Marketing to Vulnerable Groups: Testing the Effects of Peers and Incentives on Elderly Response to Home Energy Audits

R. Bruce Hutton, University of Denver
Olli T. Ahtola, University of Denver and University of Tampere
Dennis L. McNeill, University of Denver
ABSTRACT - This article presents the results of a field experiment designed to test the effects of the use of peer group members and incentives on elderly homeowners' response to home energy audits. Results provide insights into the decision processes of this important segment of society and the impact of two motivational tactics on subsequent behavior.
[ to cite ]:
R. Bruce Hutton, Olli T. Ahtola, and Dennis L. McNeill (1986) ,"Marketing to Vulnerable Groups: Testing the Effects of Peers and Incentives on Elderly Response to Home Energy Audits", in NA - Advances in Consumer Research Volume 13, eds. Richard J. Lutz, Provo, UT : Association for Consumer Research, Pages: 481-485.

Advances in Consumer Research Volume 13, 1986      Pages 481-485


R. Bruce Hutton, University of Denver

Olli T. Ahtola, University of Denver and University of Tampere

Dennis L. McNeill, University of Denver


This article presents the results of a field experiment designed to test the effects of the use of peer group members and incentives on elderly homeowners' response to home energy audits. Results provide insights into the decision processes of this important segment of society and the impact of two motivational tactics on subsequent behavior.


In the United States, energy conservation programs have historically focused on information provision and motivation. Since 1930, market price has been the cornerstone public policy for encouraging conservation. Prior to this the Residential Conservation Service Program (1978) represented a landmark attempt at making available to all consumers, through a home energy audit, information on how to conserve energy. RCS was designed to encourage and facilitate the installation of energy conservation measures and renewable energy measures by making available low cost home energy audits. Utility companies were designated the primary distributor of the audits and audit programs. Since 1981, about one million such audits have been conducted each year (Morris et al. 1983, Centaur 1983). To date, neither pricing nor the RCS program have produced optimal consumer response in terms of conservation behavior. It is hypothesized that one of the reasons for this is the lack of attention to the differing demand schedules and situations of various market segments within the energy user marketplace.

This research deals with one component of society - the elderly--that has historically had special problems adjusting to the changing energy environment, presents unique challenges to utility companies trying to meet their needs, and in which there is very little known or understood about their responses to conservation efforts. Two reports by Hutton and McNeill (1981, 1983) summarize both the plight and opportunities facing the elderly.

In their literature review on vulnerable groups, it is documented repeatedly that the elderly and poor reach differently to external stimuli compared to other segments of the population. Besides the traditional physical and monetary-related reasons why the elderly do not participate in conservation actions, particularly important is the concept of alienation stemming from financial, health, and/or educational disparities. Both endogenous and exogenous factors appear to affect elderly response.

Endogenous factors consist of the alienated individual's own personal psychological make-up and the adaptive ways in which he/she confronts the outside world. While program planners and activators can have little control over these factors, they can and should take them into consideration in the institutionalization of the programs.

The individual increases his/her alienation from society by withdrawing from or disengaging in participation in that society. Social disengagement involves decreasing emotional dependency, commitment, or involvement. The former results in what is termed an external orientation, the latter in what is termed an internal orientation. It has been shown that "externals," though withdrawn socially, are nonetheless emotionally dependent on others and have a locus of control outside of themselves. They therefore respond well to psychographic presentations enhancing self-image and abilities. "Internals," however, are not emotionally independent, and their locus of control is internal. For this reason, they are harder to reach and influence than externals. However, if convinced of the seriousness of a situation. by the means of well-organized data, they may take action based on their own assessments rather than reliance on another's advice or persuasions. In either case, the alienated individual's response may be slow in coming due to the inertia of partially or completely reversing the process of alienation which he/she has entered into.

The other category of variables impacting an elderly or poor individual's degree of alienation are exogenous factors (i.e., societal factors). Exogenous factors may also either increase or decrease the sense of alienation. Programs which have tended to be unsuccessful have usually exhibited such qualities as a particularistic orientation which segregates servicers from utilizers, poor quality control, and/or insufficient knowledge of the target group's needs, wants, and tastes.

On the other hand, the literature review also describes a number of characteristics that have led to successful programs including:

- The servicers should go more than "half-way"--the participation of the alienated individuals should be encouraged and facilitated as much as is feasibly possible at each step. At no point should their automatic participation be assumed.

- In light of limited resources that must usually be taken into consideration when planning how much effort to expend on attaining the goods leading to culmination of a successful program, involvement of associated community organizations and volunteers is the key.

- Market studies should be conducted to pinpoint the perceived problems, value systems, priorities, desires, and tastes of the vulnerable group of interest so that properly targeted communications can be addressed to the group.

- There should be a sense of shared responsibilities and shared accomplishments between the servicers and the utilizers and a minimization of manipulation of one group by the other.

- Management of these programs should also be as efficient as possible, with a high degree of flexibility, to enable rapid response to developing problems in the program, as well as to maintain a fairly rigorous quality control system. This efficiency and flexibility help ensure a good reputation for the program, which in turn gradually filters into the households which need to be reached and facilitates a greater receptivity to the program.

- Each separate individual involved in the program, from the home energy auditor to the telephone receptionist, must display an inspired attitude toward the program. The attitude is as important as the information conveyed.

- Follow-up procedures are necessary at all stages of the program. Constant reminders--or even periodic ones--maximize the likelihood of the occurrence of the desired conservation behavior.

- Personal selling seems to work better than mass appeals to low-income and elderly groups, although its cost should be weighed in making a decision as to how to reach the desired market.

- Patience is a key factor. In general, many of these groups are more withdrawn from contact with service-providing organizations because of past experiences with exploitation, causing a high degree of suspicion, mistrust, and lack of cooperation. Time is required to reverse these suspicions and that Us why even many successful grass-roots organizations demonstrate such slow-moving programs and hard-won achievements. However, with a program which has "passed the test" the rate of participation will increase.

- At all times, the positive should be stressed, benefits should be emphasized over sacrifices.

While many of these conclusions apply to other segments of the population as well, they have been shown to be particularly important in reaching the poor and elderly. The combination of understanding the target group's needs and being able to integrate those needs with program goals and available resources will increase the probability of implementing a successful conservation effort.

In response to the environment for the elderly regarding energy conservation behavior and the social responsibility of policy-makers to particular groups such as this, the previously described factors were taken into consideration in the development of this study. Two strategies were tested to improve elderly response to an RCS audit.

- Peer. Accompanying each utility company auditor would be a trained elderly person. The role of the peer would be to provide support, information, and demonstration of some conservation actions.

- Incentives. The utility company auditor would provide one or more incentives to the elderly household (e.g., installation of a hot water heater blanket) to encourage them to engage in more such actions recommended by the audit.

Both strategies, peer involvement and the use of incentives, are well-tested and proven marketing strategies for other goods and services. The use of incentives, for example, was particularly effective in encouraging homeowners to take low-cost or free conservation actions in the Low Cost/No Cost Energy Conservation Program (Hutton and McNeill, 1982).


Two primary objectives were the focus of the study:

- Gain insight into elderly household response to the RCS audit experience, including changes in knowledge, attitudes, intentions, and behavior.

- Assess the impact of two marketing strategies-peer involvement and incentives on different levels of consumer response.


In order to attribute causality to the independent variables (i.e., peer, incentives) and to assess response in a "real world" setting, a field experiment was utilized. The design chosen was a 2 x 2 factorial. Each treatment had two levels. Either the peer or incentive was present or absent. The four groups were:

- Peer/Incentive. In receiving the RCS audit through a utility company auditor, a peer was present during the audit to answer questions, install the incentive, and demonstrate other conservation actions.

- Peer. A peer member was present during the audit; however, an incentive was not provided.

- Incentive. The utility company auditor was responsible for providing the incentive, answering questions, and any demonstrating. A peer was not present.

- Control. The standard utility audit was conducted. Neither peer nor incentive was present.

Peer group personnel were recruited by the local utility from area organizations. Each peer went through several training sessions, including having an RCS audit conducted for their own home. Peers were compensated at $20 per audit completed.

The primary incentive used was a hot water heater blanket. For each home designated to receive an incentive, a hot water heater blanket was installed. In the Peer/Incentive condition, the blanket was installed by a peer. In the Incentive only condition, the auditor installed the blanket. In cases where a hot water heater blanket was already installed, other appropriate low-cost incentives were substituted.


The sample for the study was a judgement sample selected to generally represent elderly homeowners in a large western metropolitan area. (qualifications included screening for homeowners 60 years of age and older who lived in single family detached homes and had never had an audit. Subjects were recruited at random by telephone. The total sample of 120 (30 per group) was then randomly divided into four experimental groups. Thirteen subjects refused to cooperate (i.e., changed their minds before the audit could be done) resulting in a final sample size of 107 (Peer/Incentive [28], Peer [28], Incentive [26], Control [25]).

Data Collection

Two types of data were collected for the study. Consumption data was collected via utility company billing information. Data was collected approximately 10 months prior to the audit in order to have baseline consumption information. Billing data was then collected for the 10 months following the audit. Analysis centered on changes in average daily consumption between the pretest and posttest periods.

The second type of data collection was a posttest-only survey of households conducted by means of a 15-minute telephone interview. Questions focused on measurement of a range of consumer response levels including knowledge, attitudes, intentions, and reported behavior. Respondents represented a family member present during the audit.


Results will be presented by consumer response level beginning with the earlier response levels in the hierarchy. Responses are the result of the posttest telephone survey, except for the consumption analysis. Change in consumption is represented by prior minus posttest period billing data.


A series of questions were asked dealing with homeowners' recall of the audit experience. In both conditions where a peer was present, over 90.0 percent of the sample remembered the peer. However, no one was able to recall the name of the peer representative.

In both the Peer/Incentive and Incentive Only conditions, over 70.0 percent remembered that something was done to their home during the audit to help them save energy (Peer/Incentive = 71.4%, Incentive = 76.9%). Only one person in each of the other groups reported something being done. It is interesting that 25.0 percent or more of those who actually received an incentive did not recall it.


Table l reports household impressions of the audit and provides insight into the word-of-mouth potential for diffusing positive perceptions and knowledge among other members of the elderly household segment.

Prior consideration of an audit ranged from 48.0 percent in the Audit Only condition to 38.5 percent in the Incentive Only condition. Households were asked why they had not requested the utility audit. The most mentioned responses were:

- Cost,

- Felt they had already done everything they could,

- Never got around to it, or

- Already knew what they needed to do.

When asked what they learned from the audit, households gave a wide range of responses. However, the most mentioned were:

- Need for some type of insulation,

- Caulking and weatherstripping,

- General knowledge, and

- Nothing/don't know.

Large majorities in all conditions reported discussing the audit with friends or relatives. Additionally, of those that said they had discussed it, about 50.0 percent indicated their friends or relatives would request an audit in the future.

An additional question asked was whether households felt that they would be able to save energy as a result of this audit. Majorities in each condition felt that they would. However, while the three test conditions involving some combination of incentives and peers all reported over 70.0 percent feeling this way, only 56.0 percent in the Audit Only (control) condition felt they would save. However, those that did feel that they could save energy, most had no idea how much they could save. For those that did estimate their savings, the most mentioned responses ranged from 5.0 percent to 15.0 percent.



Perceptions of the Audit Process

A series of questions concerning the audit process showed that all of the conditions were generally satisfied with the audit process. Few households would change anything about the audit, including the way the information was explained and presented. Additionally, there was virtual consensus that the audit should be offered to all elderly households. Primary reasons for this feeling included:

- Increased awareness/knowledge of the little things that can save.

- Improve utility company image, and

- Good service to the elderly because it is free and they need it.

Audit Evaluations

There were two types of evaluation efforts. The two conditions (Peer/Incentive, Peer Only) that received a peer member treatment were asked to evaluate the Deer. All conditions were asked to evaluate the utility company auditor and the audit process (Table 2).

In both conditions, the peer was evaluated favorably. However, the Peer/Incentive condition produced relatively higher scores across all positive categories. The presence of the incentive along with a peer produces more favorable responses, probably because the peer has a better chance of getting involved by installing the blanket or other incentives.

The utility company auditor was rated on the same characteristics by all four conditions. As one would hope, the auditor was viewed very favorable across all conditions. So consistent pattern of differences was found.



Other attitude questions also reflected favorably on the audit experience. Only in the evaluation of the utility company were there any significant unfavorable responses. About 20.0 percent of the sample in each condition had an unfavorable attitude toward the utility.

In a final evaluation of the peer, the Peer/Incentive condition continued to view the peer more positively than the Peer Only condition. While 55.6 percent of the Peer/ Incentive condition reported "very favorable" attitudes toward the peer, only 23.l percent in the Peer Only condition did so. Also, 67.9 percent of the Peer/Incentive condition felt it was a good idea to have a peer accompany the auditor; only 53.6 percent of the Peer Only condition felt this way. For those that did feel it was a good idea, the most mentioned reasons were cuts down on time, helps the auditor, and helps in understanding what to do.

Actions Resulting from the Audit

Tables 3 and 4 present the results of both self-reported behavior and an analysis of consumption data. Subjects were first asked what one change could be made in their home that would save them the most energy. The most mentioned responses were:

- Attic insulation,

- Caulking/weatherstripping,

- Storm windows/doors, and

- Turn thermostat down.

When asked whether they had actually done the one recommended change to save the most energy, there were significant differences between conditions. The Peer/ Incentive condition was the only one to have a majority reporting taking the recommended action (59.37). The Incentive and Peer conditions were next with 38.l percent and 37.0 percent, respectively. Only 28.6 percent of the Audit Only condition responded positively. However, the Incentive and Audit Only conditions tied with the largest percentage planning to take the action in the future (83.3%). The Peer condition reported the largest percentage making other recommended changes (35.7%), and, again, the Incentive condition reported the highest percentage of intentions to take other actions (34.6%).



Table 4 measures behavior based on actual consumption. The analysis of variance table indicates that there are significant differences in mean consumption patterns between the conditions (p < .05). Analysis was accomplished by subtracting average daily consumption during the test period (following the audit) from average daily consumption during a pretreatment period (prior to the audit). Consequently, positive mean values (average daily change in consumption) reflect a reduction in consumption from the pretreatment period. In order of effectiveness in reducing consumption, the conditions rank:

Incentive .0864

Peer -.1041

Audit Only -.1852

Peer/Incentive -.4165

The Incentive condition was the only one that achieved a reduction over the pretest consumption figures. The only statistically significant difference is between the Incentive condition and the Peer/Incentive condition.




The results of this study provide several insights for both policy-makers and utility companies. Perhaps most important was the positive evaluation given the audit process. At least for the utility company involved in this study, the problem of elderly response to their audit program is not with the product. It is tore with strategies to encourage accessing the audit and then motivating the household to take action. For example, one area that shows strong potential is the development of strategies that enhance diffusion of the audit concept by word-of-mouth.

Behavior measures presented both interesting and unexpected results. The condition that was most likely to believe they would save energy was the Incentive condition, followed closely by Peer/Incentive and Peer. Subjects in the Audit Only condition were significantly less likely to believe they would save. In reported actions to save the most energy, the Peer/Incentive condition led. However, the Incentive condition reported highest intentions in both cases and was second in reported actions.

In terms of actual savings from the pretest period, only the Incentive condition showed a decrease in consumption from the pretest period. These figures most likely reflect an extremely severe winter. The Incentive condition showed an average daily saving of just over .08 therms per day. Any average percentage reduction, however, is somewhat misleading. There was extremely high variance across both seasons and households. For example, winter gas usage varies from several thousand therms per month for a household to 150 therms per month for other households. In summer, gas usage dropped to a low of 10 therms per month. Further analysis is needed to refine these figures. However, it does appear that the .08 therms-per-day saving is a significant amount.

The unanticipated result in this study was the poor showing of the Peer/Incentive condition. While it is impossible to say for sure why this condition performed so poorly, at least two possible explanations are worth considering. The first is differences between groups going into the study. For example, an analysis of classification data show the Peer/Incentive condition to be slightly less educated, older, and lower income than the Incentive condition. Another factor could have been the lack of coordination between the auditor, peer, and incentive. For the elderly homeowner this may have produced a type of environment that was not conducive to the most positive response .

In summary, several marketing and consumer behavior implications arise from the results of this study. The first is the importance of understanding the target market, particularly a segment such as elderly homeowners. Second, and most important, the product (i.e., audit process) does not appear to be the problem in explaining peer response to energy audits. Increasing response to audits and conservation among the elderly is more a function of appropriate front-end marketing efforts (i.e., pricing and targeted promotional strategies) and the utilization of positive incentives to gain follow-up action. However, more information is needed about how to supplement audits with incentive packages and other motivational tactics. This study is just one step in that direction.


Centaur Associates, Inc. (1983), 1983 RCS Evaluation Highlights: Cost-Benefit Evaluation of the Residential Conservation Program, U.S. Department of Energy.

Hirst, Eric (1984), "Evaluation of Utility Home Energy Audit (RCS) Programs," Oak Ridge National Laboratory, Oak Ridge, Tennessee.

Hutton, R. Bruce and Dennis L. McNeill (1981), "Energy Conservation and Vulnerable Groups: A Selected Literature Review," Building Conservation Service Division, U.S. Department of Energy.

Hutton, R. Bruce (1982), "The Value of Incentives in Stimulating Energy Conservation," Journal of Consumer Research, 8, 291-298.

Hutton, R. Bruce (1982), "Marketing to Vulnerable Groups: Strategies for Energy Conservation Programs," Building Conservation Service Division, U.S. Department of Energy.

Morris, L.E. et al. (1983), Residential Conservation Service Program: Summary of the Second Annual Reports Submitted to the Department of Energy by States and Non- Regulated Utilities, U.S. Department of Energy.