Too dumb to breathe

An article in Oxford University Press just cited by Forbes:

Why are men less likely than women to embrace environmentally friendly products and behaviors? Whereas prior research attributes this gender gap in sustainable consumption to personality differences between the sexes, we propose that it may also partially stem from a prevalent association between green behavior and femininity, and a corresponding stereotype (held by both men and women) that green consumers are more feminine. Building on prior findings that men tend to be more concerned than women with gender-identity maintenance, we argue that this green-feminine stereotype may motivate men to avoid green behaviors in order to preserve a macho image. A series of seven studies provides evidence that the concepts of greenness and femininity are cognitively linked and shows that, accordingly, consumers who engage in green behaviors are stereotyped by others as more feminine and even perceive themselves as more feminine. Further, men’s willingness to engage in green behaviors can be influenced by threatening or affirming their masculinity, as well as by using masculine rather than conventional green branding. Together, these findings bridge literatures on identity and environmental sustainability and introduce the notion that due to the green-feminine stereotype, gender-identity maintenance can influence men’s likelihood of adopting green behaviors.

In accord with the notion that behavior is often guided by the need to maintain gender identity, particularly among male consumers, we posit that one plausible explanation for men’s avoidance of environmentally friendly behaviors is that a pervasive mental association exists between greenness and femininity, such that engaging in green behaviors could threaten men’s masculine identity. Thus we predict that due to the green-feminine association and its effect on social judgments and self-perception, men will be more likely than women to avoid green products and behaviors, particularly when their masculinity is threatened.

While the basic thesis of our research is that men may be motivated to eschew green choices in order to maintain their gender identity, there might be boundary conditions under which men may be willing to be associated with green products and behaviors in spite of their feminine connotation. For example, prior research suggests that men may be more likely to engage in an activity that is perceived as stereotypically feminine if their masculinity has been affirmed, but that they may be reluctant to do so when their masculinity has been threatened (Gal and Wilkie 2010). Building on this research, we predict that when their masculine identity is affirmed, men will feel less compelled to reassert their masculinity through nongreen behaviors and will therefore be more open to engaging in environmentally friendly behaviors. We likewise predict that when the association between greenness and femininity is weakened (e.g., through masculine branding), men will be more likely to engage in green (vs. nongreen) behaviors.

Across a series of seven studies, we test our hypotheses that both men and women mentally associate greenness with femininity, which leads them to stereotype green consumers as more feminine than nongreen consumers (studies 1, 2, and 3), and that as a result of this green-feminine stereotype, men’s willingness to engage in green behavior is sensitive to gender-related cues that threaten or affirm their gender identity or influence a brand’s gender associations (studies 4, 5, 6A, and 6B). Together, these studies aim to provide support for our prediction that due to the prevalent cognitive link between greenness and femininity, gender-identity maintenance contributes to men’s relatively low engagement in green behaviors.

The objective of study 1 was to test for an implicit cognitive association between the concepts of greenness and femininity. To do this, we used a Single Category Implicit Association Test (SC-IAT; Karpinski and Steinman 2006) to measure participants’ implicit attitudes toward the perceived gender-affiliation and greenness of products. The rationale underlying the SC-IAT is that when asked to categorize stimuli, participants will respond more quickly when paired categories match (vs. mismatch) their subjective mental representation. For example, if participants cognitively represent green products as feminine, response latencies should be shorter when the label “Female” is paired with the compatible label “Environmentally Friendly” rather than the incompatible label “Environmentally Unfriendly.” In a separate SC-IAT, we examined the extent to which greenness is associated with masculinity.

A total of 127 university students (52.0% male; mean age = 21.42) were randomly assigned to complete either the Feminine SC-IAT (designed to test different combinations of femininity and greenness) or the Masculine SC-IAT (designed to test different combinations of masculinity and greenness). Each participant completed four blocks of trials. In each trial, the target stimulus (either the name of a person or the photo of a product) was displayed in the center of the screen and category labels were displayed in the upper right and left corners of the screen. For our Feminine SC-IAT [Masculine SC-IAT], participants categorized names as Female [Male] and products as either Environmentally Friendly or Environmentally Unfriendly. The procedures for both the Feminine SC-IAT and the Masculine SC-IAT were identical except for the gender of the names used. See online appendix A for stimuli and additional details.

Using Greenwald, Nosek, and Banaji’s (2003) improved scoring algorithm, we created an IAT D-score for each participant. The D-score is an effect size estimate that reflects strength of association between two concepts and is created by dividing differences between the mean response latencies of compatible and incompatible blocks by the standard deviation of all latencies in the blocks. IAT D-scores can range from ?2 to +2, with the direction and size of the effect reflecting the strength of associations between the target concepts and attributes (Greenwald, McGhee, and Schwartz 1998).

For the Feminine SC-IAT, the mean IAT D-score was .23, which is significantly different from 0; t(57) = 4.25, p < .001, d = 1.13. Consistent with our prediction that greenness and femininity are cognitively associated, this positive D-score indicates that participants were quicker to categorize stimuli in compatible than incompatible trials. There was no difference in IAT D-score by participant gender; F(1, 57) = .11, p = .74, ?2p < .01, suggesting that both men and women cognitively associate the concepts of greenness and femininity.

For the Masculine SC-IAT, the mean IAT D-score was .03, which is not significantly different from 0; t(68) = .56, p = .58, d = .14, and there was no difference in this D-score by participant gender; F(1, 68) = .69, p = .41, ?2p = .01. The lack of a prevalent cognitive association between greenness and masculinity suggests that the concepts of femininity and masculinity may be independent constructs rather than polar ends of a single continuum.

Study 1 provided evidence consistent with our theorizing that a mental association exists, among both men and women, between the concepts of greenness and femininity. Our theory further suggests that this green-feminine association results in a prevalent stereotype (held by both genders) that green consumers are more feminine than nongreen consumers—a stereotype that may encourage men to avoid eco-friendly behaviors. Whereas prior research has shown that women are more likely to engage in green behaviors than men, it has not examined the possibility that a person’s green or nongreen behavior might impact others’ perceptions of his or her gender identity. Our next study examines whether consumers who engage in green behaviors are indeed judged by others to be more feminine.

In this study, we aimed to show that the association between greenness and femininity observed in study 1 affects social judgments, such that those who do (vs. do not) engage in green behavior are perceived as more feminine. It is important to document this stereotype because we theorize that men will be reluctant to engage in green behaviors if doing so would cause them to be perceived as feminine. Thus study 2 examined the prevalence of a stereotype, among both men and women, that green consumers are more feminine than nongreen consumers.

Participants were 194 students (45.9% male; mean age = 23.05) simultaneously recruited from two private universities to participate in an online survey. Participants were randomly assigned to one of four conditions in a 2 (target: male vs. female) × 2 (behavior: green vs. nongreen) between-participant design. Participants read the following scenario that included an image of groceries in a plastic bag (nongreen behavior) or an image of groceries in a reusable canvas bag (green behavior): “Imagine you are at your local grocery store and see a [man/woman] leaving the checkout lane, carrying [his/her] groceries in a [plastic bag/reusable canvas bag]. Please indicate the extent to which you feel each word below describes this [man/woman]” (see online appendix B for stimuli).

Participants then used a 5 point scale (1 = Not at all; 5 = Perfectly) to provide ratings for 11 traits that were presented in a random order to each participant. Two traits (eco-friendly and wasteful) were intended as a manipulation check to ensure that using the reusable canvas bag was in fact perceived as more green than using the plastic bag. The remaining nine traits were considered to be stereotypically masculine (masculine, macho, and aggressive), stereotypically feminine (feminine, gentle, and sensitive), or gender neutral (athletic, attractive, curious). Our selection of these nine traits and their expected classification was empirically based, leveraging both prior research on individuals’ perceptions of the gender affiliations of these traits (Hoffman and Borders 2001; Holt and Ellis 1998) and our own pretesting.

Following this task, participants were asked to guess what this study was designed to test. They then reported their age, gender, gender identity, interest in dating men, interest in dating women, and indicated whether they were currently single or in a relationship.

An analysis of responses to the hypothesis-guessing question showed that 21 participants explicitly linked gender or gender stereotypes to environmental behavior. However, the pattern of results does not change when these participants are excluded, so the analysis here includes the full set of respondents. Our between-participant manipulation of green versus nongreen behavior was successful; participants in the two green conditions rated the trait eco-friendly as more descriptive of the target (M = 4.54, SD = .68, N = 94) than participants in the two nongreen conditions (M = 1.43, SD = .88, N = 100; t(192) = 27.42, p < .001). Conversely, participants in the two green conditions rated the trait wasteful as less descriptive of the target (M = 1.07, SD = .30) than participants in the two nongreen conditions (M = 2.44, SD = 1.09; t(192) = 11.68, p < .001).

To test our prediction that the green target would be perceived as more feminine than the nongreen target, we first formed a femininity index by averaging the three feminine traits (? = .80) and a masculinity index by averaging the three masculine traits (? = .76). The creation of these indices was supported by principal component exploratory factor analysis with oblique rotation (Promax), which showed that these six traits loaded on the two hypothesized factors, together explaining 71.2% of the total variance. Moreover, a confirmatory factor analysis with the same six traits on two factors yielded a chi-square residual of 413 (df = 15, p < .001), a goodness-of-fit index of .95, a root square mean residual of .07, and a comparative fit index of .95. These statistics indicate a reasonable fit to the data, and the ? coefficient of .25 provides evidence of discriminant validity. Each loading estimate was highly significant (p < .001), and the .80 and .76 reliability of the feminine and masculine factors, respectively, was acceptable (table A in online appendix H shows the results of the factor analysis). Together, these results provide support for the notion that ratings on these six traits adequately capture perceptions of a target’s masculinity versus femininity.

As expected, the femininity index differed significantly based on the target’s environmental behavior; F(1, 190) = 44.00, p < .001, ?2p = .19, such that the green target (M = 2.64, SD = .90, N = 94) was perceived as more feminine than the nongreen target (M = 1.82, SD = .86, N = 100); table B in online appendix H lists the mean ratings for each trait. Not surprisingly, a main effect of the target’s gender was also observed; F(1, 190) = 8.39, p = .004, ?2p = .04, such that the female target (M = 2.40, SD = 1.04, N = 95) was perceived as more feminine than the male target (M = 2.04, SD = .88, N = 99). The interaction was not significant; F(1, 190) = 2.42, p = .12, ?2p = .01, indicating that both male and female targets were judged as more feminine when they engaged in green (vs. nongreen) behavior. Next, we examined whether the results were influenced by participants’ gender, age, gender identity, interest in dating men, interest in dating women, or relationship status. Across all these demographics, participants seemed to hold the same stereotypes about the heightened femininity of consumers who engage in green behaviors; when these variables were included as covariates, the target’s gender and environmental behavior remained significant predictors of femininity (p’s < .01) but none of the covariates was significant (p’s > .37).

We then examined perceptions of the targets’ masculinity. The masculinity index also differed significantly across conditions, but this effect was driven entirely by the target’s gender, such that the male target (M = 1.67, SD = .74, N = 99) was perceived as more masculine than the female target (M = 1.38, SD = .63, N = 95); F(1, 190) = 8.52, p < .01, ?2p = .04. There was neither a main effect of environmental behavior; F(1, 190) = .28, p = .60, nor a significant interaction; F(1, 190) = .70; p = .40. The different patterns observed for masculinity and femininity in this study and in study 1 are consistent with prior research that conceptualizes these constructs as independent rather than mutually exclusive (Hoffman and Borders 2001).

Our claim is that because green behavior is cognitively associated with femininity, targets who engage in green behavior will be perceived not only as more green, but also as more feminine. To test this claim, we examined whether the effect of green versus nongreen behavior on a target’s perceived femininity was mediated by perceptions of the target’s eco-friendliness. For this analysis, we used the femininity index as the dependent variable and a dummy variable to indicate the type of environmental behavior (1 = Green; 0 = Nongreen) as the independent variable. The moderator was an eco-friendliness index we created by averaging the wasteful (reverse-coded) and eco-friendly traits (? = .71). Then, using Hayes’s (2013, model 4) PROCESS macro, we found that the indirect effect of environmental behavior through eco-friendliness was positive (B = .59, SE = .26) and statistically different from zero (95% confidence interval [CI], .12–1.12), thus providing evidence of the suggested mediation.

Consistent with our predictions, results of study 2 showed that consumers who engaged in green behavior were perceived by both male and female participants as more feminine than consumers who engaged in nongreen behavior. The fact that none of the covariates we measured in study 2 moderated the effect is consistent with the results of study 1 in suggesting that the green-feminine association is prevalent across both genders. We argue that this association may discourage men from engaging in green behaviors, particularly if they are motivated to maintain a macho image and wish to avoid being stereotyped as feminine.

The above is silly, but we’re the too dumb to breathe, since we just paid $13 to read this stuff. “Your order (12674109) has been accepted, please print this page for your records.”

2 Responses to “Too dumb to breathe”

  1. feeblemind Says:

    Related via Maggie’s Farm:

    Toxic Masculinity could be the reason for climate change:

  2. Bob Risko Says:

    It’s amazing the silliness that some people get up to once they’ve acquired a vocabulary that includes mean, standard deviation, degrees of freedom, null hypothesis, alternative hypothesis, etc. Statistically ‘proving’ that gender bias is behind opposition to “green politics?” My God, what next?

Leave a Reply