Congruent or conflicted? The impact of injunctive and descriptive norms on environmental intentions

Abstract

Two experiments examine the interplay of injunctive and descriptive norms on intentions to engage in pro-environmental behavior. In Experiment 1, Australian participants were exposed to supportive or unsupportive group descriptive and injunctive norms about energy conservation. Results revealed that a conflict between the group-level injunctive and descriptive norm was associated with weaker behavioral intentions: The beneficial effects of a supportive injunctive norm were undermined when presented with an unsupportive descriptive norm. Experiment 2 replicated this effect in both a Western (UK) and non-Western (China) context, and found that the extent to which norms were aligned or not determined intentions even after controlling for attitudes, perceptions of control, and interpersonal-level injunctive and descriptive norms. These experiments demonstrate that conflict between injunctive and descriptive norms leads to weaker intentions to engage in pro-environmental behavior, highlighting the need to consider the interplay between injunctive and descriptive norms to understand how norms influence behavioral intentions.

Introduction

Human behavior is a critical contributor to environmental degradation and climate change (IPCC, 2007) and, as a result, changing human behavior is also critical in addressing climate change and environmental issues. But what strategies do change agents have at their disposal? One strategy is to use structural interventions, such as economic incentives and disincentives, to promote pro-environmental behavior. However, economic strategies can have negative consequences. For example, taxes and fines may lead people to put a price on negative behavior (e.g., Gneezy & Rustichini, 2000), reasoning that they can continue to engage in the behavior as long as they have the capacity to pay. On the other hand, incentives may promote positive behavior in the short term, but it is less clear whether incentives can sustain long-term lifestyle changes (e.g., Kane, Johnson, Town, & Butler, 2004), particularly if the incentive is removed. Moreover, one can question whether economic factors or other popular strategies (e.g., a call to protect the welfare of future societies) are even the most influential determinants of human behavior. Indeed, Nolan, Schultz, Cialdini, Goldstein, and Griskevicius (2008) found that while people stated they engaged in energy conservation behaviors because of economic or environmental concern, these factors were only weak predictors of actual behavior. Rather, the strongest predictor of energy conservation intentions was the very factor that people stated was least influential: social norms. The present research investigates how two different types of social norms – the descriptive norm and the injunctive norm – interact to influence intentions to engage in pro-environmental behavior.

Social norms (i.e., group-based standards or rules regarding appropriate attitudes and behaviors) play a crucial role in shaping how individuals interpret and act in their social worlds. However, as noted by Cialdini, Kallgren, and Reno (1991) and Cialdini, Reno, and Kallgren (1990), there are two types of norms: injunctive norms and descriptive norms. Injunctive norms reflect perceptions of what others approve or disapprove of, and motivate action because of the social rewards and punishments associated with engaging, or not engaging, in the behavior. Descriptive norms reflect perceptions of whether other people actually engage in the normative behavior themselves, and motivate action by informing people about what is likely to be effective or adaptive behavior in a particular context. Injunctive and descriptive norms represent separate sources of motivation and should have separate influences on behavior. Indeed, there is evidence from both correlational and experimental research that descriptive and injunctive norms have independent and distinct effects on intentions and behavior (Cialdini et al., 1990; Kallgren, Reno, & Cialdini, 2000; Manning, 2009; Reno, Cialdini, & Kallgren, 1993; Rivis & Sheeran, 2003).

In the domain of pro-environmental behavior, the importance of injunctive and descriptive norms has been demonstrated repeatedly. Indeed, the initial studies on injunctive and descriptive norms examined littering behavior (Cialdini et al., 1990). However, injunctive and descriptive norms have been found to influence energy conservation (Gockeritz et al., 2010; Nolan et al., 2008), environmental theft (Cialdini et al., 2006), environmental conservation in hotels (Goldstein, Cialdini, & Griskevicius, 2008; Schultz, Khazian, & Zaleski, 2008), and recycling (Schultz, 1999).

In addition, research has examined whether injunctive or descriptive norms are more influential. Some researchers have argued that injunctive norms are more powerful than descriptive norms (e.g., Cialdini et al., 2006; Reno et al., 1993; cf. Manning, 2009). This is because injunctive norms can motivate behavior across a range of contexts while a descriptive norm tends to motivate behavior in the immediate context in which others' behavior occurs or can be observed. In addition, the fact that descriptive norms are often associated with a boomerang effect (Cialdini, 2003), whereby exposure to descriptive norms increases, rather than decreases, problem behavior, might suggest that injunctive norms are a more fruitful tool for behavior change (see e.g., Blanton, Koblitz, & McCaul, 2008). Indeed, Schultz, Nolan, Cialdini, Goldstein, and Griskevicius (2007) found that the potentially destructive effects of descriptive norm information could be ameliorated by including supportive injunctive norm information. This research highlights not only the power of the injunctive norm but also the impact of the combination of injunctive and descriptive norms on behavior.

In an effort to shed light on the way in which injunctive and descriptive norms influence behavior, recent research has begun to consider the nature of the interactions between these two types of norms. Some researchers have focused on the congruence of the two types of norms, whereby one type of norm can imply the other (e.g., Blanton et al., 2008). For example, if a group of people are observed to disapprove of smoking, one might infer that it is unlikely that the group members will smoke. However, it is not always the case that the descriptive and injunctive norms align. For example, a widespread descriptive norm of energy wastage may co-occur with a group's injunctive norm in favor of energy conservation. Indeed, it might be argued that by definition, attempts to change behavior are often prompted by a misalignment of descriptive and injunctive norms (e.g., “as a group, you're currently doing X, but you should be doing Y”). Thus, to understand normative influence on behavior change, it is critical to understand how conflicting descriptive and injunctive norms might interact to influence behavior.

At the most basic level, researchers have argued that a behavior is more likely to occur if people believe that the behavior is commonly done by others and if they believe it is approved by others (i.e., when the descriptive and injunctive norms align). Indeed, correlational research has demonstrated that behavior rates are highest when both descriptive and injunctive norms supportive the behavior (Gockeritz et al., 2010; Lee, Geisner, Lewis, Neighbors, & Larimer, 2007; Rimal & Real, 2003; but see Rimal & Real, 2005). Similarly, Schultz et al. (2008) found in a field experiment that a combined supportive descriptive and injunctive norm message was more effective than only a supportive descriptive or injunctive norm message.

To date, however, there has been little experimental work on the interaction between descriptive and injunctive norms that has orthogonally manipulated descriptive and injunctive norms (but see Smith & Louis, 2008). As a result, the question of what happens when injunctive or descriptive norms are in conflict has been somewhat overlooked (cf. Keizer, Lindenberg, & Steg, 2008). The little research that does exist suggests that, faced with the information that there is a discrepancy between what people approve of and what people actually do, the impact of either normative message is likely to be weakened. This may be because unaligned norms lead people to perceive that it is acceptable not to translate attitudes into action. However, research to date has focused only on political behavior (Smith & Louis, 2008) or has not explicitly compared aligned norms to unaligned norms (Keizer et al., 2008). The aim of the present research is to investigate further the interplay between descriptive and injunctive norms and to test the effects of exposure to conflicting descriptive and injunctive norms on behavioral intentions. In addition, given that all the research on injunctive and descriptive norms has been conducted in Western (i.e., individualist) cultures, an additional aim was to examine the impact of aligned versus unaligned norms in both individualist and collectivist (i.e., China) cultures.

We predict that when both descriptive and injunctive norms emanate from the same source (i.e., when they reflect the approval and behavior of the same group), unaligned norms may undermine conformity to either norm. This hypothesis may seem intuitive, but in fact the effect has not been demonstrated empirically with orthogonal norm manipulations. Moreover, existing theoretical models and empirical studies either imply the primacy of injunctive norms (Cialdini et al., 2006), or aggregate injunctive and descriptive norms into one summary normative judgment rather than considering possible interactions (e.g., Terry & Hogg, 1996). In the experiments below, we investigate the interactions of descriptive and injunctive norms on intentions. We predict that the combination of a supportive injunctive norm with a supportive descriptive norm should produce the highest level of behavioral intentions. However, we further predict that a supportive injunctive norm will only motivate behavior if others are also expected to act (i.e., when the descriptive norm is also supportive). If the descriptive norm is unsupportive of the behavior, injunctive norms may be unrelated to behavior, or even produce lower action (a backlash hypothesis, such that this condition is associated with lower intentions than even unaligned unsupportive norms).