The Transsituational Influence of Social Norms

Jurnal:

Tolong rujuk ke sumber asli

The Transsituational Influence of Social Norms

Raymond R. Reno, Robert B. Cialdini, and Carl A. Kallgren

Journal of Personality and Social Psychology, 1993, Vol.64, No. 1,104-112

Copyright 1993 by the American Psychological Association Inc

0022-3514/93/S3.00

Three studies examined the behavioral implications of a conceptual distinction between 2 types of social norms: descriptive norms, which specify what is typically done in a given setting, and injunctive norms, which specify what is typically approved in society. Using the social norm against littering, injunctive norm salience procedures were more robust in their behavioral impact across situations than were descriptive norm salience procedures. Focusing Ss on the injunctive norm suppressed littering regardless of whether the environment was clean or littered (Study 1) and regardless of whether the environment in which Ss could litter was the same as or different from that in which the norm was evoked (Studies 2 and 3). The impact of focusing Ss on the descriptive norm was much less general. Conceptual implications for a focus theory of normative conduct are discussed along with practical implications for increasing socially desirable behavior.

         

     Despite social norms having a history of long and extensive use within the discipline, there is no current consensus within social psychology about the explanatory and predictive value of social norms. Whereas social norms have been trumpeted by some as crucial to a full understanding of human social behavior (e.g., Berkowitz, 1972; Fishbein & Ajzen, 1975; McKirnan, 1980; Pepitone, 1976; Sherif, 1936; Staub, 1972; Triandis, 1977), other social psychologists have suggested that the concept may be vague, overly general, and ill-suited to empirical study (e.g., Darley & Latane, 1970; Krebs, 1970; Krebs & Miller, 1985; Marini, 1984). Social psychologists are not alone in their disillusions concerning social norms. A parallel controversy has developed in academic sociology, where enthnomethodological and constructionist critics have faulted the dominant norma-tive paradigm of that discipline (Garfinkel, 1967; Mehan & Wood, 1975).

Seeking to clarify the role of social norms, Cialdini, Reno, and Kallgren (1990; Cialdini, Kallgren, & Reno, 1991) distinguished two types. The first of these, descriptive norms, specify what most people do in a particular situation, and they motivate action by informing people of what is generally seen as effective or adaptive behavior there. Injunctive norms, on the other hand, specify what people approve and disapprove within the culture and motivate action by promising social sanctions for normative or counternormative conduct. Which of these two types of norms is focal (i.e., salient) at a particular time will direct an individual's immediate behavior, according to Cialdini et al.

     In a series of studies that examined the problem of litter in public places, Cialdini et al. (1990) increased focus on the descriptive norm by having a confederate litter in front of subjects.

Contrary to simple imitation or modeling predictions, but in line with Cialdini et al.'s norm-focus theory, subjects littered less after witnessing the confederate litter if the descriptive norm of the situation was to not litter (i.e., if the setting was clean). The fact that the littering rate of subjects who witnessed the confederate litter in a clean environment was lower than the control rate of littering was seen as support for the norm-focus model. Simply increasing salience that most others had not littered in the environment decreased littering, even when this was accomplished by exposing subjects to a counternormative act.

Even though these studies demonstrated that focusing subjects on a descriptive norm can result in prosocial behavior, the successful use of descriptive norm salience techniques appears limited to settings in which antisocial behavior is not much of a problem. That is, Cialdini et al. (1990) found that littering was reduced only when the descriptive norm was made salient in a clean environment. When the descriptive norm was made salient in littered environments (once again, by focusing subjects on what others had done there) littering increased above control levels. Consequently, although the demonstration of these effects was important for validating the norm-focus theory, the practical utility of such descriptive norm manipulations for reducing antisocial behavior seems limited.

Fortunately, however, there are conceptual reasons for believing in the greater utility of injunctive norm activation. An active, injunctive norm focus should have a pair of socially desirable effects in settings characterized by socially undesirable action. First, it should cause a shift of attention away from evidence that antisocial behavior constitutes the descriptive norm for the setting. Second, it should lead individuals to attend to a motivational construct—social approval and disapproval—that directs behavior in a socially desirable direction regardless of what others may have done in the setting. The potentially beneficial impact of such a shift in attention focus is suggested by the results of research indicating that moving a person's attention to a specific source of information or motivation will change the person's responses in ways that are congruent with the features of the now more prominent source (Agostinelli, Sherman, Fazio, & Hearst, 1986; Kallgren & Wood, 1986; Millar & Tesser, 1989; Storms, 1973). To test OUT reasoning in this regard, we chose the norm regarding littering in public places, as it afforded an instance of a clearly felt and widely held injunctive social norm (Berkowitz, 1972; Heberlein, 1971) in our culture. Moreover, attention to the descriptive and injunctive aspects of the littering norm could be fairly easily manipulated.

Study 1

     Study 1

To make the injunctive norm salient in the present series of studies, a confederate picked up a piece of litter in full view of our subjects. This manipulation was designed to communicate the confederate's objection to others' littering; as such, it incorporated the fundamental motivational component of injunctive norms: social approval and disapproval. The descriptive norm was made salient in the same manner used by Cialdini et al. (1990) in prior studies: The confederate littered in the environment, thereby drawing attention to its littered or unlittered condition and to the fundamental motivational component of descriptive norms: the responses of most others in the situation.

In addition, a control treatment was included in which the confederate merely walked by subjects to provide a constant amount of social contact. As in previous studies, the environment was prepared to be either clean or littered.

We had two main predictions: First, we expected that making the descriptive norm salient would be a successful litter reduction tactic only when the environment was clean. But, we expected, second, that making the injunctive norm salient would successfully reduce littering in both littered and unlittered environments. Finding this latter pattern would support our belief that an injunctive norm focus is effective in reducing undesirable behavior even in settings where a negative descriptive norm exists. Such a result would illustrate the advantage of Cialdini et al.'s (1990) distinction between injunctive and descriptive norms—the ability to make clear differential predictions based on injunctive and descriptive norms. Furthermore, because the application of the norm-focus model to problem behaviors relies on the effectiveness of the injunctive norm focus in such negative settings, a clear demonstration of this effect was viewed as important.

Method

Subjects

Subjects were 173 (75 female and 98 male) visitors to a municipal library who were returning to their cars in an adjacent parking lot during the daylight hours. The age of subjects ranged from the midteens to the early 70s, with 90% between 20 and 50 years of age.

Procedure

Norm salience. As subjects neared the parking lot, they encountered an experimental confederate of college age walking toward them.

In approximately one third of the instances, the confederate carried a bag from a fast food restaurant, which he or she dropped into the environment approximately 4.5 m (5 yd) before passing the subjects; this constituted the throw-down (descriptive norm salient) condition in which subjects' attention was drawn either to the clean or littered state of the environment. In another third of the instances, the confederate was not carrying anything but rather picked up the fast food bag approximately 4.5 m in front of the subject; this constituted the pickup (injunctive norm salient) condition in which subjects' attention was drawn to social disapproval of littering. In the final third of the instances, a confederate merely walked by the subject so as to provide a similar degree of social contact; this constituted the walk-by (control) condition. A second confederate judged whether a subject had noticed the littering incident and consequently had deflected his or her attention at least momentarily to the parking lot floor, or whether subjects had noticed the confederate pick up the bag. This procedure allowed us to eliminate a priori the data of subjects who had not experienced the experimental manipulation.

Existing state of the environment. For some of the subjects, the parking lot had been heavily littered by the experimenters with an assortment of handbills, candy wrappers, cigarette butts, and paper cups; this constituted the littered environment (existing prolittering norm) condition. For the remaining subjects, the area had been cleaned of all litter; this constituted the clean environment (existing antilittering norm) condition.

Measurement of littering. On arriving at their cars, subjects encountered a large handbill that was tucked under the driver's side wind-shield wiper so as to partially obscure vision from the driver's seat. The handbill carried a stenciled message that read, "This is automotive safety week. Please drive carefully." A similar handbill had been placed on all other cars in the area as well. From a hidden vantage point, an experimenter noted whether the driver littered the handbill. Littering was defined as depositing the handbill in the environment outside of the vehicle. Because there were no trash receptacles in the area, all subjects who failed to litter did so by taking and retaining the handbill inside their vehicles before driving away. In addition to recording the subjects' behavior, this experimenter also recorded the gender of the drivers because gender differences in littering have been noted in past research (see Geller, Winett, & Everett, 1982, for a review).

Analyses

The analyses in this and subsequent studies were conducted using the SPSS-X log-linear program, wherein tests for effects within dichotomous data are examined through the nesting of hierarchical models.

This technique allows the testing of individual parameters by comparing the differences in the likelihood ratio chi-square of a pair of nested models. The difference likelihood ratio is reported as a chi-square.

Results and Discussion

Although gender was not a variable of theoretical interest to us, we sought to account for and control for its influence when possible. Therefore, in addition to our theoretical hypotheses, we explored the data for gender differences. To do this, a log-linear model that included gender, norm-salience, and environmental conditions was tested first. In this model, there was a marginal tendency for women (13.33%) to litter less than men (28.57%), x² (l, N= 173) = 3.34, p < .068. Gender did not inter act with the experimental variables, however. Thus, it was not included in further analyses.

Our overall expectation was that an injunctive norm focus would reduce littering regardless of the state of the environment but that a descriptive norm focus would suppress littering only when the environment was clean. Looked at another way, we expected that (a) in the clean environment, both injunctive and descriptive norm-salience procedures would result in littering rates below control levels; but (b) in the littered environment, only the injunctive norm-salience procedure would do so.

To test our predictions, we conducted a series of planned comparisons on the littering rates depicted in Figure 1. First, we tested our prediction within the clean environment condition with a contrast showing that the combined littering rate of the injunctive (7%) and the descriptive (11 %) norm-salience conditions was significantly lower than that of the relevant control  (37%) condition, (l, N= 85) = 9.54, p < .01. Next, we tested x2

our prediction within the littered environment condition with a pair of orthogonal contrasts demonstrating that, although the littering rates of the descriptive norm-salience condition (30%) and its relevant control condition (38%) did not differ (x² < 1), their combination was significantly greater than the rate of the injunctive norm-salience condition (4%), x ² (l, iV= 90) = 11.40, p < .01. Finally, we examined our overall expectation in an omnibus contrast that combined our two predictions by comparing the three conditions we hypothesized to show higher littering rates (the throw-down and littered environment condition plus the two walk-by control conditions) against the remaining three conditions, which were hypothesized to show reduced rates of littering; the difference, 35.2% versus 7.4%, was again clearly significant, x ² 0, N = 173) = 21.21, p< .01.

The pattern of results supported the predictions made from the norm focus theory. When an antilittering norm (injunctive or descriptive) was made salient, littering rates were decreased from control conditions or from conditions in which a prolittering descriptive norm was made salient (the throw-down and littered environment condition). The alert reader may notice that subjects who witnessed the confederate throw down a bag in the littered environment did not litter at rates greater than the control conditions, as had been demonstrated by Cialdini et al. (1990). Although at first glance, this apparent failure of replication may be disconcerting, it may actually point to a limitation on the generality of the descriptive norm's influence. In the previous research (Cialdini et al., 1990), after witnessing a confederate dispose of a flyer in a littered environment, subjects were more likely than control subjects to also litter a flyer. The confederate's behavior was quite informative about what the subjects should do with the flyers. This was less the case in the present study. The confederate threw down a paper bag, not a flyer that was similar to that which the subjects would have the opportunity to litter. Thus the confederate's behavior in the present study was less informative to the subjects concerning the descriptive norm. As a result, the littering rates of subjects in this condition were similar to those of the control subjects.

Effects related to injunctive norms were quite provocative from an applied standpoint. Salient injunctive norms resulted in decreases in littering regardless of the environment's status.

This suggests that injunctive norm activation procedures can be valuable tools in the amelioration of socially undesirable behavior; even in settings where these undesirable behaviors predominate (e.g., fully littered environments, roads where most drivers speed, or political precincts with low voter turnout). These findings provide a conceptual replication of the findings of Cialdini et al. (1990) and a clearer demonstration of the greater trans-situational influence of injunctive norms relative to descriptive norms.

However, the advantage of an injunctive norm focus may not be limited to settings characterized by socially undesirable behaviors. That advantage may apply, as well, to the likelihood of normative conduct in settings that are different from the one in which the relevant norm was evoked. Because the injunctive social norms of a culture typically apply to most settings of that culture, an activated injunctive norm should continue to direct behavior—provided that it remains salient—in a novel, second setting. Study 2 offers a test of this possibility.

Study 2

If we are correct that an injunctive norm focus should transcend situational boundaries, we should expect that such a focus would lead to reduced littering even in environments other than the one in which the focus was evoked. To test this hypothesis, we conducted a second experiment in which the injunctive norms focus and control manipulations from Study 1 were implemented in two separate environments. The first environment was a parking lot similar to that used in Study 1. The second environment was a pathway along a grassy area that was separated from the parking lot. Both environments were clean.

Consequently the descriptive norm was held constant across both settings. We hypothesized that when subjects were focused on the injunctive antilittering norm, they would litter less than control subjects, regardless of whether the injunctive focus had been created in an environment that was the same or different from the one where they would have the opportunity to litter.

Method

Subjects

Subjects were 137 (75 female and 62 male) patrons at a municipal library who were returning to their cars in an adjacent parking lot during the daylight hours. Subjects' ages were estimated to range between 16 and 70, with 92% estimated to be between 20 and 60 years of age.

Procedure

Norm salience. Between leaving the library and entering the parking lot, subjects encountered an experimental confederate of college age. As this confederate walked toward the subject, the confederate either picked up a crumpled fast food bag that was lying on the ground (injunctive norm salient condition) or simply walked by subjects (control condition). A second confederate judged whether subjects noticed the confederate picking up the bag. As in Study 1, these judgments allowed us to make an a priori determination of which subjects experienced the experimental manipulation and, consequently, which data should be retained for analysis.

Same versus different environment. Subjects encountered the confederate in one of two settings. For some subjects, the procedure was the same as in Study 1: Subjects encountered the confederate as they were entering and the confederate was leaving the parking lot where subjects would have a chance to litter (same-environment condition). The remainder of the subjects encountered the confederate along a path within a grassy, landscaped section of the property (different environment condition) that was separated from the parking lot by a brick wall. Because we were interested only in the impact of the injunctive norm, both the parking lot and the pathway had been cleaned of visible litter to hold the descriptive norm constant.

Measurement of littering. Both the procedures for providing subjects an opportunity to litter and the procedures for recording their actions were identical to those of Study 1.

Results and Discussion

As in Study 1, we first estimated a model that included gender to determine whether differences in littering rates were due to subject gender. The difference in the littering rates of women (12%) and men (19.35%) was not statistically significant, nor did gender interact with norm salience or environmental conditions. Thus, it was not included in further analyses.

Examination of Figure 2, which depicts the percentage of litterers in each condition of Study 2, reveals the predicted pattern. Only one significant effect emerged—the main effect for norm salience; subjects for whom the injunctive norm against littering was made salient littered less than control-condition subjects (6.7% vs. 22%), x ² (l, N= 137) = 5.48, p < .01. Neither the main effect for type of environment nor, more importantly, its interaction with norm salience approached significance (x² s < 1). Thus, witnessing another pick up litter reduced littering tendencies in our subjects regardless of whether their observation of this act of social disapproval occurred in the same setting as or in a different setting from their opportunity to litter.

These results, when coupled with those from Study 1, are encouraging with regard to the influence of salient injunctive norms on behavior. First, an injunctive norm focus appears to be effective in those situations where behavior change is most needed (e.g., littered environments). Second, it does not appear necessary to create this injunctive norm focus within the particular environment in which abatement of negative behavior is desired. It is this latter finding that may be of most practical significance to those interested in modifying behavior by creating an injunctive norm focus. That increasing injunctive norm salience in one environment reduced littering in a second environment enhances the potential practical utility of such an approach.

This finding also has theoretical import for norm-focus theory. It highlights one of the major conceptual differences between a descriptive and an injunctive norm focus. That distinction is explored further in Study 3.

Study 3

Examination of Cialdini et al.'s (1990) conceptualization of the differences between descriptive and injunctive norms shows why the utility of these two types of social norms may vary across a variety of situations. Descriptive norms appear to be more situation-specific in the information they provide.

That is, descriptive norms communicate what others have felt to be appropriate behavior in that setting. Consequently, the influence of such norms may weaken precipitously as individuals move out of the environment in which the norms were made salient. This may be particularly true if the environments differ along a dimension related to the behavior in question.

Injunctive norms, on the other hand, involve perceptions of what is approved conduct within the culture in general, which should have substantial cross-situational relevance. That is, descriptive norms are designed to tell us what makes for adaptive or effective behavior, which can be influenced and changed by many situation-based factors. But, injunctive norms are designed to tell us what others have been socialized to approve and disapprove in the culture, which is likely to change relatively little from situation to situation; consequently, their influence should transcend environments.

This point, however, has not been confidently established by the data presented thus far. In Study 1, as the environmental conditions changed so did the descriptive norm; thus, it was not possible to evaluate the trans-situational influence of descriptive norms. Similarly, because Study 2 did not include a descriptive norm-focus manipulation, the issue could not be investigated. Therefore, a third study was conducted to test this component of the norm-focus model, as well as to replicate the results of Study 2 with regard to the influence of the activated injunctive norm.

We conducted Study 3 in the same environments that were used in Study 2. If, as we have suggested, descriptive norms primarily communicate what is typically done within a specific environment, then littering rates should drop principally where an antilittering descriptive norm is made salient and not in relatively different environments. In contrast, as was found in Study 2, making the more general injunctive norm against littering salient in one setting should have the effect of reducing litter in similar and different settings. In summary, then, we expected decreased littering (compared with control conditions) whenever the injunctive antilittering norm was activated, but we expected decreased littering when the descriptive antilittering norm was activated only if the opportunity to litter occurred in the same setting as the norm activation.

Method

Subjects

Subjects were 131 (62 female and 69 male) visitors to a municipal library who were returning to their cars in an adjacent parking lot during the daylight hours. Subjects' estimated ages ranged between 16 and 70 years, with 90% ranging between 20 and 50 years.

Procedure

Norm salience. Between exiting the library and entering the parking lot, subjects encountered an experimental confederate of college age.

In all conditions, the library grounds and parking lot had been cleaned of visible litter. The confederate's behavior was scripted to make salient either the descriptive or injunctive antilittering norm or to provide no normative information (control). The descriptive norm-salience manipulation differed from that used in Study 1. In the present study, subjects in the descriptive norm condition witnessed a confederate dispose of a fast food bag he or she was carrying by throwing it into a nearly full trash receptacle approximately 4.5 m (5 yd) before passing the subject. Although this receptacle was present in all conditions, we reasoned that the confederate's act of throwing litter into it would bring a subject's attention to evidence of what most people did with regard to litter in that particular (clean) setting—that is, that they refrained from disposing of trash improperly there. To increase injunctive norm salience, we relied on the manipulation we had used successfully in Studies 1 and 2: Subjects saw a confederate who, after passing the earlier described trash receptacle, encountered a discarded fast food bag, picked it up, and carried it away in the direction of the library.

Same-different environment. Subjects encountered the confederate in one or the other of the two settings used in Study 2. Due to the method by which the descriptive norm was to be activated in the present study, it was necessary to place a trash receptacle in each of the two settings. These receptacles were present in both settings for subjects in all conditions of the experiment. However, because the placement of the receptacles in both settings made it inconvenient, no subjects used a receptacle to dispose of their fliers.

Measurement of littering. Both the procedures for providing subjects an opportunity to litter and for recording their actions were identical to those of the prior two studies.

Results

As in the previous two studies, we screened for any gender differences in littering. Although women tended to litter less than men, 16.13% versus 27.54%, this difference was not statistically significant, x ² 0, N = 131) = 2.465, p > .10, nor did it interact with our primary independent variables. Thus, gender was not examined in our subsequent analyses.

The hypothesized effects were tested through a series of a priori contrasts. The main hypothesis was that subjects' littering would be reduced from control levels wherever (parking lot or pathway) the injunctive norm against it was made salient or when the descriptive norm against littering was made salient in the parking lot (same-environment condition). However, no such reduction was predicted when the descriptive norm against littering was made salient in an environment different from the parking lot (i.e., the pathway). Looked at in another way, we expected that (a) in the same-environment condition, both the injunctive and descriptive norm-salience procedures would suppress littering relative to control levels, but (b) in the different-environment condition, only the injunctive norm-salience-lience procedure would do so.

Figure 3 depicts the littering rates for each of the experimental conditions and shows a pattern of data that is congruent with our hypotheses. First, we tested our prediction within the same environment condition with a simple contrast demonstrating that the combination of the littering rates in the injunctive (13%) and the descriptive (17%) norm-salience conditions was lower than that of the relevant control (32%) condition, x ²  (l, #= 9i) = 3.55, p < .06. Next, we tested our prediction within the different environment with a pair of orthogonal contrasts demonstrating that, although the littering rates of the descriptive norm-salience condition (36%) and its relevant control condition (33%) did not differ from one another (x² < 1), their combination did differ significantly from the lower rate of the injunctive norm-focus condition (0%), x² (l, N=A0) = 7.62, p < .01. Finally, we tested the interaction form of our overall prediction with an omnibus contrast. The predicted interaction was assessed by combining the two injunctive norm-salience conditions with the descriptive norm-salience-same-environment condition and comparing them with the combined control conditions plus the descriptive norm-salience-different-environment condition. This contrast generated clear support for the overall prediction, x² (l, N = 131) = 8.14, p < .01.

Discussion

Interpretations of the Data

The results of the present study further support the practical advantage of focusing individuals on injunctive versus descriptive norms. An injunctive norm focus proved decidedly more robust in its impact across situations than a descriptive norm focus. Subjects who saw another refrain from littering were affected by that display only in that particular setting. On the other hand, subjects who witnessed evident social disapproval of another's littering were affected by that display in a rather different setting as well. According to the norm focus theory, this was the case because (a) subjects were focused by the re-specie displays either on the descriptive norm or on the injunctive norm, and (b) the effect of focusing on the injunctive norm is more likely to transcend situational boundaries, be-cause the injunctive norm orients individuals away from a concern about how others have behaved in a particular setting and toward a concern about what others approve or disapprove of across the culture.

A secondary benefit of the obtained pattern of results of Study 3 is that it provides evidence against an alternative explanation for the findings of the first two studies. That is, it might be argued that the conditions that produced the lowest littering rates in those earlier experiments were the conditions that made the confederate's litter-related behavior most salient to subjects. Take, for example, the procedures of Study 1. It is conceivable that a confederate who littered into a clean environment or a confederate who picked up litter in either a clean or a littered environment was a more distinctive litter-related stimulus than a confederate in the other three conditions of that experiment. Similarly, a confederate who picked up litter in Study 2 may well have become a more salient litter-related stimulus than a confederate who simply walked past subjects. Thus the outcomes of those two studies could be explained as due to subjects' negative reactions to the concept of littering whenever that concept was made prominent for them. However, such an account cannot readily explain the pattern of data in Study 3, where subjects who saw a confederate dispose of a bag in a parking lot (same-environment condition) littered less than those who saw the confederate dispose of a bag in a different environment. The confederate's behavior should have been no more salient or more positive in either environment. One might argue that the saliency of the manipulation may decay over time. This line of thought accounts for the pattern of data for the trash can (descriptive norm) condition but fails to predict the observed pattern of data in the pickup (injunctive norm) condition of Study 3. Consequently, it is more difficult for this alternative theory to explain the results from all three studies.

A closer look at Study 3 reveals the important role of social sanctions in distinguishing between the effectiveness of injunctive and descriptive norms. After all, subjects in both conditions saw another person perform a behavior in a particular setting from which they could have reasonably inferred the other's disapproval of littering. Yet, it was only when the confederate picked up another's litter that littering was generally re-diced. We think this was so because of the idea of social sanctions that is associated with injunctive norms. The behavior of the confederate who picked up another's litter was unambiguous as to its interpersonal message. It communicated that the confederate disapproved of and found littering by others (our subjects included) objectionable. In this way, subjects were focused on potential social sanctions, the central feature of in-juncture norms, which increased their awareness of the society-wide rule against littering. In contrast, subjects who saw litter thrown into a receptacle got a different message from the con-federate, that is, "I find littering objectionable within my own behavior," which did not remind them directly of social sanctions or, consequently, of injunctive norms. Instead, subjects may have remained focused on the descriptive norm of what another and similar others had done in that setting.

Although such interpretations fit well with the data, it should be recognized that they are not based on any direct evidence that the experimental procedures designed to focus subjects differentially on descriptive and injunctive norms had the in-tended effects. Because of the field character of the reported studies, it was not possible to administer manipulation checks or measures of mediating cognitive processes, as can be done in traditional laboratory research. A critic could fairly contend, then, that confident interpretation of the present three studies in terms of norm-focus theory requires another step. It remains to be demonstrated that the experimental operations we have used to focus subjects differentially on descriptive and injunctive norms are indeed effective in doing so.

Tests of Norm-Focus Inductions

To provide such a demonstration, we conducted a separate study on 70 undergraduate psychology course students at Arizona State University. Each student received a questionnaire describing the scene encountered by the subjects in each of the three experimental conditions of the present research. Thus, the students were asked to read and picture the following scenes (presented in a random order):

A college-age individual who is walking on the grounds of a local municipal library discards a paper bag by throwing it into a trash barrel, and then walks on [descriptive norm induction].

A college-age individual who is walking on the grounds of a local municipal library discards a paper bag by throwing it on the ground, and then walks on [descriptive norm induction].

A college-age individual who is walking on the grounds of a local municipal library picks up a discarded paper bag from the ground and throws it into a trash barrel, then walks on [injunctive norm induction].

The students were then asked to indicate whether, when they pictured the scene, it focused them more on (a) the extent to which other people do and do not litter [descriptive norm] or (b) the extent to which other people approve and disapprove of littering [injunctive norm].

The students' responses offered good support for the validity of our norm-focus procedures. Picturing an individual who discarded a paper bag by throwing it in a trash barrel led the great majority of students to focus predominantly on the descriptive norm, that is, "the extent to which other people do and do not litter," rather than on the injunctive norm (87.2% vs. 12.8%, z = 12.48, p < .01). The same was true when students pictured an individual who threw down a paper bag (70% vs. 30%, z = 8.37, p < .01). However, picturing an individual who picked up a paper bag led the great majority of students to focus predominantly on the injunctive norm, that is, "the extent to which other people approve and disapprove of littering," rather than on the descriptive norm (91.4% vs. 8.6%, z = 13.48, p < .01).

Analyzing these data further by McNamara’s chi-square test for dependent samples with a Bonferroni correction showed that the relative proportions produced by the injunctive norm-focus induction were clearly different from either of those of the two descriptive norm-focus inductions, McNemar x² s(l, N = 70) = 55.00 and 43.00, respectively, both ps < .001. The relative proportions of the two descriptive norm inductions were not different at conventional levels by this test when controlling for Type I error; they were marginally different, however, McNemar x² (l, N = 70) = 5.538, p < .019; statistical significance using the Bonferroni correction requires a = .05/3 = .016.

Thus, the results of the questionnaire study support well our assumptions concerning which norm, descriptive or injunctive, our experimental procedures made salient for the subjects in our field experiments. Consequently, our confidence is heightened that the norm-focus theory provides an apt interpretation of the data from the three field studies reported here as well as from the field studies reported by Cialdini et al. (1990), which used similar experimental procedures.

General Discussion

A set of predictions based on a norm-focus theory (Cialdini et al., 1990) was tested in a series of three studies. The data patterns from these three studies converge to allow four main conclusions that further clarify the distinction Cialdini et al. made between injunctive and descriptive norms. First, despite existing skepticism and criticism of normative explanations, the present data support the viability of social norms as powerful behavioral directives, consistent with Cialdini et al.'s findings. Second, at least two distinct types of social norms are effective in this regard: social norms of the descriptive kind, which guide one's behavior through the perception of how most others would or do behave; and social norms of the injunctive kind, which guide one's behavior through the perception of how most others would sanction one's conduct. Third, in contrast to descriptive norms, injunctive norms can increase prosocial action even in settings characterized by antisocial behavior. Finally, an injunctive norm focus enhances norm-congruent responding in environments similar to and different from those in which the focus occurred; descriptive social norms, on the other hand, seem only to influence behavior in the environments where they are made focal.

Although we argue for a norm-focus model, we do not discard the previous criticisms against normative explanations of behavior (e.g., Darley & Latane, 1970; Krebs, 1970). These criticisms underline the components from which the norm-focus model draws its power. According to our theory, one cannot think of norms in a general or vague sense. Rather one needs to specify clearly the type of social norm (injunctive or descriptive) at work and guarantee that it will be focal before one can feel confident in normative accounts of behavior change. In all three of our studies, for example, both descriptive and injunctive norms were present within each of the environments where control subjects were acting. Yet, it was not until a confederate's action made these norms focal that it was possible to appreciate the magnitude of their power to guide human conduct.

In addition to demonstrating the effects of norm focus, the data, particularly from Study 3, identified the main motivational component of injunctive social norms: social sanctions.

In this regard, it appears that the concept of approval and disapproval needs to be sharpened in its relation to social norms.

Witnessing another approve or disapprove of norm-related responding may only engage the full power of the relevant injunctive social norm when the observer is made to think that such approval or disapproval applies to his or her relevant conduct.

Expressing disapproval for counternormative behavior in one's own actions by visibly refraining from the action should not bring to bear on observers the full salutary impact of the injunctive social norm. Rather, this impact should flow from witnessing visible expression in others of disapproval of counternormative action. Thus, a key to the effective activation of injunctive social norms is a focus on the applicability of interpersonal sanctions to the behavior in question.

This is not to say that we think injunctive social norms function only when evaluating others are physically present to provide social sanctions. We concur with the developers (Cooley,1902; Mead, 1934) and modern proponents (e.g., Schlenker, 1980) of symbolic interaction theory that people often seek to satisfy the expectations of imagined audiences, one of which the generalized other—represents the generalized viewpoint of society. Thus, it is our view that once focused on a representative of society who approves or disapproves of another's behavior; an observer is likely to conform to the societal rules for that behavior even when alone, as long as the focus remains.

From a practical standpoint, these distinctions between injunctive and descriptive social norms should be of value, particularly to those interested in enhancing the likelihood of socially beneficial behavior through norm activation. Such individuals would be best advised under most circumstances to use procedures that activate injunctive social norms. Once activated, injunctive norms are likely to lead to beneficial social conduct across the greatest number of settings. Activating a descriptive social norm, on the other hand, is only likely to lead to socially desirable behavior in settings where most individuals already behave in a socially desirable manner.

As with most research domains, worthy additional questions remain to be answered. For instance, further research should examine the nature of stimuli that are likely to lead to a norm's salience. Likely candidates for investigation in this regard are certain factors related to the norm itself such as its cognitive accessibility, its recency and frequency of prior activation, and its degree of connectedness with other salient norms in the environment. In addition, Schwartz (1973, 1977; Schwartz & Howard, 1982) has argued persuasively that individuals possess personal norms, that is, self-based standards for conduct that flow from internalized values. The extent to which a heightened personal norm would affect the salience of injunctive or descriptive social norms is yet unexplored. The answers to these questions would further enhance our conceptual understanding of the influence of social norms, as well as sharpen our ability to use them in applied settings.

References

Agostinelli, G., Sherman, S. J., Fazio, R. H., & Hearst, E. S. (1986).

Detecting and identifying change: Additions versus deletions. Journal of Experimental Psychology: Human Perception and Performance 12, 445-454.

Berkowitz, L. (1972). Social norms, feelings, and other factors affectinghelping and altruism. In L. Berkowitz (Ed.), Advances in experimental social psychology (\o\. 6, pp. 63-108). San Diego, CA: Academic Press.

Cialdini, R. B., Kallgren, C. A., & Reno, R. R. (1991). A focus theory of normative conduct. Advances in Experimental Social Psychology, 24, 201-234.

Cialdini, R. B., Reno, R. R., & Kallgren, C. A. (1990). A focus theory of

normative conduct: Recycling the concept of norms to reduce littering in public places. Journal of Personality and Social Psychology, 58, 1015-1026.

Cooley, C. H. (1902). Human nature and the social order. New York:Scribners.

Darley, J. M., & Latane, B. (1970). Norms and normative behavior:Field studies of social interdependence. In J. Macaulay & L. Berkowitz (Eds.), Altruism and helping behavior (pp. 83-102). San Diego,CA: Academic Press.

Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavio Reading, MA: Addison-Wesley.

Garfinkel, H. (1967). Studies in elhnomethodology. Englewood Cliffs,NJ: Prentice-Hall.

Geller, E. S., Winett, S., & Everett, P. B. (1982). Preserving the environment. New "Vbrk: Pergamon Press.

Heberlein, T. A. (1971). Moral norms, threatened sanctions, and littering behavior. Dissertation Abstracts International, 32, 5906A. (University Microfilms No. 72-2639)

Kallgren, C. A., & Wood, W W (1986). Access to attitude-relevant information in memory as a determinant of attitude-behavior consistency. Journal of Experimental Social Psychology, 22, 328-338.

Krebs, D. L. (1970). Altruism: An examination of the concept and a review of the literature. Psychological Bulletin, 73, 258-302.

Krebs, D. L., & Miller, D. T. (1985). Altruism and aggression. In G. Lindzey & E. Aronson (Eds), The handbook of social psychology (3^rd ed, Vol. 2, pp. 1-71). New York: Random House.

Marini, M. M. (1984). Age and sequencing norms in the transition to adulthood. Social Forces, 63, 229-244.

McKirnan, D. J. (1980). The conceptualization of deviance: A conceptualization and initial test of a model of social norms. European Journal of Social Psychology, 10, 79-93.

Mead, G. H. (1934). Mind, self and society. Chicago: University of Chicago Press.

Mehan, H., & Wood, H. (1975). Reality of ethnomethodology. New York: Wiley.

Millar, M. G, & Tesser, A. (1989). The effects of affective-cognitive consistency and thought on the attitude-behavior relation. Journal of Experimental Social Psychology, 25,189-202.

Pepitone, A. (1976). Toward a normative and comparative biocultural social psychology. Journal of Personality and Social Psychology, 34, 641-653.

Schlenker, B. R. (1980). Impression management. Monterey, CA: Brooks-Cole.

Schwartz, S. H. (1973). Normative explanations of helping behavior: A . critique, proposal, and empirical test. Journal of Experimental Social Psychology, 9, 349-364.

Schwartz, S. H. (1977). Normative influences on altruism. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 10, pp. 221-279). San Diego, CA: Academic Press.

Schwartz, S. H., & Howard, J. A. (1982). Helping and cooperation: A self-based motivational model. In V Derlega & H. Grezlak (Eds.), Cooperation and helping behavior (pp. 83-98). San Diego, CA: Academic Press.

Sherif, M. (1936). The psychology of social norms. New York: Harper.

Staub, E. (1972). Instigation to goodness: The role of social norms and interpersonal influence. Journal of Social Issues, 28,131-150.

Storms, M. D. (1973). Videotape and the attribution process: Reversing actors' and observers' points of view. Journal of Personality and Social Psychology, 27, 165-175.

Triandis, H. C. (1977). Interpersonal behavior. Monterey, CA: Brooks/Cole.

Catatan:

Raymond R. Reno, Department of Psychology, University of Notre Dame; Robert B. Cialdini, Department of Psychology, Arizona State University; Carl A. Kallgren, Department of Psychology, Pennsylvania State University, Behrend College.

Correspondence concerning this article should be addressed to Raymond R. Reno, Department of Psychology, University of Notre Dame, Notre Dame, Indiana 46556.