration Form............. 41 Workshop Registration Form....................... 42 Individual Checklist............................. 42 Workshop Comments................................ 42 Telephone Checksheet............................. 43 Residential Telephone Survey..................... 43 Monthly Utility Data............................. 44 Summary.......................................... 44 Variable Classification and Reduction................. 47 Outcome Variables................................ 47 Process Variables................................ 50 Descriptive Variables............................ 60 Summary.......................................... 61 Analyses.............................................. 61 Comparison of Treatment Conditions on Key Process and Descriptive Variables....... 62 Main Effects..................................... 63 Relationship of Selected Process and Descriptive Variables to Natural Gas and Electricity Usage.............. 68 Condition One (Thermogram + Workshop) Process Analyses............................... 69 Multiple Regression Analyses..................... 69 'III RESULTS-UCCCCOIOIOOOOII...IO.IO...IIIIOCIIIIOIOIUIIOOI 7o Participant Attrition................................. 71 Comparison of Key Process and Descriptive Variables............................... 73 Main Effects.......................................... 80 Hypothesis 1: Treatment Impact on Natural Gas Usage.............................. 80 Hypothesis 2: Treatment Impact on Electricity Usage.............................. 84 Hypothesis 3: Treatment Impact on PostTreatment Conservation Actions............. 86 Hypothesis 4: Relationship of Natural Gas and Electricity Usage With PostTreatment Actions.......................... 91 Relationship Between Key Process Variables and Outcome, Treatment Condition Comparisons........ 93 Hypothesis 6: Number of Areas of Heat Loss...................................... 93 Hypothesis 7: Participation In Additional Information Services................ 93 Hypothesis 8: Barriers To Energy Conservation............................ 94 Hypothesis 9: Pro-Conservation Attitudes......... 94 Condition One (Thermogram + Workshop) Process......... 96 Treatment Integrity.............................. 97 Hypotheses 10-13................................. 101 Condition Four (No Thermogram) Process................ 107 iv Multiple Regression................................... Individual Cases...................................... An "Average" Participant......................... An Example of Major Increase In Usage............ An Example of Major Decrease In Usage............ summaryeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee Iv DISCUSSION-IIICCIIIIIIIIIOIIII IIIIII IIIOIIOODIIIOIIIIIC Treatment Group Equivalence........................... Main Effects.......................................... Correlation of Treatment Processes With Outcome........................................ Condition One Processes............... ...... .......... Predictor Variables................................... summarF'IUIOCIICIIOI.00.......IOUOOIUIOOIICCOICIOIIOO‘OO APPENDICES A. Thermogram Meeting Registration Form, With Response Statistics........................ 8. Sample Thermogram................................. C. Sample Thermogram Meeting Registration Form....... D. Condition One Invitation Letter and Response Card................................... E. Workshop Registration Form........................ F. Individual Checklist.............................. 6. Heat Leak Hit List................................ H. Weatherization Information Resources.............. I. How To Notes for Conditions One and Two........... J. Workshop Comments................................. K. Telephone Script.................................. L. Condition Two Cover Letter........................ M. Telephone Checksheet.............................. N. Residential Telephone Survey...................... 0. Comments and Ideas, As Solicited On Workshop Comments.............................. REFERENCESCUOOOIIOIOIOOOICOODOOOOIIOIOOIIIOCIIIOOIOI.COO... 108 114 114 115 117 118 119 119 122 126 127 129 130 134 136 137 139 142 143 144 147 148 208 209 210 212 213 221 an? and-s} LIST OF TABLES Table Page 1 Labels Used for Treatment Conditions....................... 19 2 Summary of Treatment and Data Collection for EaCh conditionIIII.IOIIOOOIIIIOI..IIIOIIIIOOII.0.00000... 45 3 Conservation Action Categories............................. 49 4 Tally of the Number of Planned and Completed "orkShop TaSkSIIIIIOIOIIIIIOIOIOIIOOIOCIIIC...00.0.0.0... S: 5 Residential Telephone Survey Summary of Questions on Service Recall, and Rated Importance.................. 55 6 Scale Construction for Barrier and Attitude Items.......... 59 7 Analysis of Variance Findings Used In Choice of Main Effects Test For Natural Gas....................... 64 8 Analysis of Variance Findings Used In Choice of Main Effects Test For Electricity....................... 66 9 Patterns of Attrition for Each Condition................... 71 10 Tally of the Number of Completers Versus Refusers of the Residential Telephone Survey (RTS), By conditionOIlllIIIIIIIIOOIIIIOOOIIIIOOOIIIIOOIIOIOOOOOIII 73 11 Summary of Respondent Demographics, Residence Characteristics, and Number of Areas of Heat Loss comparEd Between conditions-III...-COO-IIIOIIIIIIIO..III-ll ~J UI 12 Summary of Additional Respondent Characteristics compared Between condition‘fllflllOOIIOOIIIIIIIIII-00.0.00... ?6 13 Analysis of Variance and Covariance of PostTreatment Natural Gas Usage (October-April)............ S1 14 Group Means and Standard Deviations On PostTreatment Natural Gas Usage (October-April)......... S2 15 Analysis of Variance and Covariance of PostTreatment Electricity Usage (October-April)............ 84 vi 16 17 18 19 2O 23 24 '81 26 30 Group Means and Standard Deviations On PostTreatment Electricity Usage (October-April)......... Summary of One Way ANOVAs of PostTreatment Actions by conditiDnDOIOOIIIIIIOIIIII-IICIIOOllIIIIIOIIIIIOIIIOIIII Summary of A Posteriori (Scheffe’) Tests for Differences Between Conditions on PostTreatment Conservation Actions......................... Partial Correlations of Number of PostTreatment Actions With PostTreatment Natural Gas Usage (Controlling for PreTreatment Usage)....................... Partial Correlations of Number of PostTreatment Actions With PostTreatment Electricity Usage (Controlling for PreTreatment Usage, Income, and Education)............................................. Partial Correlations of Pro-Conservation Attitudes With PostTreatment Natural Gas Usage (Controlling for PreTreatment Usage)....................... Partial Correlations of Pro-Conservation Attitudes With PostTreatment Electricity Usage (Controlling for PreTreatment Usage, Income, and Education)............................................. Workshop Process Variable Means, For Each Workshop Station.................................. Workshop Process Variables Compared Between Workshop Sessions.......................................... Partial Correlations With PostTreatment Natural Gas Usage (Controlling fer PreTreatment Usage)............. Partial Correlations With PostTreatment Electricity Usage (Controlling for PreTreatment Usage, Income, and Education)llll.fli.flflll..IIIIOIII'I.IIIOUIIIIIICIICIO... Pearson Correlations of the Reported Intention To Act With Number of PostTreatment Actions................... Distribution of Responses on How NonParticipants Heard 0* the Thermogram meetingSCIIIIIIOIIIIOIIIIIIIOIII... Analysis of Variance and Covariance of PostTreatment Natural Gas Usage (October-April), First Regression Solution.................................. Analysis of Variance and Covariance of PostTreatment Natural Gas Usage (October-April), Second Regression Solution................................. vii 85 87 88 91 95 96 98 100 103 104 106 107 110 111 32 Analysis of Variance and Covariance of PostTreatment Electriciy Usage (October-April), First RagrESSion salutionICIOIIIIIIIIOIOIIIIOOOIOI.IIIIOIII 112 Analysis of Variance and Covariance of PostTreatment Electriciy Usage (October-April), Second Regression Solution................................. 113 viii LIST OF FIGURES Figure Page 1 Model of Sequential Goals.................................. 32 2 sampleThermograMOIIOOOIOI.I...III...OOOIIIIOOIOIOOOIOODII.136 ix CHAPTER I INTRODUCTION Field research on the effectiveness of energy conservation programs has often focused on single, or "one-shot", interventions, but has rarely examined programs which integrate deliberate followup components. The present research was designed and implemented to assess whether such deliberate followup components might serve to increase residential energy savings. In the first section, the problem of diminishing residential energy fuels and the need for improved efficiency in the use of these fuels are discussed. The second section suggests opportunities for positive action and the theoretical basis for interventions which were tested in this study. In the latter pages of this chapter the rationale for the selected program design and the experimental hypotheses are presented. The Energy Efficiency Challenge The Prggleg Since the turn of the century, the United States has experienced a rapid growth in the use of fossil fuels (coal, natural gas, petroleum) but only recently has it become apparent that these resources are finite and world reserves are being rapidly depleted. The most dramatic signal of this reality came with the Arab Oil Embargo of 1973-74. At this time, our immense dependency on plentiful supplies of these fuels became clear. The shock of limited fuel supplies was particularly acute in the United States since energy intensive life styles were considered normal. It has been estimated that Americans use 30 percent of world energy produced in a year, yet they account for only six percent of world population (Koenig, 1979). Thus, limitations in fuel supplies and increases in price were a sharp departure from previous tends. I A study in Michigan,‘authored by Gladhart,Iuiches,vand‘Morrisonr (1977), documented that in the two years following the Arab Oil Embargo, _c\\\\‘ oil, 81 percent for naturalhgas, and 50 percent for electricity. Even \ ’W.’ ‘- without total decontrol of fuel prices, this trend continued. By March, 1981, an average natural gas customer in Michigan could expect to pay in excess of three times the 1975 price (Skwira, 1981). Obviously, these increases had certainly outstripped average increasesin family income. If in...“ A partial result was that familieshadto spend an increasing proportion of income on utilities. Furthermore, the Public Service Commission of Michigan predicteda—IO4 percent price increase in natural gas by 1985 (Sharky, 1982): the actual increase between 1980 and 1985 was 48 percent (Energy Information Administration, 1983-85). One response to the energy problem was to simply attempt to use less energy. Perhaps the most personally compelling reasons for energy conservation were that it represented actions the average person could do, which were ready and reliable (tested and nonexotic), and which yielded immediate energy cost reduction benefits (Seven Reasons, 1981). On a policy level, it represented the least expensive method for /combating energy cost increases (Ross, no date). Even with these seemingly direct incentives only modest energy conservation was realized. During the dramatic price hikes in 1974-76, .aa~_.w~__~_h___*____fl___w__rw__fl, a2“ average homeowners only decreased utility usage by _five tg ten percent E w“._ .1. .u-‘-_.. a... “- ”‘ flh‘m" (Morrison, Keith, and Zuiches, 1978, page 9). Research revealed "little evidence that families with higher or more rapidly rising fuel prices have higher rates of reported conservation practice adoption (Gladhart, Zuiches, and Morrison, 1978, page 1)." Reasons for the lag in conservation actions were suggested by \vi/F Ef-rh\ 7 survey research (Olsen and Goodnight, 1977); apparently many people were =1_ uanZFE:a{ what could be done, how it could be_accpmplish9ds and 22122 conservation actions would be most cost-effective. While utility bills 4“"-..— 2......z provide feedback on usage and cost, their deficiency was attributed to the fact that this information was delayed and very general (Carlyle and Geller, 1980, page 9): the billing did not provide information or instruction on specific conservation actions. It was precisely this deficiency in specific information that the program tested in the current research addressed. Priorities The cost of residential fuels obviously provided some incentive to 'homeowners to find ways of conserving, and thereby reduce their fuel bills. Nevertheless, it appeared that without appropriate information _“ on conservation action alternatives, improvement in residential energy -._._ ._ --.~—._... “pa—o——— ——-—-- efficiency would be delayed. From this perspective, efforts to hasten conservation action would logically involyefleffective‘methbds of, - -- 21,1 111“. -1. -_W contact, education, and training. The question of the best approach to ._2___‘__~‘*‘_‘_-~u“~_~“-fifl this task was first reviewed from the standpoint of logistics. ”my... ___._-v--—- Until 1981, the Federal Department of Energy conducted a myriad of energy conservation programs, but budgetary and policy changes removed the federal government from its previous information and technical assistance role (Conservation: Uncle Sam Bows Out, 1981). As an increasing number of consumers found it difficult to pay rising utility bills, state government, local government, and local organizations were called upon for help. Cutbacks in service at each of these levels placed greater emphasis on coordinating programs at a community level. This notion of coordinating community based programs was also the topic of applied research. A study of innovative community programs (Pelz, 1981b) indicated that "energy programs succeeded when local Mm—mw organizations took leadership (but department officials took a back x; _ ___-__ .__«..‘.-.. — m”. seat), when other local governments supplied their experience, and when \_,__—-——- ‘F—-—.—.____.—— mow- state agencies established standards“ (page 13). It was further found .4... ~__-..1.._ _. ,,.-._v.__: ~11 that this kind of coordination was especially important with energy programs since energy issues were relatively heterogeneous, reaching all public and private sectors (Pelz, 1981a). As various types of intervention were considered. it was argued that developing energy conservation programs at a community level was desirable since the infrastructure of local organizations could be called upon for informal networks of communication and volunteer membership. Local government could offer necessary programmatic support services, and state government could offer specific technical assistance to build local energy conservation competencies. Pelz (1981a, 1981b) emphasized that state agencies best used their resources by (a) providing technical assistance during energy conservation program development and by (b) facilitating the sharing of innovative solutions between communities. The next question was: Where should a community energy conservation program effort start? At least two well known community energy programs suggested that initial efforts be made in the residential sector. In Springfield, Illinois, Al Casella (Benson, 1981) pointed out that the broadest consensus and support could be gained from helping the homeowner and renter with energy conservation. A similar experience was related by the organizers of the Fon Du Lac county energy program (Lehman, 1981). Additional support for starting with the residential sector came from detailed research by the Energy Policy Group (1981, page 87). Their basic recommendation for statewide (Michigan) conservation priority was for "retrofit (not new construction) residential conservation." Also, in an analysis of economic sector usage, Stern and Gardner (1980) reported that the estimatedflpercentage of energy use_ip the residental sector was a very close second, at 32.5 percent, compared to the industrial sector (35.9 percent). Thecgfore, the residential sector “MM qualified as a worthy initial target for conservatignflon.thacbagiswo{ Ir”, mw ---lfm..--I‘.1Pne—ufb ‘--~II.--f'"-'-" ' " K ‘ percentage of total use;_mw 7—— Thus, it was clear that some good reasons existed for developing energy conservation programs through the organized effort of communities, and for focusing on the residential sector (as a starting point). Further, for residential dwellings, there was little doubt that the biggest target for energy savings was in space heating and cooling. In the northern states the expense of home heating was paramount. Meeks (1981, page 26) as well as Stern and Gardner (1980) confirmed that heating and cooling were the largest energy users in most households. In fact, while residential energy use was increasing between 1960 and 1970, of the increase, 42 percent was in this category (Large, 1973). Research on residential energy conservation actions also supported this program direction. Survey data collected in Michigan, (Morrison, Keith, and Zuiches, 1978) revealed that the ”greatest potential energy reductions were related to space heating" (page 6). Further, an example of one concerted heating-related retrofit effort was documented in Twin Rivers, New Jersey: 67 percent of previous heating costs for a town house built in 1972, was saved from the “simple package of interior window insulation, basement and attic insulation, and plugging air leaks“ (Seven Reasons, 1981). The reader will note that all these conservation actions were one-time efficiency actions, apart from any curtailment behavior (Stern and Gardner, 1979). Clearly, using a community context for residential space heating conservation programming appeared to be an appropriate initial conservation information focus. Research findings therefore suggested the need for effective methods of contact, education and training. By addressing the problem at a community level, effective roles for state and local government could be incorporated. Within communities, the residential sector was identified as a good choice for initial program intervention since it represented potential for infrastructure-building and for major energy savings. Finally, of the various end uses for residential fuels, space heating was identified as a primary target for greater energy efficiency. Influencing Voluntary Actions T hni l x r ' Once program priorities were determined, attention turned to the selection of specific initiatives which might be effective in bringing about the desired change of more efficient energy use in the residential setting. In the field of residential energy technology it was —— —— commonplace to suggest solutions provided primarily by physical science engineering. These solutions might well have included the offering of energy conservation products such as thermal insulation or special N“ 5 __ _ __ f. . MW — _ thermostats. It was less frequently recognized that the ”my“. -._., fields of information transfer (communications and marketing), and the traditional social sciences (sociology and psychology) offered solutions which could facilitate the rapid adoption of the technology provided by the physical sciences. It was considered important that both the/physical_and_spcial technologies be incorporated in addressing the need for greater ____ rfi residential energy efficiency. From the standpoint of the residential customer (who ultimately must decide what, if anything, will be done about the energy use of their residence) not only was the availability of the energy conservation products important, but the system of information, and sufficient incentives for use of the products were also essential to an affirmative voluntary decision. While reference will be made to several types of energy conservation actions, many of which suggest the application of the physical products readily available to consumers, most of the discussion in the current report will refer to the application of information and social science technologies in the adoption of these actions and products. It was the application of these two latter technologies to the problem addressed above that was the focus of the present research. Several researchers familiar with energy conservation (Morrison, 1974: Winett and Neale, in press: Shippee, 1981) readily recognized the necessary interface between the technologies of energy conservation hardware, and social-psychological solutions for designing optimal promotional strategies. Pelz and Munson (1980) articulated this concept quite well: The distinction between technological and the embedding content of an innovation makes it hard to discuss the “innovating process" while ignoring the innovation itself. There is a compelling linkage, for example, among the technological complexity of an innovation, the power of the innovation source, and the strategy at design stage (page 17). Thus, an effective strategy would operationalize a "best mix" of these technologies. In the formulation of these strategies, the role of the social science practitioner could often be quite varied and far reaching. Five basic roles were suggested by Stern and Gardner, (1979): 1. identifying and implementing direct social strategies 2. enhancing market penetration of new technologies 3. predicting and analyzing barriers to implementation of programs 4. predicting and analyzing social impacts of programs 5. field evaluation of program effectiveness (page 47) I.‘ .’ ‘- 5'- .I- My A combination of these roles permitted the social scientist to offer help in very direct and practical ways. with regard to research on energy conservation programs, where energy conservation was best regarded as a set of product and process technologies, it became clear that these technologies could not be solutions in and of themselves; people needed to know whether or not these solutions were appropriate for their situation, and how they could be applied. Social science, with expertise in social adaptation and learning, could help fill this need (Carlyle and Geller, 1980). n r r Since the general program goal was to activate an entire community around the issue of residential energy conservation, it was important to .inEQEnnna1:_and_n:sanize_cesources and strategies which benefited from combination (Stern and Gardner, 1979). One component of such a combination strategy included elements which addressed the decision-making process a community resident might go through in deciding on a personal course of action. According to Burns (1980), the steps in this process might be the following: 1. identifying or recognizing the problem; 2. determining information already available; 3. detailing additional necessary information; 4. defining possible solutions or actions; 5. evaluating such solutions; 6. selecting a strategy for performance; 7. actual performance of an action or actions; and B. subsequent learning and revision based on the outcomes (page 11). It was reasoned that an optimal program design would clearly need to include components which provided clear, concise, and personalized 10 information to the individual participant. Because actual decision-making only occasionally follows the above sequence it was important to design a program for voluntary participant actions which offered maximum opportunity for the most accurate line of decisions. Stated another way, a strategy which addressed the multiple barriers to adoption of energy conservation decisions at each step was believed to be more effective than a singular approach (Kelman and Warwick, 1973). Beyond the organized provision of pertinent information it was also suggested that there was definite merit in making it possible for participants to exggriengg successes resulting from beneficial conservation decisions. Albert Bandura, author of the self-efficacy theory (Bandura, 1977), further pointed out the need to include exposure to performance accomplishments (successful experience), live modeling (to demonstrate necessary skills) and verbal persuasion. His conceptual framework therefore included the cognitive, decision-making components (information) which could lead to action, but it also emphasized skill acquisition (experience). These concepts provided the theoretical basis for intervention strategies. In order to be operationally valid, Thornton (1976) recommended the importance of project management design which embodied the elements of goal clarity, specific recommended action steps, and the presence of social support for the action. Although a compound program design was focused on components designed to prompt conservation actions, as Hansen (1976) pointed out, attitudes and beliefs about energy conservation might be part of the decisions to conserve. For this reason the current research included 11 measurement of energy conservation attitudes, and beliefs about the relative salience of certain barriers to conservation actions. Including the measurement of these internal processes made it possible to assess their relationship to program outcomes. F c nf r In order to be persuasive, information which could be used in decision-making needed to be clear and unmistakably pertinent to the person’s own immediate situation. In the process of providing new, unique, and very personalized information it was obviously desirable to design interventions which captured a person’s attention and optimized the likelihood of appropriate reactions. Since the challenge in program design was to assemble somewhat novel and compelling information for the participant it was necessary to incorporate the best available technology. It was concluded that the ability to show a resident of a community where, on their own residence, actual heat loss (i.e.,thermal inefficiency during winter months) was occurring, would be a rather profound means of persuading a person to take appropriate (heat) conservation actions. This then was the applied physical science technology aspect of the program design and it served as a foundation on which the total energy conservation information package was built. Thus, the physical pictures of residential heat loss were coupled with selected verbal and written information on where, when, and how to complete heat saving energy conservation actions. 12 Persuasion Aggrgggh A review of recent research on residental energy conservation, innovation diffusion, and learning theory suggested the types of program components best suited for inclusion. Six types of potential design components were analyzed. Incentiygs, Cash payments are an obvious way to induce people to conserve. In fact, it was observed that this method might even be the most effective (Leedom, 1980, page 12), yet there were two serious limitations. First, cash payments used for this purpose were, in sum, often considered to be too expensive and too demanding in administration. Second, desirable actions as a result of payment generally stopped when payments were curtailed. Thus, cash payments were ruled out due to both the expense and the short term effectiveness. ‘ ggfgrmgtigg. Provision of information was a rather broad design category; even so, a wide variety of research had shown that only modest differences in its application had very different effects. It was reported (Olsen and Goodnight, 1977; Heberlein, 1975) that there was m little or no relationshiflflhatwggnar ' in the Qggg for energy “hw “"'"‘ _ .... conservation and actual energy conservation behaviors. Simply stated, , ggfl - / HF—w information directed at conservation awareness was insufficient where conservation action was the goal. Information was identified as an effective motivator when content was simplified and specific. Jacoby, (1977) in research on information load and decision quality, concluded that too much information could actually diminish desired results. Also, Shippee (1980) suggested that if behavior was the goal of information provision, then it should be taskorgbehaviorfispecific. Another effective design consideration was the concerted effort to make_information as personalized as logistically possible (Zuiches, 1977). This was especially necessary for information related to residential energy consumption since these consumption patterns were highly varied (Shippee, 1980, page 1). From another perspective, it was found that when energy conservation was perceived as a problem with personal implications, then actual conservation was realized (Shippee, 1980). Thus, when the information reflected a personal frame of reference, active personal responsibility was more likely (Stern and Gardner, 1979, page 15). A final note about information provision is in order. In many community settings it was observed that the people who most needed the information were least likely to seek it. This was particularly true of low income and elderly residents who had small or only fixed amounts to spend on their utility bills. Previous research showed not only that personalized information should be used when possible, but face to face presentation could be superior to more impersonal modes (Kushler, 1977). Furthermore, when the recipient of the information perceived personal commonalities with the provider of the information, the message seemed to have more impact (Jeppesen, 1978). Feedback, Information specific to performance could be termed feedback. Like the cash payment incentives, the effects of feedback on utility usage could deteriorate and, of course, require constant external effort (Slavin and Nodarski, 1977). In fact, Stern and Gardner (1979) summarized their review of the research on feedback with the 14 comment that it worked best when the feedback was immediate, sustained, and in relevant language. It was also important that the feedback use a credible means of report (Becker and Seligman, in press). I The most effective way to use feedback seemed to be in tandem with some form of social commendation (Seaver and Patterson, 1976). This could be as simple as verbal praise or perhaps in the form of a window decal awarded for participation (Shippee, 1980). In fact, social commendation could be thought of as simply another form of feedback (Carlyle and Geller, 1980, page 46) in that it informed one about his/her performance. Thus, it was both possible and desirable to provide multiple forms of feedback, perhaps in sequence with the decision steps which would lead to conservation action. §gggifig behavigr, Project designs which focused on specific behavior seemed to be more successful. Where there were opportunities for social commendation, feedback, or other information provision, an association to the specific desired behavior offered needed clarity (Carlyle and Geller, 1980). In research on behavioral prompts, such as "shut off the lights," significantly more of the desired behavior was realized from highly specific references to the conservation behavior (Shippee, 1980). Ninett and Neale (in press) suggested not only that references be made to specific behavior but that it should be conveyed "at time of opportunity (page 26)"--presumably during the time when the behavior was most likely to occur (e.g., during the heating season). Small groups/sgcial ggntgxt. Even as early as 30 years before the current research, Kurt Lewin (1951) demonstrated the influence of groups on individual behavior. It was asserted that amall groups or neighborhoods carried with them the sense of cultural and community 15 characteristics (Nicosia and Mayer, 1976; Block and Nicosia, 1964) which serve as powerful sources for behavioral norm setting (Burns, 1980). Small groups seemed to create these norms through reciprocal reinforcement (Winett and Neale, in press). A review by Shippee (1980) suggested that meeting as a group might function to commit residents to the energy conservation content of workshops. In actual practice, Pallack and Cummings (1976) showed that a public (versus private) commitment was more powerful in producing lower rates of utility usage. In all such studies, it was assumed that when discussion among group members was allowed/encouraged these members were more likely to conserve (Pallack and Klienhesselink, i976; Shippee, 1980). Surely then, when residential energy conservation programs required group meetings or public assemblies, the program design would be well served to make use of these group dynamics in helping persuade individual participants. Also, research on the spread of innovations (Rogers and Shoemaker, 1971; Engel, Blackwell, and Kollatt, 1978) routinely concluded that early adoptors often influenced others through social interaction. Haberlein’s (1975) study of apartment dwellers during the oil crisis suggested that interventions which used small groups might well prompt peer monitoring of conservation behaviors. The nature of energy conservation as a technology and the power of small groups and social context were also considered. To many people, the concepts, skills, and products related to energy conservation were often not understood. Hisunderstandings about energy conservation actions had sometimes lead to the conclusion that completing some types of conservation action was perhaps "risky." Hagens (no date) found that members of small cohesive groups were more likely to take the "risk" of 16 doing conservation actions. Thus, the group atmosphere could prompt one to be more venturesome. Task-origntgtigg, Perhaps one of the most effective ways to help people adopt new behaviors like those required in some types of energy conservation was to have them "learn by doing." The argument usually followed that by using a relevant context for an action which had an instructive result, the person was able to learn not only the cgnceptg important for understanding the actions but also the skills necessary to personally accomplish the actions. For example, a series of environmental educators related positive results from teaching through direct, purposeful experience (Hammerman and Hammerman, 1968; Shomon, 1964; Swan and Stapp, 1974). Howie (1974), reported higher test scores using a supplement of "guided discovery“ (a form of task-oriented teaching) to classroom instruction versus classroom instruction alone. Also, Leitenberg (1976) offered the observation that reinforced practice was among the most effective methods for developing new behavioral repertoires. From Bandura (1977), who helped to integrate many learning theory concepts within his self-efficacy theory, a cornerstone developed on the premise that the experience of mastery (successful action) provided a powerful motive for future action. His studies suggested this experience of mastery was enhanced when external aids were removed (page 202), appropriate skills were selected, and necessary incentives were inherent (page 194). Effects of successful performance were also found to offer a significant supplement to vicarious experience (modeling) (page 197). As Gladhart, Zuiches, and Morrison (1978) noted "people need gxgerienceg from which they can discover that life can be good in an 17 energy efficient household and that some sacrifices are opportunities in disguise (page 11)." Thus, the literature provided strong support for the notion of designing interventions which would have several complementary components. Because factual information, by itself, was often insufficient to prompt action, other design components were added. Feedback, focus on specific behaviors, small groups/social context, and a task orientation had been proven effective in previous research. §HEIIEX No direct cash inggntivgg were used in the persuasion approaches tested in the current research. Instead, treatments included in the research design involved an emphasis on highly specific, personalized, fgctgal infgrmgtigp of heat loss pictures (thermograms) of homes, and the associated conservation action recommendations. This information was provided in special public meetings called Thermogram Meetings. The figgggggk, on energy conservation action opportunities, provided to attending homeowners was designed to take place immediately, during the meetings. Information was also directed at specific behavigrs (recommended conservation actions, relevant to the homeowner’s residence). Since the information was provided in a public context, the persuasive impact of the small groups/social context was intended to further encourage the desired conservation actions. Thus, these factors were inherent in the design and conduct of the Thermogram Meetings. It was further argued that homeowners would be most likely to actually take energy conservation actions when provided with a second phase of intervention, which would follow the Thermogram Meetings. To 18 this end, special followup interventions were designed and implemented which incorporated the offering of a strong task-orientation. By organizing a hands-on learning experience, the high intensity, or strongest, approach actually trained homeowners to complete conservation actions. A lower intensity, or less strong, approach simply provided written documents which illustrated and described the same conservation actions. The purpose of the program studied in the current research was to activate the whole community around the priority of residential energy conservation. In order to actually persuade residents to invest their time and money on such projects for their own homes it was argued that it would be necessary to incorporate several program features. Not only was it hoped that attention would be drawn to expert, novel and personalized information about the actual areas of heat loss and the associated heat loss remedies, but it was also deemed important to make maximum use of social-psychological solutions in the persuasion strategies. Thus, basic program design included heat loss pictures of individual residences, interpretations, and associated verbal and written information on appropriate heat loss remedies which a participating resident might complete. The main research question was: to what degree would followup program participation yield more conservation actions and resulting utility bill savings than without this participation? It was reasoned that the greater the intensity and specificity of the persuasive elements in program design which an individual resident experienced, the greater would be the likelihood of the associated, 19 appropriate residential conservation actions, and resulting utility bill savings. Further, it seemed that effective program followup alternatives would offer a desirable continuity of the information and experience gained during participation in the standard Thermogram Meeting program. Four treatment conditions, the membership of which differed only in type(s) of program participation, were tested for main effects (OUTCOMES). Those in Condition One (C1) were participants in heat loss picture/information meetings plus a followup hands-on workshop, Condition Two (C2) were participants in heat loss/information meetings plus a followup of mailed information, Condition Three (C3) were participants only in the heat loss/information meetings and Condition Four (C4) were those persons not participating in any of the above. The treatment condition labels and abbreviations used in subsequent references are given in Table 1. Table 1. Labels usep fpr Trgptmgn; Qpnpitipns Condition Abbreviated Label Content Reference Condition One C1 Thermogram + Workshop Condition Two C2 Thermogram + Mailed Info. Condition Three C3 Thermogram Only Condition Four C4 No Thermogram (Control) When ‘Thermogram’ is used to describe treatment content, the counterpart of energy conservation information is also implied. Hypotheses The hypotheses covered five areas of interest: main effects (impacts or outcomes); key process and descriptive variables; relationships between process/descriptive variables and main effects; selected intervention processes; and predictors of main effects. Thus, in addition to the study of the degree to which the treatments had an effect on natural gas and electricity usage, the research also examined important treatment processes. The primary hypotheses regarded treatment OUTQOM§§. The current research was primarily interested in the impact of the treatment conditions on PostTreatment usage of natural gas (heating fuel). Because electricity was the next most common energy source used (primarily for appliances and lighting) in most homes, impacts on this energy use were also examined, but were considered less important. For both types of energy usage, impacts it was assumed that the reductions would be associated with some type of behavioral or structural conservation actions, therefore, they were monitored. Hypotheses 1-4 addressed these issues. Hypothesis 1: It was predicted that reduced usage of natural gas (the major heating fuel) due to conservation actions would be greatest for Condition One, next for Condition Two, followed by Condition Three and least for Condition Four. Thus greatest effect was predicted for participants with exposure to greatest program intensit , specificity, and continuity. Differences between these experimental conditions were 21 expected to be statistically significant. Hypothgpis 2: Because reduction in usage of electricity was not a major focus for program impacts it was predicted that the four conditions would not have significantly different changes in usage of this utility. Hypothggig g: In conjunction with the expected outcomes listed under Hypothesis 1, it was predicted that the number of self reported -conservation actions (both those which were completed after program participation and those which were said to be planned) would be statistically different between the four conditions, and that the average number for the respective conditions would be, in order of magnitude: most for Condition One and in descending order for the other three groups (Condition Four having least). Hyppthesig 4: It was predicted that the number of self reported conservation actions (both those which were completed after program participation and those which were said to be planned) would evidence a significant, negative relationship to the amount of natural gas usage, but not to the amount of electrical usage. Other hypotheses were proposed for various PROCESS and DESCRIPTIVE data to be collected on program delivery records, questionnaires, and a survey. The main purpose of these hypotheses was to provide answers to research questions about the responses and characteristics of participants. These hypotheses were included to provide the potential for more complete interpretation of the main effects results, and a fuller picture of key interventions. Major hypotheses (5-13) for this set of variables follow. 22 Hypothgsig S: No differences between conditions were anticipated for most participant demographics, characteristics, or residence characteristics, however there were three exceptions. First, those attending the Thermogram Meetings, (C1, C2, and C3), would report greater usage (than C4) of the meeting-promoted services. Second, conditions would differ regarding reported barriers to conservation (least concern in C1 to most concern in C4). Third, treatment conditions would differ on pro-conservation attitudes (such attitudes strongest for C1 to least in C4). flypgtpg§1§_§: It was hypothesized that for participants in the Thermogram Meetings, (C1, C2, C3), the number of areas showing thermographic heat loss would show a significant, negative relationship to the subsequent amount of natural gas usage, but the relationship with electrical usage would be nonsignificant. fiypothggis : For all treatment conditions studied in this research it was hypothesized that participation in one or more of the information services (i.e., Energy Fair, RCS, or Energy Hotline) would show a significant, negative relationship to the amount of natural gas usage, but the correlation would be nonsignificant for electricity usage. Hypothesis 8: It was expected that the reported salience of barriers to energy conservation would be significantly, and negatively related to the amount of natural usage, but the relationship to the amount of electricity usage would be nonsignificant. Hyppthesis 9: It was also hypothesized that the reported degree of agreement with pro-conservation statements (energy conservation attitude items) would be significantly and negatively related to the amount of natural gas usage, but not related to the amount of electricity usage. flypothgsis 19: For Condition One it was predicted that the reported number of actions done during the followup workshop would show a significant, negative relationship to the amount of natural gas savings but not to the savings on electrical usage. Hyppthssis 11: It was expected that for persons in Condition One reported “usefulness" of three workshop content areas would show a significant, positive relationship to the amount of natural gas savings but not to the savings on electrical usage. Hypothesis 12: For Condition One participants it was predicted that the reported degree of intention to act on information from the three workshop content areas would show a significant, postive relationship to both the number of actions completed after participation and a negative relationship to the amount of natural gas usage, but not to the savings on electrical usage. Hypothgsis 13: Also, for Condition One, it was hypothesized that completion of one or more workshop tasks would be significantly related to the report of one or more like actions being later completed at the participant’s'home. Hypothesis 15: The ability of selected PROCESS and DESCRIPTIVE variables (identified in the analyses of Hypotheses 1-13) to predict the natural gas and electricity usage was tested using ppttiple regression analyses. This series of analyses were exploratory, and were intended to investigate the relationship of key variables to utility usage - outcomes. These fourteen hypotheses formed five basic groups. First, Hypotheses 1-4 addressed tests for main effects. Second, Hypothesis 5 compared treatment conditions on key process and descriptive variables. 24 Third, Hypptheses 6-9 tested for the relationship between selected process and descriptive variables on natural gas and electricity usage. Fourth, special attention was given to the the most intensive treatment, Condition One (Thermogram + Workshop), in Hypotheses 10-13. Fifth, Hypothesis 14 covered multiple regression analyses which explored potential predictors of the outcomes of natural gas and electricity usage. CHAPTER II METHOD This chapter reviews the design and implementation of the field research used to test the practical application of concepts discussed in Chapter I. Initial comments pertain to the setting in which the research was completed and also the logistical foundation required to organize for the Residential Program. Following this, six sections describe the basic elements of the research method: Participants and Sampling, Experimental Design, Procedures, Instruments, Variable Classification and Reductions, and finally Analyses. seem Research Context The research design was established in conjunction with program development undertaken by the state agency having the mandate for state energy conservation programs (The Energy Administration, Michigan Department of Commerce). As evaluator for such programs, the author expanded the original pilot program to include special followup treatments. Both the original Thermogram Meeting program and these Followup Treatments were studied using the research design addressed in this study. Thus, program development required for the present research INJ U! 26 represented an enhancement to basic planned research for a pilot program. City Selection The city selected for this research was chosen at random for inclusion in pilot program evaluation research. The pilot program was known as Community Energy Management (CEM). Selection guidelines for this program included an acceptable city size (population between 10 and 50 thousand) and the requirement that the city rank in the middle 50 percent of Michigan cities on a published (economic) Need Index (Department of Commerce, 1981). The Need Index was used since it offered a metric for the economic development CEM was designed to support in the form of energy dollar savings available for local commerce. Ci r' t' n Using the above selection criteria, a small city (population: 11,763; dwelling units: 4,878, 1980 Census) on the eastern shore of Lake Michigan was identified and invited to participate in the CEM program (described below), and its city council accepted. Some basic characteristics of this city may form a useful frame of reference. Residential utility customers in this city were served by a small natural gas company serving several lower Michigan counties and by a municipal electric company. Natural gas was clearly the most popular heating fuel (97.2 percent of the homes-~see Appendix A for source). Average usage per customer was 146 mcf in 1981. Local people described the housing stock in the city as primarily single family dwellings with very few apartment complexes. Residential 27 rental properties were also primarily single family dwellings. Organization for the Residential Prooram Overall CEM Program The CEM program was designed to prompt community energy conservation by offering initial program support for rapid implementation action (versus lengthy conservation planning) programs. It was thought that introduction of programs with high visability, rapid development, and relatively rapid benefit realization would hasten the development of community interest in continued conservation efforts. Thus, CEM offered programs in three economic sectors: residential, municipal, and commercial/industrial. Local program development started with the residential sector program during the summer of 1981. In developing the residential program, the state agency’s technical assistance was provided by two liaisons. These liaisons helped the city organize and orient a steering committee of local people who guided program development. The steering committee was charged with guiding programs in all three sectors but started with the residential program and recruited a subcommittee to work on its details. By the end of September, the major part of the residential program was in place and it began offering services to local residents. Rgsigential Program State agency liaisons and the local subcommittee used a standardized program model. Liaisons brought three basic program resources from which the subcommittee devised its program: (1) a complete set of land-based thermograms (heat loss pictures described below and pictured in Appendix B) of all residential structures in the city, (2) organizational assistance, and (3) training for volunteer recruits in the interpretation of individual thermograms and the relating of residential conservation information. The local subcommittee used these elements and local resources to structure and conduct their own residential program’s initial aspect, a series of (free to the public) Thermogram Meetings. Before describing the Thermogram Meetings, the state agency’s program resource contributions will be discussed: First, the thermograms, then organizational assistance, and finally, a few words on volunteer training. Thgrmggrams During the winter of 1980-81 plans were made and executed to complete a heat loss survey of all residential structures in the city. This procedure involved using heat sensitive equipment mounted in a van which scanned building fronts and recorded the heat loss data on magnetic tape. Specialized equipment was later used to decode magnetically stored data into serial black and white images of the scanned building fronts. Various shades of grey showed the locations and relative amounts of radiant heat loss from the houses thus pictured. When catalogued and indexed, this library of pictures represented highly personalized, graphic feedback on heat loss to the residents who were offered the opportunity to see them. Much of the work completed by the residential subcommittee was focused on devising a way to get local residents to come to view their thermograms, starting in September, 1981. Organization Assistance Liaisons were trained to organize a series of small, neighborhood specific, (to take advantage of small group dynamics), Thermogram Meetings to be held at public buildings (mostly elementary school building auditoriums). Technical assistance was provided for dividing the city map into meeting areas (about 200 dwelling units each), schedule meeting times, and coordinate publicity. Multiple-source publicity was also employed and it included newspaper, radio, and flyers (which were hand delivered two days before a neighborhood’s scheduled Thermogram Meeting). Volunteer Trainipg Before the series of Thermogram Meetings were conducted, the steering committee recruited local volunteers. These volunteers were trained to properly interpret the thermograms and to provide the pertinent information on opportunities for residential energy conservation. This instruction was fortified with specific brochure-length publications on energy conservation actions and also information on a Residential Conservation Service (home energy analysis) offered by the natural gas utility company. All the above information was covered in eight hours (four, two hour sessions) of training for the volunteers. With this instruction as the basis for their expertise, volunteers could then offer this information on a neighbor to neighbor basis at scheduled Thermogram Meetings. 30 Thermogram Meetings Residents were informed of the Thermogram Meetings through many sources. In contrast to the breadth of the public notices, the information was highly specific in that residents could come only to the meeting scheduled for their neighborhood. This schedule was intended to have at least three beneficial effects. First, the order of the 27 meetings which covered the entire city was randomized so that the schedule would not be biased regarding which neighborhoods would have first access, and it also maximized potential benefits of word of mouth publicity. Second, meetings were planned to be decidedly small gatherings in which residents could receive an unhurried, personalized interpretation of their thermogram and also observe their immediate neighbor’s interest in residential conservation (a group dynamics effect). Third, the schedule permitted a sustained (versus momentary) community exposure to the issue of residential conservation. Thus, because the Thermogram Meetings were a public service opportunity which was to be voluntarily attended by residents, these meetings were designed to make good use of the limited exposure. Meetings were scheduled for Tuesday and Thursday evenings and on Saturday mornings. When residents arrived, they were asked to complete a Thermogram Meeting Registration Form (see Appendix C). Then, they were asked to help a volunteer. stationed near an indexed city map, find the location of their residence and the associated thermogram strip number was recorded on the form. With this in hand. the resident found a volunteer interpreter who located the appropriate thermogram strip. In the conversation which followed, the interpreter asked about the house 31 and then interpreted the thermogram. Thus, the interpreter had an opportunity to point out specific heat loss problems and the variety of remedies which could be taken to reduce them. Possible remedies included referral to specific publications given to the resident before leaving, and also the RC8 audit program which could offer a detailed analysis of energy conservation actions specific to his/her house. While the resident was not allowed to take home a copy of his/her thermogram, conservation publications and information on the RC5 program could be taken home. If the resident wished to sign up for the RC8 audit, this was offered at the same meeting. Thus informed, each resident left the meeting. A Moge; for Follow up A simple model of residential program goals was devised by the experimenter. The first goal in the intervention process involved promotional efforts which prompted local residents to attend a Thermogram Meeting. Second, while at the meeting these residents would receive specific, personalized information about energy conservation actions. Although these goals seemed necessary, they did not seem sufficient to promote widespread adoption of actual conservation action. As noted in the research discussed in the Chapter 1, information alone frequently had a limited effect on subsequent behavior. To promote actual conservation behaviors a third goal was proposed. The graphic representation of this series of three planned goals is depicted in Figure 1. Program Requirement Goals (Benefits to Resident) Promotion ----9 Attendance Meetings --9 Information Follow Up --9 Conservation Action Figure 1. Model of Seguential Gosls This model therefore included an additional requirement designed to close the gap between motivated good intentions and completed residential conservation actions. The model also highlighted the need to focus on potential barriers to action, such as the reluctance to invest time and money on unfamiliar, (perhaps even perceptually "risky"), actions. Further, it addressed the possible lack of experience with the manual skills necessary to complete conservation actions or just understand what was really involved in completing them. For these reasons, the experimenter conducted the current research to investigate the relative value of two types of followup. Each type of followup treatment is discribed in the section entitled Procedures. Before this, participants and sampling procedures are discussed. Participants and Sampling Of the 3,397 households invited to the Thermogram Meetings 1,035 were represented at the meetings. Thus, 30 percent saw their thermograms when given a specific time and location for their Thermogram Meeting. This massive response was a clear indication of interest in saving on utility bills and the technological novelty of heat loss pictures of personal residences. («I (A The 1,035 participants were considered to be the population from which random selection and assignment to type of followup could be completed. Because procedures and information content of each Thermogram Meeting was routinely standard for all who came, these meetings were considered to be equivalent. Random selection and assignment to followup treatment invitations was blocked by meeting group. In other words, selection and assignment resulted in representation of households from all meetings in each followup condition, and the number from each meeting was a proportion of the total attendance at the meeting when compared to the attendance across all meetings. gypgripgptgl Qgsigp The design for the current research was a nonequivalent control groups design (Cook & Campbell, 1979) with four conditions. The independent variable was type of program treatmept and the dependent variables were nat ral as usa , electric t a , and pumpgr pf gopsgrvation actions. Natural gas and electricity usage was measured before and after the program treatments. Number of conservation actions was primarily measured after the program treatments. The time periods for which the focal natural gas and electricity usage data were collected was arbitrarily defined in terms of the heating season months, October through April; the pre-program period was the 1980-81 heating season, and the post-program period covered the same months in the 1982-83 heating season. The conservation promotion programs took place during the 1981-82 heating season. Reports of conservation actions were analyzed with particular attention to activities completed after the program period. Procedures Four program conditions were compared in the research. In the first three conditions all had been participants in the Thermogram Meetings and therefore they had the same exposure to the information resources of these Thermogram Meetings. In the first condition, (C1), participants also received a followup hands-on workshop experience, in the second condition, (C2), participants received a telephone call and subsequent mailing of additional conservation informatidh, and in the third condition, (C3), no follow up was provided. A fourth condition, (C4), included persons from households which had not attended the Thermogram Meetings. Permission for release of utility data for the above households was primarily secured through a formal, signed release which had been part of Thermogram Meeting Registration Forms (described in the section below entitled Lnstruments . For those not providing this release in Conditions One, Two, and Three, and for all of Condition Four SUbjECtS, the natural gas and electric utility companies provided the same data but without labels identifying the specific customer. Condition One: Thgpmggpgm + Wgrkshgp Invitations to attend a hands on workshop were mailed to 143 households previously represented at Thermogram Meetings. Four workshops were planned, each with a capacity of 25 household representatives, thus a 70 percent acceptance rate was expected. Based on actual response to the invitations only three of the workshops were held with a total of 60 households (42 percent) represented. A copy of the invitation and mail-back reply card are shown in Appendix D. Participants were instructed to assemble at a public building on a Saturday morning. There, orientation and registration took place prior to bus transport to the workshop site. Then, participants were given a training outline and were given opportunity to ask questions prior to the workshop session. Each workshop was held at the home of a senior citizen. This arrangement provided the senior with no-cost installation of energy conservation materials in exchange for use of the house as a hands-on training site. The Director of the local senior center provided information on this opportunity to members. After interested seniors were identified, the experimenter and a subcommittee representative visited the senior’s home to see if the house needed the items which were to be the object of instruction. After agreements were made with the selected senior hosts, the measurements for materials were completed and the material contributions were solicited from local materials merchants who had indicated interest. (These merchants were pleased to provide these donated materials in view of the increased sales attributable to Thermogram Meetings). Results from the 1,035 Registration Forms collected at the Thermogram Meetings included indications of which categories of information were most desired by participants. In rank order, they were the following: 36 34.3 2 Foundation Insulation 25.3 X Window/Door Modifications 17.4 Z Caulking and Weatherstripping 12.7 2 Wall Insulation 8.3 Z Attic Insulation 1.5 2 Financing Energy Conservation PFOjECtS From these findings the top three categories were selected by the residential subcommittee as the best focus for workshops. These results were also shared with workshop participants during orientation. Prior to the short trip to the workshop site, the participants were given an overview of the instruction to be covered, learning station rotation pattern, and timetable. Workshop Registration Forms (see Appendix E) were completed at this time and each household representative was assigned a starting learning station. When participants arrived at the workshop site, they were guided to one of the three learning stations. A total of 25 minutes were allowed at each station before participants rotated to the next station. During each session, each participant was instructed to use the Individual Checklist instrument to record tasks done (see Appendix F). Activity at each station is discussed below: Station I (Foundation Insulation): At this station, participants were to install paper-faced batt insulation in the box sill area of the of the foundation on the home. Each participant was encouraged to complete two of five tasks: (1) measure (2) clean and fill openings (3) cut batts (4) insert batts (5) staple batts When all participants had the opportunity to complete the two behavior 37 criteria, the station instructor was to lead a discussion and ask for any questions participants might have. Station II (Window/Door Modifications): At Station II participants constructed two alternative window treatments. The respective tasks for each window treatment follow: Insider Storm Window Foam Board Shutter (1) measure (1) measure (2) cut wood (2) cut foam board (3) glue/nail 3) cover (optional) (4) rough cut plastic (4) edge tape (5) staple (5) foam tape (optional) (6) trim plastic (6) install (7) edge tape (8) foam tape (9) install When all had the opportunity to complete three of the tasks under one of the window treatments, the instructor would proceed to discussion and a question and answer period. Station III (Caulking and Weatherstripping): Participation at this station involved active location and remediation of areas of unwanted air infiltration. Since these actions involved relatively little time to complete, participants were requested to attempt two tasks under each of the two actions below, and then go directly into a short presentation of types of material which could be used for these jobs. The two categories and associated tasks were: Caulking Weatherstripping (1) load caulk in gun (1) measure (2) clean crack or opening (2) cut weatherstripping to be filled (3) install (3) run bead This presentation included a discussion of the Heat Leak Hit List (see Appendix G) and weatherization information resources (see list in 38 Appendix H). Any questions participants had were to be addressed during the last portion of the rotation. When all participants completed the three station rotation, (approximately 1.5 hours, total) they boarded the bus and returned to the public building meeting site. Prior to leaving, participants were given copies of how-to notes for each station activity (see Appendix I) and were given an opportunity to sign up for an RCS home energy analysis. Prior to being dismissed, participants were instructed to complete the Workshop Comments (see Appendix J) questionnaire, and when this was completed, it was collected along with the Individual Checklists. Qpnggtigp Two; Thgpmggrpm + Mailed Informstipp For a second group of Thermogram Meeting participants, followup involved provision of: (1) the same written material dispensed in Condition One procedures and (2) an emphasis on brief conversation (see Script, Appendix K) including multiple references to and opportunities for neighbor to neighbor information sharing. During each telephone call the operator indicated that he/she would send a personalized packet of energy conservation information which would exceed material available at the original Thermogram Meetings. This packet included a cover letter (Appendix L) and the same series of how-to notes (Appendix I) used in Condition One workshops. Emphasis on neighbor to neighbor contact was operationalized both in what was said during the telephone contact and in the cover letter to the packet which was subsequently sent to the participant. During the telephone contact, the participant was referred to by name and the specific community meeting which they attended was also mentioned. 39 Also, in the cover letter to the information packet, participants were appraised of opportunities to help their neighbors take advantage of prOject offerings: by indicating willingness to have what they had done to save energy serve as a local case study, and by telling neighbors about a schedule of second-chance thermogram showings. Throughout this packet, personal reference was in evidence, including hand addressing and signatures. dition Three: Thermo ram Onl (No Follow U ) Participants in this group were not offered any subsequent followup : opportunities. They were, nevertheless, similar to persons in Conditions One and Three in that they had attended a Thermogram Meeting. F r- No h rm ram ( n rol No Pr ram onta t) Households represented in this group received no program contact. These households had not been represented at the Thermogram Meetings, and they were not offered followup programs. They were simply randomly selected from the local telephone book, cross-checked against Thermogram Meeting Registration Form records, and assembled as a comparison group. Emmi: 2 Conditions One and Two and the Thermogram Meetings for Conditions One, Two and Three offered three general program features: provision of a noncommercial service which was cost free to the participant, use of community voluntarism, and inclusion of the planned social inducement of seeing neighbors participating. The research design allowed comparison of two types of followup and two control groups. Thus, the research was 40 devised to help assess the degree to which followup was useful and, if so, which of the two kinds was most effective. The procedures used in all treatment conditions, other than Condition Four, were designed to make the best use of the "physical technology“ of the Thermograms by providing program treatments which incorporated important features of "social science technology". The social science features used in Conditions One, Two, and Three were delivered to the participants in a consistent manner by using two control functions of program design and management: prganizatignsl 1221253253 and volunteer training. Within the context of these methods of assuring quality control, the highly specific, personalized, factual informgtipp of the heat loss pictures (thermograms) could be reliably associated with the verbal fgsgbagk on energy conservation opportunities. In other words, since the thermograms indicated specific types of conservation action, the volunteer interpreters could be trained to suggest appropriate, specifip pehavigrs which would have relevance to the individual homeowner. Furthermore, the ssstl gpgppsisgstsl_gpgtgyt of the Thermogram Meetings was designed to provide an arena in which local people could experience the encouragement of others as they considered decisions about future energy conservation actions. In Condition One, the followup hands-on Workshop included an additional "social science technology" feature: learning through task-orientatigp. This enhancement permitted participants to learn by actually pptpg selected conservation actions. In contrast to this, Condition Two participants simply received information which was equivalent to that provided in Condition One, but it was in the form of 41 Mailed Information. Thus, Conditions One (Thermogram + Workshop), Two (Thermogram + Mailed Information), and Three (Thermogram Only) involved procedures which incorporated these social science features, and therefore they were inherent to the treatment design. Instruments Data collection occurred at three critical times: the first time (T1) was during the Thermogram Meetings, the second (T2) was during the experimental followup interventions and the third (T3) was about seven months after (T2). References to these time periods are found in the description of the instruments, and Table 2, which follows the discussion of instruments, provides a summary of treatments and data collection dates. We The Thermogram Meeting Registration Form (see Appendix C) was completed by participants in Conditions One, Two and Three during the Thermogram Meetings (at T1). It served to collect participant identification, demographics, responses to publicity methods, a utility data release, and records on the the exchange of information. As the schedule of meetings proceeded, early tallies on this information gave feedback to organizers regarding staffing needs and best information emphasis. The current research required only the information from (i) the checklist of heat loss areas, (2) the checklist of information requests, and (3) the checklist of prior energy conservation actions. 42 Wgrkshop ngistrgtipg Form The Workshop Registration Form (see Appendix E) used at the Condition One workshops (during T2) served to confirm participant identity. When more than one representative of a household came to the workshop, an additional registration form was completed. n 'vi 1 k i t The Individual Checklist (see Appendix F) served as a self reported behavioral checksheet for the Condition One Workshop participants. The checksheets provided a written statement of behavioral goals for each work station at the workshop. Participants at these workshops were directed to read these recommended behavioral goals and recognize that before the workshop was completed they would be asked to place a check mark next to the tasks they actually helped complete. This instrument was therefore intended to have some incentive value (an implicit request for social compliance), but more importantly it provided a record of actual hands-on activity. Immediately following the on-site workshop activities (T2) participants marked on these checklists which tasks they had actually performed (versus tasks they had only observed). W rk h t Printed on the back side of the Individual Checklist another series of questions, entitled Workshop Comments (see Appendix J), were introduced to Condition One participants. These questions provided a means for measuring the impact of the workshop experience for these participants. For each of the three "learning stations" referred to in 43 the description of the workshops, the first series of questions requested a rating of workshop usefulness. The second series of questions inquired about the degree to which the person had the intgntigp tp apt on the conservation actions suggested at each of the respective learning stations. Finally, an open ended question also asked for comments on the workshop experience. Telgphpps Checkshpgt In the process of inviting participation in Condition Two (during T2) a simple record keeping sheet, called the Telephone Checksheet (see Appendix M), provided space to record progress on the Telephone Script (see Appendix K). The Telephone Checksheet served as a recording mechanism which would document, paragraph by paragraph, the standard verbal delivery of the Telephone Script text. The procedure for verbal delivery of the Telephone Script was straightforward. After it was confirmed that the person was, in fact, included on the Condition Two sample listing, the text of the Telephone Script was read. The Telephone Script included information about the Condition Two followup service (a special packet of written publications to be mailed to the participant), and the associated plans for later survey contact. Acceptance or rejection of the offer for this service was then recorded. Resigential Telephone figrvgy Seven months after T2, (at T3), the Residential Telephone Survey (see Appendix N) was completed by all the available household representatives which had been originally included in the research (all 44 four conditions). While the general content was used with all respondents, some selected questions were presented according to the condition in which the respondent was included. Seven content areas were included in this survey: confirmation of participation, measurement of conservation knowledge, report of conservation actions, report of residence characteristics, responses referring to "economic multipliers“, report on relative salience of barriers to conservation actions, report of energy conservation attitudes, and specification of respondent demographics. Mgptpgy ptglgty fists Monthly utility bill data was obtained from both natural gas and electricity utilities. PreTreatment and PostTreatment comparison periods were for the same span of months for natural gas and electricity usage, (i.e., October-April). Special base (nonheating) load and weather adjustments for the natural gas data are discussed in the following section. Man To clarify the correspondence of time periods, instruments used and participant groups to which they were applied, Table 2 is provided on the following page. Please note that treatment(s) and data collection are listed for each condition. 45 Table 2. ummar of Treatment and Data Collect' n f r ach C n ' ' Time Span sinusitis-p.92: Trgatment; Thermogram Meeting September-November 1981 Workshop February-March 1982 Co 1 - Thermogram Meeting Registration Form Workshop Registration Form Individual Checklist Workshop Comments Residential Telephone Survey Natural Gas Data: Base Load Data Heating Season Electricity Data: Comparison Season Qppgttipn Two Trsgtment; Thermogram Meeting Mailed Information pats Collgptipn: Thermogram Meeting Registration Form Telephone Checksheet Residential Telephone Survey Natural Gas Data: Base Load Data Heating Season Electricity Data: Comparison Season September-November 1981 February-March 1982 February-March 1982 February-March 1982 October-November 1982 June-August 1980 and June-August 1982 October 1980-April 1981 and October 1982-April 1983 October 1980-April 1981 and October 1982-April 1983 September-November 1981 February-March 1982 September-November 1981 February-March 1982 October-November 1982 June-August 1980 and June-August 1982 October 1980-April 1981 and October 1982-April 1983 October 1980-April 1981 and October 1982-April 1983 46 Table 2 (cont’d.) Cpndition Thpsg Trgatment: Thermogram Meeting Qstp Collgction; Thermogram Meeting Registration Form Residential Telephone Survey Natural Gas Data: Base Load Data Heating Season Electricity Data: Comparison Season Cond’ ' n F r Mimi. (None) W931. Residential Telephone Survey Natural Gas Data: Base Load Data Heating Season Electricity Data: Comparison Season September-November 1981 September-November 1981 October-November 1982 June-August 1980 and June-August 1982 October 1980-April 1981 and October 1982-April 1983 October 1980-April 1981 and October 1982-April 1983 (Not Applicable) October-November 1982 June-August 1980 and June-August 1982 October 1980-April 1981 and October 1982-April 1983 October 1980-April 1981 and October 1982-April 1983 47 Table 2 also clarifies which months were actually included in the three data collection periods mentioned at the begining of the discussion of instruments. As can be inferred from the forgoing, time periods and respective labels are as follows: September through November of 1981 was T1, February and March of 1982 was T2, and October and November of 1982 was T3. Dates given for the natural gas and electricity data represent the months for which data was collected; this data was necessarily pptained in the months following the tabled time spans. Variable Classification ang ngggtigp The instruments designed for the current research provided a broad base of data. The following section of this chapter provides a system of organization for this data, placing emphasis on the practical classifications needed in later discussion of the analyses used to test the hypotheses. Variables included in these analyses are discussed below under three classifications: outcome (dependent) variables, process variables, and descriptive variables. Qgtcppe Variablss “lit a . Collection of residential utility use data involved both natural gas and electricity billing statements for all households in the research. For both natural gas and electricity usage data the PreTreatment (heating) season was defined as including the months of October 1980 through April 1981 and the PostTreatment (heating) season was defined as October 1982 through April 1983. Thus, October 1980-April 1981 usage was compared to October 1982-April 1983 usage. For the natural gas data, two adjustments were made to insure comparablity: (1) 48 the average monthly base load (summer average) was subtracted from each month’s usage, thus leaving the heating load usage, and (2) this usage was corrected for weather differences (divided by Heating Degree Days for each season). This computation produced the figure for the hundreds of cubic feet usage per heating degree day (ccf/HDD), corrected for heating load only, a direct index of weather-corrected natural gas heating usage for the heating months. Monthly electrical usage data was simply summed for the same two comparison periods. RTS survey questions about fuel use confirmed that none of the residences in the research used electricity for the main heating fuel so it was concluded that electrical consumption would not require weather (HDD) correction. The sum of PreTreatment usage was compared directly to PostTreatment usage. Epgrgy ggnservatgon actions. The Residential Telephone Survey (RTS) provided a large data set on energy conservation actions. Some of this data was collected solely for use by the sponsoring agency, the Energy Administration, but all essential components were included in the present research. Responses on the conventional residential items (the first 17 on the list) were considered necessary for this study; the last six items were omitted. The data on the month the action was completed, and relative material quantities were used solely for a separate response validation study. These 17 residential conservation actions were organized into three logical groupings. Table 3 indicates how the items were grouped. 49 Table 3. Conservation Action Catggories Category Actions —— Space Heating Caulking Weatherstripping Turn Down Thermostat Setting Closing Off Unused Rooms Clock Thermostat Tune Up Furnace Derate (BTU) Furnace Automatic Flue Damper Attic Insulation Wall Insulation Foundation Insulation Storm Windows Window Coverings Water Heating Reduce Use of Hot Water Turn Down Water Temperature Water Heater Wrap (Insulation) Lighting Reduce Use of Lighting ALL ACTIONS (All the above) Clearly, conservation actions related to space heating were most heavily represented. Water heating items were often discussed during thermogram interpretations so the most frequent action recommendations were included. It was decided that only one item under Lighting actions lended itself to a categorical response. For each of the items in the CONSERVATION ACTIONS section of the survey the respondent was asked whether the action had been done before, or after the program treatment time period, or whether it was planned for the future. The later two response categories were included as 50 outcome variables; as described below, the former (action done ‘Before’) were analyzed as a separate category, (see Dgscriptive Varigplgs). Responses to the RTS actions questions were primarily used to gauge the gross quantity of conservation actions, thus data was reduced to simple tallies of the number of actions in each status category, and combinations of these categories. The resulting action status categories were ‘Done’, ‘Planned’, ‘Done and Planned’, and the combination of ‘Done’, ”Planned’, and ‘Done and Planned’. Propsss Vsrtgplgs Beyond the examination of program outcomes, the current research provided extensive data on the processes involved in the delivery of each of the treatment conditions. This was considered important since this body of information would be used to better explain outcomes. Collection of process data took place throughout the research timeline, and as such, discussion of this data was organized by time period. Separate headings are given for PreTreatment, During Treatment, and PostTreatment collections of process data. PreTreatment. Three months before any of the followup treatments were scheduled, the Thermogram Meetings were offered. During these meetings the Thermogram Meeting Registration Forms were completed by participant homeowners. This data included names and addresses necessary for the sampling for Conditions One, Two, and Three. The local telephone book, cross referenced with the assembled Thermogram Meeting participants list, was then used as a resource in sampling for Condition Four. 51 In addition to the participant identification data, the Thermogram Registration Form was also used as a means of collecting responses to (1) a checklist of heat loss areas, (2) a checklist of categories of information in which the participant was interested, and (3) a checklist of prior energy conservation actions. For the first two questions the sum of check marks were defined as (1) an index of the need for heat loss remedies (energy conservation), and (2) a measure of interest, respectively. For the checklist of prior actions no data reduction was necessary, each item was retained as an individual variable. Observation of the process of Thermogram Meetings suggested that although these two checklist items were intended to be separate, they were not treated as separate by the volunteer interpreters who completed the responses. During several of the Thermogram Meetings it was observed that the volunteer interpreters would often record the heat loss shown on the thermogram and then, without actually asking the homeowner about areas of interest, they would just assume parallel check marks for areas of interest. Effectively, this meant that frequently only the first (heat loss) question had face validity, the interest checklist was completed such that it just "mirrored” the heat loss checks. To confirm the extent of this lack of item discrimination, parallel categories under respective items were checked for the percentage of exact agreement. Results showed that when attic heat loss was checked and/or when foundation heat loss was checked, homeowner interest was also checked under these categories (100 percent agreement). Where heat loss from walls, or windows and doors was indicated, homeowner interest was also very frequently checked (80 and 73 percent exact agreement, respectively). Since both observation and 52 statistical analysis indicated this lack of separation between item responses, heat loss data was considered both more important and had higher response rates, therefore homeowner interest responses were dropped from subsequent analysis. Qgring treatment. Only in Conditions One (Thermogram + Workshop) and Two (Thermogram + Mailed Information) were any followup treatment given. Special data was collected to monitor the followup contacts. For Condition One households, the Workshop Registration Form provided a means of checking that attendees were indeed among those randomly assigned from the Thermogram Meeting Registration Form list. For Condition Two households, the Telephone Checksheet documented that the person contacted, (and which subsequently received the mailed information), had been among those randomly assigned from the Thermogram Meeting Registration Form list. These records were monitored continuously during the treatment procedures; in this way sample composition was confirmed. The Telephone Checksheet was the only instrument used with Condition Two during this time period, however, for Condition One two other instruments were used. This was considered desireable because the followup treatment offered to Condition One participants was far more complex than the simple information mailings for Condition Two. The first instrument, the Individual Checklist, provided data on the degree to which the intended hands-on feature of the workshop design was actually experienced by participants. Since each participant at the workshops reported activity or observation for each task listed under each workshop station these actiVities were easily monitored. A count of the number of hands-on tasks planned for each workshop station and 53 the average number of these tasks which were actually completed is given in Table 4. Table 4. Tall of the Number of Planned and Com 1 ed Worksh T Number of Planned Average Number of Tasks Completed Tasks Station 1 5 1.03 (Foundation Insulation) Station 2 15 0.15 (Window/Door Modifications) Station 3 6 0.25 (Caulking/ Weatherstripping) TOTALS 26 1.43 (Stations 1,2, and 3)] From Table 4 it is clear that although all Condition One participants observed the demonstration of conservation actions at each of the three stations, relatively few participants engaged in the suggested hands-on activity. For each station, the sum of the number of tasks completed was used as an index of workshop involvement. The second instrument used during the Condition One workshops was the Workshop Comments page. Participants in Condition One were directed to complete the instrument, Workshop Comments, shortly after all participants had rotated through the three workshop stations. The first group of three questions on this page requested a rating of gsefulnpss for each of the respective workshop stations. For this set of questions data was also summarized as the average of the usefulness ratings. The 54 second group of three questions requested a self report of the degree to which the person had the intention to act on the topical information given at each respective workshop station. Here the summary variable was defined as the response "definitely yes", and the sum of "probably yes" and “no“. (The latter two response categories were considered similar in that they suggested a lack of strong intention to act). Comments and suggestions were also collected via an open ended question on the Workshop Comments instrument. The majority of persons attending these workshops did respond to the question and the reader will find all comments reproduced in Appendix O. PostTrgatment. All PostTreatment process data was collected using the Residential Telephone Survey (RTS, see Appendix N). In the first part of the RTS, a series of questions were designed to inquire about which locally available energy information resources had been used by the respondent. If use of an energy information resource was reported, other questions, respective to the particular resource, were asked with regard to the respondent’s rating of the resource. As indicated in Appendix N, the RTS included questions not only about specific program treatments intended in this research (the four treatment conditions), but also other energy conservation information services which were known to be available during the same time period. By asking this relatively complete set of questions concerning energy conservation resources which were then available to the public, it was hoped that the process of energy conservation promotion. which happened during the treatment period could be better understood. Thus, for persons in each treatment condition, a few questions were asked which had direct relationship to program treatment (i. e., their respective experimental condition), as well as questions which referred to programs which were concurrently available to the whole community. Table 5 offers a summary of the content of these questions and the experimental condition groups to which they were addressed. Table 5. Residential Telephone Survey Summary of Questigns Qn §grvige Recall, and Rated Importance Experimental Condition RTS Question Groups Asked Category Question Content This Question Service Recall Q1 Thermogram Htg. One Two Three Four Q3 Energy Fair One Two Three Four Q5 Workshop One Q7 Information/ Mailing Two Q9 RCS (sign up) One Two Three Four Q10 RCS One Two Three Four Q13 Hotline One Two Three Four Service Importance Q2 Thermogram mtg. rating One Two Three Four Q4 Energy Fair rating One Two Three Four Q6 Workshop rating One Q8 Information/ Mailing rating Two Q11 RCS rating One Two Three Four All RTS respondents were asked if they remembered going to the Thermogram Meetings (Question 1), and the Energy Fair (Question 3), an exhibition of locally available energy conservation products and services. If the respondent recalled taking part in an event they were requested to rate the importance of that event in helping them take Q conservation actions (Questions 2 and a, respectively). Condition One 56 respondents were also asked a similar pair of questions about the workshop treatment (Questions 5 and 6); and Condition Two respondents were asked the similar question pair about the additional mailing of energy conservation literature (Questions 7 and 8). A similar question group, Questions 9, 10, and 11), referred to whether or not the household had had an Residential Conservation Service (RCS) audit (a complete on-site residential audit, offered for a fee of ten dollars, through the gas company). Question 13 inquired about use of the state-wide Energy Hotline (a toll-free information service). For Questions 2, 4, 6, and 8, the importance of services in helping thg pgrspn tp takg consgrvptipn aptipp were rated by respondents using a Likert-like, five point scale. Coded values ranged from 1 (very important) to 5 (not at all important). Thus, Questions 4, and 11 obtained the ratings respective to the Energy Fair, and the RC8 Home Energy Analysis. Persons surveyed in Condition Four had not, by definition, attended the Thermogram Meetings. For these persons a special version of the RTS included three key questions. In this version of the survey Question 5 inquired as to whether they had heard about the Thermogram Meeting, and if their answer was "yes", Question 6 asked about their sources for this information, and Question 7 inquired about their reason for not attending. Descriptive data collected in this manner provided useful findings on the process of nonparticipation. It was thought that answers to these questions might be used to improve program promotion. The remainder of process data was obtained through four question categories in the Residential Telephone Survey (RTS): a question about attribution of improved energy conservation knowledge (Question 16), a 57 series of questions related to the “economic multipliers" (impacts of conservation on the local economy, Questions 32-35), a series of questions about barriers to personal conservation actions (Questions 36-46), and a series of questions about specific conservation attitudes (Questions 47-58). The question about improved knowledge (016) was addressed only to those who received some form of intervention treatment (Conditions One, Two, and Three). All respondents were asked the questions concerning economic multiplier information, barriers to conservation, and conservation attitudes. Both the barrier and the attitude questions provided a wealth of process data, consequently they were examined in relatively greater detail. Prior to the RTS survey, but after the Thermogram Meetings, a test-retest pilot evaluation had been completed. This evaluation, performed with a separate series of telephone surveys in April 1982, and again in October 1982, confirmed that these items had very high test-retest reliability. Pearson correlation coefficients averaged .37 and were all significant at p=.05 or less. Only the barrier and attitude items, on the RTS, were subjected to data reduction. For both item sets the strategy was the same. This strategy for scale construction combined empirical (Cronbach, 1970) and rational (Jackson, 1970) methods. First, all items were examined for instances in which 60 percent or more of the participants used only one numeric code category. Items not passing this criteria were dropped from further analysis. Next, remaining items were grouped according to rational similarity, and then were subjected to examinations of item-total correlations and coefficient alpha (Cronbach, ibid.) 58 until items with greatest relationship were determined. The solutions to these scale construction trials are displayed in Table 6. One barrier item scale, and two attitude item scales were thus formed. Three attitude items proved to be independent, they were retained as single items, termed “singlets“. 59 Table 6. Scale Construction for Barrier and Attitude Items Scale Statistic Item RTS Construction Used, Category Question Content Solution Value Barriers Q36 cost 36 + Q37, Spearman- Q37 payback (cost/ Brown coef., benefit) .47 Q38 know what to do first dropped NA Q39 proper materials selection dropped NA Q40 know how to complete dropped NA actions 841 obtaining a loan dropped NA Q42 uncertainty about the dropped NA duration of future residency Q43 comfort of household dropped NA members Q44 amount of work required dropped NA Q45 family cooperation dropped NA Q46 house construction limits dropped NA Attitudes 647 individual effort is a dropped NA significant contribution Q48 favor conservation vs. dropped NA more power plants Q49 emphasize government Q49 + Cronbach’s spending on conservation 054 + alpha, Q54 need stronger government Q55 .64 regulations (need Q55 government should let free government market work (item involvement) reflected) Q56 energy shortage is real Q56 + Q57 Spearman- (energy Brown coef., crisis is .62 real) Q57 energy crisis is a hoax (item reflected) Q50 favor environmental singlet NA protection Q51 I am an innovative person singlet NA 052 would spend money for singlet NA conservation even without direct payback benefit 60 Qescriptive Variables All of the descriptive variables were obtained from the Residential Telephone Survey (RTS), and were retained as item singlets identified under four categories. Rggpondgnt demographics. The last five questions (Q59-Q63) on the RTS provided data on respondent characteristics. In previous research placement of these questions at the end of the survey seemed to increase the response rate; at this point the respondents were frequently at ease in providing this individual information. These variables, age, education, income, income change, and sex, were considered essential in describing the heads of household in this study. R i c h t ' ' . Since experimental conditions were developed to examine relative impacts on household heating requirements it was important to have information on residence characteristics. A considerable series of questions (Q17 - Q31) provided this information. Of these, questions 28 and 29 were included in the survey to help explain possible extremes in utility data. Engrgy knowledge rating. The treatment conditions were devised to test effects of specific strategies for promotion of residential energy conservation. One question (Q15) was included in the RTS to record self-rating of knowledge about residential energy conservation. Persons in all conditions were asked this question. Actions before Thermogram Meetings. All of the conservation actions under the CONSERVATION ACTIONS section of the RTS which were reported as completed before the Thermogram Meetings were designated as a descriptive conservation action profile for the individual household at 61 that time. As with the responses of ‘Done’ and ‘Planned’ , which were described as outcome variables, these responses of what was done ‘Before’ were reduced to simple sums of the number of actions. mm r The research plan, which was executed over a three year period, enabled a long term examination of the effects and processes involved in the four treatment conditions compared in this study. Comparisons of main program ogtcomgs were made possible through the collection of energy consumption (natural gas, and electricity) and energy conservation action data. To better explain findings about these main program effects, data on the progesg of service delivery and the ways in which it was received were also collected. Lastly, the dpggriptivg data provided the means to examine respondent characteristics which might be related to program processes and outcomes. Analxseg Discussion of the analyses required in the testing of the respective hypotheses (see the last section of Chapter I) are organized by five gagic groups. The order in which the analyses are discussed parallels the order in which the results are presented in Chapter III. The only substantial change from the sequence used in the discussion in Chapter I is that Hypothesis 5 is discussed before the main effects hypotheses (1-4). Thus, the order is as follows: (a) comparison of treatment conditions on key process and descriptive variables (Hypothesis 5), (b) main effects analyses (Hypotheses 1-4), (c) relationship of selected process and descriptive variables on natural 62 gas and electricity usage (Hypotheses 6-9), (d) special analyses (Hypotheses 10-13) of the processes essential to the treatment condition, C1 (Thermogram + Workshop), and (e) the multiple regression analyses (Hypothesis 14) used to examine potential predictors of the outcome variables. The choice of analyses was, of course, directly related to the research design. In the current research random assignment to condition was not possible, therefore a (quasi-experimental) nonequivalent control groups design was employed. As Cook and Campbell (1979) have indicated, with this design, treatment condition participants cannot be assumed to be equivalent since the ideal of random assignment is absent. The alternative then, was to assemble the comparison groups in such a way as to maximize the opportunity for equivalence on key variables, and then collect data which would test for equivalence on the desired dimensions. In the current research, participants were matched by neighborhood prior to invitation to the followup treatments; it was intended that this would improve treatment condition equivalence on residence characteristics (gas and electricity usage) as well as respondent demographics and characteristics. ngpgrison pf Trgatment Conditions on Key Process and Descriptive Variables This set of analyses, (Hypothesis 5), were studied first in order to better understand any treatment condition differences which might have bearing on the main effects analyses. The key process and descriptive variables included all Respondent Demographics, all Residence Characteristics, the Number of Areas of Heat Loss, and a separate grouping of additional Respondent Characteristics. In all, 35 63 variables were analyzed for differences between treatment conditions. For the continuous level variables a series of one-way analyses of variance, (ANOVAs), were performed. Discrete level variables were examined using Chi-square analysis. Main Effegtg Foremost among these tests were Hypotheses 1 and 2, which were concerned with main effects for natural gas and electricity, respectively. Since the equivalence of the treatment conditions, prior to treatment, could not be assumed, a simple ANOVA was discarded as a suitable statistical test of condition differences on natural gas and electricity savings. Alternately, an analysis of covariance, ANCOVA (Cook and Campbell, 1979), was considered to be a better choice since it could be used to covary PreTreatment differences in natural gas and electricity usage. The appropriateness of the ANCOVA approach (for both natural gas and electricity main effects) was investigated in a three part process (Cook and Campbell, idid.). First, the variables which were presumed to be the most likely influences on PostTreatment natural gas and electricity usage were tested for PreTreatment differences between treatment conditions. Three variables were selected: (1) PreTreatment usage level of the respective utilities, (2) Income, and (3) Education. A one-way ANOVA was used to test for PreTreatment differences. Second, to be considered as viable covariates in an ANCOVA analysis the potential variables also had to be significantly (p<.05) related to the PostTreatment usage of the respective utility. These tests used Pearson correlations. 64 Third, the regression of PreTreatment utility usage on PostTreatment usage had to meet the requirement of approximating a linear function. Scatterplots of these regressions were thus completed for the aggregate of the four treatment conditions and for each treatment condition, separately. Figgipgg fpr the ngtpgal gap uspge ANCOVA. Results of the one-way ANOVA tests for PreTreatment natural gas usage, Income, and Education are shown in Table 7. Table 7. A a ' f V ' Fi in se n h i f ' T F N Group Means Variable C1 02 C3 C4 F ratio F probability PreTreatment Natural Gas Usage (heating load only, in ccf per heating degree day) .14 .15 .15 .14 1.59 .19 Income 3.00 2.93 3.06 2.43 2.56 .06 Education 3.28 3.22 3.18 2.45 5.57 .00 x t p<.05 Note: The scale for INCOME is l=under $10000, 2=$10001 to $20000, 3=$20001 to $30000, 4=$30,001 to $40000, 5=$40001 and up. The scale for EDUCATION is 1=grammar school, 2=high school, 3=some college, 4=college graduate, 5=post graduate work or degree. PreTreatment natural gas usage was computationally adjusted so that only heating load (the amount of natural gas used for heating) was used in 65 the analyses. This was done by substracting the prior summer’s average monthly usage (the base load) from each month of heating season usage. The resulting figure was then corrected for weather by computing it in terms of hundreds of cubic feet (ccf) of natural gas per heating degree day (HDD). From Table 7 it was clear that only the variable Education differed significantly between groups. Thus, it was tentatively considered as a covariate in the ANCOVA for natural gas. Next, correlations with the PostTreatment usage of natural gas were examined. As one would suppose, PreTreatment usage was highly correlated to PostTreatment usage (r=.80, n=238, p=.000). On the other hand Income (r=.11, n=178, p=.09) and Education (r=.05, n=205, p=.24) were not significantly (p<.05) related to PostTreatment natural gas usage. Finally, all treatment conditions had linear functions for the regression of PreTreatment natural gas usage on PostTreatment usage. These findings suggested that the data met this major assumption for use of ANCOVA. Thus, on the basis of these preliminary tests, the choice of main effects analyses for natural gas was an ANCOVA with PreTreatment natural gas usage as the only covariate. PreTreatment natural gas usage was retained as a covariate of PostTreatment usage since this ANCOVA would have greater precision in testing main effects than the default, an ANOVA. Although the variable Education showed significant differences between treatment groups, it was not included as a covariate since it was not significantly correlated to PostTreatment natural gas usage. 66 Finding; for thg electrigity gsagg ANQQVe. A similar set of analyses guided the decision on an appropriate main effects test for electricity. First, were the results of the one-way ANOVA tests for PreTreatment electricity usage, Income, and Education. Table 8. Anplyig Of Variance Findings Used In Chgice Of Mgin Effgcts Tgst Fpr Electrigity Group Means Variable C1 C2 C3 C4 F ratio F probability PreTreatment Electricity 3529 3999 3178 3241 4.46 .01 x Usage Income 3.07 2.91 3.06 2.43 2.79 .04 1 Education 3.28 3.22 3.18 2.44 5.51 .00 3 Note: The scale for INCOME is 1=under $10000, 2:010001 to $20000, 3=$20001 to $30000, 4=$30,001 to $40000, 5=$40001 and up. The scale for EDUCATION is 1=grammar school, 2=high school, 3=some college, 4=college graduate, =post graduate work or degree. From the results in Table 8, it was concluded that all three variables showed significant differences between groups and were therefore probable candidates as covariates in an ANCOVA on electricity usage. The question of significant correlation (p<.05) of these variables with the PostTreatment usage of electricity was next resolved. As with the natural gas correlations, PreTreatment usage was highly correlated to PostTreatment usage (r=.86, n=232, p=.000). Unlike the natural gas 67 findings, however, both Income (r=.53, n=169, p=.00) and Education (r=.27, n=l94, p=.00) Egg; significantly (p<.05) related to PostTreatment electricity usage. These data also passed the final test for ANCOVA appropriateness: treatment conditions had linear functions for the regression of PreTreatment electricity usage on PostTreatment usage. Thus the electricity usage data also met this ANCOVA requirement. This series of tests suggested that an ANCOVA with PreTreatment electric usage, Income, and Education as covariates was an appropriate choice of analysis. Even though Income and Education were correlated (r=.50, n=177, p-.000) it was still considered useful to include both as covariates since they presumably measured different (and not perfectly related) characteristics, both of which were related to electricity usage. H he ' n m r P Tr n n rv ' ' In preparation for the analysis of PostTreatment conservation actions the question of PreTreatment differences on this dimension was addressed. Two independent sources were used: the Thermogram Meeting Registration Form data on prior actions, and the RTS survey data which included information on which conservation actions had been done before the Thermogram Meetings. Of the thirteen action categories on the Thermogram Meeting Registration Form (C1, C2, and C3 only)‘ggpg_of the Chi-square tests showed significant differences (p<.05) between the three conditions which attended these meetings. The RTS, having data for all four treatment conditions, provided more complete analysis of PreTreatment conservation action differences between groups. RTS findings were 68 similar: for all three action categories (i.e., Space Heating Actions, Water Heating Actions, and Lighting Actions) no significant differences were detected between conditions. Thus, the evidence was convincing that the treatment condition groups did not show any significant differences on PreTreatment conservation actions. With these findings in mind, the main effects test for PostTreatment conservation actions was essentially a post-only examination of treatment condition differences. This involved a series of one-way ANOVAs on the number of self-reported conservation actions in several categories. Hypgthesis 4I :elationship betweep PoetTreatmen; gpneervagipp ' n an n r ' ' . The analysis for the degree of relationship between PostTreatment conservation actions and both natural gas and electricity usage outcomes was determined by the choice of analyses (ANCOVA) for both of these two indicators of treatment impact. Partial correlations were used for this hypothesis: For the correlation with PostTreatment natural gas usage, the previously identified covariate of PreTreatment natural gas usage was ‘partialed out’, or controlled. Similarly, the correlation with PostTreatment electricity usage used a partial correlation with the three covariates, (PreTreatment Electricity Usage, Income, and Education), being controlled. .Belationehip of Selecged Process and Deecriptive Variebles to Natural Gas and Elegtricity Usage All of the analyses for the hypotheses in this grouping required partial correlations. Partial correlations with PostTreatment natural gas usage controlled for PreTreatment natural gas usage, and partial correlations with PostTreatment electricity usage controlled for 69 PreTreatment electricity usage, Income, and Education. Process variables included in these correlations with natural gas and electricity usage outcomes included Number of (thermographic) Heat Loss Areas, reported Use of Information Services, and the responses on the survey items for Barriers to Conservation and Pro-Conservation Attitudes. These partial correlations answered the research questions posed by Hypotheses 6-9. Condition One (Thermogram + Workshop) Process Analyses For both Hypothesis 10 and 11 the appropriate analyses were partial correlations. Hypothesis 12 required a combination of analyses. First, a Pearson correlation was used for testing the relationship between intention to act on workshop recommendations and subsequent number of actions completed. Second, the relationship between intention to act on workshop recommendations and natural gas and electricity usage was tested with partial correlations. Again, all partial correlations were patterned after those described for Hypothesis 4. Last, Hypothesis 13, the test of relationship between workshop tasks and the later completion of same-category energy conservation actions, required a Chi-square analysis. Multi 1 Re r s 10 An s The foregoing hypotheses helped identify the groups of variables to be tested as predictors of PostTreatment natural gas and electricity usage. Consistent with the analytical approach in Hypotheses 1-13, the basis for these multiple regression analyses was analysis of covariance, ANCOVA (Nie, Hull, Stienbrenner, and Brent, 1975). CHAPTER III RESULTS Chaoter III presents the findings of the current research. The order in which the results are presented is the same as described in the latter part of Chapter II, however three sections have been added. These three sections are Pergipipeng Aggrigipp, the findings for WW and W- 0* the eight sections in Chapter III, the first two concern issues of treatment condition equivalence. The section on Participant Attrition is presented first, and is followed by the Comparison of Key Process and Descriptive Variables. A section on Main Effects, and another on the Relationship Between Key Process Variables and Outcome, detail the findings of greatest importance. Next, the results on Condition One (Thermogram + Workshop) Process, and the added section on Condition Four (No Thermogram) Process offer a closer examination of important dynamic features of the treatment conditions which were designed to offer the most intensive intervention (Condition One), and the least (no program contact) intensive condition (Condition Four). Then, a section on Multiple Regression indicates the findings on the study of potential predictors of the PostTreatment indices of natural gas usage, and 70 71 electricity usage. Finally, to better describe the impacts and processes as they were experienced by individual participants, the last section provides several Individual Cases. Eargicipang Atgrition The goal of the sampling procedures was to obtain four treatment conditions of approximately 60 participants each. Three steps were necessary in this process: random selection, invitation to followup treatment, and acceptance of followup treatment. Table 9 shows the size and composition of the samples for the four treatment conditions. Table 9. Patterns gf Attritipn fpr Eeeh Qpnditipg Steps In Establishing Samples Condition Random Invitation to Acceptance/ Final Selection Follow Up Rejection Sample Of Invitation C1: Thermogram + 143 143 60/79 60 Workshop, C2: Thermogram + 56 56 56/0 56 Mailed Info. C3: Thermogram 63 -- -- 63 Only C4: No Thermogram 67 -- -- 67 (Control) Note: Of the 143 people invited to C1, four later moved away, prior to data collection. It was clear that for C2, C3, and C4 participant attrition did not exist. For C1 the situation was different, to obtain the final sample of 60 participants, 143 were randomly selected and invited to a hands-on workshop. Of these, 79 refused the invitation. This raised the question as to whether or not these 79 might be different from the 60 C1 “acceptors“ on an important characteristic other than invitation acceptance. The most important criteria was PreTreatment usage of natural gas and electricity, the major outcome variables. In the interest of addressing the concern raised by this question (self-selection differences) a random sample of 35 (the maximum number available from the respective utility companies) out of the 79 refusers was drawn and the necesssary natural gas and electricity data was obtained. Student’s T-tests were performed to test for potential differences on PreTreatment utility usage. No significant (p<.05) differences were found for natural gas (T-value=-1.0, df89l, p=.32) or for electricity (T-value--.78, df=82, p=.44). The next point at which treatment condition participants might differ was whether or not they completed the RTS, the telephone survey which collected the majority of the process and descriptive data. Table 10 provides the breakdown of those who completed the survey versus those who refused it. 73 Table 10. Telly pf the Number of Completers Versus Refusers of the Reeidengial Telepngne Survey (RTS)I By gongitigp Condition Completers Versus Refusers C1: Thermogram + 51/9 Workshop C2: Thermogram + 52/4 Mailed Info. C3: Thermogram 57/6 Only C4: No Thermogram 51/16 (Control) Again, a series of Student’s T-tests were done for the completers versus the refusers in each treatment condition. Of the four t-tests for PreTreatment natural gas usage, and the four t-tests for PreTreatment electricity usage, pppe were significant (p<.05). This supported the generalizability of the RTS survey findings to RTS nonresponders. Thus, the data on participant attrition suggested that there were no self-selection biases with regard to natural gas and electricity usage, or the completion of the telephone survey. These findings strengthened the interpretation of main effects for the treatment conditions since this potential source of bias was adequately addressed. Cpmperieon of gey Prpceee ang Qeegrgpgive Vacieglee Prior to testing the treatment conditions for main effects, the analysis plan called for the review of key process and descriptive data 74 with regard to any significant differences between conditions. No differences between conditions were anticipated for most participant demographics and characteristics, or residence characteristics. Specifically, it was intended that the sampling procedures, which matched the condition participants by neighborhood, would improve condition equivalence on these factors. The assessment of group differences on Respondent Characteristics, Residence Characteristics, Number of Areas of Heat Loss, and other self-reported Respondent Characteristics therefore provided the means by which the relative degree of group comparability could be presented. Tables 11 and 12 summarize the findings. 75 Table 11. Summary of Respondent Demographics, Residence Charecgeristigs, and Numper of Areas of Heat Lpss Compared Begween Qonditipns Statistical Degrees of Statistic/ Characteristic Procedure Freedom, df Value Respondent Demographics: Age ANOVA 3 F ratio/ .66 Education ANOVA 3 F ratio/ 5.57 It Income ANOVA 3 F ratio/ 2.56 Income Change Chi-Square 6 Chi-Square/ 2.67 Sex Chi-Square 3 Chi-Souare/ 5.52 Residence Characteristics: Own/Rent -- -- (all pup) Billed for Heating Cost -- -- (all é:£.9£ll§fl) Type of Heating Fuel -- -- (all gee_gee§) Type of Water Heating Chi-Square 3 Chi-Square/ 3.99 Number of Stories ANOVA 3 F ratio/ .76 Number of Bedrooms ANOVA 3 F ratio/ 1.46 Square Footage ANOVA 3 F ratio/ .57 Type of Siding Chi-Square 12 Chi-Square/ 4.34 Daytime Thermostat Setting ANOVA 3 F ratio/ 3.11 X Nighttime Thermostat Setting ANOVA 3 F ratio/ .72 Change in Heated Space Chi-Square 3 Chi-Square/ 5.98 Type of Change in Heated Space Chi-Square 6 Chi-Square/ 19.6 It Number of Areas of Heat Loss Indicated On Thermogram ANOVA 2 F ratio/ 6.85 xx * p<.05 It p<.01 Note: Comparisons for each characteristic include all four experimental conditions. Exceptions exclude Condition Four and therefore show 2 degrees of freedom. 76 Table 12. ngmary of Additional Respondent Characteristigs mepared Begween Conditione Statistical Degrees of Statistic/ Characteristic Procedure Freedom, df Value Self Reported Use of: Energy Fair Chi-Square 3 Chi-Square/ 25.05 it Residential Conservation Service (RCS) Chi-Square 3 Chi-Square/ 25.91 xx Energy Hotline Chi-Square 3 Chi-Square/ 1.96 Self Rated Importance of: Energy Fair ANOVA 3 F ratio/ 1.34 Residential Conservation Service (RCS) ANOVA 3 F ratio/ .11 Thermogram Meeting ANOVA 2 F ratio/ 1.89 Self Rated Improvement in Energy Conservation Knowledge ANOVA 2 F ratio/ .99 Self Rated General Energy Knowledge ANOVA 3 F ratio/ .33 Percentage of Conservation Materials Purchased Locally ANOVA 3 F ratio/ 1.24 Dollars Spent on Conservation ANOVA 3 F ratio/ .82 Loan Taken Out for Conservation Chi-Square 3 Chi-Square/ 3.17 Barrier to Conservation (Cost) ANOVA 3 F ratio/ .18 Energy Conservation Attitudes: Need Gov’t Involvement ANOVA 3 F ratio/ .31 Energy Crisis Is Real ANOVA 3 F ratio/ 1.07 Favor Environ. Protection ANOVA 3 F ratio/ .81 I Am Innovative ANOVA 3 F ratio/ 3.2 I Would Conserve Without A Direct Payback ANOVA 3 F ratio/ .25 t p<.05 Note: Comparisons for each characteristic include all four experimental conditions. gxgeptione exclude Condition Four and therefore show 2 degrees of freedom. 77 It was apparent, then, that since treatment conditions were not significantly different on most of these measures, the sampling strategy was largely successful in obtaining equivalent comparison groups. Hypothesis 5 had however predicted three exceptions to the anticipated group equivalence. Findings for these three exceptions (in Table 12) are discussed first. First, significant group differences were hypothesized for self-reported use of meeting-promoted information services. At the top of Table 12, the results indicate that, in fact, there were treatment group differences on the use of the Energy Fair, and Residential Conservation Service (home energy audit), but not for the Energy Hotline service. Furthermore, the percentages of affirmative answers on use of the Energy Fair and RCS program were generally in the order which had been predicted. For use of the Energy Fair percentage ‘yes’ responses were, in order from C1 to C4: 51.1 percent, 24.0 percent, 16.4 percent, and 10.2 percent. Similarly, the ‘yes’ responses for the RC5 service were: 51.0 percent, 38.0 percent, 41.1 percent. and 5.9 percent. (A slightly higher percentage of the C3 group used the RCS service than did the C2 group). Although the CEM program promoted use of all these ancillary services, use of the Energy Hotline (toll free energy conservation information) service was minimal: three or fewer participants in each treatment condition used the Hotline. Second, Hypothesis 5 had predicted treatment group differences on perceived barriers to energy conservation actions. The statistical test of this prediction, on the single scale, COST, did not, however, indicate significant group differences. t appeared that the various treatment combinations inherent in C1. C2 and C3 did not influence an 78 appreciable reassessment of the economic value of conservation investments. Third, group differences were expected on Energy Conservation Attitudes, yet none of the attitude items except ‘I Am Innovative’ showed significant differences. Closer examination of this item revealed that on a scale from 1 to 5, (with 1 corresponding to "Strongly Agree"), the group means were 2.84 (CI), 3.20 (C2), 3.33 (C3), and 3.35 (C4). Thus, it seemed that the trend was in the direction of stronger treatment interventions being related to greater agreement on self perceived innovativeness. A posteriori testing (Scheffe’ proceedure) of this trend between conditions however indicated no significant (p<.05) differences between any pair of treatment groups. The most conservative summary was that differences in energy conservation attitudes did not appear to be associated with type of treatment. According to Hypothesis 5, other significant differences between groups were not anticipated. Contrary to this prediction, Table 10 shows four significant differences. First on this list was the between-group difference on Education. This variable was a 5 point continuum with 1 equal to grammar school and 5 equal to post graduate work or degree. Mean values for treatment groups were 3.29 (CI), 3.22 (C2), 3.18 (C3), and 2.45 (C4). A posteriori analyses (Scheffe’ procedure) confirmed that C4 (No Thermogram), had significantly less formal education than the other groups (C1, C2, and C3). Therefore, it was noteworthy that the control group (C4), which had not attended the Thermogram Meetings, tended to have less formal education than those in the conditions who had elected to participate in the treatments. 79 Second, Table 11 showed significant differences between treatment conditions on the Daytime Thermostat Setting reported by participants. Group means, in degrees Fahrenheit, were, from C1 to C4, 62 degrees, 64 degrees, 63 degrees, and 65 degrees. A Scheffe’ test indicated that C4, the No Thermogram group, was significantly higher than C1 (but not the other two conditions). This suggested an important difference between these two groups, but it was interesting that no significant differences were found for the Nighggime Thermostet Setting. In light of the findings that treatment conditions did not differ on most residence characteristics, it seemed that control group participants, without the benefit of the energy conservation information gained by those in the other conditions, simply chose to keep their thermostats at higher temperature settings. Third, the Type of Change in Heated Space was also significantly different between groups. Close review of the summary statistics indicated that in CI 16.3 percent of the group had decided to close off rooms which did not need to be heated, whereas in the other three treatment conditions less than 2 percent did this. Originally, this question had been included in the telephone survey (RTS) as a way of detecting reasons for sharp reduction in natural gas (heating) usage level. In retrospect, it appeared that it was best construed as an additional measure of treatment outcome. Last was the significant between-groups difference for the Number of Areas of Heat Loss Indicated On the Thermograms. Since C4 people did not attend Thermogram Meetings, this variable only referred to C1, C2 and C3. Mean values for this tally of detected heat loss areas were 1.3 (C1), 1.9 (C2), and 1.1 (C3). The a posteriori Scheffe’ indicated that 80 C2 residences showed significantly (p{.05) more areas of heat loss than either C1 or C3. At least as important as the significant effects above were the variables for which no significant differences were found. Clearly, the majgrity of the variables listed in both Tables 11 and 12 were in this category. It was particularly useful to recognize that nearly all of the Residence Characteristics, in Table 11, were not significantly dissimilar. This was further evidence that the sampling strategy was effective in producing initial equivalence across the comparison groups. 'n Eff t This section includes the findings on the three outcome criteria of natural gas usage, electricity usage, and the number of PostTreatment conservation actions. The discussion presents findings for natural gas and electricity first. Hypotheses 1: Treatment Impact on Nature; gee U539: Of the two types of utility usage, the treatments were designed to primarily effect a reduction in natural gas (heating) requirements for. the participating homeowners. As shown in Table 11, all participants in the current research were homepage55, with obvious responsibility for heating bills, and natural gas was the heating fuel. Findings on the outcome for natural gas usage are shown in Table 13. 81 Table 13. Analysie of Veriance and Covariance of PoetTreatment Natpral Gas Usege (chober-April) Source of Variation Sum of DF Mean F Squares Squares Covariates PreTreatment Usage .3846 1 .3847 391.39 xxx Main Effects Condition .0042 3 .0014 1.43 Explained .3889 4 .0972 98.92 it! Residual .2290 233 .0010 it: p<.001 These results indicated that PostTreatment usage of natural gas (which was adjusted for nonheating usage rates [base load], and weather differences [Heating Degree Days]) was determined for the most part by PreTreatment usage. Treatment condition was not a significant factor in affecting PostTreatment usage. Thus, the hypothesis about significant (p<.05) natural gas savings was not realized. The ANCOVA analysis also provided group means and standard deviations on PostTreatment natural gas usage (base load and weather corrected). Table 14 provides these treatment condition mean values and standard deviations, in base load corrected hundreds of cubic feet of natural gas per degree day (ccf/HDD). Table 14. Groug Means and Standard Deviations Qg PostTreatment Natural gas Usage (October-Agril) Statistics, Statistics, Unadjusted Adjusted for Covariates Condition N Mean SD Mean SD C1: Thermogram + Workshop 58 .1136 .0441 .1182 .0326 C2: Thermogram + Mailed Info. 54 .1372 .0506 .1305 .0293 C3: Thermogram Only 62 .1280 .0534 .1231 .0333 C4: No Thermogram (Control) 64 .1177 .0531 .1240 .0292 Grand Mean = .1238 Note: The single covariate was PreTreatment Natural Gas Usage. The first column of means and standard deviations are for PostTreatment natural gas usage, with no adjustment for the covariate; the second column of means and standard deviations include adjustment for the influence of the covariate. This table indicates that the most intensive treatment combination (Thermogram + Workshop) was associated with the lowest PostTreatment usage of natural gas. Thus, although the differences between conditions were not statistically significant, the mean value for C1 was rank ordered lowest. as predicted. Although it was anticipated that C2 (Thermogram + Mailed Information) would have had the second lowest PostTreatment usage, followed by C3 (Thermogram Only), the results showed ouite a different order of effects (Table 14). On the basis of the adjusted mean values, it appeared that those attending Thermogram Meetings alone (C3) did only slightly better than the control group. The fact that C2 (Thermogram + Mailed Information) had the highest mean value for PostTreatment usage was especially puzzling; however, another finding for C2 suggested an explanation. Participants in Condition Two (C2) had been found to have significantly more Areas of Heat Loss on their Thermograms. This fact suggested that the residences of those in C2 might have been in greater need of conservation actions. These results were also translated into the dollar equivalents for the heating season. Using the adjusted means (in ccf/HDD) for each condition, a price of 0 .519/ccf, and the 5727 Heating Degree Day heating season (PostTreatment), the average heating expenditures for each condition were computed as a simple product. The dollar equivalents for each condition were as follows: $351.33 (Cl), $387.89 (C2), $365.89 (C3), and $368.57 (C4). Thus, in comparison to the C4 (control group) average, C1 (Thermogram + Workshop) spent $17.64 less, C2 (Thermogram + Mailed Information) spent $19.32 more, and C3 (No Thermogram) spent $2.68 less. Obviously, the first year net differences from the control group were not substantial. Consideration of other factors such as 1) cumulative natural gas savings over the subsequent years, 2) increasing cost of natural gas, and 3) the potential for additional treatment-motivated conservation actions did, however, suggest that longer term evaluation of natural gas usage impacts may look better over time. 84 ngothesis 2: Treatment Imgact gn Electricity Usage Hypothesis 2 anticipated no significant main effect for PostTreatment electricity usage. No significant main effect was expected since the treatments were primarily aimed at space heating, where natural gas was the fuel. Results of the ANCOVA for electricity follow, in Table 15. Table 15. Anal 1 f Va ia e an v r‘ PostTreetmegt Electricitx Usage (October-Agril) Source of Variation Sum of BF Mean F Squares Squares Covariates PreTreatment Usage 171561370 1 171561370 414.77 it! Income 115312 1 115312 .28 Education 1646227 1 1646227 3.98 3 Main Effects Condition 747019 3 249006 .60 Explained 264110760 6 44018461 106.42 Residual 67007784 162 413628 3 g<.05 :xt g<.001 Clearly, Table 15 findings confirmed the expressed expectation. In this instance both PreTreatment Usage and Education were major influences on PostTreatment Usage, but the effect of treatment condition was nonsignificant. 85 Mean values and standard deviations for PostTreatment Electricity Usage are listed in Table 16. All tabled values are in terms of kilowatt hours (kwh). Table 16. rou Means and Standard Deviations n PeetTreetment Eleetrgeitx geage (Octgger-Agcilz Statistics, Statistics, Unadjusted Adjusted for Covariates Condition N 'Mean SD Mean SD C1: Thermogram + workshop 39 3465.90 1408.74 3403.31 635.16 C2: Thermogram + Mailed Info. 45 3822.47 1312.52 3439.06 643.76 C3: Thermogram Only 48 3159.75 1165.08 3416.46 521.98 C4: No Thermogram (Control) 37 3384.08 1713.15 3583.36 722.40 Grand Mean = 3455.98 Note: Covariates include PreTreatment Electricity Usage, Income, and Education. As with PostTreatment Natural Gas Usage (Table 14) the most intensive treatment condition, C1, had the lowest (best) average on PostTreatment usage. Unlike the PostTreatment Natural Gas Usage mean values, both C2 and C3 did better than the control group, C4. Thus, although the prediction of nonsignificant group differences on PostTreatment Electricity Usage were supported by the results, the gattern of mean values showed that all the treatment conditions (C1, C2, and C3) had 86 lower mean usage than the control, C4, and C1 had the lowest PostTreatment electricity usage. The dollar equivalents for the adjusted mean electricity usage (in kilowatt hours, kwh) during the heating season for each condition illustrated the practical meaning of the findings to members of each treatment condition. Based on a price of $.0571/kwh, average dollar equivalents for each condition were: $194.33 for C1 (Thermogram + workshop), $196.37 for C2 (Thermogram + Mailed Information), $195.08 for C3 (Thermogram Only), and $204.61 for C4 (No Thermogram). In comparison to the average electricity cost to those in C4, C1 participants spent $10.28 less, C2 participants spent $8.24 less, and C3 participants spent $9.53 less. In summary, all three of these groups realized savings, relative to the control group, but these first year net savings were rather small. The addition of these savings over subsequent years, consideration of the increasing cost of electricity, and the potential for additional conservation actions would, however, suggest potential improvement in the overall treatment effects over time. nggtheeis 3: Treatment Imgact gn PostTreatment Conservation Actions It was established, in Chapter II, that treatment conditions did not differ on.£§eIreatment actions, therefore findings on PostTreatment actions were not biased by prior differences. Having established this PreTreatment equivalence, the question was: Did the conditions differ on the amount of PostTreatment energy conservation actions? Tables 17 and 18 provide the answer. In the process of studying these tables, the reader may wish to refer back to Table 3, in which the contents of each category of action are defined. 87 Table 17. Summarx of One Way ANOVAs of PoetTreatment Actions p! angition Group Means PostTreatment Action By Category C1 C2 C3 C4 F ratio Space Heating ‘Done’ 1.48 1.57 1.26 1.05 1.42 ‘Planned’ .64 .63 .50 .28 2.84 t ‘Done and Planned’ .3 .31 .10 .03 6.75 it! ‘Done’, ‘Planned’ and ‘Done and Planned’ 2.45 2.52 1.85 1.35 5.70 $1: Water Heating ‘Done’ .90 .63 .73 .25 7.71 :3! ‘Planned’ .12 .17 .15 .06 1.18 ‘Done and Planned’ .03 .06 .00 .00 1.52 ’Done’, ’Planned’, and ‘Done and Planned’ 1.05 .85 .87 .31 8.63 :3: Lighting ‘Done’ .09 .11 .03 .05 1.22 ’Planned’ .03 .02 .00 .00 1.34 ‘Done and Planned’ .02 .04 .00 .00 1.45 ‘Done’, ’Planned’, and ‘Done and Planned’ .14 .17 .03 .05 3.19 1 ALL ACTIONS ‘Done’ 2.47 2.31 2.02 1.3 3.79 1: ‘Planned’ .79 .81 .65 .34 3.75 it ‘Done and Planned’ .38 .41 .10 .03 7.42 it: ‘Done’, ’Planned’, and ‘Done and Planned’ 3.64 3.54 2.76 1.71 8.77 XXX t p<.05 xx g<.01 tit g<.001 Table 18. Summer 88 A Posteriori (Sch ffe’) Tes for D'ff r n e d'ti n n P T m n Status of Conservation Actions ‘Done’ ‘Planned’ ‘Done, ’Done’, and ‘Planned’, Categories Planned’ and ‘Done and Planned’ Space Heating -- no C1,C2>C4 C1,C2>C4 differences water Heating C1,C3>C4 -- -- C1,C2,C3 >04 Lighting -- -- -- no differences ALL ACTIONS C1>C4 C1,C2>C4 C1,C2> C1,C2>C4 C3,C4 Note: All differences between conditions shown in this table are predicated on the occurrence of one-way ANOVA tests where g<.05. Nonsignificant (9}.05) differences are noted by (--)e 89 Table 18 reports results of a posteriori (Scheffe’) tests which, in parallel with Table 17, complete the picture of these conservation action findings. Entries in Table 18 which indicate ‘no differences’ refer to instances in which the one-way ANOVA on Table 17 was statistically significant (g<.05), but the rather conservative quality of the associated Scheffe’ test suggested that comparison of.ee;;§}of treatment conditions were nonsignificant (g<.05). The combination of Table 17 and Table 18 offer considerable explanation of the actions which are presumed to have resulted from the treatments. First, in all action status groupings under ALL ACTIONS (Table 17) the treatment conditions were significantly different. Group means suggest that the groups which received treatment (C1, C2, and C3) did consistenly better than the control, C4. The associated a posteriori (Scheffe’) tests for significant differences between treatment groups (Table 18) revealed that C1 did significantly better than C4, and C2 did similarly well in all action status categories except the one labeled ‘Done’. Each action category could then be considered individually. First, although treatment group means (Table 17) were uniformly higher for Space Heating actions than those for C4, it should be noted that this was true in all action status categories excegt actions which had been ‘Done’. In reference to Table 3 it can be seen that these actions included relatively more expensive actions, a potential reason for deferring action. The findings were somewhat different for Water Heating. Treatment groups tended to complete these actions (’Done’) at a significantly higher rate than the control group, C4. 90 Lighting action results indicated that only in the combined action status category was the ANOVA significant. The Scheffe’ test however suggested no significant differences between treatment condition comparisons. It may be useful to reiterate that the Lighting action category included only one action, reducing the use of lighting. Moreover, reduction in electricity usage was not the goal of the treatments. Although these results offered the necessary summary statements about the impact of treatments on subsequent conservation actions, one additional series of special analysis was also completed. This analysis simply examined the possibility of treatment condition differences on the action categories which were the specific focus of the followup treatments in C1 (Thermogram + Eggtengg) and C2 (Thermogram + Meiteg Leigeeetteg). Chi-square results for differences on the three focal conservation actions were as follows. First, PostTreatment fogngatigg gnegtatign showed a significant (g<.05) effect, and the percentage of ‘yes’ responses, by treatment condition, were in the expected order: 42.9 percent (C1), 32.0 percent (C2), 19.6 percent (C3), and 11.8 percent (C4). Second, the PostTreatment combined category of installed of storm doors, storm windows, and other window coverings also showed a significant difference between treatments, and the percentages of ‘yes’ responses were 30.6 percent (C1), 36.7 percent (C2), 21.4 percent (C3), and 9.8 percent (C4). Thus, C1 and C2 did the most of this set of window and door retrofitting, as one might predict, but C1 did not outperform C2. Third, and last among these special analyses, was a similar test for treatment effects on caulking and weatherstripping. Treatment condition effects were not significant, as is represented by the comparison of 91 percentages of ‘yes’ responses: 61.2 percent (C1), 56.0 percent (C2), 50.0 percent (C3), and 41.2 percent (C4). nggtheege 4: Reletionshig of Netural Gas and gtettrtegtx Usege With PgetTreatment Actions It was predicted that the number of self reported conservation actions would show a significant, negative relationship to PostTreatment natural gas usage, but not for PostTreatment electricity usage. Tables 19 and 20 reflect the results of the appropriate partial correlations. Table 19. P r ' C rr ‘ f N r P stTr tme t ' W' h P t Na ural Ga r ' PreTreeteegt geege) Types of Subsequent Conservation Action Sum of Sum of All All Sum of Space Hater All Heating Heating Lighting Grand Actions Actions Actions Total PostTreatment Natural Gas Usage -.05 -.03 -.09 -.06 t g<.05 Note: Number of cases was 235. 92 Table 20. Partia rre ations of Numb r f P s Tr Actigne With PgetTreetment Electrteitx Usage (Controlling fg; PreTreatment UsageI IncomeI and Education) Types of Subsequent Conservation Action Sum of Sum of All All Sum of Space Water All Heating Heating Lighting Grand Actions Actions Actions Total PostTreatment Electricity Usage -.02 -.04 -.04 -.03 3 g<.05 Note: Number of cases was 164. All eight partial correlations were in the expected direction (higher number of conservation actions associated with lower utility usage), but these relationships were not significant (p<.05). In other words, while the relationship between number of conservation actions and reduced usage of natural gas and electricity were in fact negatively associated, the correlations were not statistically significant. This finding does not necessarily mean the causal link between conservation action and energy savings was not important. The lack of statistical significance for these relationships could also be interpreted as an indication that the number of conservation actions may not be as important as the individuelizeg cgnftggcetign of specific conservation actions and a home’s particular conservation needs. 93 Relationshie Between Kex Process Variablee and QetggmeI Treetment Condition Comgarisons This section concerns the results for Hypotheses 6, 7, 8 and 9. For each of these hypotheses the general question was the degree to which a key process variable was related to PostTreatment natural gas and electricity usage. All of these hypothesis tests required partial correlations; for the correlations with PostTreatment natural gas usage the only variable controlled for was PreTreatment natural gas usage, and the correlations with PostTreatment electricity usage controlled for PreTreatment electricity usage, Income, and Education. H h ' - N r f Ar f H t Only those in C1, C2 and C3, by definition of treatment condition, had data on the Number of Areas of Heat Loss. (For each of these treatment conditions the Thermogram Meeting Registration Forms served to record the necessary information). Hypothesis 6 predicted a significant, negative relationship with PostTreatment natural gas usage, and a nonsignificant relationship with PostTreatment electricity usage. Results of the correlation with natural gas usage were significant, but the direction was gositive (r=.15, df=235, p=.01). This suggested that a higher Number of Areas of Heat Loss was related to subsequent higher usage of natural gas, exactly counter to the prediction. On the other hand, results for correlation with electricity usage were nonsignificant, as predicted (r=-.01, df=164, p=.46). ngothesis 7: Particigation In Additional Information Services Hypothesis 7 offered the expectation that participation in one or more of the information services (Energy Fair, RCS, or Energy Hotline), 94 which were promoted in the CEM program, would be inversely related to PostTreatment usage of natural gas, but no significant relationship would exist with PostTreatment usage of electricity. For both natural gas and electricity the correlations were as expected. A significant, inverse relationship was in fact discovered for the natural gas correlation (r=-.12, df=235, p=.03), and for electricity the relationship was nonsignificant (r=-.07, df=164, p=.18). Apparently, use of the recommended information services helped people increase space heating efficiency, and thereby reduce natural gas usage. 0 h sis 8: Barr‘ r T n r Conservation Perceived barriers to energy conservation were also considered to be a potential influence on natural gas usage, but not for electricity usage. The only barrier to be tested for these correlations was the two-item scale, Cost. (Other items had been omitted due to insufficient variance). Partial correlations showed no significant relationship for either natural gas usage (rs-.04, df=200, p=.28) or for electricity usage (r=-.07, df=164, p=.20). Not only was Cost not significantly different between groups (Table 12), but it was also not significantly related to PostTreatment usage of natural gas and electricity. nggthesis 9: Pro-aneervation BSEitUQQS It had been reasoned that Pro-Conservation Attitudes might have some influence on PostTreatment natural gas usage, but they would have no influence on PostTreatment electricity usage. Tables 21 and 22 list the findings which tested these assertions. The first two attitude 95 categories were multiple-item scales and the remaining three were item singlets. Table 21. Partial Cgrrelations gf Pro-Conservation Attitudee With PostTreatment Naturel Gas Usage (Controlling for PreTreatment Qeage) Attitude Category Correlation Coefficient Need Gov’t Involvement -.05 Energy Crisis Is Real -.03 Favor Environ. Protection .05 I Am Innovative -.08 Would Conserve Without A Direct Payback .03 t g<.05 Note: Degrees of freedom (df) for all correlations was 198. 96 Table 22. Partial Correlations of Pro-Conservation Attitudes MW (ControllinL—for EceT :eetmegt geege, Inggme, and Educational Level) Attitude Category Correlation Coefficient Need Gov’t Involvement -.08 Energy Crisis Is Real -.02 Favor Environ. Protection -.06 I Am Innovative -.01 Would Conserve Without A Direct Payback .00 8 g<.05 Note: Degrees of freedom (df) for all correlations was 163. The results were unequivocal: gene of these correlations were significant. Thus, the assertion that Pro-Conservation Attitudes would be inversely related to PostTreatment natural gas usage eee_ggt supported, but the expected nonsignificant relationship for electricity gee supported. Condition One (Thermogram + Workehgg) EEQSEEE Condition One (Thermogram + Workshop) was intended to be the most intensive treatment combination. In addition to the information provided at the Thermogram Meetings the followup intervention was designed to encourage participants to get involved in the completion of specific conservation actions at the hands-on workshop. Thus, before discussing the relationships between workshop characteristics and specific outcome indices (Hypotheses 10-13) the first part of this 97 section will address the degree to which the workshop sessions (3 groups of participants, on three separate days) were delivered as planned. With the knowledge of the degree of conformity of the workshop treatment to the intended plan, the interpretation of the relationships of its essential processes to the desired outcomes could become more meaningful. Treetment Integrity The issues of how consistent and complete the delivery of each intervention (in this case, hands-on workshops) was has been presented by Sechrest and Redner (1979). This is an issue of t;eeteegt_igteggity. Integrity of treatment may be especially important when interpreting treatment outcomes which are statistically nonsignificant; that is, information on the experienced treatment may be helpful in guiding later improvements. The current research had been designed so that information on treatment integrity could accompany other major findings, and thus increase interpretability of the results. The Individual Checklist and Workshop Comments instruments provided data on (1)the number of hands-on tasks actually completed during the workshops, (2) ratings of the usefulness of the workshop, and (3) statements of intention to act on workshop recommendations. This was collected for each of the three content-specific workshop stations. Table 23 displays the mean values. 98 Table 23. Workshog Process Variable MeansI For Each Workshog Station Workshop Process Variables Number of Usefulness Plan To Station Hands On Rating Take Action Task (tallied (1=very useful (lsnot planned number) to 2=planned) 5=not useful) #1. Foundation .98 1.38 1.50 Insulation #2. Window/Door .16 1.63 1.56 Modifications #3. Caulking and .26 1.64 1.63 Weatherstripping OVERALL 1.40 1.55 1.56 (sum) (average) (average) Note: Number of cases was 58 for 54 for "Plan". "Tasks", 56 for ”Ratings", and 99 Several features of this table are noteworthy. First, it is apparent that relatively few hands-on tasks were completed, and between the three workshop stations the Foundation Insulation station showed greatest activity. Second, ratings of the workshop station content was generally quite high, and the Foundation Insulation station was most highly rated. Third, results were about evenly divided on planned versus nonplanned action. The lower the mean values on this dichotomous variable indicated the tendency for rejection of the recommended actions. Thus, workshop participants seemed least inclined to report the intention to take Foundation Insulation actions, and were relatively more inclined to plan actions on Window/Door Modifications and Caulking and Weatherstripping. Nevertheless it is important to remember that these were responses collected immediately following the workshop station rotations: it was entirely reasonable to expect that some participants later reconsidered these plans for conservation actions. Next, the issue of consistency between workshop sessions was considered. Table 24 indicates the findings for these comparisons. 100 Table 24. Workshog Process Variables Comgared Eetween Workshop Sessions Process Variable F ratio F probability Number of Hands On Tasks Foundation Insulation 18.22 .00 t Window/Door Modifications 0.13 .88 Caulking and Weatherstripping 4.18 .02 t OVERALL 4.53 .02 t Usefulness Rating Foundation Insulation 1.42 .25 Window/Door Modifications 1.43 .25 Caulking and Weatherstripping 1.78 .18 OVERALL 1.75 .18 Plan To Take Action Foundation Insulation .02 .98 Window/Door Modifications .2 .76 Caulking and Weatherstripping .94 .40 OVERALL .27 .77 t e<.05 Note: All one-way analyses of variance were computed with 2 degrees of freedom. 101 For both the Usefulness Ratings and Plan To Take Action findings no significant differences were found, but for Number of Hands On Tasks there were significant differences. When the three workshop sessions were compared, hands-on activity for Foundation Insulation and Caulking/Weatherstripping were different between sessions. A Scheffe’ test indicated that, for the Foundation Insulation station, participants in the first session did significantly (g<.05) more than those in subsequent sessions. For the Caulking/Weatherstripping station those in the first workshop session did significantly more hands-on tasks than did those in the second session. In summary, these findings suggested that although the workshops were well received, the available indicators revealed less hands on activity than planned and some important variation between sessions. This is further discussed in Chapter IV. Hygotheses 10-13 All hypotheses under this heading dealt with the relationship of the hands-on workshop process data with the outcome data. Thus, the treatment integrity issues addressed in the foregoing section provided some useful insights about the workshop process, and therefore helped explain results in this section. Hypotheses 10-12 centered on natural gas and electricity outcomes whereas Hypothesis 13 concerned specific categories of PostTreatment conservation action. Significant relationships were predicted for all tests except those with the electricity outcomes. Because of the similarities in Hypotheses 10-13. tabled results have been consolidated. For each workshop process variable (i.e., 102 Number of Hands On Tasks, Usefulness Rating, and Plan To Take Action) the correlations with PostTreatment usage are given separately for each workshop station. For example, the first entry in Table 25 indicates a partial correlation coefficient of -.15 for the relationship between PostTreatment natural gas usage and the Number of Hands On Tasks reported for the Foundation Insulation workshop station. Table 25 displays the partial correlations regarding natural gas outcomes, and Table 26 shows these correlations for electricity outcomes. 103 Table 25. Partial CorrelationeyWith PoetTreatment Natural Gas Usage (Contrglling for PreTreatment Usage) Workshop Process Variables Number of Usefulness Plan To Station Hands On Rating Take Action Tasks (tallied (1=very useful (1=not planned number) to 2=planned) =not useful) #1. Foundation -.15 .24 X -.09 Insulation #2. Window/Door .19 .17 -.22 Modifications #3. Caulking and -.14 .12 .03 Weatherstripping OVERALL -.11 .22 -.13 (sum) (average) (average) I g<.05 Note: Number of cases was 55 for "Tasks", 53 for "Ratings", and 51 for "Plan". 104 Table 26. Partial Correlations With PostTreatment Elegtcig Usage (Controlling for PreTreatment geege. Inggme. and Educat‘ona v ) Workshop Process Variables Number of Usefulness Plan To Station Hands On Rating Take Action Tasks (tallied (1=very useful (1=not planned number) to 2=planned) 5=not useful) #1. Foundation .14 -.04 .14 Insulation #2. Window/Door .03 -.08 .03 Modifications #3. Caulking and -.44 XX -.15 .21 Weatherstripping OVERALL .04 -.12 .17 (sum) (average) (average) It g<.01 Note: Number of cases was 34 for "Tasks", 33 for "Ratings", and 31 for “Plan". 105 Table 25 findings will be reviewed first. The predicted direction of these correlations were negative for columns 1 and 3, and gositive for column 2. The results generally showed the expected direction of these relationships. The only significant (g<.05) relationship suggested that high ratings of the Foundation Insulation workshop station were related to low PostTreatment natural gas usage. Table 26 shows the correlations for electricity usage outcomes. Although the hypotheses anticipated no significant correlations for PostTreatment electricity usage, one relationship was significant. This correlation suggested that a higher number of hands-on tasks at the Caulking/Weatherstripping workshop station was related to lower PostTreatment electricity usage. Since this did not appear to represent a direct, logical relationship, it was suggested that this finding was either spurious, or some unknown moderator variable may have been responsible. Last in this series was Hypothesis 13. Here the prediction was that completion of one or more workshop tasks would be significantly related to the report of one or more PostTreatment actions in the same category. Since the workshop focused on teaching conservation actions related to residential space heating, this was where subsequent actions were expected. Table 27 indicates the correlations with space heating actions, as well as two other action categories. 106 Table 27. Pearson Correlations of the Regorted Intentign Tg Act With Number of PgstTreatment Actione Types of PostTreament Conservation Action Sum of Sum of All All Sum of Reported Area of Space Water All Intended Action Heating Heating Lighting Grand Actions Actions Actions Total (1=not planned, 2=planned) #1. Foundation .14 -.14 -.11 -.07 Insulation #2. Window/Door .04 -.04 -.15 -.11 Modifications #3. Caulking and .22 X .18 .12 .11 Weatherstripping OVERALL .18 -.01 -.07 -.03 t g<.05 Note: Number of cases was 55 for row 1; 56 for rows 2 and 3; and 54 for row 4. 107 As one would predict, all correlations with PostTreatment Space Heating Actions were positive. Based on the only significant correlation in this table, it appeared that when participants indicated an intention to do caulking and weatherstripping at home, it was associated with a report of more PostTreatment Space Heating Actions. Condition Four (Ne Thermggcemz-Ecggesfi Although not directly related to the research questions of the current study it was considered useful to ask a series of questions on the telephone survey (RTS) which were addressed only to those in Condition Four. The first question was simply: if they had heard of the Thermogram Meetings, how had they heard about them? Table 28. Distri uti n Of Res onse n How NonPar i ' ant Heegg Of The Thermggram Meetings How Heard Count Relative Frequency (Percent) Newspaper 18 27.7 Flyer 1 1.5 Friends 1 1.5 Neighbors 3 4.6 Relatives 2 3.1 School 1 1.5 Energy Fair 1 1.5 Businessmen 1 1.5 Newspaper + Radio 3 4.6 Newspaper + Television 2 3.1 Newspaper + Flyer 1 1.5 Newspaper + Flyer + Relatives 1 1.5 (No Data) 30 46.1 100.0 0* UI TOTAL 108 From this table, it should be emphasized that 53.9 percent gig hear about the meetings. Obviously, the local newspaper was the most important source by which nonparticipants had heard about the Thermogram Meetings, but it was also noteworthy that the total of multiple source responses (such as Newspaper + Radio) represented a category with the next highest percentage, 10.7 percent. Apparently, Thermogram Meeting publicity had reached the majority of nonparticipants. The other question was: why didn’t you attend the Thermogram Meeting? A relatively high percentage said they didn’t recall the reason for not going (46.1 percent): this seemed reasonable since it was roughly one year after the Thermogram Meetings that this survey question was asked. For the remainder of those surveyed the responses were, in rank order, 18.5 percent ‘bad timing’, 16.9 percent (miscellaneous), 10.8 percent ‘not interested’, and 7.7 percent ‘conflict with work schedule’. Thus, relatively few people who had not attended the Thermogram Meetings were disinterested. H II. 1 E . The final series of analyses (Hypothesis 14) explored potential predictor variables for the PostTreatment natural gas and electricity usage outcomes. Since these main effects had been tested within an ANCOVA framework (Nie, et al.,1975), where PreTreatment usage and other key covariates (Income and Education) were essential elements, the exploration of potential predictor variables, in addition to the previously selected covariates, also involved an ANCOVA approach to the tested multiple regression solutions. In short, variables which had been identified as potential predictors in Hypotheses 1-13 were included 109 with previously identified covariates in a stepwise regression with no entry criteria; covariates and other potential predictors were entered into the regression analysis simultaneously. Thus, the ‘other potential predictors’ were placed in direct competition with prior covariates. To be selected as potential predictors, variables had to meet the following criteria: (1) be a continuous level variable, (2) show significant (p<.05) difference between treatment conditions, and (3) have a significant (p<.05) correlation to the respective PostTreatment natural gas or electricity usage level. For each set of analyses the list of potential predictor variables (in addition to those previously identified as covariates in the main effects analyses) included four possiblities: Daytime Thermostat Setting, Information Services Use, Total Number of (‘Done’ or ‘Planned’) Conservation Actions, and (for C1, C2, and C3 only) Number of Areas of Heat Loss. Both dependent variables (PostTreatment Natural Gas Usage, and PostTreatment Electricity Usage) had two multiple regressions. One multiple regression was done including the latter variable, Number of Areas of Heat Loss, and this was for participants from C1, C2, and C3. The other multiple regression excluded this variable and was completed for all participants (C1, C2, C3, and C4). Results are presented first for the prediction of PostTreatment natural gas usage. Variables which met the three part criteria for inclusion were Daytime Thermostat Setting, and Number of Areas of Heat Loss. The ANCOVA multiple regression solution for the three-group analysis is revealed in Table 29. 110 Table 29. Analysis of Variance and Covariance of EestTreetment Nateral Gae Usage (October-Agril)I First Regression §gletign Source of Variation Sum of DF Mean F Squares Squares Covariates PreTreatment Usage .23 1 .23 220.28 it: Daytime Thermostat Setting .00 1 .00 .19 Number of Areas of Heat Loss .01 1 .01 5.27 # Main Effects v Condition .00 2 .00 1.08 Explained .25 5 .05 49.17 31$ Residual .15 149 .00 t g<.05 It: p<.001 For those who attended Thermogram Meetings (C1, C2, and C3) the number of areas of heat loss which were identified to the participant seemed to be a significant predictor of PostTreatment natural gas usage. The raw regression coefficients for the series were .81, .00, and .01, for the respective covariates above. The .01 regression coefficient is postive, however, when one would predict it to be negative (i.e., higher number of areas of heat loss associated with subsequent lower gas usage, due to conservation). A similar multiple regression, which omitted Number of Areas of Heat Loss, was completed for participants from all four treatment conditions. 111 Table 30. Analysts of Varience and Covariance gf PeetTreatment Natural Gas Usage (October-Agril), Second Regression Solution Source of Variation Sum of DF Mean F Squares Squares Covariates PreTreatment Usage .33 1 .33 324.06 tit Daytime Thermostat Setting .00 1 .00 .23 Main Effects Condition .00 3 .00 1.31 Explained .37 5 .07 71.52 it: Residual .20 198 .00 ttt p<.001 Here the only added covariate, Daytime Thermostat Setting, showed no significant contribution to prediction. The raw regression coefficents for the two covariates were .84 and .00, respectively. For PostTreatment electricity usage the list of potential predictors included five variables. The varaiable Income was dropped since, in Table 15, its contribution to the main effects ANCOVA had been nonsignificant. The other two original covariates were retained, and three variables which had met the three part criteria for inclusion (i.e., Information Services Use, Number of Conservation Actions, and Number of Areas of Heat Loss) were also added. Table 31 reports the resulting multiple regression solution. 112 Table 31. Anatyeis of Variance and Covariance gf P s r a m nt c r' i sa e ( to r-A ril) First Re ression fiotgtion Source of Variation Sum of DF Mean F Squares Squares Covariates PreTreatment Usage 155958150 1 155958150 345.68 38* Education 3012277 1 3012277 6.68 33 Information Services Use 867 1 867 .00 Number of Conservation Actions 34571 1 34571 .08 Number of Areas of Heat Loss 833672 1 833672 1.85 Main Effects Condition 50537 2 25268 .06 Explained 191537120 7 27362445 60.65 *4! Residual 63163772 140 451170 :: g<.01 tit g<.001 Clearly, of the three added covariates none were significant predictors of PostTreatment electricity usage. Raw regression coefficients were .79, 129.17, -5.43, 5.54, and 69.22. The final multiple regression was completed for participants from all four treatment conditions, and therefore the variable Number of Areas of Heat Loss was omitted from the list of covariates. As in Table 31, none of the added covariates showed a significant ability to predict PostTreatment electricity usage. Table 32. Analysis of Variance and Coveriange of PostTreatment Electritity Usege (October-Agril), Second Regression Solution Source of Variation Sum of DF Mean F Squares Squares Covariates PreTreatment Usage 240815920 240815920 520.16 it: Education 3609634 3609634 7.80 at Information Services Use 36579 1 36579 .08 Number of Conservation Actions 96192 1 96192 .21 Main Effects Condition 615830 3 205277 .44 Explained 285705120 7 40815018 88.16 xx: Residual 86110840 186 462962 #3 g<.01 33$ p<.001 Thus, for the multiple regression analyses little new information was learned. The only exception involved the regression analysis for PostTreatment natural gas usage, in which Number of Areas of Heat Loss was indicated as an important predictor, but the sign of the regression coefficient was opposite of what had been predicted. One possible interpretation was that this was an artifact of some other important participant or residence characteristics (perhaps lower income residences showing more heat loss). 114 Individua; Casee Thus far in this chapter, the results were obviously based on data gathered for treatment condition statistics. As a brief departure from type of content, this section provides a review of selected case studies which illustrate the dynamics and effects of the treatments on individual garticigante. Three examples of energy savings outcomes are provided: the first example was selected to depict an "averege" participant, the second indicates an instance in which energy usage appeared to inerease dramatically, and the third example describes a case in which a substantial energy usage geegeeee resulted. All energy savings were calculated as a difference between PreTreatment and PostTreatment usage: natural gas savings reflected the base load corrected (heating only) totals. In each case, actual participant data is quoted. Ag "Averege" Partigigent A participant in the C1 (Thermogram + Workshop) treatment was selected to describe a typical participant, and average conservation activity and results. He was a retired male, over 55 years of age, with an annual income of between $10,001 and $20,000. He and his wife lived in a three bedroom house (1750 square feet). During the Thermogram Meeting the thermogram of this man’s house showed noticable heat loss from the walls and foundation. Survey (RTS) data indicated that prior to the Thermogram Meeting, a fair amount of weatherization had already taken place. The home already had some wall insulation, storm doors and windows, caulking and weatherstripping, and 115 it was reported that the residents turned down the thermostat at night, had a furnace tune-up, and had closed off rooms which did not require heating. After attending the Thermogram Meeting and the Workshop, additional home energy conservation actions were also completed. To reduce space heating requirements, they decided to install a solar space heating module, reduced nighttime thermostat settings below previous levels, and did some additional caulking and weatherstripping. Water heating conservation actions included a reduction of hot water usage, reduction of hot water temperature, and insulation of the water heater. All of these actions were intended to reduce natural gas usage. Among these actions, only the caulking and weatherstripping had been recommended at the Workshop. Actual savings of natural gas was 107 ccf or about $56 and the electricity savings was 422 kwh or $24. Taken together, the total energy bill savings was $80 during the first year following the Thermogram Meeting and Workshop. While these results were fairly typical, not all participants in the research had energy savings performance within this general range. An Examgle of Majgr Inereese In ”539: The second example was taken from data collected for a participant in condition C3 (Thermogram Only). In this instance however, utility usage actually increased. This example showed how some changes in natural gas and electricity usage were unexplained, even with the availability of extensive descriptive and process data. 116 The respondent selected for this case study was a male, over 55 years of age, retired, with an annual income of $20,001-$30,000. He and his wife were the only occupants of their house (three bedrooms). Before attending the Thermogram Meeting this homeowner had done several energy conservation projects. All the prOjects he mentioned in the RTS survey which were done beforehand were related to space heating. He had completed several simple actions including turning down the thermostat temperature setting, closing off unheated rooms, and weatherstripping. Furthermore, the walls and attic had been insulated. When he and his wife looked at their thermogram at the Thermogram Meeting it indicated considerable heat loss from the foundation and doors. Following the Thermogram Meeting and Workshop this homeowner completed both space heating and water heating actions. Four of these actions should have logically contributed to lower demand for natural gas heating fuel: they caulked, added a storm door, installed some inside window coverings, and insulated the foundation. The only other action was for water heating, it involved reducing the water temperature. With these conservation actions being completed after the Thermogram Meeting and Workshop one would expect energy savings to result, but this was not the case. Natural gas and electricity usage tncreased. For natural gas the increase was 280 ccf or $145, and for electricity it was 1452 kwh or $83. Thus, the total increase in utility bills, due to increased usage, was $228. None of the survey data suggested any reason for this anomaly; in particular the respondent did not mention any long vacations or periods of illness which might suggest 117 a partial reason for a relative increase in usage. In past studies the researcher had, however, occasionally noted dramatic effects from unreported actions. For example, some residents confessed to turning up the thermostat after weatherizing their home, since they felt they “could afford it now“. Also, while it might not explain all of the increase in electricity usage, the addition of major appliances such as arc welders, large space heaters, or deep freezers could account for some of the increase. In sum, without more intensive data collection, (perhaps even on-site inspections), some changes in natural gas and electricity usage may not be adequately explained. An xam l of Ma' r ecr ase In 5 The last case study provided an example of a large reduction in utility usage. This household had participated in the C2 (Thermogram + Mailed Information) treatment. In this instance the respondent was a male, over 55, retired, and had an income of less than $10,000. He lived in a two bedroom house (1120 square feet) with his wife. Prior to going to the Thermogram Meeting, and receiving the Mailed Information, the attic and walls of the house had been insulated, and they had done some weatherstripping. Their thermogram showed heat loss problems around the foundation and windows. After the C2 treatment, the residents apparently resolved to reduce their natural gas bills. They reduced their usual thermostat setting, had a furnace tune-up, installed an automatic flue damper on the furnace, installed new windows and doors, and caulked. All these items reduced heating fuel (natural gas) use, and among them it was noted that the latter two items had been recommended in the Mailed Information 118 packet. Beyond this, two actions reduced natural gas usage for water heating: a reduction in hot water temperature, and an automatic flue damper for the water heater. In sum, all these actions should have reduced natural gas usage. The natural gas usage reduction realized by this couple was 415 ccf or $215, and electricity usage was only slightly more than before, 13 kwh or $1 worth. Total utility savings were therefore $214. These case studies provided an additional perspective to the reported findings. While most participants experienced changes in their natural gas and electricity bills, these examples illustrated the great diversity of homeowners with regard to combinations of conservation actions taken before and after the planned treatments. Because of this great diversity, large sample research, such as the one described here, is very desireable. Also, these case studies suggest that policy about conservation program design may do best to focus on individual needs (versus average needs) where it is feasible. m ar Results presented in this chapter have covered issues of treatment group comparability, main effects for the treatments, and the investigation of important treatment processes, and their relationship to main effects. Also, the discussion of case studies highlighted the importance of attention to program needs of the individual. Chapter IV offers a discussion of these results presented in this chapter. CHAPTER IV QISCQSSION Chapter IV reviews and discusses the major findings of Chapter III. Thus, the six sections cover treatment group equivalence, main effects, correlation of treatment processes with outcome, Condition One processes, predictor variables, and a summary. Tr tment rou uival nce It was particularly useful, in the current research, to determine the degree to which treatment groups had parity on important indices. The research design was, of necessity, a nonequivalent control groups design, that is, participants could not be randomly assigned to treatment. Thus, comparison groups were matched on geographic neighborhood with the intent of improving comparability on important characteristics, both on the residents and the residences. Participant attrition was found to be a factor only for C1 (Thermogram + Workshop), and this was due to the fact that some of those invited to the hands on workshop refused the invitation. Also, each treatment condition had some participants who did not complete the Residential Telephone Survey. Because these subgroups represented the object of questions about potential self selection biases they were 119 120 compared with those who did participate. These comparisons, on PreTreatment natural gas and electricity usage showed no significant differences. This finding suggested that although self selection represented a possible threat to the interpretation of participant group differences the data showed that this was not an issue in the current research. Treatment groups were also compared on key process and descriptive data. Of the 35 variables, including the categories of Respondent Characteristics, Respondent Demographics, Residence Characteristics, and Number of Areas of Heat Loss, 28 showed no significant difference between treatment conditions. For the eeyeg variables for which there were significant treatment condition differences, tflgee,were predicted (Hypothesis 5). Specifically, the results showed that those who participated in Thermogram Meetings did make greater use of the Energy Fair and RCS information serVices than did the control group, C4. The between-groups comparison on the conservation attitude ‘I Am An Innovative Person’ was also significant, but a Scheffe’ test indicated no significant differences between treatment condition pairs. Four variables which had not been expected to be different between conditions were, in fact, significantly different. Since it was desireable that the treatment conditions be equivalent, these differences were worthy of attention. Discussion of the findings for each follow. For the variable Education it was found that C4 participants had significantly less formal education than C1, C2, and C3 participants. During the choice of analyses this variable had already been selected as a logical choice for a covariate. The Daytime Thermostat Setting was also different between groups. An a posteriori Scheffe’ test indicated C4 people had significantly higher settings than did C1. Since this data was collected months after the Thermogram Meetings, one explanation could be that C1 people responded to the thermogram by reducing the thermostat setting. Thus, it was not clear for how long the thermostat setting had been at the reported temperature, and no data was collected on the associated dates of initiating this setting. The appropriate variable for testing a causal effect of the treatments was an item on the CONSERVATION ACTIONS portion of the RTS, ‘Turn Down Thermostat Setting’, which was included in the Space Heating actions category (Table 2). Thus, the result for Daytime Thermostat Setting was somewhat ambiguous, but the ‘Turn Down Thermostat Setting’ item, which was included in the main effects tests (Hypothesis 3), was designed to test the relationship between condition treatment and subsequent thermostat setting behavior. Next among the list of items with significant differences for treatment condition was Type of Change in Heated Space. In this comparison Cl participants apparently closed off more unheated rooms in their home than did other groups. Again, this could be construed as an outcome of the treatment condition (in this case, the most intensive treatment combination, Thermogram f Workshop). It should be noted that this same group, C1, also had the least PostTreatment usage of natural gas: the fact that they closed off more unheated rooms than did those in other treatment conditions suggests that this action may have contributed to lower natural gas usage. The possibility that this was influenced by the treatment condition was nevertheless better tested by the CONSERVATION ACTION item on the RTS, ”Close Off Unused Rooms’. 122 The last of these variables was the Number of Areas of Heat Loss, where the significant treatment condition differences were, by definition of treatment, limited to C1, 2, and C3. Here, C2 (Thermogram + Mailed Information) had shown significantly more areas of heat loss than C1 or C2. One could infer from this that C2 might have greater motivation to take conservation actions than the other two groups, but the main effects analyses (see Table 14), did not show lower usage of natural gas (heating fuel) for C1 over C2 and C3. In sum, nearly all of the analyses which tested for important between-group differences supported the goal in the research design of group equivalence. The main effects analyses were chosen and configured to address other concerns about between-group covariance with the utility data. Main Effects The primary focus of the treatments tested in the current research was to educate and persuade the residential community to take conservation actions which would reduce the use of the main heating fuel, natural gas. The Thermograms, information, and follow-up treatments centered on heat loss remedies. The results did not however indicate that the treatments made a statistically significant difference in PostTreatment natural gas usage. Group means for the four treatment conditions (adjusted for PreTreatment natural gas usage) nevertheless showed that Condition One (Thermogram + Workshop) did do better than both C3 (Thermogram Only) and C4 (No Thermogram, Control). C2 had the highest mean value for PostTreatment usage, a fact which was puzzling since this condition had the highest Number of Areas of Heat Loss on the thermograms. Initially, it seemed logical that more perceived heat loss would motivate greater conservation actions, and the resulting reduction in natural gas usage. Further, the between group comparisons on Income (Table 11) and the perceived barrier of Cost (Table 12) were not significant. From this perspective, the reason for the C2 mean PostTreatment natural gas usage being higher than C2 and C3 was unexplained. An alternative hypothesis asserted that C2 homes had more Areas of Heat Loss and this simply suggested that these residences needed more significant work done to remedy heat loss problems. Because space heating actions are frequently more expensive than water heating or lighting actions, C2 participants may have deferred taking Space Heating actions to a later date, when the necessary money for these actions might be available. The PostTreatment electricity usage had not been predicted to be significantly different between conditions. Since all participants heated with natural gas, electricity conservation actions were considered to be effected only by association to a general willingness to take energy conservation actions. The results simply confirmed the expectation of no effect. Even so, the group means (adusted for the covariates) showed that all the participants in Thermogram Meetings (C1, C2, and C3) did better than the control, C4. It was interesting that Cl had the lowest PostTreatment electricity usage, the same rank order as for natural gas. The theory behind the design of the treatments had suggested that attendance at the Thermogram Meetings was a necessary. but not 124 sufficient motivator toward appropriate energy conservation actions. Obviously, without these actions fuel usage would be expected to stay the same. Therefore, the followup treatments were introduced to see if the addition to Thermogram Meetings would result in the desired target behaviors, space heating conservation actions. Based on the Scheffe’ tests for significant differences between pairs of treatment conditions several significant and important effects on conservation actions were revealed. For the composite of positive action status categories (‘Done’, ‘Planned’, and ‘Done and Planned’) there were significant differences for both Space Heating Actions and Water Heating Actions. For Space Heating Actions C1 and C2 did significantly more actions than did C4 (No Thermogram, Control). For Water Heating Actions C1, C2, and C3 did significantly more than C4. Thus. the general finding was that the order of effects in terms of conservation actions were realized as had been expected. An examination of the component action status categories (Table 17) showed that the ‘Done’ category had higher representation with the Water Heating Actions than was true for the Space Heating Actions. Conversely, Space Heating Actions were more likely to be represented in the categories where the actions were planned. but not yet completed. Comparison of the two categories of action (Space Heating versus Water Heating, Table 3) suggested that the possible reason that more Space Heating Actions were not yet done was because they are generally more expensive. Both C1 (Thermogram + Workshop) and C2 (Thermogram + Mailed Information) had followup treatments which emphasized three special sets of space heating conservation actions, namely, Foundation Insulation, 125 Window and Door Modifications, and Caulking and Weatherstripping. It was intended that this focused emphasis on certain highly effective residential heat loss remedies would result in more action in these areas following the respective treatment interventions. The special series of analyses which tested for between-group differences on the desired results revealed that significant treatment condition differences were found for Foundation Insulation and Window and Door Modifications, and for both of these action categories C1 and C2 did, in fact, do better than C3 and C4. No significant treatment effects were realized for Caulking and Weatherstripping actions. These findings support the conclusion that the followup treatments prompted the desired conservation actions in two of the three focal categories. One interpretation for the lack of effect in the Caulking and Weatherstripping category might be that this category of conservation action is very common, therefore many people may not have considered it novel enough to be considered or remembered. Because the other two areas included do-it-yourself actions frequently not already known by homeowners, the novelty may have attracted their attention. and subsequent action. It should be remembered that treatment conditions were found to have no significant differences on PreTreatment conservation actions. and this finding was consistent for data from two independent sources. In general, the treatments had good success in prompting appropriate energy conservation actions. but the actual natural gas and electricity usage was not significantly affected. Interestingly. Number of PostTreatment Conservation Actions was not significantly related to PostTreatment usage of either natural gas nor electricity. Mean values 126 for natural gas and electricity usage did nevertheless suggest that effects were in the expected direction, especially for the most intensive treatment combination, C1 (Thermogram + Workshop). (Potential reasons for the lack of more significant effects on PostTreatment natural gas usage for C1 are more thouroughly discussed under Condition One Precesses . Cgrrelatign of Treatment Processee With Outgome Four treatment processes were considered to be importantly related to PostTreatment natural gas usage, but nonsignificant as regarded PostTreatment electricity usage. For all four of the processes, which are discussed in this section, the results did show no significant relationship with electricity usage. Correlations with natural gas usage provided some interesting results. For both the perceived barrier to conservation of Cost, and for the series of five Pro-Conservation Attitude items the relationship with PostTreatment natural gas usage was nonsignificant. Thus, perceived cost of conservation and important attitudes about energy conservation had little to do with natural gas usage. The lack of association with perceived Cost was particularly surprising since it is often cited as a key barrier to conservation. It may be that perceived Cost would be a significant decision factor if the question were to have asked about specific conservation actions. The other two treatment process variables, Number of Areas of Heat Loss and Participation in Additional Information Services did have a significant relationship to PostTreatment natural gas usage. The correlation,with Number of Areas of Heat Loss (C1, C2. and C3 only) 127 indicated a significant relationship but the correlation was positive (i.e., a greater number of areas of heat loss was associated with higher PostTreatment natural gas usage!). One possible interpretation was that when a residence requires a lot of energy conservation retrofitting it simply takes longer to get the job done. The correlation between use of Additional Information Services and PostTreatment natural gas usage was significant and negative. When people used one or more of these services (including the Energy Fair, RCS audit, and Energy Hotline), as had been promoted in the CEM program, the subsequent natural gas usage was less. Thus, it appeared that promotion of these services may have contributed to some of the natural gas savings. This variable was included in the multiple regression analyses to test for its relative importance in predicting natural gas usage. Condition One Processes Condition One (Thermogram + Workshop) was considered to be the treatment combination with the most intensive or persuasive impact on participants. The workshops had been designed so that local people could conduct it, and to the extent that the workshops could be judged successful this field setting pilot test might suggest its readiness for more general application in other CEM cities. Because it was hoped that the workshops would provide a significant contribution to increased energy conservation actions and subsequent reduction in natural gas usage, the current research investigated the dynamics of the workshop and the relationship of these dynamics to the desired outcomes. 128 Two types of investigation were included: tests of treatment integrity and tests of relationshig of key workshog grocesses to outcomee. The findings related to the treatment integrity issue suggested that the workshop treatment was not perfectly administered. Participants did indicate generally high ratings on the usefulness of workshop content, and they also reported a modest amount of intended conservation action in the areas covered during the workshop. But the number of hands-on tasks completed was both much less than had been planned and the amount of hands-on activity was significantly different between the three separate sessions. The author’s observation was that participants felt a little shy of being one of the few to actually aid in the demonstrations (do some of the hands-on tasks). Also, it became clear that time was inadequate for everyone to complete hands-on actions at each workshop station. The parts of the workshop which were relatively more successful did however, suggest some potential mainstays to the overall followup treatment design. The Foundation Insulation station achieved the highest number of hands-on tasks, and it was most highly rated among the three stations. The tests for the relationship of key workshop processes to the desired outcomes further emphasized strong features of the workshop design. High ratings (usefulness) of the Foundation Insulation workshop station were significantly related to lower PostTreatment natural gas usage. The correlations of the reported Intention to Act on workshop recommendations with PostTreatment conservation actions in the Space Heating area showed a consistent positive relationship. Even so, only one correlation was statistically significant: when participants 129 indicated intentions to do the Caulking/Weatherstripping actions, this tended to be related to significantly more PostTreatment conservation actions in the targeted category of Space Heating. Thus, the strongest features of the workshops tended to be the Foundation Insulation station, and in some respects, the Caulking/Weatherstripping station. It was observed that participants in the Thermogram Meetings were consistently surprised to see substantial heat loss from foundation areas, and relatively few homes had insulation in this area, therefore this seems to be a logical focus for future followup workshops. It would seem to further improve the participation rates in coming to such followup workshops if this followup activity were offered only to homeowners which had substantial foundation heat loss. Also, with a single content area (Foundation Insulation) the opportunities for more uniform hands-on activities would be less constrained by available time. With these improvements the impacts on PostTreatment natural gas usage might show statistical significance. Predic r V r' It had been considered useful to explore the possibility that selected variables might prove to be significant predictors of PostTreatment natural gas and electricity usage. If identified as significant predictors potential implications for treatment design might be recognized. Unfortunately, the multiple regression solutions provided little eee_information about predictors of natural gas and electricity outcome. 130 Summary This research succeeded in testing the relative effectiveness of two forms of followup treatment. In combination with Thermogram Meeting information, the Hands-On Workshop and Mailed Information followup treatments did prompt significantly more of the targeted conservation actions. As anticipated, results for PostTreatment electricity usage were not significantly affected by the treatments. The lack of significant treatment effects was also indicated for PostTreatment natural gas usage outcomes, but the egge; of effects suggested that the treatments had some of the desired impact. The investigation of the strength and integrity of the followup treatment designed to be most intensive suggested that, with some improvements, the Hands-On Workshops might effect significant reductions in natural gas usage. It was considered useful that the content of the workshop be narrowed to exclusive attention on Foundation Insulation, and that the promotion of these workshops could be emphasized for groups with identified need for this information. Furthermore, hands-on workshops might be designed for improved access and more immediate continuity with the Thermogram Meetings. Although the logistics would be more demanding, these workshops might be more successful if they were offered immediately following the standard Thermogram Meetings. If demonstration models were available in adequate numbers, many more people might take advantage of this second phase of the Thermogram Meeting. This would also eliminate a separate invitation process for workshops and would take place while the interest in the Thermogram information was at its peak. 131 The fact that C2 (Thermogram + Mailed Information) participants had significantly more areas of heat loss indicated on their Thermograms seemed to suggest that these residences simply had much more need for space heating conservation actions. In spite of the fact that the graphic demonstration of major heat loss might have been motivating toward the most effective conservation actions, it might also have influenced the resident to defer action on many of the major heat loss remedies. Thus, one programmatic change might be to integrate attractive financing alternatives. Because C2 participants were different from C1 and C3 with regard to Number of Areas of Heat Loss the need for improvements in research design controls was also suggested. If future research included matching of conditions on generic categories of heat loss, comparison groups would achieve greater comparaability. This would then allow a better test of treatment differences on outcomes. The current research also demonstrated the value of monitoring intervention processes. This category of data not only permitted the monitoring of treatment integrity (a factor which is commonly ignored), but it also provided valuable information for insight on ways to improve both the treatment and future research. Equally important were the tests for treatment condition equivalence. These findings were useful in addressing questions about self-selection differences and also shaped the content of exploratory analyses. While the treatments discussed in this manuscript had been explicitly designed to operationalize and incorporate "social science technology“, which was coupled to the ”physical technology" of the Thermograms, future program designs could involve even more social 132 science technology, especially in the area of community dynamics. In the current research, the components of social science technology of fectea; igformatton were available from the interpretation of the individual Thermograms, verbal feedback on energy conservation opportunities was part of the interpretation process, and eeeeiitg gehavigrs were recommended by the volunteer interpreters. Because the Thermogram Meeting logistics always involved small groups of local residents who talked with volunteer interpreters and each other, it was assumed that the small groups/social context of this environment would enhance the likelihood of social support for decisions to take appropriate residential energy conservation actions. Thermogram Meetings were even held during the winter, when heating bill issues would be most salient to homeowners. The quality and consistency of these program design features were perhaps best exemplified in the C1 (Thermogram f Workshop) condition, where participants were to benefit from the small group setting and the special opportunity to learn energy conservation skills by participating in them (tesk-arientetion) at the Workshop. Nevertheless, as shown by the very low rate of actual hands-on activity at the Workshops, even very carefully considered program design can have shortcomings in practice. It was considered important that the current research include measures of key program processes. In particular, the Workshop process data was used to examine treatment integrity issues. That which had not been included, however, were procedures and measures which could examine the relative importance of the community context and the small group dynamics which were supposed to be an inherent, operationalized part of the program design. Subsequent research (Jeppesen. 1985), which had 133 been initiated during the writing of this manuscript, broached these issues. An example of findings from this research was that Thermogram Meeting participants tended to spread the word of the meetings to an average of four other people. While these findings are preliminary, they suggested the value of having more information about the importance of community dynamics and social processes. Research which emphasizes the salience of the community context, and the relevance of local neighborhood dynamics, can make major contributions. It may often identify those program components which are suitable for direct intervention by program staff, and those components which can rely on a participant’s decision to take appropriate actions. For example, a program could provide full service attic insulation (no paperwork or installation work required of the resident), but might require that the resident do a complete job of exterior caulking as a prerequisite for the attic insulation. To the extent that the social mores of a neighborhood permit this combination of efforts to weatherize a home, it could then become an improved program strategy. APPENDICES Appendix A Thermogram Meeting Registration Form, With Response Statistics l 137 REGISTRATION FORM Please answer as fully as you can. This will help us to give you information which fits fir situation. Type of radiance? (Circle one) Single fully resino- Duplex Mart-ht labile ho- ” or mt? (Circle one) can bet If you rent a you pay heating bills? (Circle one) in: ho lein heating full (Circle one) lateral Gas Pal 0‘" Wood Coal Electric Pi'ooana Solar mn‘t know Iaiat kind of things have you all-em done to save on energy costs? (Check all that apply.) ’ Installed ailing insulation Reduced the aunt of hot Heated with :olay ___Installed wall insulation "m" 3“ _ownu driving mm Installed store winaIa/aors —w.g= hot water Switched to m car Headwatrippodlcaulhed had . furnace a...” (land carbon/usa transit Set back turnout ”“1““ h.“ m Other (mlain briefly) __M:cad In. liming Heated with wood it: did you hear about this pm? (Check all that apply.) ___lhupapar __Flyer __School ___Telavision _Frionds ___looth at a public event __ladio _hai9nbors __Speabar ___pg.m ___l|alativas __Chwrch NEST: As the Grand Haven Energy Conservation Organization (ECO) Project avelops we would like to know if we are actually helping people save on energy bills. To help us answer this question we need your written parlisaion to obtain a copy of your energy bills. If you are willing to help us in this way please comleta the following and sign your nan below where it says "Your Signature.“ Mich utility canny or cowanies provide energy to your home? (Please write the no. of the my or dealer in the appropriate place below.) Natural Gas Electric Fuel Oil PM“ I authorize the release of infomtion on the mint and cost of energy purchased from the above canonies and/or haler: from January 1978 to January l984. I understand this intonation will be used by the Grand Haven ECO Project to see if it is helping people save on energy bills. Your Signature IF YW WAIT TO WRITE mam YOUR CWTS 0R IEAS USE SPACE LABELED "YOUR CMNTS" ON THE BACK OF THIS FORM. 138 Your-1 COMMENTS » Verifies-dco—etsareiqortant. li'youwould lihetosherutheepleesewritetheedownontheblanh lines bear. HELpEn's wonKspAca ‘:.:.:"1:‘.:::2.'°:.. ' m: W of structure shows heat loss fro-(Check all that apply.) __pttic Foundation Other (enlain) __""3 ___llinrhhn __Jaeerells more AS: I. bet hind of inibraation inurests you net? (Chech all that apply.) ___Attic inulation ___Other (enlain) __hll insulation ___Fouatioe insulation ___Iin¢hrlaor udificatio. _ milking“ weather-stripping ___Finucing energy conservation Pm 2. If I- could avelop a free unstretiu workshop on the inflation you are interested in would you lihe us to let you kno- about it? (Circle one.) Yes D 3. heuld you be interested in rah-tearing so. of your tin to help with this Energy Conservation Orgnization mectl (Circle one.) Yes D 4. before cuing to this mating had you signed in for a hichi Gas Utilities RCS (Residential Conservation Service) Energ Analysis? (& the person a copy of the brochure and the mu sip-up card. then circle one of the responses below). Yes lo M If the pun answered "no' to P4 and W in signing-w: 1. Help thu fill out the rem-st card. keep it and indicate that we will nail it Lo; then. Then check one below. on 2. If they went to think about it. give than the PCS brochure and sigma card. Then. check one below. _person took infomtion ___person filed out must and left it to be nailed Appendix 8 Sample Thermogram $— b. or Sam le Thermo ram Figure 2. Appendix C Sample Thermogram Meeting Registration Form Appendix D Condition One Invitation Letter and Response Card 139 YOU ARE INVITED . . . To attend a free "hands— on" workshop on home weatherization on Saturday, at _ , o 'clock. This workshop will give you an opportunity to ask questions, get instruction and practice installation of weatherization materials. We hope this workshop will help people learn how to complete these actions with a minimum of cost and effort. Currently, the Grand Haven Energy Conservation Organization (ECO) is offering this workshop on a trial basis to a small number of people who registered at the ECU, "thermogram", meetings. Part of the project will be to see how much interest there is in such workshops and how helpful we can make them. So, next summer we plan to call people who come to the workshop and ask them questions which will give us some idea about how helpful the workshop was. This telephone survey will be brief and confidential. And if you'd like to know what the survey showed we will gladly send you the results. If you want to attend this workshop, please return the enclosed pre-addressed and post-paid reservation card. Transportation will be available from the Senior Citizens Center, Columbus and Fifth Street. Please plan to be at this location 15 minutes before the above scheduled time. Mark us down on your calendar. 140 YES, I plan to attend the workshop on home weatherization on Saturday, at o'clock. I will be at the Senior Citizen Center fifteen minutes prior to above time when transportation will be available. (Signature) ENERGY CONSERVATION ORGANIZATION CITY HALL GRAND HAVEN, MI 49417 141 WANT TO SEE YOUR THERMOGRAM ? Although you may have missed the opportunity to see the heat-loss picture (thermogram) of your home last Fall, you can still see it. Schedule of dates, times, and location January 26 7:30 pm ....... Loutit Library (lower level) February 23 7:30 pm ...... Loutit Library (lower level) March 23 7:30 pm ......... Loutit Library (lower level) April 27 7:30 pm ......... Loutit Library (lower level) May 25 7:30 pm ........... Loutit Library (lower level) We hope to see you there! GRAND HAVEN ENERGY CONSERVATION ORGANIZATION Appendix E Workshop Registration Form Instructions: Name: 142 GRAND HAVEN ENERGY CONSERVATION' ORGANIZATION HANDS-0N ENERGY CONSERVATION WORKSHOP REGISTRATION FORM Welcome to our hands-on (learn by doing) energy conservation workshop for home weatherization! Please print your name, address, and telephone number in the space provided below. If you came with another resident of your home please ask for one of these registration forms for them too. When you have it filled out hang on to it until the workshop is completed. Address: Phone Number: Appendix F Individual Checklist 143 INDIVIDUAL CHECKLIST INSTRIETIONS: As you have seen there were three training stations at the workshop” site. Each station offered different activities for stopping heat loss. We hope you had a chance to practice sons of these actions at each station. Please fill out this checklist, indicating which actions you personally gi_d_ and which actions yoqubserved. Hhen tallied these checks will give us an idea of which actions people tend todo nest. - ' ‘ ‘ STATION 1 : FOUNDATION INSULATION did or ggserved observed :did :observed _did _observed did (__pbserved Check (V) Etions Goal Measure did Clean and fill openings Cut Batts do at Insert Batts least Staple Batts m STATION 3 : HINOON ANO DOOR MODIFICATIONS Insider Storm Window Goal Actions Measure Cut wood Glue/nail Rough cut plastic Staple Trie plastic Edge tape FOal tape Insert STATION 3 : CAULKING AND HEATHERSTRIPPING least three . Check (1) __did '__pbserved __did __pbserved _did _observed did observed did observed did observed did observed did I observed did observed Caulking Actions Load caulk in gun Clean and Goal do at fill opening least Run bead Ewe Check _di d _observed _did _observed __did _observed :did _observed Foal Board Shutter Actims ' Goal Measure Cut fbam do at board Co least ver three Edge tape --- Foam tape Install Weatherstripping Actions Goal Measure Cut weather- do at str1pp1ng least Install Egg k _did _observed __did __pbserved _did _observed _did _observed __did _observed __did ‘__pbserved I M __did __pbserved __did _ob served __did __pbserved Appendix 6 Heat Leak Hit List 144 THE HEAT LEAK HIT LIST .r‘ n In . , ,1 - . ;—a.I&‘-'I 145 (::> BASEBOARDS - PUSH THIN STRIPS OF _ UNFACED FIBER GLASS IHSULATION UNDER aAstsOAROs. <9 ATTIC - In mane Is NOT A TOP PLATE _ oven EXTERIOR wALLs. STUFF UNFACED risen sLAss Down between THE srups. STAPLE rOLrtTHvLene oven The TOP. @ CRIRNEY OR FORRAQ FLUE - THE wOOp rnAnIue or The ATTIC rLOon Is IOXED OUT AROUND THE rLus on Chrnncv. Tweet Is usUALLr A GAP THAT Is NOT INSULATED. STurr unrAcep risen GLAss In TuIs GAP. ALTHOUGH rIaen eLAss Is nOT rinsrnoor (IT wILL Chan AT SCOOP.) THE rLua on cwInwcv Is nor LIktLr To cxceee ZSOPF. UNLsss soneTwIne Is TennIaLv wnono. (El) ATTI TRAP -» InsULATa TH! aacx or The pope. Ir vOU SELpon use IT. seAL TH! anes wITh DUCT TAPE. (5) ATTIC STAIRWAY DOOR - Imam: m AACA wITu FIBER GLASS OR IthLATIon aOARO. NEATHERSTRIP THE SIDES TwonoucaLv. .' DOOR TD UNHEAE §PAC§ - Sucu AS A honor. GARAGE. OASEHCNT. INSULATE UNHEATED SIDE. veATHERSTRIr ALL eases. HEATING AND COOLING QUCTS - STUFF INSULATION In THE GAP WHERE DUCTs PENETRATE CEILINGs. SEAL JOINTS wITH DUCT Tare. wear OUCTS wITH FIBER GLASS aATTs. PLUMBING VENT - STUFF GAP wwene IT — PENETRATES THE ATTIC OR OUTSIDE wALL. SILL PLATE - CAULK CRACK BETWEEN SIUL'SLATE'ANO FOUNDATION. INSULATE INSIDE OF BASEMENT OR CRAwL SPACE HALLS. OUTDOOR U TER FAUCET - CAULK ARouxo " . OPzflifiG ON CUIOID: Aha INSIDE OF HALL. 1455 (::> ELECTRICAL CABLE - CAULk HHERE CAaLE * ENTERS HOUSE. ON INSIDE AND OUTSIDE. NEAR THE FUSE aox.,flQI INSIDE THE FUSE aox. ® ANTENNA CABLE - CAULk HOLE INIERE ANTENNA CAaLE ENTERS THE HOME. STUFF FIaen GLASS ON THE INSIDE OF THE HOLE. (I?) TELEPHONE CABLE - CAULK HHERE THE THIN HHITE CASLE ENTERS THE HOUSE. (§:) ELECTRICA SHIT /SD - ON . OUTSIDE HALLS. LITTLE OR NO INSULATION IS aEHINp boxes. INSTALL INSULATINs GASKETS (ANAILASLE Fnon HAaprRES) SEHINO THE-COVER ' PLATE. WHOLE HOUSE FAN - IN sunnea IT I: m coven OPENING av HAAINS A PLUG Peon INSULATION aoano. SEAL EDGE: HITH TAPE. ROOM AIR CONDITIONER - CAULA spots. covER IT INSIDE. OUTSIOE. on aOTH HITH SIx-HIL POLYETHrLENa. SEAL HITH TAPE. BATH EXHAUST FAN - NAAE SURE THE OPENING CLOSES TIGHT'HHEN NOT IN USE. (IE) KITCHEN AND STOVE FAN - COVER OP- ENING FROM INSIDE WHEN FAN IS NOT IN USE. (::) CLOTHES DRYER VENT - CAULK Anouuo EDGES. KE SURE IT CLOSES TIGHTLY. PUT ON A MAGNETIC CLOSURE. VENT HARN AIR TO THE INSIDE. FIRFPMCE“ - WHEN THERE Is A FIRE. HARH AIR IS SUCkED UP THE CHIHNEY. INSTALL TIGHT FITTING GLASS DOORS TO PRC- VENT THIS. MAKE SURE THE DAHPER FITS TIGHT. IF THERE ISN’T A DAHPER. HARE ORE OUT OF NON-FLAHHAELE MATERIAL SUCH A; CEHEHT ASBESTOS. Appendix H Weatherization Information Resources Appendix I How To Notes for Conditions One and Two InfOrmation on Information on Information on Information on 148 INFORMATION PACKET CONTENTS FOUNDATION INSULATION Insulate Your Basement Walls Solving Moisture Problems with Vapor Barriers and Ventilation WINDOW AND DOOR MODIFICATIONS It' s Curtains for Heat Loss Calculating Energy Savings from Window ModificatiOns How to Build An Insider Storm Window How to Build A Foam Board Insulated Window Shutter CAULKING AND WEATHERSTRIPPING Weatherstrip Your Doors And Windows The Heat Leak Hit List other ways to save on utility bills Common Sense Energy Tips 65 Ways to Save Natural Gas Where to get more information Energy Extension Service: Who We Are and What We Do What else can you do? 1. Share a schedule of ongoing thermogram meetings with neighbors (tw9_copies are enclosed) 2. Mail in the Grand Haven ECO INTEREST CARD (one enclosed) Extension Bulletin 11“: In the Bank or Up the Chimney INSULATE YOUR BASEMENT WALLS A MODERATELY EASY DOW-YOURSELF PROJECT Install‘r‘XJ'SmdSSIongthawallstobeWAdd “fiber blanket inmlationbetwesnthefum‘qsuipa udlhilwithwaflbotdorpeneilhg. l 2. lull-.fl ‘ 3. Myduynqlenuw-dufi 4. Tapas-us s. “WONWMMW . w Zfi 7. www.mu\ Safety I. mumbling 2. Ityoau-gh-fibuocrockwool,wecgloveauda Materials What you'll need I R7(2-2)iuadi)3attorbhnketu-Iladon ginfiber withevapochmiedbuy polyethylene ifyoua'tpsbuuorbhnhnvithavaporbenh) ”J W... «— 3. Drywallorpanelling d. Waterproof paint,ifnmy @ g l-Iowmuoh I. Hedd- avua'heiflt above Ibuomdofdn wensyouhnndioieeflaaaandaddmfeetl'haa Iran-a the laugh of the rails you intend to unable. Ndplydntwoflpuastodemmbow mmfeetolundauoeianeaded. (haidItIXUenngPuae x I 2. Findtbelinearfonofsmdayou’uneadbymu- piyingIhalengIhofIhewallsyou‘mIendIoiml- stelry(6). (B) X (I-Igfll) ' (III. fl.) (8) X ' 3. Thearosofwellcoveringequahthebaemntwafl ”tamarind: ofwafl you intend Iofinidl. (WXllengehl‘arae x I MICHIGAN DEPARTMENT OF COMIC! e ENERGY ADMINISTRATION 150 Installation Prep-ation Manama:mmnoomgmywmtwalkfmmewmdoumde.flithaedymufls unmwmmmdmmtomiduh-Abdonyou'ngoingtomflhoubsoom madhefl'eeuve. m.-- thhebotmphsatotheflooranheb-oflhewefl mumm'wmwgu with MOULpIGG .‘I‘I-le‘. nil? 'Hl W. WED [nauseodiamallpieoeofinnfladonabovethefiu- Imflfinflwallboardorpmellingoverinafladonand ringandagaimdiesintounfleuchesfllandbandjoist. fun-ins. INSULATE YOUR '3 151 TWO OPTIONS AVAILABLE (I) Do-ls-Vouraafl: Install batt or blah! insulation madlluwflhmdprhntuofywaaflwhy sfladcvaporbanierdownonlheaavvlspacaeenh. (2) mule-hadrflyowcrawlspaospreneu mamflngspacaproblemyou mire-Ito “Wammbmmkl‘oryoum ooatraotorvrlprobablyfolo'ansathodsilfllloh do-iI-yorusall‘methodrbscribedbalew. Milling- paumiflngdlffamhevehinpioebothmhob uddlbwyoundrichisbataarjeapepédforadvioaoe howtonlacuoooinctor. nova: Mo‘eddluad-eh-eun-h _byrefien-dA‘JIn—garln-fin Henna-Inela- Inn-ee- Tools a. ”mag-“n... I. lint-dufi ‘ """' 2. l-bavyrhasysbuaorlholeuahi‘a at: _ 3. Tmllfithg Safety 4. ”tablet-afloat I. mug-um“ Tom‘s-mien 2. km‘aheuhflwheaworhm enhghfiherorrockwool 5' 7"“ @\ 3. mugs-mm 6. Ductorluh‘ngTade’wlde) s. Immunodelwireuralloflwat’oued Materials How much What you'll need Lula-3)!" dick)bhnhtso{rockwoolorgh- flmmamm 2. Six mi polyethylene plastic Iolay oncanh for vapor barrier(nnl'sarcamaemrvof1hickma) g 3. I/2"Xl- [12" stock for millngstripast Ihasilland a! Immune. ( muplyilnmtofledwululalioelae (MIXIIIeien+3’I-area x +3' - 2. Deiominetheareatobacoveredbythevqor beniabyfimingtllaresofyourcranrlsma (lervng)X(widllI)-aree x - Yourneyhaueiodlvideyouraeurlmoainuuv-i m-mmuaddupmeuwu. (W) x X X _— x ___ - TOTAL (width) ' I“ The Iota) length of nailing strips required equals the length of wall Io be insulated Installation Wrillbamt.anditwon'tkupyourhoueas mltwilalao-euthatwoodeumanbenthathold upmhmaewillbevet.xndtluy’llmt.l’mperven- maxim will prevent both of these problems: Lumen-loaniep-tdywmm “(moumifmfrmymfiamm Wit).adymmwlwutldtuyuposdblo— theairmovimthroudlitfmyoufunaeeismugh ventilationinwimernywhawmwlspaavenmkeep themahutinvlnter.openinunm.lfthenareno mmtheblaweronyourfumace3or4timesdur- mgthemmertokeeptheairinthemwlspecefrom gettingtoodarnp. rerr/y ,r’éw/I/ mamethajouunnpualldtohnlm m'tmmmmdmmmm mmmmwmmmdu upthMMMthsllz"XI-Il2’uha invinter. Note: Ywfinuamymiumm airfrouthaaawlspaa. Ifao.aomeoftheventadtould baleftopen. CheckvlthyourloaIHUD/FHAofflee. ~ ~ \ I' 4‘»..- l‘r‘ ‘ u ‘ ‘35.” \ IA . humanism-astute 153 TWO OPTIONS AVAILABLE LOO-IT-YOURSELP Installbetteorblanketsbetweenmefloorjoieubyen- pllr'wiremeitordtiekenmeothehotmol‘the jo’etsandslidlngtbehemorblnketainonmpol‘the mammmup. nejoblsquleeaeytodoinmeueeu‘youue inlet!“ over a call gees these my be acne nobles: Mummm.ht add MMMMpmuhmnpothly-d nir- Mmlhorjoiupedq-thlsmflvoek Mildew Ié‘oru"joinspedq.lfyouhln ‘Mammflnuvilhemm m-dfim'undmemdmfl. IWIMALLID Seepe'eé. Safety manner-mm “Mdmmummm [h-flbuorruekwool mmmflm e. Keeplidtusndallvireeoffvetmtnd P:- 9.. Materials What you'll need I. III (3".3W') hettaor hIenketaorrockvoolord. fillet. “(fly with foil fed“ (See Intuition). 2. Viewfinchiehnviredmlem Mfumhmm. Howmuch Determine the ares to be iuuleud by min. thelengthendvidthendmuluplyingtopttheene. (lengthIthidIhI-III (——.IXl—l-—_ Youmayfinditneeeeearytodmdethefloorinto smelletereeeendeddtlwm. llmflthMidthl-ene (_JX(—J-_ l—lxl-e -—+ totin- -— (.Sllw reel lure-of insulation (.9ll__l- mum-mofwmmeehordtickenwin . \ EOE-5‘9.- ‘\ «a a. Smuevel‘oeeenddthejfis-dwkou. mumnmmdmmuam *whmehnehoebpdthem.mlk vhhnmdmdhuothaitm'tbe mmmnthMH-m wmddnmuumm 154 ”mmmemm.uwmm mfmwheuwthemidflleufi-fi "mummdmflw.6etfoi- l’aeedhhheifyouafitvlnhtheflgeee www.3mmmdmatuyep “mmdemeo-d’hea Doe'lhhekeulhdunu'mfoefu- WM "InfiehnkaptheCMuulq'WyHUD. 155 FACT SHEET Afl-Z-S-d UNITED $1.~-E$ @ WWI oemtum ENERGY CONSIVATION IN THE RURAL HOME SOLVING MOISTURE PROBLEMS WITH VAPOR BARRIERS AND VENTILATION When you install insulation—or “weatherin” m . pint. water stains in the atiie. or an extremelympaavl space. Trapped moistuie invites deay and inserts. Moisture whith gets into insulation also increases the rate nl heat loss; that-tore. you should osntrol moistureasanessentialpartolyourotvnenergy tonservation plan. During the heating season. warm indoor air holds more moisture than cold outdoor air (fig. I). 'Htis crates vapor pressure inside. which constantly forces water vapor out through walls and ceilings as it seeks lower moisture levels outside. When moisture levels within walls. suits. or crawl spaees become high. the water vapor tends to condense on cold stubs. In most structures maisture an esape to the outside. but if motsture moves into the walls. ceiling. or crawl space faster then it an esape to outsifl air. the moisture wtll build up. Here are three things you can do to control moisture buildup: (I) control humidity in the house: (2) install vapor barriers tn walls. floors. and ceilings: and (3) ventilate attics and travel spas-es. CONTROL HOUSEHOLD HUMIDITY In tnld climates. set your indoor mole for relative ht-sidity iii the winter nohigherthan'ss toio patient. When outdoor temperatures are 20'? or Iosver. mince the humidity to less than 35 percent. Although a higher humidity might be healthier and might improve the performance of your heating sys- tan, it muld ause serious condensation problenu in your home. When condensation on insulated glass windows. you know that the relative humtdt'tv ts definitely too high. You 21H to the morsture level inside your home by bathing. cooking, and doing laundry. These 31‘"th an raise the humidity level too high. Exhaust lane in baths and kitchens will help eliminate Ihls moisture before it spreads mimigl; the house. Clothes dryers he vented to the outdoors. ll high humidity persists. you mtgnt t‘Olllldl‘f using a dehumidifier. VAPOR BARRIERS AveporWieaaymrsrisIthatMeIe-I themeemeatolmoisturelromapoiatolhighvm wallorceilingtotheotmhhveporMnIMB 'plaeerloahoththeimhmiloatmotafl'm tram-eminently. “momuuquu—u pitia'ttthemperasuresetvhiehmkmtioaom) mmmammdthemityflhsvm 156 doreduoemoisturelevelsnearthesidingoroutside nlLThismyhelppeventpedil‘dpian Wtheeieamhmtloes. MWMammficmfinh heqtnrulyusediaitevvellaTaheolthehsehiag Wale-”heathehdourtheebo‘sttmm “wwwpmuwem Mmmmmelthm thiebomoieusrecaaeneerthevallshetsveenaditiirt- iaginailadmhlmhemTliisissomli'mgtome-s 'Aaothercnmmonlyusedvmormnsriaaewhaifl- misplyethyleneilminlargerolla'flteflrlis mplesil' mauairottsl' yoverthetriside' ‘ lacedsturhmver' mbmdedflagjoutueridontopotloorjoiers oeei'aaasvlspaoe.Suchailmlimtheadvartmgeol “Mummyppaeewheeholesm mlaoper'mpsuehmtrmvendelsaiol cameh-eholesshouldhsatwymmm mmuhmisutrelmapmpom'hle. It isdilorlt maid a vomit-trenching mat-ohh'hmmhsesevenlmd m’I-mdm'atoamlkmdthiemyeeneas- shim m l-rrirr ilyou miatsin reasonable hot-eholdhtmiditim. l-lotveser. theonlytvaytohe ua-mmuuusmw mdtewelhmqplyoewmidingorothertkwsll ombudsmmmaaulspamaamflly hrvendeaorrghtoarrymoieuuemsohsrriereia mmd'm'maotmaidal-mrirnoa walk rim: Mummmhmmni-slm 157 Amier' eoerreeotnioesture‘ inhouaeetanheelrmr' '- 1 J mascodeshmIdheatIr-t l some hotbmhlfleqmekaolhmlvhereams mmhhmimrmmmhMD vaultdthslmm ,4 l Imus. munmuwhum energy dispatch Energy Extension Service Energy Administration Michigan Department at Commerce PO. Box 30228 Lansing. Michigan 48903 (517) 373-0480 158 IT‘S CURTAINS FOR HEAT LOSS liindows can account for up to one-third of e hae's heat loss. depending on the percentage of window space and how adequately the he is insulated. To understand how heat loss occurs through windows end ways to reduce window heat loss. an understanding of the following terns is necessary: convection. radiation and conduction. finvection is the transfer of heat by currents of air of rent densities. liens eir molecules expend. becm ligtter. and rave upward while the cooler ones becae heavier and sinit dowrnerd creating an air flow. Infiltration inserts cold air costing into the has air leaks m frestes end sashes creating a draft. is causes convection currents around winder areas. (A three by four foot window with Ills-inch leek around it is like having a hole in your well the size of e grapefruit!) You can test for infiltration around windots usingons onetheof following methods: 1. Hold a lighted reach or candle up to the suspected drafty area. If the flaws flickers or is blown out. are losing valuable energy. love careful holding e fleets near curtains. shades. or plastic. These mtsriels may be flee-able .) Halts a draft guege by attaching a piece of tissue or csslophaneto clothes hrenge rowithtape rpins. IloId it‘up to the suspected drafty tree and watch for it to b ietion is the transfer of heat in mes which are mittei sby a ter- objects; carpet. furniuire. wells. and people. rerient hast flowsto wi.ndote sconetcted by the glass and3 winder fem. end radiates outside. Qnduction is the transfer of heat throuyt solid objects. For exup e. eat is conducted through a spoon in a hot cup coffee. Heat is conducted at different rates througt different tutorials. For energy savings. the best window freees end sashes should be poor heat conductors. lint menu: Heat can be tramfereed in three darteeene eeye. Convection Ls that: is eeane by saying "hot air.- rises.” conduction alerts the: heat is transferred through neeertals tit-selves. Radiation easi- heat waves. ditch elueye travel tron melee objects to cooler ones in a space. Energy Hotline . . . 1-800-292-4704 159 R-Values R-Values measure the resistance of a material to heat flow. The higher the l-Value. the better the thermal protection against heat loss or gain. A well- insulated wall has a heat retaining ability around R-l? (depending on wall thickness and insulation quality). A window with double-glazing and loose drqiery has an R-Value of only 1.9. For the total R-Value of a particular window. add R-Values of all materials: panes of glass. window treaonents. and deed air spaces created. (U-Values measure the anount of energy transmitted. a‘reciprocal of the. R-Value: U . l/R or R - l/ll.) R-VALUBS measure the resistance of building neteriala to mm the flow of heat by conduct ion. The higher tmw&iin the R-Valoe. the lean wag-a heat transfer. / Conplete window R-Value mus intonation is available Erna Michigan's Energy Adniniatration Clearinghouse. Ask for “Calculating hergy Savings tron window Modification.“ publication ”01. —- m The primary considerations to reducing heat loss frat windows are caulking and w‘eitherstripping to reduce infiltration, and adding window panes to help reduce conduction. radiation. and convection. Adding air-tight thermal treatments or improving your existing drapes or shades can also vastly increase energy savings. Caulk around imvable parts of the exterior window frame. Do not caulk the Etta- of storn windows because condensation escapes there. Caulking is the most inortant and least expensive window treabnent. Heatherstrig around the movable window parts. Weatherstripping products range a se -sticking adhesive-backed vinyls to higher-priced spring metal strips that need to be nailed into place. There are durable easy-to-apply plastic types now on the market, too. Installing a lock on the window will nuke the seal even tighter. Add at least one window gne if you have only a single pane. Double-glazing is somet mes pre erred over triple-glazing because it allows more solar heat gain. especially on south-facing windows. (The nost widely recolmiended window treaonent is a combination of double-glazing and movable window insulation.) Exterior storm windows used to be the mast common glass addition. but insulated g ass 5 a so becming popular. Insulated glass is comprised of two or three panes of glass welded or sealed together with caulk. A dead air space or vacutn is created between the glass panes adding to the insulation value. The seals can deteriorate. resulting in condensation between the panes. The higher quality seal that is used. the longer the guarantee of the insulated glass. Plastic can be used as an exterior storm window, either by tacking the plastic to the window frame or by building a wood frame for the plastic. 160 gfigrior storm windows. are available frat several manufacturers. Some are permanent y nsta led and others snap on and off. You may wish to build a lastic covend wood frame interior storm window which can be easily renoved. _ most useful and inexpensive way to add a pane is simply to tape plastic over the interior of windows that are not opened during the heating season. than doing this. you may want to leave the bottom untaped. with plastic overlapping the windiw sill: Then the plastic can be ripped off in a hurry in case of e urgency such as fire. - ln-bebveen plastic store windows are another possibility to reduce heat loss. A frame covered with plastic can be used on the upper portion of a double-tong windiw. or a full sheet of plastic my be applied. than purchasing new winmws. consider the reterial of the window frame for its conduction rate. Since metals are good heat conductors. some new metal-fraud windows are being made with a vinyl gasket between inside and outside natal sections. This is called “thermal-bras? construction. Thermal window treements or window insulation should fit properly. If there are air leaEs, Wir Ffectiveness as insulators will be drastically reduced. Leeks say allow condensation to form on the window. then coinidering window insulation. think about the practicality of each window treated. For sample. north- and west-faci windows would be top priority: Their orientation toverds sun and wind eahe th- ose are heat than east- or south-facing windows. South-facing windows can actually gain heat fro the sun on a winter my. Types of window insulation inclua the following: ades include linen shades. quilted shades. and roll-up shades. The shades are usua y caprised of layers of thermal anterial such as fiberfill. plastic. or reflective plastic acting as a vapor barrier. and outer layers of fabric. The Inn shades have quilting rings tied to the back where strings are attained to fold or roll up the shde. wilted shades roll up into a valence and usually have a tight-fitting frne. Roll-up shades are tied up by ties attached to the top of thewindow frue and thebottaof theshde. Theshades can alsobe fitted with velcro attached to the shade and window frame or a hinged frne clam made of wood for a tight seal. ‘ saw-eases meanness-u. ml l_m 0:“ design for loans Sindee. Notice the weathentripped side clue to create a tight seal between the shade and window fr-e. Quilting rings attadiad to the back of the shade allow it to be raised or lowered with a poll cord. See the bergy Administration Clearinghouse liindoI- bibliography for are denigr- for window Q...“ shades. A good. tight-fitting shade can vastly increase window energy savings it closed at night. Pill 161 multiple-layer shades and multiple shade systons are also on the market. Multiple-layer shades are constructed with four or five layers of reflective materials separated by spacers that flatten as the shade rolls up. A motor is available frat at least one manufacturer that will raise and lower the shade. but the water's cost makes it practical only for large window areas. Multiple shade systh have three plastic roll-up shades mounted in the same frame. One shade is transparent. one is reflective. and one is heat-absorbing with perforations to allow viewing. The shade can provide insulation. reduce infiltration. allow st—er sun control. or allow winter passive solar gain. One shade wads of fiberglass can be used on the interior or exterior of the window. The exterior version can be protected from wind damage by a wind sensor which autonatically rolls up the shade during heavy winds. gngerior shutters include sliding thermal shutters. hinged thermal slutters. and pop-in shutters. Sliding thermal shutters are attached to an overhead dowel and slide over the window M in use. They are constructed with insulation hard or filled with fiberfill. fibe less. or cellu ass. and they. include a plastic vapor barrier within a wooden frue. Pop-in shutters can be constructed with a high density cardboard covered with foil, and separated by a wood frame to create an insulating air space. The wood frees is edged with weatherstripping to insure a tight fit within the window frue. Interior folding shatters are also * ’ available. They 3 are node of a rigid polyurethane foam core 3 between birch L ' . 1 od nels _ i Edi own or i There are many different options for shutter designs. 162 and cue with a wood frua waetherstripped with flexible fan for a tight fit. gmrior shuttelr‘; are either hinged or roll shutters. hinged shutters are cons ru insulation sandwiched between plywood and sheet natal:- ‘ They . can be hinged on.the top. sides or botta of the shutter with the inside facing covered with a reflective material to maximize your home's solar heat gain when the shutters are opened. The outside of the shutter can be either stained or painted to ntch the house. A cable connected to the shutter will allow operation fro inside the home. There are two different types of roll flutters: one is - constructed with ninerous horizontal slats and the other has only a few sections hinged lilae a garage door. in nels are caprised of a single pieceof insulation board held in place y ate or eegnetic clips: or fiberglass or fiberfill in a wood frus. locked in place by four bullet catches. They can be covered with decorative fabric or posters for use as wall hangings. or they can be covered with burlap for use as bulletin boards. That way they can hang on the wall when not in use. rather than havi to find storage space for tho. Another type of pop-in panel is a trans ucent sandwich panel reds of bro sheets of translucent fiberglass bonded to an alt-inc grid core strucbire. roe-tn runs are - weds to fit over “.0.“ ‘ the inside window / true. They are held in place with _ nether-striping or. like in this ”is. with eepetic stripe like a refrigerator door. Several option are available for pop-in panel dwai'. They can be eede with a wood it. filled with innlatiq nterials and covered with a posts: on the inside. or covered with burlap and used as a bulletin board. Another window insulation method is a product which blows polystyrene beads bebween rulti-pened windows. The beads are stored in a storage bin and can be controlled automatically by a thermostat or annually by pushing a button. A pup and rotor blow the beads into place or vacuum that into the storage bin when not in use. The same concept can be used by manually pouring pecking beads between window panes. 163 window blankets and ‘ thernal curtains ' contain insulation quilted to or sandwiched between layers of fabric.' They are either hang on a track and folded to one side idien not in use. or they can ban on a conven- tional curtain rod. Reflective fabrics say be used on the outside to reduce solar heat in sun. They can be sealed in the sa- menner as existing curtains. Draperies and shades can be greatly increased in effectiveness by sealing the top. bottom. and sides to reduce convective air flow. The top can be sealed by attaching a piece of insulating fabric or a valence can enclose the rod or dowel. The sides can be sealed with snaps or velcro attached to the drapery or shade and the window frame. Draperies can also be sealed at the botta. using a valence or by weighting th- to fit snug on the sill or floor. Sealing drapery gaskets are available for purchase as wall as drapery liners made of eliminized polyester to help control si—ar heat gain. pursues can create 1 j a flow of cold air near the window if they are not sealed '0'“ ”owners at the top or bottom. On a cold day. hold your hand near the bottom of your drapes. Can you feel cold air sinking into the room? These botton seal options. pictured. can help cut down on unwanted air circ- ulation next to your windows. Similar treatments are possible for the curtains' top. Melting a good seal between the curtain and window ire. will $51 save a lot on heating and cooling bills. A BAR WITH BRACKETS valence can be used ? howoecmrod at curtain tops to I prevent air from -y ‘M‘ioflm convecting around the - *W‘icml curtain. past the m CLAMP STRIP window. and into your coon. 164 Insulatigfi window films are another method of insulating windows. These p ast c ms are 9 u to windovs and reflect radiation from roan temperature surfaces back into the row. Some films are absorptive allowing some of the Osorbed solar energy into the room and others reflect most of the solar radiation. Heat mirror films are still being developed and are not yet widely available. These films reflect radiant heat back inside through the glass while at the same time allowing solar radiation to enter. They are applied to the outer side of the interior glass to prevent wear and tear froa the inside. Avoid condensation missus. . . Condensation can be a problem with winww treatments. If fog or frost forms on your window. it could damage the window frame and window insulation. If you nake sure that the window insulation fits properly and includes a vapor barrier. and provirn adequate ventilation for your bathrou and kitchen. condensation is unlikely to form. If it has. you may have too ruch Midity in your home. gar Suigg... It costs more in electricity to extract a unit of heat in the slower than it costs in gas or oil to add a unit of heat in winter. windows collect the sun's heat. adding to the cooling burden. Hethods of controlli solar heat gain include reflective films and shading devices. Reflective fi n that can be removed and reused is more feasible in cl inates like ours where it is advantageous to use solar heat gain during colmr seasons. Adjustable canvas awnings. shade screens. and metal louver-ed screens are on the rerket. too. They can be folded or rooved than solar heat gain is desired. You may wish to construct a rooden support frame to hold boards for shading. They can also be ruoved to let the sun shine in. 1 .0. Many of the reterials that can be used for asking window treeinents are flameble.. Plastics. insulating materials. and fabrics are often flaawrable. and may release toxic files into the air in the event of a fire. Please use caution rhen using any cabustible materials in your home. You may want to plan to use non-combustible materials for windne near your range or oven. or other heat sources. For sure information. . . This publication is intended to familiarize you with the many types of window treatments presently availdrle. For product and manufacturers' information as well as directions to make your own window insulation treeunents. some excellent sources are Movable Insulation by Hilliam K. Langdon (Rodale Press. Eneus. PA 18049; $14.95}. or Tfiemal Shutters and Shades by willirmr Shurcliff (Brick House Publishing Carpany. 31 Essex Street. mom. HA 01810; $12.95). . . The drawings and diagrans in this publication are reprinted frat Movable Insulation. (c) 1980 by iiilliam K. Langdon. Permission granted by Rodale Press. nc.. Ernaus. PA 18049. Energy Extension Service Energy Administration Michigan Department of Commerce 9.0. Box 30228 Lansing. Michigan 40909 (517) 373.0480 165 MCULATITG ENEmY SAVINGS W HIDOH MIFICATIQLI§ If you are planning to modify your windove "tos veoenergy. to estimate savings for neww i.ndows youca nuse this publica- tionto tocheck how each energy you might save. The Hunt of heat energy conducted through vim is measurd in units" of heat flow. and these values for different window types rt in 'U-factors.‘ The U-factor measures the transmittancg of heat in 8111' per square foot of witnaw area. per ur. par degree Fehrenireitnrepresenting the ffert-gnce between inside and outsim conditions written mm (Mr) (r-n. Since heat always flows fru the rmrmer side of a window toward the cooler side. windore can be considered to be heat-losers or orienation toward the sun du ng different unths of the year. This publication is not intenad to answer questions drout passive solar heating use of windows to gain heat during cold mouth: but it will provide useful information for consiaring has t from windows-«dis regarding their solar orientation." In “general. though. the better your fight heat loss in the winter ninths thebe better they will fight heat gin during the sumer. The following calculations will provid close estimates for the energy-saving potential of various winaw modifications. avi li-fectors for different windrw types are nasured An experimental conditions where the inside and outside t-para ratores can be closely ronitored. Once you know- the U-factor for a given window type you can uti-ate the total ennui heat loss vie conduction using this formula: ll-velue x 2‘ hours x d. d. x window area - BTU heat loss per year. The 'd.d." stands for annual degree daysua valu‘e representing averap climatic conditions for different locations- window area should be expressed in square feet. To find the U-value for various window types. consult the following chart. or ask a windaw nmnufacturer or distributor for test results from an independent testing laboratory. (Page 3 of this publication lists average annual agree-days for each Michigan county; page. 2 shows average u-values for many window types.) -' W is short for British Thermal Unit. One BTU is the «mount of heat energy required to raise the tuperauire of one pound of water one degree Fahrerheit--or approximately one kitchen match worth of heat energy. "For nore information about passive solar energy. contact the Energy Hotline. Energy Hotline . . . 1-800-292-4704- 166 . 12.2.9532. To calculate the savings likely to result from window modifications. apply the forrnale once to calculate annual fuel use (in BTU/year) for your present windows. Then apply the formula again using U-values based on planned window alterations. If you are planning to add shades. shutters. or insulated blinds. then you will have to consider the average lumber of hours per day they might be in place. To include shatters. shades. etc. in your calculations. Just make a fraction-min hours per day/Zim-and add that factor in the forwaala (see example on page 4). If you find an 'il' value for shades. curtains. etc. instead of a U-factor. than use the reciprocal of the il-factor to calculate the U (i.e., R - IN or U - l/R). Figrig infilgrggign. . . The U-velue formula. above. is to calculate savings based on heat losses due to conduction of heat throupr the window raterials thuselves. Another important energy emanation consideration is the heat loss due to infiltration of cold air into your hue. Sosa window infiltration occurs around the window frame itself, and this is where weatherstripping is applied. That kind of infiltration is measured in cubic feet per minute of air per linear foot of window frame (cfm/ft). and that's information that you should be able to get fru window manufacturers or distributors for new windows. For your present windows. you can reduce infiltration by applying weatherstripping to seal the Joints where window frmres meet. and at the better: and top of double-hang winmws. Caulking can be applied to seal windows rmich will not be mummwmmumww Mflnhwhemtrmflmhummrmt.~(nraven. ma—mmmmmmmum-nm—amm‘umu I“ m nae-mun u 0* “a M“ lineman—triple “guru-deals sinus“ 1.10 moron-s 0.. Guinness 0.. ”but... 0.31 merca- 0.. ”pp-r“ piers-staircase ase mm may” up tat-autos- 0.0 in.“ otherness 040 sedans.“ 0.. Might Pun“ lesdin.“ 0.. “he.” “no fiance. M "'01. "3 eras-vim ins mien-teens one ..:.= a ass-straws an mourns-near cu " mm our restaurant. on MI—MMm—fl Cass-ammo.“ Parac— I" a w m Typ-(WUV.esIum-AMIQT~P~ “a u “‘9‘- ‘3 swam when mm- Mum“: oauom a: Dee-muse Che to fies-e maniacs 05in Illll “Home m.:spaosiow 0.50 Adel. 1oo LN in) mm mad-mm 0.” 0.! 0.” e-@ 0.40 mean-mow- 0.” 0.. 0.” e-0.40 waus-n-eonor- m 110 no sound 03‘ SliourgPsmoDeore aimless: WoadFrerne 0.8 LN - iixiinainthiohoilh “Fr-no mo 1.10 - mm 0.53 ifixizxdimtniokmitn ”divider 0.51 MWieouedhu-thewwcuem” MOO!!- ammnmmmummu W 115 MW Wise 0.70 167 MIQHIGM m aroma; occngg-oers a; This map incMes average annual degree— days for each Michigan county. ‘ magnum The concept of degree-days is used to define average clinte conditions. The orders on this ‘ chart are 30-year averages as reported by weather bureaus throughout the state. The Mr 0f degree-days in ( one 24-hour period is equal to the difference between the average tmerature for that day and 65‘ F. The 'everage" ( tqerature is figured by reading the high and low teno- erature for a day. and divid- ing by two--to find the mean temperature. iihen that mean falls below 65'. then heating degree-days are counted. The annual order of degree-days is an accumulated total "W for a year. ['5 6400 6400 “[700 6600 168 A greater share of infiltration around windows occurs between the rough-frame and the finished window frame. For the best protection against infiltration. you should caulk all around the outside of the window frame. and insulate any open spaces in the rough frame. if possible. (See diagrea on page 5.) Fgr gale. . . Suppose my home is in Kalkaska Coun and I have 240 square feet of window area. Pom square feet already has storm ndow. but I want to estimate the savings fro adding storms to the rest of the windows. and I want to check the potential savings fro using shutters on half my window on winter nights. First I want to calculate the present heat loss in BTU/year. Fro the chart on page two. I find that the single-pane windows' U-value is 1.10 (part A of chart). I'm adjusting that factor for wood sash windows that are 80! glass (part0). mrltiplying by .90. For the windows with storms. I'll use .50 (fro part A) and .95 (fro part C). Annual degree-days for Kalkaske County--fro the rep in page 3--equals 8000. Applying the formula for my present situation I have: U-velue for single glass - 1.10 x .90 - .99: window area - 200 ftzz U-value for store window - .50 x .95 - .475; window area - 40 ft . (Roder. the fomla for analog heat loss in BTU/year is: U-value x 24 hours x degree-days x window area (in ft ) - BTU heat loss/year). Meat loss fro single glass ' .99 x 24 x 8000 x 200 I 38,016,000 BTU/year. heat loss fro store window - .475 x 24 x 0000 x 40 - 3.648.000 BTU/year. Present total for house (add each sag-nt's heat loss) - 41.664.000 bTU/year. 1f 1 add storm window to the mining 200 square feet, I'll have 240 square feet. all with U-value of .475. or: heat loss for all storm window - .475 x 24 x 8000 x 240 - 21.800.000 BTU/year. Savings (present total minus all storm window total) - 19,776,000 BTU/year. how 1 want to calculate the savings fro using shatters" on half the winow. eight hours per day (in addition to the storm window). The shutters have an R-value of 5. Il-value for the stoma wincbw is 1/0. or l/.475 - 2.105. Adding the R-values. 1 find the total R-value will be 7.105 when the shutters are in place. The total U-value will ba21/7.105 or .1407. The heat loss for 8 hours per day for ear simttard windows (120 ft ) will bez" I'Shutter's on“ 120 ft2 - .1407 x (8/24) x 24 x 8000 x 120 - 1.080.576 BTU/year. For the other 16 hours each day the heat loss will be: (Continued on page 6) *AZtuaTTy. since the shutters will be in place at night. rmen it's coldest. the actual savings will be greater than the formula shows. Rumour that these figures are estimates. Your actual savings will vary depending on local weather conditions. on the location and orientation of your home toward the sun and wind. and on the condi- tion and operation of your heating systur. "Shades. blinds, etc. would be calculated in the same manner. Add R-values to find the total for the window syston. 169 If possible. insulate the space between the . around (For more information about frame and rough irurne (white space in apply caulking outside ask for E58 Publication e19: Weatherstrip Your Doors and Windows.) house. Also. finished window infiltrating air will enter the the edge of window frames where they meet the siding materials. caulking and weatherstripping. drawing). That's where most Rough Framing 1 t 1;. ~ i 7 .‘ 29:. gorse} u .. “www.cvomrs. , a Sr. Elfibgflnfifid 3.. .. 343. ., 009.51.. a , www... .., _onwon...new...”o..£..r.n..o.a.,.s43 a.» . the. 214.7... .. .1, . n us. Shrub: .Am . . Ci? fiww. “.Nwa.” raq L44 . s ,_ 1...... “m1 .. . isn’t”. ..._..v i‘. £15..» “Jyoti! 'W-mn . a": .efav‘r w... ;- «flawe- a t. E . a «Sumo .. m... .J z . .. o. ... . ".4... - » u o m “5.1.1.4 ti. 7......lmlmhfi 11.17.14. 2, .9133! ..-. .L..3w.....,....r._..~m¢.r.Z.» W, a .1. t .. .J’ »V_ . 7.... n... 170 'Shutters off“ 120 ftz - .475 X (16/24) x 24 x 3000 x 120 - 7,295,999 BTU/year. The rest of my windows will be 'shutters off' for 24 hours, or: llindows with no shutters - .475 x 24 x 8000 x 120 - 10,944,000 BTU/year. The total for the house using shutters is (adding subtotals): 19,320,575 BTU/year. Savings fros the shutters is the difference between the heat loss for all store windows and the heat loss for the louse with shatters. or 2.567.425 BTU/year. Compared to the way the house is now. the savings would be 22,343,425 BTU/year. In orchr to calculate the dollar savings these BTU represent, check EES Publication '93: llhich Fuel _t_g Choose. . Mt fracas... Steel, alt-inn, wood, and vinyl are the must own materials used to wake window franss. Steel and lllllfll- will conduct much more heat than wood or vinyl. but you shouldn't let the heat loss of the frame material be your only consideration when buying new windows. Also think about how long the fraees will last. how each eaintenance they will require. and how they will look. For exaeple. wooden frames will conduct less heat than alumina, but the wooden ones will require regular painting idrlle alusintn will not. You say not be able to find out accurately how long each window type is expected to last, but the unufacmrer or distributor might provide you with news and addresses of satisfied custaers you could tall: with. A guarantee or warranty is another good assurance of product durability. I'l'his material was prepared with the support of the 0.5. Department of Energy (DOE) Grant llo. 80-77-6-01-5902. However, any opinions, findings, conclusions. or recal- mendations expressed herein are those of the authors and do not necessarily reflect the views of DOE.“ 171 HOW TO BUILD AN INSIDER STORM WINDOW An Insider Storm Window is a simple wood frame and plastic film ‘window'trsatment. It helps to seal off drafts of cold air coming from.the window'and in addition to offering some window insulation its design does not restrict sunlight from entering your home. MAIBRIALS'AND TOOLS YOU'WILL NEED hams doors e 3/4 inch.wood e wood saw (a miter box and . ripped to 1 inch. saw set up is nice but not wide from.wood a . absolutely necessary) bit longer than window'height . e wood glue yard stick or measuring tape pencil are finishnails' elcnife e staples for staple e hammer .9un e duct tape staple gun .e foam.tape weather- stripping e flexible plastic film (comes in a roll. should be a little wider than window frame opening) e wooden corner supports (3/4 inch wood ripped to 2 inches wide, then ~ cut into triangular 2- '\ pieces. as shown to the right) 2- READ THROUGH ALL INSTRUCTIONS BEFORE STARTING THE PROJECT. THIS WILL HELP SAVE TIME AND HELP YOU AVOID MISTAKES. 172 STEP ON! : MEASURE “ensure the width and height of the window frame opening where the insider can fit against a flat surface on top. bottom and sides. Tips: measure the window frame opening at more than one place along the flat surfaces for both . width and height dimensions. this is a good idea since some window frames may be warped or irregular. Remember to subtract.about one half of the thickness of the foam.tape weatherstripping .frcla width and height measurements (see Tips under are! 2163!). ESP 1'0 8 CU! WOOD Cut wood in lengths needed for the overall dimensions determined in err: 08!. but take into account the way the wood pieces will be fitted together as shown here. In this example. top and bottom pieces (A) are'full width measure. side pieces (B) are each two inches shorter than full height measure. and the support piece (Cl is two inches shorter than top and bottom pieces. . .rhen cut wooden corner supports as shown on page one. Tips: Make sure yours cuts are square so joints fit without gaps. Assemble wood pieces on the floor to double check that assembled measurements add up to needed overall width and height. {‘C 173 M=W Assemble wood pieces by gluing where wood surfaces meet, then nail joints as shown: C - ” “ A, “I 0' ‘_ _ then m ‘ Li ‘ ‘4 l l glue and nail ' glue and.nail main pieces wooden corner supports Tips: Determine where you want the support piece (C in 8T2? THO) before gluing and nailing - you.may want it to parallel a cross piece in the existing window. STEP POUR : ROUGH CUT PLASTIC Lay assembled wood frame on the 2. floor. roll out plastic next to ~q the frame. then cut plastic four inches longer than the frame height. This will give you a two inch border on top and bottom. Then. if plastic is folded as it comes off the roll. unfold it and cut it so you have a two inch border on each side of the frame. Cut two pieces of plastic with these dimensions. \. T198: Think about the dimensions 2.,4 .---------—-. of the plastic as it comes off the roll; you may be able to find a way to cut out the pieces you need with less waste than you would have if you did it as suggested above. 174 STEP FIVE : STAPLE With frame flat on the floor, or on a sturdy work surface, staple one sheet of plastic to each side of the frame. 17? 9 3 10 18 21 23 1: 15 s 7 1 2 ‘ .9 3 1: 16 22 2: 19 11 4 12 20 Tips: work in a well lighted. clean area. Light reflecting on the plastic will help you see how much and where to stretch the plastic over the frame before stapling. A clean area will insure that lint and other unwanted debris will not be trapped between the plastic surfaces. Staple from.the center of opposing sides outward (this is illustrated by the series of numbered staples in the drawing above). complete stapling plastic on one side before stapling plastic on the other side. STEP SIX 8 TRIM.PLASTIC Trhm off excess plastic about 1/8 inch in from.the edge of the frame. Do this on all edges of each side of the frame. rout lines—.1 l Vi J l . i 1 cut . . lines l i I i Tips: Make cuts with a knife against a straight edge as a guide. By cutting l/8 inch in from edges the plastic will not pucker as the duct tape is applied to the edges of the frame (STEP SEVEN). 175 STEP SEVEN : TAPE EDGES Seal the exposed wooden edges of the frame with duct tape such that the tape overlaps onto the trimmed edges of the plastic on each side. - Tips: Apply a length of tape so that 1/3 of its width is stuck to the front side. hart, fold and press the second 1/3 of the tape's width against the outside edge. finally, press the remaining 1/3 of the tape's width against the back side edge. Repeat this taping process on the other three edges. When taped on all edges the inside air”space (between plastic surfaces) will be air tight. creating a 3/4 inch 'desd air' space. STEP EIGHT : APPLY FOAM TAPE Apply self adhesive foamrbacked tape on two outside, adjacent edges. foam.tape Tips: As stated in STEP ONE wood frame measurements must leave enough room for about one half of the thickness of the foam tape when it is applied to one side and either the tap or bottom edge. ghe foam tape helps the insider to fit snugly within the window rams. 176 STEP NINE : INSERT Insert the finished insider storm window into the window frame opening. Tips: when fitting the insider into the window frame opening push the side with foes tape.on it in first. This will help the foam tape stay in place while you swing the opposite side into position. Some type of pull knob or tape tab can be secured to the front of the frame to aid in easy removal. TEE FINISHED PRODUCT When finished the insider will, by itself, have an R value of l but you can add another 31 to its overall installed R-value for the space of trapped air between the insider and the window glass (if this distance is 3/4 inch or more). In sum.you.will have a window with added insulation value without sacrificing natural day lighting or the view outside. 177 HOW TO BUILD A FOAM BOARD INSULATED WINDOW SHUTTER A foam.board. pop-in window shutter is a very simple way to reduce heat loss from windows. In addition to helping to seal off cold drafts coming from.windows, it has a fairly high Refactor. The most common use of these shutters is as a supplement to closing drapes at night. although they can also be left in place during the daytime. When considering use of foam board for shutters you should also know that covering themwwith some type of fireproof material is a necessity: if the foem.board should catch fire toxic fumes would be released. .A fireproof covering helps prevent this possibility. MATERIALS AND TOOLS YOU WILL NEED MATERIALS moss e 4 foot by 0 foot e utility knife insulating foam board. 3/4 inch a yard stick or thick measuring tape a duct tape e pencil e fireproof covering e decorative covering a glue (if needed for coverings) e foam tape weatherstripping READ THROUGH ALL INSTRUCTIONS BEFORE STARTING THE PROJECT. THIS WILL HELP SAVE TIME AND HELP YOU AVOID MISTAKES. 178 STEP ONE : MEASURE lteasure the width and height of the window frame opening where the shutter can fit against a flat surface on top. bottom and sides. Tips: neasure the window frame opening at more than one place along the flat surfaces for both width and height dimensions. This is a good idea since some window frames may be warped or irregular. Remember to subtract about one half of the thickness of the foam tape weatherstripping from.width and height measurements (see Tips under STEP rIVS). Also. allow for the thickness of any coverings. STEP TWO : CUT POAN BOARD cut out the piece of foam.board based on the measurements mode in STIP our. Tips: Place and old board or some type of work surface protection under the foam.board where you will be cutting through. Use a sharp knife blade and make cuts clean and square. Measure in from the edge of the 4 by 8 sheet of foam.board. pencil in cut lines, than with the knife against a straight edge as a guide. make the cuts. STEP THREE : COVER Glue or tape shutter coverings in place. Tips: You may want to do STEP POUR before this step, depending on your choice of coverings and whether or not you mind having a taped border around the shutter. 179 __smP POUR = W Seal the shutter edges with duct tape such that the tape overlaps onto the covering ed es on each side. Tips: Apply a length of tape so that 1/3 of its width is stuck to the front side. Next. fold and press the second l/3 of the tape's width against the outside edge. finally, press the remaining l/3 of the tape's width against the back side edge. nepeat this taping process on the other three edges. STEP FIVE : APPLY POAH TAPE Apply self adhesive foamrbacked tape on two outside. adjacent edges. ‘,¢"5 foam.tape -~\i Tips: As stated in STEP ONE shutter measurements must leave enough room for about one half of the thickness of the foam tape when it is applied to one side and either the top or bottom edge. The foam tape helps the shutter to ii: snugly within the window :rame. 180 STEP SIX : INSERT" Insert the finished shutter into the window frame opening. Tips: When fitting the shutter into the window frees opening push the side*with foam.tape on it in first. This will help the foam.tape stay in place while you swing the opposite side into position. Some type of pull knob or tape tab can be secured to the front of the frame to aid in easy removal. THE FINISHED PRODUCT . When finished the shutter will. by itself, have an R value equal to that indicated on the foamlboard packaging. To this you can figure another R1 for the space of trapped air between the shutter and the window glass (if this distance is 3/4 inch or more). In sum. you will have a very effective and attractive means of saving on window heat loss. lsurrslon Bulletln 1100: In the Bank or Up the Chimney? WEATHERSTRIP initialilanlli In»: am am AN EASY DO-IT- YOURSELF PROJECT You can wathemrlp your doorsevsn ifyou're not an experienced lmndymmi. There are several types of weethmstnmfordoorieachwithiumlevelof effecuveneadmabilhyanddepeeofinetafladondlffi- silty.Selectunongtheoptioasp‘vestheoasyouredb bestforyotLTheimtallationsmthesameforthetwo sitsandtogofadoor.withadifferenl.moredluable ‘msfordltlueshold. “" 43:9 “anthems. _ Tapem I M-mnmlymeymmiavisblewlwn mammaliamureefiecdveosdoonthan whdowe. Emanuele, Haedeaw. qum Suwanee-easy tomeullveiblewlwemeufledaot varydiuable. Mes—nelson: mupyagsimtlwcloeeddoor. .' 2. WWflmmwr-z Task / 1:» View A Tape means 1‘ limmm’ls. Tinnupe M-mytolnstsll.vlsiblewhenlnsulled. durable. ludaelse-nailstflpmudy againstdomonthecseing 4. Sales med Took / Tin w J 1D? View Harm. nails. 7 Tape hie-me firm—easytoinstaflJnVisibIswhenW extremdy durable. Insulation — cut to length 3 and lack in place. Lift outer edge of strip with screwdriver .. after main. for better seal. MICHIGAN DEPARTMENT 0! COM.“ 0 ENIGY ADMINISTRATION Notat‘l'heeemethodsmelmderthmllthroughi I. Maud-wide: memo-anus: ”wheadofdoorfirm: amlemrlpoadoor.fenmle uheedttheah'mgeaibd' door: amlemposjenb. Ides-doctflnflylock academies-lees fl. 182 0. Door These: Tosh Scewdriver, Hacksaw, Plane. Tyson‘s leduelles-ueefliwithwoodesthreehholdthetienot eorn.vsry “MNW(mumm door). —9 [W 5“ VI“ hmfledes-remowedoor-d uinnqukdmnoestod'bea- tom. Cuttodoorwidth. ls- stallbyslidhnvinyioetasd fateawithacsewe. C Hail-Mead“ M \I M—wmtmuuapdonflypod womb ed. m eh wed. notespoeedto podflehmm'. w-WRWWamJilot Wbrdoitvyoureelfinetdlsnoaiflemdoeeby neceusplldwdlmndymml. I. men-em Test Screwdriver. Had sew. new. Tape meme m some way m-mMMbaoW woedeaosebmoetdlfllmdttohmdlvhyl e-rbstrephoenpum'eavailable. W-mdooranduhnrequired od‘bottoin. lottomdiouldbveaboet 1/8" sealapnlstvlnyLbsrrebevsliecutianflt foropslung. Iii 7.” Task Sores-drive. Hack sew. Tape ins-Ire ”A _' W-umfiflforflatthledlholdnmaydngoa arpetormg. W-cutsweeptofitllloinchinl’romthe edpsot'thedoor.Somesweepeanlnstalledonthe Mandsonwoutside.Clwckinstructions for your Muller type. mlmm I m ‘r v- ‘ ' Evdweeisn - very difficult to install. exceptionally good weather seal. lneelleeise - should be installed by a skilled carpenter. Weatherstrip Windows 183 lintallbymovin'sashtotheopenpoeitionandslidin; stripinbetweenthesadlandthedtannelJ‘acltinplaoe htothteuinpDonotooverthepulleysintheupper channels. 2 lnstallstrlpthafullwidthofthesafltonthebortomof thelowerndlbottornnilandthetopoftheuppersaali topru'l. a m —_, OUTSIDE £7 CLOSED Then attach a strip the full width of the window to the upper sash bottom rail. Countersink the nails slightly so they won't catch on the lower sash top rail. Rolled vinyl Nail on vinyl stripe on doublfltun‘windows assltown. Aslidin'windowismuchthesameandanbetreatedas a double-hung window turned on its side. Casement and tilting windows should be weatherstripped with the vinylmiledtothewindowasingsothat.athewindow shutthoontpressestheroll. Wharton“: L.——-—-—"’" OUTSIDE Z \ Install adhesive backed foam. on all types of windows. only where there is no friction. 0n double-hung wm- dows. this is only on the bottom (as shown) and top rails. Other types of windows can use foam stnps in many more places. Installation of Caulking 184 Drawuapodbadofc-rlkwllltahallttlepnefloe. Hutmernptarnayheabittn-leahesuretltebead overlapbotlisidesforati'htsed. ,. y m; -- -.. arr-r: was-a :25- ‘ :u- .‘ tx‘av...“ -' n} “3:599:3‘“ f . ._ ¢ '4 . i' ééiaétéfl fin?" Awide bead my benecewy tomaltesurecaullt adherestobothsides. 1‘ "at"; " 1‘“ .I'.~ . ._i‘\ E('af»\"-ot 4“!- ' all" Fill umwidecraehllketheuatthefl(wheredie hou- meets the fundatbe) with oahn. flan fill. baboon strips. etc.) - angina-m ' -. ‘ - ;‘ ”IFS-”3,”; ‘7 h 3“" 'Jv‘,‘ . n5..nh!.t.'h- \I\\“ -- V ammo? \3 >>-. Caulking compound also comes in rope form. Unwind it and force it into cracks with your finprs. You can fill extra long cracks easily this way. 185 CONTRACTOR INSTALLED Tripletrschconihinatioe (windowsaedsaeenhtonn wiadowsarsderipedforinstallatlonovsrdoublehun' Theyuepsrnimentlyinaralledandcsnbe windows. openedsnytlntewitltsscrsadldintophoefor humbled. Yummsfewdell-suosto 1550!th thewhdows lemuaesitwllbeeefitohasstheespfiuie- dyeuwhdoufayommnwiflconm. Tl. w wfl hues! tlhw'ndowswh-e mummmltwflmw hon-“daystosfewwsiatonahupyouordu befontheupplumtoinnalthen. herallatloeWtahsle-tlmmedaydependluon hownI-tywiadowsueinvolsed. Two very important “mud.“tomahmthemlhfloeis done. supposed to reduce ail-hp uouedwindowefll dqthoftlnnetalm(ndttrecks)attheflaef thew‘sidowadtln quditynlhesa bi'differenceinlwwwulstornwindowscudothh. Canperssnsraltypssbefotsdecidiu. Had—em:1‘hequalityoflochandcetcheshna dlrectel‘fectondurabilityandisspodindieetorol‘ overallconstruetionquallty. INSTALL COMBINATION 186 NORMALLY CONTRACTOR INSTALLED YucatanefewdollarsflOltolSioftliepurchaee price)byiaetallln¢doorsyoutself.&rtyou1|needsonn Mmmmm.mmmm hint-tamitwilbeeee‘ertoltevsdieeuppiierinstal ”doors“. Themwifistmalthedoonwh-eyeu w-tmdoasineflled. ltwiukeanywherefrotn te—sldeystoafewwedatofleupyo-ordc ”commuters-town Inflation mmumwm. leforetbisnafltl-veehemtlndoorsoperete snootllyadclaeuditly.aieekforcraeksanundthe janhudmahmtliesealiseeaim‘dttupo-ihle. Abe. remove lid replace u- M puels (whdoweedeceenhomhsutstheyfitproperiyand withawnthcddttml. Sebctionaludgingauality DeerlleifisAmilllinish(pln'nalurninurn)willoxidize. Repair-calm "In (hamster UprheCht‘nsuey"byHUD. WMMmmmm ummmmwm Memdflersnoeinbwwdlmdooucando ummmbfimm mmmmdmmm mohashouldbeevsluatedsinceitc-luvsadlreet dfectoedntahiltysndisapodindlatorol‘overdl unality. mm:8tormdoorsofwoodorstedc- aleobepurcheedwithintlieslneprloennpaethe duniinunivariety.1'heyliavsthesaneqeditydlder~ scanddtoiddbe munchies betweendooreofsirnilarqudltybutdlfl'ermn-tlialis ptinuilyuptoyourownpersoealm. Steve at Mlehlgee Willi-O.” 00"“ WW efCenmeree 190 To cut your utility coats and help conserve Michigan’s energy supply EM?! 6 > Erie Admin'etratlon w @> Mlchlgan Department of Commerce 191 Thu following energy conservation erasures are designed to provide Michigan residents with quick and connon-senae ways to save enurgy in the hone. W e Lower your thernoetat to 65 degrees during the day and 55 degrees at night. e leep windows near your thernostat tightly closed. Otherwise your furnace will keep working after the rest of the roon.has reached a cenfortahls tqsrature. e If you do not have storniwindows. cover windows with clear plastic sheeting. You'll seal out the cold and reduce heat loss. e Dust or vacuun radiator surfaces. Dust and grins inpede the flow of heat. e flake sure there are no ohetructions. such as furniture or draperies, around heating air vents inside the house. e Open draperies or shades on the sunny side of the house and let the sunshine in. Otherwise, keep draperies and shades closed to help keep warn.air in. Always close draperies and shades at night. e Close off unoccupied rouse. ledroons or other roons which are unoccupied for long periods need not he heated to people-confort levels. e Keep your fireplace danpsr closed unless you have a fire going. An open danpsr in s bd-inch square fireplace can let up to 81 of your heat out the chinney. e To lessen heat loss when a fireplace is in use and the furnace is on: - Lower thernoetat setting to 50-55 degrees. - In the roon where the fireplace is located, close all doors and warn air ducts. e For confort in cooler indoor tenperatures, use the heat insulation of all-warn clothing. Dressing wisely can help you retain natural heat. - Hear closely woven fabrics. They add at least a half a degree in warnth. - For wonen. Slacks are at least a degree wsrnar than skirts. - For nan and wonen. A light long-sleeved sweater equals almost 2 degrees in added warnth; a heavy long-sleeved sweater adds about 3.7 degrees. 192 We leap thersostat at 78°? or shove when you are hose. and set it substantially higher if you are going to he away for a large part of the day. Turn off the air-conditioner if you are going to be away fron hose for sore than 2‘ hours. ' Keep windows and outside doors closed. lenind your fanily not to hold doors open and allow wern.air to rush inside. he sure to turn off lights not in use - the heat produced by lighting suet he rssoved by your air-conditioner. ‘ Don't position heat-producing devices such as lanps and TV sets beneath a well-sounted thernostat for a central cooling systen. Class or replace filters. Clogged filters sake your systen work harder and less efficiently. Vsntilate high noisture areas such as hathroosh laundry tons and kitchen. lunid air sakes you feel werner then dry air. Turn off window air-conditioners in unused rouse. leap doors to unused rouse closed. loop draperies closed on the sunny side of your hose. leetrict the use of dryers. ovens and other heat-producing equipner Hhensver possible. use this equipseet during the cooler hours of nursing and evening. vent your clothes dryer outside. Otherwise it punpe heat and noisture into your hone. Don't forget your ”solar clothes dryer." Sunndried clothes swell greet and cost nothing to dry. Hear light-weight and light-colored clothing. natural fibers like cotton and lines are generally cooler than synthetics. On entressly hot days, serve salads or cold cuts rather than hot seals. Drink plenty of cool liquids. They really do help cool you. without air-conditioning.... he sure to keep windows and outside doors closed during the hottest hours of the day. Use window fans to cool the house when it's cool outside. 193 HA ener savers e Check the tenperature on your water heater. Host water heaters are set for 1609!, or higher, but you say not need water that hot. unless you have a dishwasher. A setting of 120°? can provide adequate but water for seat fanilies. (If you are uncertain about the tank water tenperature. draw sons water fron the heater through the faucet near the bottoe and test it with a thernoneter.) a Don't let sedinsnt build up in the button of your hot water heater. Sedinont lowers the heater's efficiency and wastes energy. About once a south. flush the sedinsnt out by drawing several buckets of water fron.the tank through the water heater drain faucet. e Linit the length of your showers. Showers can use less hot water than bathe. but take care not to "soak” under the shower head. a Always use cold water when it will do the job as well as hot. e Iaplaca worn washers on leaky faucets. e dripping hot water faucet leaking at the rate that would fill a b-ouncs teacup in ten sinuses can waste over 1600 gallons of hot water per year. e no not leave water running while shaving. brushing teeth, etc. o Turn off faucets prosptly after use. a no not wants but water on a garbage disposal. host operate better with cool water. e Use hot water during off-peak hours when possible. Off-peak hours are 10 p.s. to 6 a.s. W Cooking e Preheat your oven only for baked goods. It is generally not necessary to preheat the oven for seats, casseroles. etc. Load innediately when pro-heating tenperature has been achieved. a Preheating is unnecessary for broiling. The broiler of your range does not require preheating, no setter what you've heard. a Make use of night or early norning baking or roasting and aepgrate this electrical load densnd fron that of other cooking equipnant. a Start baking with products needing the lowest tenperature. e When possible, use low tenperature roasting. e Never use your oven to heat your kitchen. This is expensive and unsafe because ovens are not designed for space heating. 194 Don't ”peek." Instead, cook by tins and tenperature. Use a neat thorn-star when roasting to prevent over or under cooking and senses shrinkage. Use a tiner to tine all precise cooking operations. Tining prevents loss of heat through repeated openings of the oven door or by "peeking under the lid” during surface cooking. Place utensil on the proper size surface unit. If the unit is toobigfore-ellpen, heetisweeted. In the preparation of vegetables, rice, pasta or puddings. use ste- cookere (if you have thu). They are speedier ad need only enough power to nintain the us. up to pressure. lever line your oven with alt-int- foil. It can interfere with cooking and fuse to the heating el-snt of an electric oven, thereby reducing oven efficiency. Do not place foil on the enerackaeyourfood. butontheeeparste rackbelow. Leave -ischor-rsof speceonallsidee forproperaircirculation. lsverboilweterieanopenpu. iisterwillconetoaboil fasterdueeleesenerginakettleorcoveredpsn. Oee high heat setting to bring water to a boil or to start cooking foods with water. then reduce the heat to desired lower setting. but don't set an electric surface unit on "high" if you're Just warning as it.. “in cooking with electricity. get in the habit of turning off burners several dilutes before the alloted cooking tine. The heating elenent will stay hot long enough to finish the cooking without using sore electricity. Use swell electric pens or ovens (if you have ch.) for snail seals rather than the kitchen range or oven. They use less energy. Use pressure cookers and sicrowsve ovens (if you have thu). They can save energy by reducing cooking tine. Keep range-top burners and reflectors clean. They will reflect heat better. Expand the fanily nenue to include stews and other single-dish seals that can he prepared in a slow cooker or crock pot (if you have one). let cold foods and sandwiches more often. _— 195 rs and Food Freezin e Don't keep your refrigerator or freezer too cold. Recon-ended tanperstures: 38 to 60 degrees for fresh food conpsrtnant of the refrigerator: 5 degrees for the freaaer section. (If you have a are freaaer for long-tern.storage. it should be kept at , however.) ' a lake sure your refrigerator door seals are airtight. Test than by closing the door over a piece of paper or a dollar bill so it is half in and half out of the refrigerator. If you can pull the paper or bill out easily. the latch any need adjuetnent or the seal any need replacing. e If possible, locate your refrigerator or freaaer away free heat-producing equip-ant. such as the range, and out of direct sunlight. e insure proper ventilation. Haintain adequate clearance, as recs-dad by the senufacturer. fron walls and/or cabinets. e leap condenser coils clean. If dust or dirt is allowed to secs-slate, operation will be inpaired. e Defrost freaaer when llfi inch of frost has accunulated (on a annual-defrost nodal). The front buildup causes the cooling systen.to work harder. a Cool very hot foods for a short tins at roon tenperature before placing in the refrigerator. but don't let food stand for too long-hacterial growth can sake it unsafe. e Label all food cleanly and lsgibly. This elininates confusion and facilitates quick rasoval of food. e Place sore frequently used food itsne in the front. a Store products loosely to allow good air circulation. s Proper wrapping of foods helps prevent excess frost forsetion on sides and coils. e Hake a neural list of the things you need before you open the refrigerator or freaaer door, than take out as nany itene as you can at one tine. Dishwashing Studies show that a dishwasher uses less but water than washing dishes by hand. however. further savings can he ends in the way you operate it. e Always wait until you have a full load before rhnning your dishwasher. e Use the "short cycle” or "light wash" if your dishwasher is equipped with one. s Dee only dishwasher detergent. Other cleaning agents can block the washing action. causing overflow and possible danage to the appliance. e lewovs ensues food before placing dishes in the dishwasher. e Check the filter frequently to he surc it's not clogged with food. a Turn off the drying cycle. After the tines cycle is cowpleted, turn off your dishwasher and open the door so your dishes can air-dry. They will dry quickly and you save electric energy used by the heating elenent. a Do not use your dishwasher to warw.platee. e If your dishwasher has a filter screen, clean it often. a Dee dishwashers during off-peak hours when possible. Off-peak hours are 10 p.n. to 6 a.s. daily. ‘ESQE£EI_2¥£EELJILJ!EEE a Don't leave but water running while washing dishes. e lines with were water. W m a flesh clothes in were or cold water. rinse in cold. e Till washers (unless they have swell-load attachwents or variable water levels). but do not overload then. s Des the-suds sever if you have one. It will allow you to use one tub full of hot water for several loads. a Don't use too such detergent. Oversudsing sakes your nachine 196 work harder and use were energy. Pre-eoak or use a soak cycle when washing heavily soiled garwents. You'll avoid two washings and save energy. Hash during off-peak hours (10 p.n. to 6 a.s.) when possible. 197 93922—22131 e If you have space and weather pernits. hang clothes to dry in the sushins or air. a fill clothes dryer. but do not overload. e keep the lint screen in the dryer'clean. Lint inpedes the flow of air in the dryer and requires the wechins to use were energy. a Dry your clothes in consecutive loads. Stop-end-start drying uses are Clergy since the dryer met reach the desired tqerature each tine you begin. e Separate drying loads into heavy and lightweight it.. Since the lighter ones take less drying tine. the dryer doesn't have to be on as long for these loads. a lfdryingthef-ilywaehtakessorethanoneloed.leeveswall lighnreight it-e intil last. Tousey be able to dry thn, aftaryouturnthepoweroff.withtheheetretainadbyaarlierloades e Dee heated inter only in the washing cycle. a Mmmmhthedryer. e Invaituewhendryerstopstoavoiduwnecseeerywrinkling which-yrequirepreeeingtor-ive. a Dry during off-peak hours (10 p.n. to 6‘a.n.) wh- possible. 3231—31 s have clothes that will need ironing frow the dryer while they are still d-p. There's no point in wasting energy to dryth-thoroughlyiftheyonlyheve tabedanpenedagais. a First iron those fabrics that require lower tenperaturee and work up to those requiring higher heat. An iron heats faster than it cools. e Turn off iron five sinutae or so before all clothes have been ironed, and finish ironing with the heat stored in the soleplate. e Always turn off the iron when work is interrupted by telephone or doorbell. --- a Do all your weekly ironing at one tine. 198 ens in the DATHROOH Take showers rather than tub baths. It takes about 30 gallons of water to fill the average tub. A shower with a flow of 6 gallons of water a winuneuees only 20 gallons in S winutes. Consider installing a flow restrictor in the pipe at the shower hand. These inexpensive, easy-to-inatall devices, restrict the flow of water to an adequate 3 to 5 gallons per winuta. 80H! LIGHTING Indoor Lightigg e Spend sore tine in the ease roow*with other fawily washers. You can share the use of the ease lighting and entartainwent. ”Light-sons" your boss and save electricity. Concentrate lighting in reading and working areas and where it's needed for safety (atairwalls, for exewple). Induce lighting in other areas. but avoid very sharp contrasts. Induce overall lighting in non-working spaces by rewoving one bulb out of three in wultipla light fixtures and replacing it with a burned-out bulb for safety. laplace other bulbs throughout the house with bulbs of the next lower wattage. Use one large bulb instead of several swell ones in areas where bright light is needed. lead new lawpe? Consider the advantages of those with three- way switches. They sake it easy to keep lighting levels low when intense light is not necessary. Use the brightest setting only for reading or activities that require sore intense light. Always turn three-way bulbs down to the lowest lighting level when watching television. You'll reduce the glare and use less energy. Dee fluorescent lights whenever you can; they give out were lunena per watt. for exawple, a 50-watt fluorescent lanp would save about 140 watts of electricity over a seven-hour period. These savings. over a period of tine. could sore than pay for the fixtures you would need to use fluorescent lighting. __ Consider fluorescent lighting for the kitchen sink and counter- top areas. Consider installing solid state dinners or hi-low switches. They sake it easy to reduce lighting intensity in a roow and thus save energy. 199 Contrary to popular opinion, you will use less energy by turning an incandescent light off and than on again, even a fur winutes later. than you will by leaving it on continually. leap bulbs and fixtures clean. Accuwulation of dust can lower lighting levels. Control window brightness to your best advantage. Use daylight when possible. At night. cover windows with light colored draperies or shades to reflect artificial light back into the IOU-b Install tine switches. Leaving lights on day and night while you're away frow.hows is wasteful and expensive. W Outdoor safety/security lighting that is nornelly turned on at sight can be put on a photo cell or tiner so lights will go off entowetically and not waste power if sowsone forgets to turn than off during the day. To reduce power usage, use swell-sired wercury vapor bulbs or fluorescent tubes. W Ivan though these itewe are swell energy users individually. you can save considerable energy through care in their use and operation. Don't leave your appliances running when they're not in use. leap appliances in good working order so they will last longer, work were efficiently and use less energy. use appliances wisely; use the one that takes the least awount of energy for the job. for exawple: Toasting bread in the oven uses three tines sore energy than toasting it in a toaster. A popcorn popper uses less energy than a unit of your range. Portable electric heaters should be therwostatically controlled. Lflnit their use to tenporary heating. These units are not designed for full-tine heating operation. When the extra heat is not needed or no one is in the house, turn then off or "_ unplug thew. Rechargeable appliances generally use sore energy than those that operate directly frow.the electrical outlet. energy V dispatch Energy Extension Service Energy Administration- Michigan Department of Commerce PO. Box 30228 Lana-n9, Michigan 48909» (517) 37341480 200 65 WAYS TO SAVE NATURAL GAS Tie mm mm or 1379-3) Atacama: Tl-E Merton’s macs av mm as. A arm-mm; an. THAT is m Innismax eon my. Inmmrsosnecnmlmmasxsmmm- trematmormxcm. Smweavausm mum. mummmmats. Ocuvmnmnr wwwmmaasts mmnmmm'mmmmwm rrnu matrimonial. Yuiciaasntnwnemnam- srarmfineuensmeofiumanmnemm mmmnteatenn'ana. ‘ TmmmrmoSmsmawmasm recommnucnwxnncwwmsmm mmmmonnoo). .Tevnainosaionstmiss mmvanunuwnu.‘ lknnulmi'rmtctsimmm mm otter TD save mmuneasmtsmusutsseis. Yunurttm mus my mmmmmmumwnmu IF you mo norm. ,- Evanrmcts m.mmsttnnmorcousava- Timmnntswmasatuavmxmmos ' 'nasu'mtctmemncmatmmmsrm. BEGEM’IVEAM PRSEVBQE. BECAUSE “1R WATICN WES HILLSAVEYGJMEYINTI-ELWMNOiflffi-ECOLNIRY WMITSWESNMMGAS. Energy Hotline . . . 1-800-292-4704 14. 15. 16. 17. 18‘. 201 HEATING Check your attic to see if your home needs insulation. Contact an insulation dealer, your local building inspector, or your county extension service agent if you need advice about insulation. Buy attic insulation by R-value, not by thickness. Recommended R- values are R-ZG to R-38, regardless of the type. Insulate floors over unheated spaces, such as garages and crawl spaces. Insulate, or increase the insulation, in your attic or top floor ceiling to at least R-26. Insulate your exterior walls if you live in a very hot or very cold climate. Call in a contractor for this service. Caulk and weatherstrip doors and-windows to reduce fuel use. Install store windows: combination screen and storm, single-pane store, or clear plastic file taped or stapled to the window frame. Add storm doors to your house if you live in a very hot or very cold c mate. Lower thermostat settings to 65'F during the day and SS'F at night. Dress warmly if you are cold. Let the sun shine in during the day to were the house; close draperies and shades at night to hold in the heat. Ask your gas utility or Hichigan's Energy Clearinghouse about the savings potential of conservation devices for gas furnace. Have your furnace checked once a year to make sure it is as efficient as possible. Ask your gas utility how to turn off the furnace pilot light during the sooner; make sure you turn it back on when cold weather comes. If you are buying as gas furnace, look for one that has an automatic flue damper to reduce heat loss when the furnace is off. Do not set the thermostat at a warmer setting than normal when you first turn the heat up; the house will not warm up faster. ___ Clean and replace the filter in your forced-air heating system about once a month for better system efficiency. Check the ductuork for a forced-air system, especially at connection points. Fix leaks with duct tape or caulking. 19. 21. 31. 32. 33. 34. 35. 202 HEATIIG WON'T.) Do not heat rooms that you are not using; close themloff and save energy. Close your fireplace damper when you are not using the fireplace so that warm room air does not escape up the chimney. Install glass doors on your fireplace to reduce heat loss up the chimb' ney. You can still enjoy the fire's warmth. - ‘ amine men Do not waste hot water by letting faucets drip or by running water needlessly. Install flow restrictors in your showers to reduce hot water flow to about three gallons per minute. Install aerators or spray heads in hot eater taps to reduce the flow. Do as much household cleaning as possible with cold voter. Use cold water rather than hot to operate your sink garbage disposer. Make sure the temperature in your gas water heater is no more than IZO'F (140'F if you have a dishwasher). buy a water heater that has thick insulation on the shell, or...add insula- tion to the outside of your present water heater. Insulate your hot water pipes if they are not adequately insulated where they pass through unheated areas. Flush out the bottom of your water heater about once a month to reduce sedi- ment build-up that lowers heating efficiency. Be sure your dishwasher and washing machine are full (but not overloaded) when you turn them on. Do not use the rinse-hold feature if you have one on your dishwasher. Buy a dishwasher that has an air-power or overnight-dry setting oF’both. Let the dishes in your dishwasher air dry by turning it off and by opening the doors at the beginning of the drying cycle. Hash clothes in warm or cold water as much as possible and rinse them in cold vater. 51. 203 HOMES APPLIANCES Reeove clothes from your clothes dryer as soon as they are dry; fill, but do not overload your dryer. ' Keep your refrigerator at 38‘oo40'F for the fresh food compartment, 5°F for the freezer compartment. . Keep the temperature in a separate freezer at 0‘? for long-term storage of food Hake sure the seals on the refigerator and the freezer are airtight. If they are not, replace the gaskets. _ Defrost manual-defrost refrigerators and freezers before the frost builds up to more than one-quarter of an inch. If you buy a self-defrosting refrigerator or freezer, buy one that has a power saver switch to turn off the defroster's heating element. Turn off decorative gaslights or replace them with electric ones. COOKING Buy energy-efficient appliances and keep them in good working order. Do not leave them running when they are not in use. If you are buying a new gas oven or range, look for one that has an electro- nic igniter instead of a pilot light. Hake sure the pilot lights burn with a blue flame for maximum efficiency. A yellow flame means an adjustment is needed. Use lids on pots and pans for faster cooking time and less energy use. Adjust burner flames to the pan size so that you do not heat the air around the pan. ' Plan your meals so that your oven is filled every time you use it. Keep top range burners and heat reflectors clean. COOLING Install a whole-house ventilating fan in your attic or upstairs window to draw cool air from the outside through your home. Use a ventilating fan when the temperature is 82‘F or below to cut down on air-conditioning use. 204 COOLIIG WON'T.) Set the thermostat for your air-conditioner at 78’F or higher and dress for the warmer temperature. Do not set the thermostat at a cooler setting than normal when you first turn your air—conditioner on; it will not cool faster. In humid weather, the 'low' fan speed on your window air-conditioner removes moisture more efficiently than the 'high' setting. Turn off your room air-conditioner when you leave a room for several hours. Keep lights low or off during the day to keep heat build-up at a minimum. Place lamps and TV sets away from air-conditioner thermostat. Their warmth triggers more cooling than necessary. Buy the smallest, least powerful air-conditioner you need to cool the space you have for the climate in which you live. Clean or replace the air-conditioner filters at least once a month so that cool air can flow better through your home. Insulate ductwork in your air-conditioning system, especially ducts passing through the attic or uncooled areas, to prevent cooling loss. Draw shades or draperies during the day to keep the house cool naturally; use awnings for the same reason. Cook and use other heat-generating appliances in the early morning or late evening to help keep the house cooler. Close off rooms that are not in use to avoid wasting energy to cool them. Never run the air-conditioner when windows or outside doors are open. Use the kitchen, bath, and other ventilating fans sparingly if your air-conditioner is on so that cooled air is not blown away. DID YOU KNOH . . . ' If 10 million gas-heated homes with inadequate insulation were properly : upgraded, we would save about 300 billion cubic feet of natural gas each year, or about 8% of the total demand for natural gas for home heating. If every gas-heated home were properly caulked and weatherstripped, we would save enough natural gas each year to heat almost 4 million homes. If you reduce the setting on your gas hot water heater from high (140°F) to nor- (120‘F), you could reduce the gas it uses by 18S. 205 If you buy a gas oven or range having an electronic igniter system, you could cut the amount of gas used by your oven by 471 and the gas used by the top bur- ners by about 53!. If storm windows and doors were added to 10 million of the gas-heated homes that need them, we would save enough natural gas to heat another 1.6 million homes. If heating temperatures in every gas-heated home were lowered 6 degrees, the gas saved could be used to heat an additional 4 eillion homes in winter. If you do not use the rinse-hold feature on your dishwasher, you could save 3 to 7 gallons of hot water every time you wash dishes. . If you turn off just one decorative gaslight, you could save $40--SSD a year. Eight gaslights burning all the time use as much gas as it takes to heat a whole house for a winter heating season. If you fix a faucet that is leaking a drop every second, you could save as much as GD.gallons of hot (or cold) water a week. If you insulate in your attic or top floor ceiling to at least R-ZD, you could save 5 to 301 a year on heating and cooling. If you insulate floors over unheated spaces, you could save about DI on heating and cooling costs. If you live in a very hot or very cold climate and you insulate your exterior walls, you could save 16 to 20s a year on heating and cooling. If you lower thermostat settings to 65°F during the day and 55°F at night, you could save about 31 of your fuel costs for every degree you reduce the average temperature in your home for a 24-hour period, or about I: for each eight hour, one-degree set back. If you have your gas furnace properly adjusted, you could save up to It! in heating fuel use. If you raise the average temperature in your home by G'F, you could save between 12 and 47 percent in cooling costs, depending on the length of your cooling season and the air-tightness of your home. The Energy Administration Clearinghouse has more than 250 free Publications about energy conservation and renewable resources. If you need further information or have additional questions, please contact the Energy Clearinghouse. Thank you for your interest and concern for Michigan's Energy Future. 206 Energy Extension Service , ENERGY ADMINISTRATION MICHIGAN DEPARTMENT or COMMERCE WHO WE ARE. . .AND WHAT WE DO The Energy Extension Service Clearinghouse operates an information service which is available to all Michigan residents — a toll-free ENERGY HOTLINE. A division of the Energy Administration/Michigan Department of Commerce. the Energy Extension Service is supported by a grant from the 0.5. Department of Energy. The EES Clearinghouse has a variety of energy information and materials about conservation. renewable resources (solar. wind. water. etc.). new technologies. and community and financial assistance. The EES Clearinghouse staff is ‘m—cail' to help Michigan residents with many kinds of energy questions. For example. we currently have over 200 different publications available to interested citizens. inciuding: Which Fuel to Choose (ase) Conservation Dollars (#229) The Energy-wise Home-buyer (#55) Do-lt-Yourseif Insulation Packet (#32) Wood Packet (#42) Conservation Packet (#41) Solar Energy Packet (#33) Single copies of these items and a complete list of energy information can be requested by calling the... energy hotline 1-800-292-4704 or by writing the... Energy Extension Service Clearinghouse Energy Administration/ Michigan Department of Commerce PO. Box 30228 Lansing, MI 48909 Please feel free to call or write the EES Clearinghouse staff with any energy questions. requests. or ideas that you may have. We look forward to hearing from you. 207 WANT TO SEE YOUR THERMOGRAM ? Although you may have missed the opportunity to see the heat-loss picture (thermogram) of your home last Fall, you can still see it. Schedule of dates, times, and location January 26 7:30 pm ....... Loutit Library (lower level) February 23 7:30 pm ...... Loutit Library (lower level) March 23 7:30 pm ......... Loutit Library (lower level) April 27 7:30 pm ......... Loutit Library (lower level) May 25 7:30 pm ........... Loutit Library (lower level) We hope to see you there! GRAND HAVEN ENERGY CONSERVATION ORGANIZATION Appendix J workshop Comments 208 WORKSHOP COMMENTS Your coments on this workshop will help us improve it. Please rate the usefulness of each workshop station shown below. Indicate your rating by circling m of the five numbers. VERY SDNENHAT NOT USEFUL . USEFUL USEFUL , l. The station on FDLIiDATIDN INSULATION: l 2 3 4 5 2. The station on moon AND DDDR MODIFICA- TIONS: l 2 3 4 5 3. The station on CAULKING AND NEATHERSTRIP- PING: l 2 3 4 5 During the next six (6) mnths do you think you will do some of the actions shown at the workshop stations? Please circle 93 answer for each of the areas listed. Any other cosments are also welcomed. 4. Are you planning to do someUNDATIM INSULATIM? Definitely Probably Cos-ants: Yes Yes No 5. Are you planning to do some NINDDli AND NOR mDIFICATIfliS? Definitely Prryibably Consents: es Yes No 6. Are you planning to do some CAULKING AND NEATHERSTRIPPING? Definitely Probably Cements: Yes ‘ Yes No If you have any other conlnents or ideas you would like to share please do so in the space provided below: Thank you. Appendix K Telephone Script 209 TELEPHONE SCRIPT Hello, this is . I am calling to let you know that the Grand Haven Energy Conservation Organization (the group that put on the "thermogram" meetings) is mailing out additional energy conservation information to a few of the people who attended the meetings back in September, October, and November last year. Do you recall the meeting you went to? I see from our list that you went to the meeting at school. (Brief discussion). Anyway, I will send you this packet of additional information including short publications on different things you can do with window treatments, foundation insulation and how to find and fix places where home heat can leak out. These are all pretty low cost options. And, all the publications are free. . He also thought you might like to know that 30 percent of the city turned out to those meetings. There was a lot of interest but not everybody had a chance to see the thermogram of their house. In case you know somebody who-would like to see their thermogram, I'll include a couple of cards showing when they can come in to see them over at the Loutit Library. Well, I'll send you that information today. We are sending out a sample of this type of information to just a few people who attended the thermogram meetings to see how useful it is. Next summer we plan to call people who receive this additional information and ask them about the usefulness of it. This telephone call will be brief and confidential. Oh yes, if you want to know how useful the information was to others who get these additional items, we will be glad to send you a copy of the results. Hell, I guess that's it -- if you have any questions, our number is 842-3210. I hope the information will be helpful. Thank you. Bye. Appendix L Condition Two Cover Letter 210 February 6, 1982 Dear Energy Conscious Resident: Enclosed is the information packet which we promised to send you, as well as a schedule of ECO thermogram meetings. Please share the schedule with friends and neighbors who would still like to see infrared pictures of their homes. The enclosed postage prepaid card will give ECO permission to release written information concerning how you saved energy. If you would be willing to share this kind of information or if you could help with some aspect of Grand Haven's ECO program, please sign this card, make a note and put it in the return mail. Sincerely, Jerry Brochu, Chairman Enclosure 211 WANT TO SEE YOUR THERMOGRAM ? Although you may have missed the opportunity to see the heat-loss picture (thermogram) of your home last Fall, you can still see it. Schedule of dates, times, and location January 26 7:30 pm ....... Loutit Library (lower level) February 23 7:30 pm ...... Loutit Library (lower level) March 23 7:30 pm ......... Loutit Library (lower level) April 27 7:30 pm ......... Loutit Library (lower level) May 25 7:30 pm ........... Loutit Library (lower level) We hope to see you there! GRAND HAVEN ENERGY CONSERVATION ORGANIZATION Appendix M Telephone Checksheet Page Contact No. of Date 212 TELEPHONE CHECKSHEET Time Paragraph Covered A B C D E F Notes/Comments Appendix N Residential Telephone Survey 213 ‘ .Do "”'-"‘-'NOT'" WRITE' IN THIS ‘AREA‘ m, 0000000000 m m: “m m m" 0000000000 W cm: AID M) 0000000000 Mmqutuatammmlm( ). luau 0000000000 mufiyaMcquch-dmoaflm 0000000000 2:: “.333: ”W33" u“ M. g“. ; L a r o: o «In: to you to ‘ cut 1 bun; "us, We" to 5 being "no: a 01.1 1.0:“: " 000000000 1. h'” m 110000000000 1.00mwpmumdchehm Lu: ( )7 m1 ”2““? . 0000000000 2. (1! W) lo- m no that 3% mum in , 0000000000 mp1... you to uk- cm actual? 0000000000 L-myt-cz-WMmMI-t Swami-c a 000000000 3. unnummlmudn( )Luc( )1 m - 1 I! - 2 -—-) 0000000000 b. (HW)hmn-thhcurmnm’u to m m nut-q? 0000000000 m 1.: m Jill-e can but 1.: s-Iac a an 1.: (6.1. 5. no you null pin to u h- m m I 000630000er at” or m e! um ”at? m 1 a 2 0000000000 . (u '70-”) I" M 1.: ——) 0000000000 . m I. m m to 1. haul-1.03.: 3m: 2 s-uc n 0000000000 way an “no: 1.: a: :11. in: (c.z 7 n. w . -——§ 0000000000 59 . you m - militia-J. .1113: at m use“. in "bu-:7 o: m at an. yen; 0000000000 '13 - 1. no - 2 __. 0000000000 I. (I: W) I" w an the autumn u mag-1:: you when: mm «mm 2 0000000000 l-vuy 201-can: u" m 1'" “ 11'... 0000000000 9. In. 1W t ( ) ms (mumwcafié.m mm.) a... mu mm“: 0000000000 m ' 1 m ' .2 0000000000 1.0. LEW-3.; MchhmmMy-ummonym 0099000000 “3 ' 1 m ' z __, 0000000000 u.(12”yu")laumt :mmmmmzuu man; you to uh.- srgnzmunatm mum? . 0000060000 bury 1‘: lumen-n: 3m: our. “no: Lap: he: a: 1.11. Ln: 0000000000 12. (H‘yu")flh1ch :hdidthcydomlo-m “nu? gum-:2.- 1: baton or an: rant-21mg, 0000600000 nu, In: t, t .) Munch 20:: . C m 2 DIGI‘J.’ COD! 00600630060 [0006390000 “SUI: on Alum knoll/(Manualw mm 15 h"?! hurl M. Ifiuumg “ta hogan 50.1“" U DO ‘ ‘ NOT WRITE 214 IN THIS - - AREA .. \\\§\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ \\\\ \\\\\\\\\\\\ - ‘-_‘W 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 W' F—- ns-x no-z rum 1.5. hmtalmmwuchumhut-mu e and: 0! ch- :bmp- noting? man-m “.mwrrm. ~— l-Iu.¢muwz-!u.m meant-lb (catamaran m.twuuamummmmuymm camammhmmnumdum bun. unawammmmampmuu ”.1wflsbwmammmotm mwmu-mmmmmum. mmmmmmm. man an: M an use“: datum-z. Maintain. U ml Io! ' ] Wan: ".‘T F—Ii‘ 1": Wang (MM) 6 Walkman I (mm) ‘ qund '17 L::; scorn d, - 1 Madden“ W Installs-ulna : .4... “.5. I undercutting: W Wands“ 13. En. you «0: called the con-Em may mum more 1‘. (1! “7“") m and: 41.4 you can‘t—m (m: flu 1.: baton or at": mum. bah a: July. law. and) no. 2 4131.: lunch/yr. «a (W) 1.5. mamo£1:om.mmuu¢mmu,m “mm-maulmofhaumomg? “-—-——_-_—__-_-_—~——————————-—---—--—--——-—-———--H_-n-—u 35 .DO '- - NOT‘WWRITE ’ IN THIS AREA mm: (unmannxuua (nutununuhfl umu ann mu mocha 1'0 no can nun: hummus Id u/ a“; ' ‘5‘. .b Iva-h. _. -. ’ -. “1:... ~12... “La -.—' \dqb I-fi' __i __6 ._J -—~o h—~« .—mq _h -Q -~ I u \ I \ l fl ~0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 GOGGOGGOGO C‘f“;‘-‘C‘G‘G‘C’K‘fl‘@ MSU u an Alf-rm: Acton/Equal Mum ham cl I‘ll” bl. luv at Sun unruly “mung 216 DO NOT WRITE IN THls- - AREA -.\-_\..\.5.5333303335.5.5.).>.>.>.>.>.>.>.>A.>.>.>.>.>A.>..\.>.>§..\..\..\.3..\.3..\.3.>..\._\..\.>.> W‘ (m man can.) 0000000000 “”3 “m "4' duo duo noun: u m: 0000000000 m. .2 _‘m a,” mm m 0000000000 0000000000 J .succhu to an wood has: __2 _2 __2 0000000000 0000000000 ] mum son: «.1 .tt. «.3. 3pm Inn: at t of @®®@©®©®©0 _anliuw 0000000000 1 M no. at at 0000000000 0000000000 ] and.“ a. 0000000000 "m" m 0000000000 ] _ m m, 0000000000 "" “"m 0000000000 ] 0.... 0000000000 @®®®@®©®®® (mm W) ©®®®®©©®®® mmtumm‘mmammcum @®®®®®©®®® ”than. You: mnunmmmmmmw any you my be m. 0000000000 ©O®®®®®®®® 1.7. no you an or run: you: tom“? 0000000000 .— 1...“ 2.0,, 0000000000 1,, n. m m m1... mm 0000000000 4.__ 1.,“ 2.... 0000000000 n. 0000000000 4-—1-u.mu 5.. 2.,“ on 34m be“; s-lloctric 6-Ptapm 7-Sou: 8900f : know ©©©®®®@©.® 20. In your can: hence; an or clan-1:: 0000000000 0000000000 :000000000 mummsu-hucn final? h 19de Ga 24106:“ Main: 0 NJ IN THIS AREA O - " NOT’ ,- WRITE I‘ll. SUI!!! lflllll: (Qatar curvuy unlbot) t Page m 3 3 (mm W car.) 21. Ian ally scottn- doc. you: incid-Ic- hav- 7 MIL-1.5 acacia. 2-2-2.5 station 3-3-3.$ start-- 22. ill: typo a! still. Ins-31.1 1- tI-callnl an 70-: tutti-Inn! 19.000 2----:--:, J-Ilullnln. b-vtlyl J-och-I ---)' u.Inunnuuwucuuucuocuudon-u-huuch 26. 27. 42:2! ‘l'lifll tail-III: (I) 1- ch: Ills-I'Iils Chill-It‘! ousting do you not during eh- d-csunn tat-Inns: (I) [ luv III, halted-n as. that. tn you: rustic-3. ' (nu-bur) d yun.knau 1:, Ian: 1- th. I'DIIIIIIS. h-Itgg IQIISI (outta. your :nltdnne-I I! at don't know - 9999 squat. (on: luv. you .4404 to a: rIIflVId any button span. a: your rostd-ncn in ch. 113: cue yours? l-Iddld Z-rI-av-d J-nochin. 044.4 or :Ilnv-d ‘--” 28. (I! add-d or rulnvod) About can: month an. ch- chlnga undo? (PIDHIT: “:3 t: botoru at alto: thanksgiving, XIII, Enncnr?) (nonch)(ycnt) (NOTE: 0:. SIII 2 4131: cod. as for 1::- 12) 0000000000 0000000000: 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 “SUI: an Alarm Acton/(MW m H In") filo by." Sulr (bums-l. Huumq DO NOT WRITE -3.§.§.)..\.).).).).).).)_\.\}.)AAAAAWAAAA.5.3.3.).3. - .).§.‘>.)..\.).).}1) 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 218 IN THIS - - AREA .. 29. Caymtudlaymts.mhulongvnnnt1m0tilhul: muuucmmummmmmmz (Ft-1t. tupmc «men. including mud-tn data (month: worm/you) - - - - - - - - - §—— Wlhmz : 30. mmummuuv‘mmmr : - I (“03) : 31. hm‘motmmu.mmu. Ink: mmmnmummuz . (m). (IIIIIILIIIS) aluminum-nautmm 2.1...“ ( ) )mthnl-tyu: 32. “tumoIthnda-m I I Wy.h¢un¢hdfly~w¢dtmmtmum cmntionmuyonuntunnd? Donut Runyontnknnoutnloantaranyofthommcmor- «tion projocts sinc- ( )2 JP!“ 2-uo 33. (1.! "yu") when? l-Locnl bah/{W institution 311th coqany Z-Nonlocu. financinl institution é-tnvutnnnt sourc- S-Muutnncn progrn 219 .. DO - '- NOT WRITE IN THIS AREA m; 0000000000 m m: 0000000000 um, m m, 0000000000 0000000000 0000000000 ‘ 0000000000 (mun 1'0 Gamma) ht- nun: ‘ “13000000000 laml'sgatngtatasdsustatrusansvuth IIva t fin tabla. W attics. Plssss carat-{y «1.1:: I!“ 0000000000 yan strongly spas, , mm. W..........:::. m “m " 'w’ 0000000000 36. ns cast at nasty swing m hss hspt - Ira tsung 0000000000 m was. 3: i g 1‘} “5, 0000000000 —’ 37. :“umammmmum 9999999990 usa]: ts Inns 2!. m.” "9‘ "' “'1" 0000000000 u t 0 D 9 —’ 0000000000 3‘. umuuummmnm . “ ‘ arson n 0000000000 39 a. ' 00000000001 . «nanny tn sum. or buying ths nu: ma. ant-tints h- int n Ira- tsfln' W m. 0000000000 8A A U D SD ' so. lat m ha- ta do ths Wu stuns usa: hss 99999999991 int n 2:. than. m sun-:3 ‘ u D 83 9999999990 —" 00000000001 u. m1- uu ohm. . 1o- u- an .- tta- 0000000000 m . 8A A U D SD —’ 0000000000 ‘2. Mamba-langtvmstsyntnypnssnt tssidsnns has kapt - It: nun. manna scttan 9999999999 A U D SD 63. Cancun to: tbs contort at hausahald _bsrs hss kqt an —" 9999999999 tr. tskin. W nuns. SA A u D 30 —D 0000000000 lob. thsmtatuathtsqursdhnsksptnstratsung tansstystian nations. 9999999999 SA A U D 51) —> 0000000000 6.5. has caapststian mun-ad fra- fs-lly odors hss kspt ss tra- tsnn. tusatystian sntians. 9999999999 8 ‘ ‘ a D s” —+ 0000000000 66. has hs bans. 1s boil hss «again. .31.... ‘ ““" " m' M" 0000000000 3‘ ‘ a D SD —§ 0000000000 MSU a so 4"th Mum/(oust Guyana-w m H h". I“! “I ””0th .0. luv I” SlJls‘ (its 220 no NUT WRITE IN THIS -' - AREA. - \\\§\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ - 5A A-“““AAA‘A‘ AHA- A HMM AAA--- WES-0000‘ :m cmsnvmm mu) 0000000000 s.- 1 would m. to ssh you} opinion «on: ssvstsl suns-cs mmtwem-ummrm. A618,“! 0000000000 .nzucnn as snow: mucus nah ssch «ac-3c. *4]. mutmndshyMMcom-astgc‘m 0999999990 A .30: inset u out nsta prob].- Si 2 3 3 3 A D SD 00990999991 ‘8. t tsun- casssrvtsg my rubs: can b01141." ass 1.“! 3A A U D SD 0000000000 ""— m ’ 69. his m should sp-d s t :hsu 0000000000 4""— .m up: a. nsta mm m ° 3A A D D 3D 009999999 50. I - Ins tonsils casts cute-um mum ch. V b 0000000000 ' """"""' “ ‘ ° ” s” b 1 as o: vhs cuss product 0000000000 *4” Jun 322.. $21....“ 0: m3; 1: °' SA A D D D 0999990990 ' g3. 37. I In“ assu- m my 0099900000 quag- S-xsrg m, t 32:; quiz}: 0000000000 “ ‘ ' ” " , . I usa-nus sells: ms to is h- nsta 0000000000 P” m... "' 99999999901 IA ‘ D D SD . as m sssls es III-cs W m as ®®©®©®©®©® A m nsta m. 8A A D D SD ©®®®®®<9®®® as. Gauss-s m say as: a: an pum- .4 u: an hss 0000000000 m a“ ‘h' b“ ”m“ °" “" ""u" SA A D D an @®®®®®®®©® . Its use.“ dds nsta shrugs 1s tssl. SA A D D an F 5. 0000000000 37. nsustyeaststsursot-ctlmbcdl- 0000000000 ¢-— ”‘u" “" “ ‘ u D a 0.4! 0000000000 21:: gaunt-3;: ~1- bv mac- to am 0000000000 a . u n :1: 0000000000 mm“) m. m w mu m “m "TF1 0000000000 ‘3 ”mm 3m; m ! 0000000000 «__52- As- 0000000000 .._29- man... 0000000000 +_6.%-‘ 1m- 0000000000 +_6_Z- 1m- Qum— -00000000a +__g. 3s: Appendix 0 Comments and Ideas. As Solicited on Workshop Comments 221 Appendix 0 Comments and Ideas, As Solicited on the Workshop Comments Instrument *I am building a system for solar hot water. After this change, in about 4 to 6 months, I plan to start other things. *would like more information. *I would have liked to see installation of weatherproofing fully exposed basement walls (walkouts) and remodeling ideas as far as insulation of existing upstairs walls (plastered, etc.). Excellent demo--Thank You!! *Well run. --Like to see more of these workshops and similar things done by the community. *I used 1/8 inch plexiglass sheets for basement storm windows. *Hould like more information on older home with crawl space. *Excellent idea! I'd like to attend another workshop! *Article in Detroit Free Press stated that you shouldn't leave your storm window up on a south facing storm door because if the inside door is completely sealed and the widow is up, moisture can form in between and warp both door frames. *This was really worthwhile. *Thank you! *Thanks, hope more people can do this. *Didn't show how to weatherstrip or caulk windows, just did doors. *Very good--in areas covered. *would like to have observed caulking and window weatherstripping. *This program is excellent. My major concern has been basement and bedroom windows. *Fine workshop! *Good ideas--Some I wouldn't have thought of. *Excellent program. Persons were knowledgable and willing to assist. *(The presentor) did not know prices or availability of materials needed for doing the Job. ' 222 *More time would be needed at each stop to actually do the work, even in a group of only eight people. *workshop very good. *Best result was when placed an extra insulation blanket in the attic. *I thought the workshop was very useful for me because I have a very old house and am just learning how to accomplish some of the things I want to do. *Hould like information on weatherstripping large double entrance doors--between doors. *Thank You. *Very helpful in my future plans. *Good ideas. Informative. Helpful suggestions. *I feel it was very worthwhile and although there wasn't enough time to get involved with everything, your prepared displays assisted in making the demonstrations more meaningful. A real fine program. *Horkshop where, for the cost of material plus fee, would actually supervise the making of window panels, quilts etc. would be helpful (and pop in frames for basement). I hope I can follow the instructions. I can't always. Very worthwhile experience! REFERENCES REFERENCES /”Bandura. A. (1977 . Self-efficacy: Toward a unifying theory of behavioral change. Psvghological Review. §g<2), 191-125. Becker, L. 3., a Seligman, C. (1978). Reducing air conditioning waste by signaling it is cool outside. Personalitv and Social Psychology Bulletin. fiiZ), 412-415. Bem, D. J. (19?O). Beliefs, attitudes, and human affairs. Belmont, California: Brookstole Publishing. Benson, J. (1981). Getting on the right path: Constituency building through energy planning. Alternative Sources of Ener ". 22, 20-24. Brim, D. G. Jr., Glass, D. C., C., Lavin, D. E a Goodman, N. (l?62). Personality and decision processes. Stan rd, California: Stanford University Press. ‘4’.- D“ Brown, L. A. (1?7B). The i. ' t'on d ffus' - " cubl' “0" context {Discussion Paper No. 58). Columbus, Ohio: Department of Geography, The Ohio State Universit Burns, B. A. (1?80, November) The relevance of behavioral and social models to the studv of consumer ener v dec1sior malin behaviors. Golden. Colorado: Solar Energy Research Institute. {SERI/RR-722- 341). \/Carlvle, J. reducin (Technical :eport, Title a Geller, E. S. (1980). Behavioral approaches to idential ener ~ consumption: A critical review HEA Grants for Communitv Service and T L Continuing Education in ‘firginial. Blacksburg, virginia: virgini State University. Pol,techic Institute a..d 1e figflul, ué4-Sé5. Coot, T. -., a Campbell, D. T. {1979). Quasi-enoerimentat-or: Design and analgsiz issues for field settinos. Boston: Houghton Hif‘lin Company. NJ [‘3 (A 224 Cront bach, L. J. (1970). Essentials of ._. holocical testin . (Zrd ed.). New York: Harper a Row. Energy Information Administration. Office of Oil and Gas, U.S. Department of Energy. (1983-I985). Natural Gas Mont Iv. washington, Energy' Policy Group. (1981, June). The potentigl role of solar energv and conservation in residential and commercial buildinos in Michioan (Technical report for the Michigan Energy Administration, Department of Commerce). Ann Arbor, Michigan: Center for Research on Utilization of Scientific Knowledge, Institute for Social Research. The Universitv of Michigan. Engel, J. F., Blackwell. R. D., S Kollatt. D. T. (1973). Consumer behavior. (3rd ed.). Hinsekle, Illinois: The Dryden Press. Engel, J. F., Kollat. D. T., a Blackwell, R. D. (1973).,fippgpmg: behavior. (2nd ed.). New York: Holt, Rinehart, and Ninston. Gladhart, P. M., Zuiches. J. J., a Morrison. B. M. (1977, June). Impapts of rising prices upon residential energy consumption. attitudes and conservation policv acceptance. Unpublished paper. Institute for Family and Child Study, College of Human Ecology, Michigan State Universitv, East Lansing, Michigan. Gladhart, P. M., Zuiches, J. J., a Morrison, B. M. (1978). Impacts of rising prices upon residential energy consumption, attitudes, and conservation pol icy acceptance. In S. Markov (Ed.), ner 011: in the United States: Social and behavioral dimensions. New York: Praeger. Glock. C. Y.. a Nicosia, F. M. (1964, July). Uses of sociology in studving consumption behavior. Journal of Marketing. 51-54. Haberlein. T. A. (1975). Conservation information: The energv crisis and electricitv consumption in an apartment complex. Eneroz Systems and Policv. 1. 105-117. Hagens, E. A. €19TT. September). Interface of behavioral svstems and enoineeri np systems. Unpublished paper. College of Environmental and Applied Sciences. Governor’s State Universitv, Park Forest South. Illinois. Hammerman. D. R., a Hammerman, w. M. (Eds.). (1958). Sutdoor education. a book of readings. Minneapoli Burgess Rublishing Co.. Hansen. E. (197$. September). Psychological theor1es of consugmer: choice. In R. Eerber (Ed.), Selected asoe:ts o° corsumer behavior. Washington. D. C.: U. S. Government Print: g Df*.ce. a"): Lei—i Haves. S. C., a Cone, J. D. (1977). Reducing residential electricity energy use: Payments. information. and feedback. Journal of Applied Behavior Analvsis. L2. 425-435. Howard, J. A. (1963). Marketin : Executive and buver behavior. New York: Columbia Universitv Press. Howie, T. R. (1974). Indoor or outdoor environmental education? The Journal of Environmental Education..§, 32- -36. Jackson. D. N. (1970). A sequential system for personalitv scale developement. In C. Spielberger, 9 r ent ton ’ rommun't' v . New York: Academic Press. Jacobv, J. (1977). Information load and decision oualitv: Some contested issues. gpurnal of Marketing Research. XIV. 569’577. Jeppesen, J. C. (1978). Organization and service: A correlational study of social services for the elderlv. Unpublished master’s thesis. Michigan State University. Jeppesen. J. C. (1935, March). Neighborhood open house survey: Lan nsinp sample findings. Unpublished manuscript. Energy Administration, Michigan Department of Commerce, Lansing, Michigan. helman, H. C., 1 Warwick, D. P. (1973). Bridging micro and macro approaches to social change: A socio-osvchological perspective. In G. Zaltman (Ed.), Processes and phenomena of social chanpg. New York: wiley. Koenig, H. E. (1979). Sha in our enero' future Foundation for community initiatives. Unpublished manuscript, Center for Environmental Qualit‘, Michigan State' 'nivers itv, East Lansing, Michigan. Kushler, M. A. {1977). Alternative modes of conducting outreach to low income elderly: An xperimental examination. Unpublished master’s theseis, Michigan State University. Large. D. B. (Ed. ' i1973). Hidden cons “r'ation. washington. D. '\v/Leedom, N. J. {1930). Energy conser-ati approach. Unpublished mas er’s thesi Lehman. S. (1981). Fond Du Lac Countv snagga nanaoement ol:n for the I m 1 1990’s; Strategies cf economic develocment through local energv j _- w self- reliance. Fond Du L:c. miscons1n: Ford Du Lac Sountv Commission. v}, Hznj bnn' of hahavl I. i f ! IJ 1 _ :i .J 0 11’ Cl. 1 a" [I [I1 :0 m p...- m 2' D U (1 ll“? [.4 .h M .2 L4 TI 1 II] ”M5 1.4. Lewin, H. (1951). Field theorv in social science. New York: Harper. Lovins, A. B. (1977). Soft energv paths: Toward a durable peace. Cambridge, Massachusetts: Ballinger Publishing. Meeks, C. B. (1981. Summer). Household energy use and conservation. Familv Economics Review. 26-28. Morrison, B. M. (19T4). The importance of a balanced perspective: The environments of man. pan-Envirpnmeng Svstemg. g (3). 171-178. Morrison, B. M. (1975). The closino energv resource environment: 3 multi-disciplinarv research team approach to the sudy of a complex problem. Presented at the American Home Economics Association, 66th Annual Meetings, San Antonio. (Available from Institute for Family and Child Study, Michigan State University, East Lansing, Michigan). Morrison, B. M., Keith, J. G., a Zuiches, J. J. (1973). Impapt§_pp, h useh I per v n um tion- An m ir' . t1 ' f Mi families. Report to the National Research Council’s Committee on Nuclear and Alternative Energy Systems, National Academy of Science. (Available from Institute for Family and Child Study, College of Human Ecology, Michigan State Universitv, East Lansing, Michigan). Nicosia, F. M., a Mayer, R. N. (1976. September). Towards a sociology of consumption. Journal of Consumer Research. 3, 65-75. Nie, N. N., Hull. C. H., Steinbrenner, N., a Brent, D. H. (1975). Statistical package for the social sciences. (2nd Ed.) New York: McGraw-Hill. Office of Intergovernmental Relations. Michigan Department of Commerce. (1981, June). Michioan communitv indicators. Lansing, Michigan. Dlsen. M. E. a Goodnight, J. A. (1977). 9 'a s e t f ponservation. Seattle, Washington: Battelle Human Affairs Research Center. (NTIS No. Pb-226-29). Pallack. M. 8., a Cummings, W. (1976). Commitment and voluntarv energy conservation. Personality and Psvchologv Bulletin. 2(1), 27-30. Pallack, M. S. a Klienhesselink, R. R. (1976, Winter). Polarization of attitudes: Belief inference from consonant behavior. Perspnalitv app Psvcholoqv Bulletip. ;, (I), 55-58. Pelz, D. C. (1930). Use of information in local oovernment innovation; fippg ppelimipapy aspegts. Prepared for DRSA/TIMS meeting, Colorado Springs. (Available from Center for Research on Utilization of Scientific Knowledge, Institute for Social Research. Universitv of Michigan, Ann Arbor. Michigan). 227 Pelz, D. C. (1981). Information use in urban innovations: Actorg and ghappglfi. Presented at the Annual Meeting of the American Psychological Association, Los Angeles, California. (Available from Center for Research on Utilization of Scientific knowledge, Institute for Social Research, University of Michigan, Ann Arbor, Michigan). Pelz, D. C. (1991). "Sta ino" effects in ado tion of urban ' n v ' . Presented at the Evaluation Research Society, Austin, Texas. (Available from Center for Research on Utilization of Scientific Knowledge, Institute for Social Research. University of Michigan, Ann Arbor, Michigan). Pelz, D. C., & Munson, F. C. (1980). A framework for organizational innovating. Presented at the Annual Meeting of the Academy of Management, Detroit, Michigan. (Available from Center for Research on Utilization of Scientific Knowledge, Institute for Social Research, University of Michigan, Ann Arbor, Michigan). Rogers, E., 1 Shoemaker, F. F. (1971). Communication of innovations. New York: The Free Press. Ross, M. H., a Williams, R. H. (no date). Drilling for oil gng gag in our 1 'n . Unpublished manuscript available from Marc H. Ross, Department of Physics, University of Michigan, Ann Arbor, Michigan. Seaver, W. B., & Patterson, A. H. (197é). Decreasing fuel oil consumption through feedback and social commendation. Journa f Applied Behpvipr Anaglysifi. 2, 147-152. Sechrest, L., a Redner, R. (1979). Strength and integrity of treatments in evaluation studies. In National Institute of Law Enforcement and Criminal Justice, How well does it work: Review of criminal iustice evaluation. U.S. Government Printing Office, on. 19-62. Seligman, C., Kriss, M., Darlev, J. M., Frazio, R. H., Becker. L. J., 1 Pryor, J. B. (1977). Predicting residential energy consumption from homeowner’s attitudep. Unpublished manuscript available from Center for Environmental Study, Princeton Universitv. Seven Reasons. (1981, June/July). Energy Conservation Bulletin: t r of the Ener v Conservation Cpalitigp. ;(1), pp. 1:3. Sharky, J. (1982). PrOjECtEd natural gas price . Presentation to the Energy Administration, Michigan Department of Commerce. Available from the Michigan Public Service Commission, Lansing, Michigan). Shippee, G. (1980). Energy consumption and conservation psychology: A review and conceptual analysis. Environmental Management. 5. 297-314. Shippee, G. (1981). Energv availability as a survival problem and thg role of psychology. Division Thirty-four Newsletter, American Psychology Association. 36-40. Shomon, J. P. (19é4). Manual of outdoor conservation education. New York: National Audubon Society. Skwira, G. (1981, March I). Reagan’s plan would double gas bills in 1982, Detroit Free Press. pp. 136A. Slavin, R. E., A Wodarski, J. S. (1977). Using group contingencies to reduce natural pas consumption in master metegeg apactmgpta (Rgpgpt No. 232). Baltimore, Maryland: Center for Social Organization of Schools, Johns Hopkins University. Slavin, R. E., Wodarski, J. S., a Blackburn, B. L. (1978). A rou contingency for electricitv conservatipn in master metered appartments (Report No. 242). Baltimore. Maryland: Center for Social Organization of Schools, Johns Hopkins University. Stern, P. C., 1 Gardner, G. T. (1979, August). Psychologiggl regggcgn and energy pgligy. New Haven, Connecticut: Institution for Social and Policy Studies, Yale University. Stern, P. C.. a Gardner, G. T. (1980). A review and critioue of energy research in psychology. Social Sciencg Review. 3(1), 1-62. Swan, J. A., a Stapp, W. B. (Eds.). (1974). Environmental educatign: Strategies toward a morg livgblg fggurg. New York: Sage Publications. Taylor, L. D. (1975, Spring). The demand for electricity: A survev. Bell Journal of Economics and Management Sgignge. §(I), 74-110. Thornton, D. W. (1976). Self-management: Commitments ang ggptcagtg (Unpublished training checklist. No. 76-3). (Available from Dozier W. Thornton, Department of Psychology, Michigan State University, East Lansing, Michigan). Wicker, A. W. (1969). Attitudes versus actions: The relationship of verbal and overt behavioral responses to attitude objects. Journal of Social Issues. 2;, 41-78. Winett, R. A., a Ester, P. (1983). Behavioral sc1ence and energv conservation: Conceptualizations. strategies. outcomes, energy policy applications. Journal of Economic Psvcholoox. 3(3-4). '30’.""? £— -..' d..- I Winett, R. A., & Nietzel, M. (1973). Behavioral ecology: Contingency management of residentaial energy use. American Journal of gpmmunitv Psycholoov. T, 123-133. n 229 Zuiches, J. J. (1977). Changing family energv behavior through infra-red heat loss evaluation (Interim report on Project EA-77-X-01-2118). East Lansing, Michigan: Institute for Family and Child Study. Michigan State University.