Download Normative Social Influence is Underdetected and more Schemes and Mind Maps Social Psychology in PDF only on Docsity!
Normative Social Influence is Underdetected
Jessica M. Nolan University of Arkansas
P. Wesley Schultz California State University, San Marcos
Robert B. Cialdini Arizona State University
Noah J. Goldstein University of Chicago
Vladas Griskevicius University of Minnesota
The present research investigated the persuasive impact and detectability of normative social influence. The first study surveyed 810 Californians about energy conservation and found that descriptive normative beliefs were more predictive of behavior than were other relevant beliefs, even though respondents rated such norms as least important in their conservation decisions. Study 2, a field experiment, showed that normative social influence produced the greatest change in behavior com- pared to information highlighting other reasons to conserve, even though respondents rated the normative social influence produced the greatest change in behavior compared to information highlighting other reasons to conserve, even though respondents rated the normative information as least motivating. Results show that normative messages can be a powerful lever of persuasion but that their influence is underdetected.
Keywords: social norms; social influence; pro-environmental behavior; social inference
Authors’ Note: Funding for this study was provided by a grant from the Hewlett Foundation (2001-
- and by National Science Foundation Graduate Research Fellowships provided to the fourth and fifth authors. Our appreciation goes to Kimberly Brown, Allen Risley, and Lori Large from the Social and Behavioral Research Institute and Azar Khazian for their help on Study 1. We also want to acknowledge the work of our field research team on Study 2: Veronica Briseno, Dulcinea Contreras, Matt Dorlaque, Reginald Hartfield, Edgar Medina, Laura Murphy, Leezel Nazareno, Rene Quiroz, Ronald Tilos, Monica Tinajero, and Christina Wade. Study 2 was conducted while the first author was at California State University, San Marcos, as part of her master’s thesis. Portions of this article were presented at the annual meeting of the Society for Personality and Social Psychology, Palm Springs, CA,
- Address correspondence to Jessica M. Nolan, Department of Psychology, University of Arkansas, Fayetteville, AR 72701; e-mail: jmn03@uark.edu.
Normative social influence is potent and widespread. The cumulative findings from the research on nor- mative social influence are clear—witnessing the actions of other people has a powerful effect on behavior (Asch, 1956; Berkowitz, 1972; Darley & Latané, 1970; Deutsch & Gerard, 1955; Milgram, Bickman, & Berkowitz, 1969; Sherif, 1936; for a review, see Cialdini & Goldstein, 2004). It can lead people to say things they know to be untrue (Asch, 1956), to use illicit drugs (Maxwell, 2002), or to fail to respond to an imminent threat (Latané & Darley, 1970). In these situations, it seems clear that the direct personal experience of wit- nessing another person act can be influential in one’s own actions (Terry & Hogg, 2001; Turner, 1991). More recent research has shown that direct observation of others is not required for normative social influence to have its effect. Instead, communicating a descriptive norm—how most people behave in a given situation—via written information can induce conformity to the communicated behavior (Parks, Sanna, & Berel, 2001; Von Borgstede, Dahlstrand, & Biel, 1999). For example, Schultz (1999) found that households that received normative information describing the amount recycled by an average neighborhood family increased both the amount and frequency of their subsequent curbside recycling behaviors. Similar results were found in a hotel setting where normative messages increased towel reuse by more than 28% (Goldstein, Cialdini, & Griskevicius, in press). The use of written normative information has also become common practice in attempts to reduce heavy drinking among college students (e.g., Haines & Spear, 1996; Perkins, 2003).
Detecting Social Influence Having established the tenacity of normative social influence, researchers have now begun to question and speculate about the extent to which people are able to detect the influence of social norms on behavior (Cialdini, 2005; Schultz, Nolan, Cialdini, Goldstein, & Griskevicius, 2007). When choosing to engage in a given behavior (e.g., Cialdini, Reno, & Kallgren, 1990) or reporting an opinion (Griskevicius, Goldstein, Mortensen, Cialdini, & Kenrick, 2006; Von Borgstede et al., 1999), do individuals recognize the real or imag- ined presence of others as a causal antecedent? Is con- formity to normative social influence the result of a conscious or nonconscious influence on behavior?
Nonconscious Influences on Behavior In the past 25 years, substantial attention has been given to the study of nonconscious influences on behavior (Bargh, 2006). Much of this research has used subtle activation, or “priming,” of a concept followed by subsequent observation of the participant on a behavior related to that concept (e.g., Bargh, Chen, & Burrows, 1996; Shariff & Norenzayan, 2007). This research has produced unexpected results, showing that subtle, imperceptible primes can produce strong and perceptible changes in behavior. In the realm of social norms and conformity, research has shown that activating the goal of going to the library, which is associated with the situational norm of being quiet, led participants to lower their voices (Aarts & Dijksterhuis, 2003). Similarly, participants who were primed with words related to conformity (e.g., adhere , agree , comply ) were subsequently more likely to conform to the opinions of confederates who gave very favorable evaluations of a boring task (Epley & Gilovich, 1999). The observed behavior of other people may also be processed nonconsciously. For example, participants mimicked a confederate who was either rubbing their face or shaking their foot but were unaware that they had done so (Chartrand & Bargh, 1999). Influence is also considered nonconscious when the stimulus is perceived but is not evaluated as influential
conservation. Study 1 also provided an initial test of the actual factors influencing participants’ conservation behavior. In Study 2 (a field experiment), we extended existing research on normative social influence by assessing participants’ awareness of the extent to which different messages affected their behavior. Study 2 also provides a direct test of the accuracy of the causal explanations elicited from participants in Study 1.
STUDY 1
The goal of this first study was to conduct a preliminary investigation into the extent to which people’s beliefs about what motivates them to conserve energy correspond to the factors that correlate with their self- reported intention to conserve. We wanted to know what a priori beliefs people held about why they con- serve energy and to examine the relative weight that participants would ascribe to social norms as a factor in their decisions to conserve energy at home. To address these questions, we surveyed a diverse sample of California residents regarding their energy conservation behaviors, motivations for conserving energy, and relevant normative and nonnormative beliefs.
Method Participants. The survey data reported here were part of a larger survey of energy conservation beliefs, motivations, and actions among Californians. Data in that larger survey were collected quarterly for a 3-year period. The data reported below are from random-digit-dialing interviews with 810 participants obtained between October 2003 and January 2004, when the addition of certain survey items allowed for a test of the hypotheses of the present study.
Materials. Survey items were designed to measure self-reported efforts to conserve energy, perceived rea- sons for conservation, beliefs about the broad benefits of energy conservation, descriptive normative beliefs regarding energy conservation, and demographics. Self-reported intention to conserve was measured by the question “How often do you try to conserve energy?” ( never = 1, sometimes = 2, frequently = 3, almost always = 4). Along with this item, we also included a question about the perceived extent to which various factors, including the descriptive norm, motivated participants to conserve. To better understand the naive psychology of conserving energy, we selected three commonly used arguments that are employed in public appeals to con- serve energy (see Goldstein & Cialdini, 2007): saving money, environmental protection, and social responsibility. We also included an item that asked participants about the importance of social norms in their decisions to conserve energy. The questions read, “In deciding to conserve energy, how important is it to you... ” (a) that using less energy saves money, (b) that it protects the environment, (c) that it benefits society, and (d) that a lot of other people are trying to conserve energy. Responses were made on a 4-point scale ( not at all important = 1, somewhat important = 2, very important = 3, extremely important = 4). In addition to measuring the reported reasons for conserving energy, we also asked about their beliefs related to energy conservation. That is, we asked about their broad beliefs regarding energy conservation, not whether these beliefs motivated them to act. This would allow us to look at the relationship between beliefs and intention and to provide an initial assessment of the accuracy of participants’ naive beliefs. The questions were as follows: (a) How much do you think conserving energy will benefit society? (b) How much do you think conserving energy will protect the natural environment? (c) How much money do you think you can save by conserving energy in your home? and (d) How often do you think your neighbors try to conserve energy? Responses were on a 4-point scale ranging from 1 ( not at all ) to 4 ( extremely ).
Descriptive normative beliefs were measured with three items: “How often do you think your neighbors try to conserve energy” ( never = 1, sometimes = 2, frequently = 3, almost always = 4), “How often do you think residents of your city try to conserve energy?” and “How often do you think Californians try to conserve energy?” The items were averaged to create a scale score; Cronbach’s alpha for the three normative belief items (neighbors, city residents, Californians) was .79 ( M = 2.58, SD = .63). Demographic items included gender, age, ethnicity, education, income, and household size. One potential limitation in our study is the use of single- item measures. However, single-item measures may be less problematic than is often thought. For example, previous research on job satisfaction has shown that single-item measures are often as accurate as multi-item scales (Wanous, Reichers, & Hudy, 1997).
Procedure. Survey data were collected with the collaboration of the Social and Behavioral Research Institute (SBRI) at California State University (CSU), San Marcos. Data were collected using Computer Assisted Telephone Interviewing (CATI) software and were stratified by region of the state (Northern, Bay Area, Central, Los Angeles, Southern). The interviews lasted an average of 13 min. The response rate was 40% and the cooperation rate was 48% (see Council of American Survey Research Organizations, 1982, for description of these measures). Surveys were conducted in either English (88%) or Spanish (12%).
Results and Discussion
Before analyzing the data, we examined the representativeness of the sample by comparing the age, gender, ethnicity, and income distributions of the sample with the 2000 California Census. Of the demographic data we collected, there were no substantial deviations from California census data.
Naive psychology of energy conservation. The survey contained a series of items about the perceived reasons for energy conservation. The items read, “In deciding to conserve energy, how important is it to you.. .” These items reflect a respondent’s perceptions, or causal beliefs, about why they engaged in conservation activities. As shown in Table 1, the most highly rated reason for conserving energy was environmental protection ( M = 3.41, SD = .75), followed by benefits to society ( M = 3.17, SD = .77), saving money ( M = 3.07, SD = .76), and other people are doing it ( M = 2.93, SD = .83). A one-way, within-subjects ANOVA revealed significant differences across the four reasons for conservation, F (3, 2400) = 87.17, p < .001; all means were significantly different from each other ( p < .01). These results, which suggest that people are motivated to conserve energy out of a concern for the environment or future generations, is consistent with research showing that people tend to generate causal theories that are self- serving (Kunda, 1987). That is, people see themselves as conserving because it “saves the environment” or “ensures a happy future for children,” but they are less likely to believe that the behavior of others would have an influence on their own conservation behaviors. Regardless of their accuracy, naive explanations of behavior can have an influence on self-reports (Malle, 1999). As follow-up analyses to the stated reasons for conservation, we examined the relationship between conservation efforts and beliefs about saving energy: saving energy saves money, benefits future generations, protects the environment, and other people (i.e., my neighbors) are saving energy. Correlation coefficients showing the strength of the relationship between each of these beliefs and reported levels of conservation are shown in Table 1. As shown, the strongest predictor of energy conservation was the belief that other people are doing it ( r = .45, p < .01), despite the fact that it was rated as the least important motivating factor. The other beliefs about energy conservation were only weakly correlated with behavior: saving money ( r = .03), environmental protection ( r = .06), and benefiting future generations ( r = .23, p < .01).
TABLE 2: Final Regression Weights, t Values, and p Values for All Predictors of Self-Reported Intention to Conserve
Predictor Beta (standardized)
t Value p
Demographics Age .21 5.71. Gender –.01 –0.15. Income –.06 –1.52. Education .01 0.26. Language of survey –.09 –2.03.
Naive explanations for conservation
Environmental protection .17 3.56. Benefit to society –.00 –0.05. Saving money .15 3.66. Other people are doing it .04 1.01.
Descriptive normative beliefs .26 6.89.
The addition of naive explanations for conservation behavior on the second step increased the explained variance to 15%, with saving money and environmental protection as significant predictors. The addition of descriptive normative beliefs to the prediction equation increased the explained variance to 21%. The significant predictors in the final equation were age, with older participants reporting more conservation than younger ones; language of the survey, with English-speaking respondents conserving more than Spanish-speaking respondents; saving money; environ- mental protection; and descriptive normative beliefs. Table 2 shows the final regression weights for all of the predictor variables included in the regression analysis. By controlling for the naive explanations of conservation behavior, the degree to which descriptive normative beliefs are predictive of behavior in the final step represents an influence on behavior that is not recognized by the respondent. Consistent with the kinds of appeals used in popular discourse, our participants reported that several factors influenced their conservation behavior, such as the desire to save money, save the environment, and benefit society in general. Of interest, they reported that their normative beliefs had the least impact on their overall conservation behavior relative to all other motivations. Yet, despite the perception that other people’s behavior was least influential on their decision to conserve, beliefs of how often their neighbors tried to conserve showed a strong correlation with respondents’ own reported conservation efforts.
STUDY 2
Our initial study allowed us to acquire a rich set of data. Unlike in similar research of this nature, our large and wide-ranging sample consisted of people who differed on many important demographic dimensions, including socioeconomic status, age, ethnicity, and gender. Not only did this representative sample enable us to examine the impact of social norms across these different demographic variables, it also gave us more confidence in the reliability and generalizability of our findings. Of course, the correlational nature of the study did not allow us to establish causal relationships. To further examine the perceived influence of normative information, we conducted a second study using an experimental design. This time, we wanted to use normative information to change behavior and then to examine perceptions of the degree to which the normative information was motivational. Once again, our participants consisted of a sample of California residents, except that this time we provided household members with a specific appeal urging them to conserve energy in the home. Three of the appeals used a nonnormative message based on one of the three reasons for saving energy that our Study 1 participants identified as most influential: to protect the environment, to benefit society, or to save money. A separate appeal simply communicated a descriptive norm indicating that the majority of the recipient’s neighbors conserved energy—the same type of normative information that respondents in our first study considered the least motivational but showed the strongest relationship with their reported efforts to conserve. The effects of these four appeals were compared to a control condition that included an information-only appeal. Besides manipulating the type of information that participants received, Study 2 improved on our initial study in another important way: The dependent variable was participants’ actual energy use in the home as indicated by their electricity meter readings before and after the intervention. Having access to participants’ meters not only reduced the possible effects of self- presentation and memory bias but also provided a direct measure of the behavior of interest. With regard to participants’ actual energy consumption, we hypothesized that the descriptive norm condition would be superior to all of the other conditions at motivating energy conservation. That is, households in the descriptive norm condition would show the lowest level of energy consumption following the intervention. We had a separate set of predictions related to participants’ awareness of the influence of the messages on their energy conservation behavior. Although we expected the descriptive norm condition to be most efficacious, we also expected, based on the results of Study 1, that participants would rate the normative message as least influential. We also predicted that consistent with the naive explanations provided in Study 1, participants would rate the social responsibility and environ- mental messages as most motivating but that this verbal report of motivation would not correspond to actual conservation. In summary, we aimed to accomplish three goals in Study 2: extend the research on normative social influence, assess participants’ ability to detect the influence of normative information, and test the accuracy of naive psychology-based explanations of energy conservation.
Methods
Participants. Our field study began with a starting sample of 981 households in the middle-class
neighbor- hoods of San Marcos, California.^1 Of those 981 house- holds, 509 participated in a postintervention interview (52%). Included in the present study are 371 house- holds from the sample of interviewed households (73%) that reported seeing and reading the doorhangers that were distributed during the intervention. There are several possibilities for why people did not report seeing the doorhangers. First, respondents may have seen the doorhangers but forgot about them at the time of the interview. Second, the person who responded to our interview may not have been the same person within the household who discovered the doorhangers. A chi- square analysis showed that there were no significant differences in how many people saw and read the doorhangers across the five conditions,
energy con- servation for each condition. At the end of the interview, respondents were offered a free compact fluorescent light bulb and were asked to sign an authorization form granting permission to access their household energy bills from the local utility company. If no contact was made after three attempts, a mail survey with a cover letter describing the experiment was left in a self-addressed, stamped envelope at the respondent’s home along with a light bulb.
Meter readings. In addition to collecting self-report data on energy conservation during the interviews and requesting access to household energy bills, the researchers read the electricity meters for households with accessible meters. Electricity meters were recorded four times during the study. The first meter reading was taken the week prior to the intervention, the second meter reading was taken on the same day that the first doorhanger was distributed, the third meter reading was taken the same day the fourth doorhanger was distributed, and the fourth meter reading was taken 1 month following the delivery of the last doorhanger. By comparing each reading to a subsequent one, we were able to calculate an average daily kilowatt use figure for the baseline, shorter-term (1 month from baseline), and longer-term (2 months from baseline) periods. In Study 1, we learned that naive beliefs about energy conservation supported causal explanations that appealed to concern for future generations, environmental protection, and saving money. Assessment of the actual energy consumption of households in these different conditions provided a direct test of the accuracy of peoples’ naive causal explanations, allowing us to determine whether it was true, as people reported, that an appeal to social responsibility or environmental protection would be most effective at promoting energy conservation.
Results and Discussion Detecting normative social influence. Participants were asked “How much did the information on these doorhangers motivate you to conserve energy?” with responses ranging from 1 ( not at all ) to 4 ( extremely ). Participants in the descriptive norm condition reported that the messages were the least motivational ( M = 1.76, SE = .10). Pairwise comparisons showed that these scores were significantly lower than for participants in the environmental condition ( M = 2.07, SE = .08, p < .05) and social responsibility condition ( M = 1.99, SE = .09, p < .08) but not significantly different from the self-interest condition ( M = 1.86, SE = .08) or the information-only condition ( M = 1.94, SE = .11). This pattern is similar to that found in the survey data reported in Study 1, wherein environmental reasons and social responsibility were identified as the two reasons that people believed were most influential to conserve energy.
Meter readings. To assess the reliability of our meter readings, two research assistants were assigned to read the same meter on multiple occasions (4% of all readings). The independent readings correlated at r = .999. In addition, our measure of electricity use correlated with data provided by the local utility company at r = .964 during the short term (June) and at r = .992 long term (July). The correlations between our measures and SDG&E were for those households that granted permission to access their bill ( N = 187). This high correlation with data provided by the utility company supports the validity of our meter readings measure. Meter data were converted to kilowatt hours used per day. Use during the short term correlated at r = .61 with long-term use.
TABLE 3: Short-Term and Long-Term Energy Consumption Adjusted for Baseline Energy Consumption
Energy Consumption in Average Daily Kilowatt Hours (kWh) Short Term Long Term
Condition M SE M SE
Environmental protection 16.89 .39 16.89.
Social responsibility 17.52 .41 17.52. Self-interest 17.45 .40 17.45.
Social norm 16.10 .44 16.10. Information control 17.36 .45 17.36.
Our second step was to establish that the stimulus of interest, in this case the normative information, had an effect on behavior. Of the 371 households that we inter- viewed, 271 produced usable meter data.^2 A 5 (condition) X 2 (meter readings available: yes or no) ANOVA showed that there were no differences between those included in the meter reading analysis and those who were not on the measure of how much the doorhangers motivated them to conserve energy, F (1, 369) = .62, ns. The meter data provided the dependent variable in a series of analyses designed to test the impact of our persuasive messages. To account for differences in baseline electricity use across groups, we conducted a one-way ANCOVA with baseline use as a covariate and the short-term meter data as the dependent variable. The baseline covariate was significant, F (1, 265) = 1025.33, p < .001. In support of one of our major hypotheses, a planned contrast revealed that participants in the descriptive norm condition ( Madj = 12.97, SE = .44, N = 46) used significantly less energy short term than did participants in the combined other conditions ( Madj = 14.17, SE = .20), F (1, 268) = 5.99, p < .05. Table 3 provides the means and standard errors for each of the five conditions, adjusted for baseline energy consumption. The pattern of means for long-term use was similar, with the descriptive group consuming the least amount of electricity, after controlling for baseline use. Again, we conducted a one-way ANCOVA with baseline energy consumption as the covariate and long-term energy consumption as the dependent variable. The baseline covariate was significant, F (1, 264) = 309.16, p < .001. As shown in Table 3, the pattern of means for long-term energy consumption was similar to short- term use, with the descriptive norm condition ( Madj = consuming the least amount of energy compared to the other conditions. However, the planned contrast of descriptive norm versus the combined mean of the other conditions was not significant, F (1, 268) = 1.42, p = .24. The nonsignificant contrast for the long-term energy data suggests that the impact of the descriptive norm message had begun to erode in the 1 month following the intervention. The increase in energy consumption from short term to long term reflects seasonal changes in the weather, independent of our intervention. Overall, these results are inconsistent with the naive psychology-based explanations of conservation endorsed in Study 1. In summary, despite the private nature of conserving energy in the home, normative social influence had a direct impact on residents’ conservation behavior. Meter readings showed that a descriptive normative message—a message merely containing information about the conservation behavior of the majority of one’s neighbors— spurred people to conserve more energy than did the control message or any of the three other messages that contained appeals that are traditionally accorded motivational power. That is, the three messages using arguments based on motivations that were rated as being the most influential in Study 1 fared worse at spurring conservation behavior than just providing people with normative information about their neighbors’ conservation efforts. Furthermore, the experimental nature of Study 2 con- firmed that the positive relationship between descriptive norms and behavior in Study 1 was
term “your community” in the descriptive norm message may have made this message more personally relevant, and therefore more likely to be centrally processed, compared to the messages in the other conditions (Petty & Cacioppo, 1986). Although we cannot rule out this possibility entirely, it seems unlikely. First, research on majority influence has shown that presenting consensus information reduces the degree to which messages are processed systematically (Erb, Bohner, Schmälzle, & Rank, 1998). Second, if the messages appeared more personally relevant, then we might also expect to see that more people attended to the descriptive norm messages, compared to the other conditions, which we did not. Third, results similar to ours were found in a comparable study on recycling that used common language across conditions (Schultz, 1999). Future research is needed to better understand the basic question of how normative messages are processed in a real-world setting as well as the specific question of whether part of the effectiveness of normative messages can be attributed to an increase in personal relevance. In addition to their important theoretical contributions, these results have other practical implications as well. It is common practice for program designers to conduct focus groups in the process of designing their behavior change programs. The problem is that asking people what they think would influence them may not provide good data on which to base solutions. In fact, our results suggest that people hold incorrect beliefs about what motivates them to conserve and may not be able to predict which strategies will be the most effective. Taken together, the results from the current studies show that normative information is a powerful but underdetected form of social influence. As in previous studies (e.g., Schultz, 1999), we showed that normative information was a highly effective way to motivate a change in behavior. Also in line with previous research on nonconscious influences on behavior (e.g., Bowers, 1984; Nisbett & Wilson, 1977), we found that people were unable to identify the true cause of their behavior. Although participants in the environment and social responsibility conditions were most likely to say that they had been influenced by the persuasive message they received, their energy consumption did not differ from the control group, whereas the descriptive norm condition did differ. These results suggest that people relied on their a priori theories of what should motivate some- one to conserve energy. In conclusion, although people may not believe that the behavior of others should motivate them to conserve energy, their behavior was powerfully influenced by it nonetheless.
APPENDIX
SAMPLE INTERVENTION MESSAGES USED IN STUDY 2
Descriptive Norm : Join Your Neighbors in Conserving Energy. Summer is here and most people in your community are finding ways to conserve energy at home. How are San Marcos residents like you conserving this summer? By using fans instead of air conditioning! Why? In a recent survey of households in your community, researchers at Cal State San Marcos found that 77% of San Marcos residents often use fans instead of air con- ditioning to keep cool in the summer. Using fans instead of air conditioning—Your Community’s Popular Choice! Self-Interest : Save Money by Conserving Energy. Summer is here and the time is right for saving money on your home energy bill. How can you save money this summer? By using fans instead of air conditioning! Why? According to researchers at Cal State San Marcos, you could save up to $54 per month by using fans instead of air conditioning to keep cool in the summer. Environmental Protection : Protect the Environment by Conserving Energy. Summer is here and the time is right for reducing greenhouse gases. How can you protect the environ- ment this summer? By using fans instead of air conditioning! Why? According to researchers at Cal State San Marcos, you can prevent the release of up to 262 lbs of greenhouse gases per month by using fans instead of air conditioning to keep cool this summer! Using fans instead of air conditioning—The Environmental Choice.
Social Responsibility : Do Your Part to Conserve Energy for Future Generations. Summer is here and we need to work together to conserve energy. How can you conserve energy for future generations? By using fans instead of air conditioning! Why? According to researchers at Cal State San Marcos, you can reduce your monthly demand for electricity by 29% using fans instead of air conditioning to keep cool this summer! Using fans instead of air conditioning—The Socially Responsible Choice. Information Only : Energy Conservation. Summer is here and the time is right to conserve energy. How can you conserve energy this summer? By using fans instead of air conditioning!
NOTES
- Households were selected using 1999 and 2000 census data to represent high, medium, and low ethnic diversity areas with a median household income ranging from $50,000 to $72,000. High diversity was defined as an area that was less than 50% White, medium diver- sity was defined as an area that was between 50.1% to 74.9% White, and low diversity was defined as an area where more than 75% of the population was White.
- Meter readings were only taken from households that had a publicly visible electricity meter. Electricity meters that were not visi- ble upon natural egress to the home, or that could not be accessed because of the presence of a gate or a dog, were not read.
REFERENCES
Aarts, H., & Dijksterhuis, A. (2003). The silence of the library: Environment, situational norms, and social behavior. Journal of Personality and Social Psychology , 84 , 18-28. Aebischer, V., Hewstone, M., & Henderson, M. (1984). Minority influence and musical preference: Innovation by conversion not coercion. European Journal of Social Psychology , 14 , 23-33. Alvaro, E. M., & Crano, W. D. (1996). Cognitive responses to minor- ity- or majority-based communications: Factors that underlie minor- ity influence. British Journal of Social Psychology , 35 , 105-121. Asch, S. E. (1956). Studies of independence and conformity: A minor- ity of one against a unanimous majority. Psychological Monographs , 70 , 1-70. Bargh, J. A. (1992). Does subliminality matter to social psychology? Awareness of the stimulus versus awareness of its influence. In R. F. Bornstein & T. S. Pittman (Eds.), Perception without aware- ness: Cognitive, clinical, and social perspectives (pp. 236-255). New York: Guilford. Bargh, J. A. (1999, January 29). The most powerful manipulative messages are hiding in plain sight. Chronicle of Higher Education , p. B6. Bargh, J. A. (2006). Agenda 2006: What have we been priming all these years? On the development, mechanisms, and ecology of non- conscious social behavior. European Journal of Social Psychology , 36 , 147-168. Bargh, J. A., Chen, M., & Burrows, L. (1996). Automaticity of social behavior: Direct effects of trait construct and stereotype priming on action. Journal of Personality and Social Psychology , 71 , 230-244. Berkowitz, L. (1972). Social norms, feelings, and other factors affect- ing helping and altruism. Advances in Experimental Social Psychology , 6 , 63-108. Bowers, K. S. (1984). On being unconsciously influences and informed. In K. S. Bowers & D. Meichenbaum (Eds.), The uncon- scious reconsidered (pp. 227-273). New York: John Wiley. Chartrand, T. L., & Bargh, J. A. (1999). The chameleon effect: The perception-behavior link and social interaction. Journal of Personality and Social Psychology , 76 , 893-910.
Youth and Adolescence , 31 , 267-277. Milgram, S., Bickman, L., & Berkowitz, L. (1969). Note on the draw- ing power of crowds of different size. Journal of Personality and Social Psychology , 13 , 79-82. Nisbett, R. E., & Bellows, N. (1977). Verbal reports about causal influences on social judgments: Private access versus public theo- ries. Journal of Personality and Social Psychology , 35 , 613-
Nisbett, R. E., & Wilson, T. D. (1977). The halo effect: Evidence for unconscious alteration of judgments. Journal of Personality and Social Psychology , 35 , 250-256. Parks, C. D., Sanna, L. J., & Berel, S. R. (2001). Actions of similar others as inducements to cooperate in social dilemmas. Personality and Social Psychology Bulletin , 27 , 345-354. Perkins, H. W. (2003). The emergence and evolution of the social norms approach to substance abuse prevention. In H. W. Perkins (Ed.), The social norms approach to preventing school and college age substance abuse: A handbook for educators, counselors, and clinicians (pp. 3-18). San Francisco: Jossey-Bass. Petty, R. E., & Cacioppo, J. T. (1986). Communication and persua- sion: Central and peripheral routes to attitude change. New York: Springer-Verlag. Pronin, E., Molouki, S., & Berger, J. (2007). Alone in a crowd of sheep: Asymmetric perceptions of conformity and their roots in an intro- spection illusion. Journal of Personality and Social Psychology , 92 , 585-595. Schultz, P. W. (1999). Changing behavior with normative feedback interventions: A field experiment on curbside recycling. Basic and Applied Social Psychology , 21 , 25-36. Schultz, P. W. (2002). Knowledge, information, and household recy- cling: Examining the knowledge-deficit model of behavior change. In T. Dietz & P. Stern (Eds.), Education, information, and volun- tary measures in environmental protection (pp. 67-82). Washington, DC: National Academy of Sciences. Schultz, P. W., Nolan, J. M., Cialdini, R. B., Goldstein, N. J., & Griskevicius, V. (2007). The constructive, destructive, and reconstructive power of social norms. Psychological Science , 18 , 429-433. Schultz, P. W., & Zelezny, L. (2003). Reframing environmental mes- sages to be congruent with American values. Human Ecology Review , 10 , 126-136. Shariff, A. F., & Norenzayan, A. (2007). God is watching you: Priming God concepts increases prosocial behavior in an anony- mous economic game. Psychological Science , 18 , 803-809. Sherif, M. (1936). The psychology of social norms. New York: Harper. Sherif, M. (1937). An experimental approach to the study of attitudes. Sociometry , 1 , 90-98. Terry, D. J., & Hogg, M. A. (2001). Attitudes, behavior, and social context: The role of norms and group membership in social influ- ence processes. In J. P. Forgas & K. D. Williams (Eds.), Social influence: Direct and indirect processes (pp. 253-270). Philadelphia: Psychology Press. Turner, J. C. (1991). Social influence. Pacific Grove, CA: Brooks/ Cole. Von Borgstede, B., Dahlstrand, U., & Biel, A. (1999). From ought to is: Moral norms in large- scale social dilemmas. Goteborg Psychological Reports , 29 , 1-19. Wanous, J. P., Reichers, A. E., & Hudy, M. J. (1997). Overall job sat- isfaction: How good are single-item measures? Journal of Applied Psychology , 82 , 247-252. Wilson, T. D., & Brekke, N. (1994). Mental contamination and mental correction: Unwanted influences on judgments and evaluations. Journal of Personality and Social Psychology , 116 , 117-142.
Received June 18, 2007 Revision accepted December 12, 2007
