Abstract
Several known risk factors for nonsuicidal self-injury (NSSI), such as negative emotionality and deficits in emotion skills, are also associated with masculinity. Researchers and clinicians suggest that masculine norms around emotional control and self-reliance may make men more likely to engage in self-harm. Masculinity has also been implicated as a potential risk factor for suicide and other self-damaging behaviors. However, the association between masculinity and NSSI has yet to be explored. In the current study, a sample of 912 emerging adults from two universities in the Northeastern United States completed a web-based questionnaire assessing adherence to masculine norms, engagement in NSSI, and known risk factors for NSSI (demographics and number of self-injurers known). Stronger adherence to masculine norms predicted chronic NSSI (five or more episodes throughout the life span) above and beyond other known risk factors. Adherence to masculine norms was related to methods of NSSI. Clinical implications are discussed, including discussions of masculine norms in treatment settings. Future research should examine what specific masculine norms are most closely linked to NSSI and other self-damaging behaviors.
Nonsuicidal self-injury (NSSI) is the deliberate and self-inflicted damage to one’s body tissue (e.g., cutting, burning) with the absence of suicidal intent (Nock, 2010). Current evidence indicates that NSSI is alarmingly high among adolescents and emerging adults with prevalence rates ranging from 37% to 68% in clinical populations and 22% in nonclinical populations (Muehlenkamp & Gutierrez, 2007; Whitlock, Eckenrode, & Silverman, 2006). Research has focused on identifying factors that are associated with and which increase the likelihood of engaging in NSSI. These factors include gender, ethnicity, sexual orientation, and the number of self-injurers known (likely related to social contagion). These risk factors–specifically, identifying as female, Caucasian, and/or bisexual–increase the likelihood of engaging in NSSI (Gollust, Eisenberg, & Golberstein, 2008; Gould, Wallenstein, & Davidson, 1989; Hawton, Rodham, Evans, & Weatherall, 2002; Whitlock et al., 2011). That NSSI is influenced by the behaviors of others through the process of social contagion, potentially speaking to the perceived normativeness of these behaviors, has been well studied and documented (Heilbron & Prinstein, 2008; Whitlock, Muehlenkamp, & Eckenrode, 2008; Whitlock, Powers, & Eckenrode, 2006). Transient risk factors such as depressed mood, negative affect, negative emotionality, deficits in emotion skills, and self-derogation are also well-known risk factors for NSSI (Klonsky & Muehlenkamp, 2007). Though not previously explored, several of these factors correspond with what is known about masculine norms.
Defined as the dominant set of norms and behaviors that allow men to maintain power and privilege (Connell, 1987; Connell & Messerschmidt, 2005; Mahalik, Good, & Englar-Carlson, 2003), masculine norms broadly include dominance, violence, anti-femininity, emotional control, and self-reliance. Adherence to these norms has been associated with many forms of negative emotionality, including depression (Addis, 2008; Good & Wood, 1995), aggression and hostility (Cohn & Zeichner, 2006), and poorer overall psychological well-being (Alfred, Hammer, & Good, 2014). Theoretical work on masculine norms suggests that these norms are learned through policing and fear-based learning which can, in turn, bring about increased risk-taking behaviors (Addis, Mansfield, & Syzdek, 2010). Researchers and clinicians have argued that additional aspects of masculine gender socialization may predispose some men to engage in suicidal and self-damaging behaviors, including the desire for emotional control and self-reliance (Green & Jakupcak, 2015).
At the same time, masculinity is associated with low- and high-lethality self-damaging behaviors, ranging from decreased help-seeking for psychological and physical problems and increased risk-taking behaviors (e.g., overconsumption of energy drinks or alcohol, unprotected sex, reckless driving, and broad health risk behaviors) to attempted suicide (Addis & Mahalik, 2003; Houle, Mishara, & Chagnon, 2008; Iwamoto, Cheng, Lee, Takamatsu, & Gordon, 2011; Mast, Sieverding, Esslen, Graber, & Jäncke, 2008; K. E. Miller, 2008). Despite this body of evidence linking masculinity to increases in low-lethality risk-taking behavior as well as with high-lethality suicide attempts, the association between masculinity and NSSI has not been examined.
Extant research on gender differences in self-harm suggests that men are more likely to engage in some methods more often than are women, and that many of these behaviors appear to be consistent with traditionally masculine presentations. Studies examining sex differences in self-harm report that men are more likely to burn, self-hit, bang one’s head against objects, punch walls or other objects, and engage in generally risky behaviors (e.g., driving dangerously; Claes, Vandereycken, & Vertommen, 2007; Sornberger, Heath, Toste, & McLouth, 2012; Whitlock et al., 2011). Though these methods appear to be motivated by impulsivity and aggression, rather than self-directed violence performed for the regulation of negative affect (Klonsky, 2007), these methods largely meet criteria for NSSI. One study reported adolescent males were more likely to self-harm in response to boredom and thinking self-harming would be fun (Laye-Gindhu & Schonert-Reichl, 2005).
Though adherence to masculine norms is associated with increased self-damaging and, potentially, suicidal behaviors, an overarching theory for this phenomenon has yet to be clearly articulated in the literature. Examining literature from both the study of men and masculinity as well as self-harm, several lines of research converge to offer a potential explanation.
Nock and Prinstein’s (2004, 2005) four-factor model (FFM) provides a framework for better understanding the links between adherence to masculine norms and self-harm. The model identifies four functions of self-harm: (1) automatic positive reinforcement, (2) automatic negative reinforcement, (3) social positive reinforcement, and (4) social negative reinforcement. The term automatic refers to reinforcement such as physical sensations and emotions that naturally arise when one engages in NSSI. Prototypical NSSI is thought to be maintained by automatic negative reinforcement (e.g., cutting to alleviate negative feelings). It appears that relative to women, men’s self-harm is more likely done to experience a rush or high and more often occurs in the presence of others (Klonsky & Glenn, 2009; Whitlock et al., 2011). This notion, that many men’s self-harm behaviors are more likely to be maintained by social and positive rather than automatic and negative reinforcement, may explain differences in the presentation of these behaviors.
Adherence to masculine norms seems to be associated not only with various forms of negative emotionality (Good & Wood, 1995; Magovcevic & Addis, 2008), but also the ways in which men experience, express, and respond to these emotions. Adherence to masculine norms is linked to features of emotional dysregulation in men, including alexithymia (Cusack, Deane, Wilson, & Ciarrochi, 2006; Jakupcak, Osborne, Michael, Cook, & McFall, 2006; Levant et al., 2003; Levant et al., 2006) as well as fear and purposeful avoidance or suppression of vulnerable emotions (e.g., sadness and depression, anxiety or fear; Jakupcak, Salters, Gratz, & Roemer, 2003; Wide, Mok, McKenna, & Ogrodniczuk, 2011), and avoidance of negative affect (Green & Addis, 2012). This literature suggests that some men, as a function of masculine norms, may struggle to experience and express a wide array of negative emotions.
Communicative properties of self-harm may be of particular importance given some men’s difficulties communicating thoughts and feelings. Expanding on the FFM, Nock (2008) describes social functions of NSSI that are in line with masculine norms and which may be particularly relevant for men; self-harming to communicate strength and fitness is said to ward off potential aggressors and may be useful in attempting to establish or maintain dominance in male peer groups. Self-harm may also be a means by which to indicate affiliation with a group, with Nock (2008) providing the example of fraternity brothers branding their skin and becoming “blood brothers” through other self-damaging behaviors. This form of bonding through violent or damaging behaviors is thought to be associated with adherence to masculine norms (Addis, 2011) and may serve to communicate dominance or act as leverage for group inclusion. Indeed, there is evidence to suggest that men are more likely to engage in self-harm to join a social group (Laye-Gindhu & Schonert-Reichl, 2005) or to demonstrate strength (Claes et al., 2007).
These works, in combination, suggest that men, as a function of adherence to masculine norms, may struggle to experience and express emotions through prototypical means and may use self-harm as a means by which to communicate a variety of messages. Findings on self-harm in men appear consistent with what is known about the effects that masculine gender socialization can have on some men’s responses to psychological distress: acting out in response to distress or emotional confusion with anger, aggression, violence, and other externalizing behaviors. These same factors (e.g., emotional confusion, anger, aggression, violence, externalizing) may also explain why some men’s self-harm behaviors present differently than women’s. However, whether and to what degree masculinity is associated with engagement in these methods of self-harm has yet to be empirically tested.
The present study sought to determine whether adherence to masculine norms was associated with NSSI above and beyond the robust predictors of NSSI (identifying as Caucasian, as bisexual, as female, the number of self-injurers known, negative affect, and depressive symptoms). The study examined whether adherence to masculine norms was associated with engagement in particular self-harm methods. The study used a sample of both women and men given that identifying as female is a known risk factors of NSSI. Participants were administered measures of masculinity to both men and women since many of the factors thought to be associated with masculinity and NSSI (negative emotionality, deficits in emotion skills, self-derogation, emotional control, and self-reliance) are not unique to men. Many studies have assessed masculinity in women (Mahalik et al., 2003; Owen, 2010) and there is evidence to suggest that findings with regard to masculinity in men can be generalized, with caution, to women (Smiler, 2006). Given the links between masculinity, self-injurious, and suicidal behaviors, adherence to masculine gender norms was expected to be associated with chronic NSSI. An association between adherence to masculine norms and engagement in NSSI was anticipated, but it was not expected to remain significant after controlling for the effects of known predictors of NSSI. It was expected that self-harm methods most in line with masculine norms (e.g., punching walls or objects) would be associated with greater adherence to these norms.
Method
Participants
Participants were a convenience sample of 912 emerging adults between the ages of 18 and 24 years who were recruited from two college campuses in the Northeastern United States. The first university was a small liberal arts university (University 1). A total of 728 students from University 1 completed the survey. The second school was a large state university (University 2), and a total of 184 students completed the survey at University 2.
Measures
The Deliberate Self-Harm Inventory (DSHI)
The DSHI (Gratz, 2001) is a 17-item self-report questionnaire designed to measure deliberate self-harm behavior that occurs without suicidal intent. Respondents are asked whether they have engaged in particular forms of NSSI. If respondents endorse an NSSI method, they are then asked follow-up questions about the number of times engaging in that method, the age when the respondent first started, the length of time the respondent has been self-injuring, and whether or not hospitalization was required after self-injuring. The measure has strong test–retest reliability over a 2- to 4-week period (.68) and good internal consistency (α = .82; Gratz, 2001). Cronbach’s alpha was calculated at .66 for the DSHI in the present study, which is considered acceptable (Kline, 2013). An additional item was added to the DSHI asking participants if they had ever “punched an object (wall, locker, floor, window, etc.) causing serious tissue damage (bruising, broken bones, lacerations, torn ligaments, etc.).” This item was added to better index forms of NSSI engaged in by men and had minimal impact on reliability (α = .63).
The Conformity to Masculine Norms Inventory–22-Item Version (CMNI-22)
The CMNI (Mahalik et al., 2003) is a 94-item measure assessing the degree to which individuals endorse masculine norms. The CMNI includes 11 subscales that represent several facets of the masculinity domain: Winning, Emotional Control, Risk-Taking, Violence, Dominance, Playboy, Self-Reliance, Primacy of Work, Power over Women, Disdain for Homosexuals, and Pursuit of Status. Respondents rate their agreement with statements on a 4-point Likert-type scale (strongly agree, agree, disagree, strongly disagree). Subscales are summed to produce an overall score. Higher scores indicate a greater degree of conformity to masculine norms. The test–retest reliability for the CMNI for men is reported at .95 whereas internal consistency ranges from .72 to .91 for each subscale and .94 for the measure as a whole (Mahalik et al., 2003). The 22-item version of this metric was used, which is composed of the two highest loading items for each of the 11 subscales. The CMNI-22 correlates with the full 94-item CMNI at r = 0.92. A previous study reported the theta coefficient for the CMNI-22 at .64, and was calculated, rather than Cronbach’s alpha, to manage problems of heterogeneity found with scales that have multiple factors (Mahalik, Burns, & Syzdek, 2007). Cronbach’s alpha was calculated at .66 in the present sample.
Number of Self-Injurers Known
Participants were asked to report on the number of people they knew who engaged in self-injury. The survey item read, “How many people do you know that self-injure?” Participants’ responses were not limited to a given range or scale and they were free to enter any integer.
Positive and Negative Affect Schedule (PANAS)
The PANAS (Watson, Clark, & Tellegen, 1988) is a measure of the participant’s positive and negative affect. Participants were asked to report the degree to which their feelings corresponded with the emotion words provided. The measure consists of 20 emotion-words with 10 positive affect words and 10 negative affect words. Previous research has reported internal consistency between .84 and .90 (Watson et al., 1988). Cronbach’s alpha was calculated at .86 for the negative affect items in the present study’s sample.
Harvard Department of Psychiatry/National Depression Screening Day Scale (HANDS)
The HANDS (Baer et al., 2000) is a validated, 10-item measure of depressive symptoms. Previous research has reported good internal consistency (α = .87; Baer et al., 2000), as was observed in the present study (α = .87).
Procedure
A total of 2,353 students at University 1 in the study’s age range received an email asking if they wanted to participate in an online study about “stress and coping.” The email included a link to the study’s online survey, which informed prospective participants that they could discontinue participation at any time and that they would be entered into a raffle to win one of two $50 cash prizes. Prospective participants who clicked the link were directed to an online version of the study’s consent form and were required to indicate consent before starting the survey. The survey was designed such that participants could only complete it one time. Students at University 2 received the same email but were recruited from a psychology research pool. Participants privately completed the same web-based survey in a research laboratory. All participants across both sites took between 20 and 30 minutes to complete the survey and were presented with contact information for their university’s counseling center in the event that they found any of the questions upsetting. The study survey was built in Checkbox (Checkbox Survey, Inc.), a web-based survey design program. The study survey was hosted and managed by University 1’s technology services department. All participants provided consent to participate in the study, and the study was approved by the institutional review boards at each university. The authors observed a response rate of 38.7% at University 1 and approximately 21% at University 2 (the size of the research pool fluctuated at University 2 throughout the course of the study).
Analyses
One case was removed from analyses as a result of data cleaning and screening as data from this participant suggested a dishonest pattern of responding. A series of two-tailed bivariate correlations and logistic regressions were conducted to address the study’s research questions. Odds ratios (ORs) and adjusted odds ratios (AORs) are reported for all logistic regression analyses, as are statistical significance of relations, effect sizes, and 95% confidence intervals (CIs) as recommended by Cohen, Cohen, West, and Aiken (2003), Cumming and Fidler (2009), and Wilkinson and the APA Task Force on Statistical Inference (Wilkinson, 1999). Cutoffs described by Cohen (1988) are used to describe the strength of correlations and effect sizes. Only the total score for the CMNI-22 was examined as this measure was not developed to examine the subscales of the full 94-item CMNI. Additionally, the subscales of the CMNI-22 have not been tested. All analyses were performed separately for both samples, but since all results were in the same direction and had similar effect sizes, results from the combined samples are reported here.
Participants who reportedly engaged in NSSI six or more times, regardless of method, were categorized as chronic self-injurers. When examining specific methods of NSSI, participants who had engaged in NSSI six or more times using that specific method were categorized as chronic self-injurers for that method. The number of self-reported functions of NSSI increase significantly after six episodes (Whitlock, Eckenrode, et al., 2006), suggesting an increase in the strength of phenomena maintaining NSSI behaviors and potentially suggesting greater chronicity (Nock, 2010).
Results
The overall sample of 912 participants were primarily Caucasian and heterosexual (M age = 19.97 years, standard deviation [SD] = 1.44, 620 or 68.0% female, 292 or 32.0% male, 724 or 79.4% Caucasian, 71 or 7.8% Asian/Pacific Islander, 41 or 4.5% Hispanic, 27 or 3.0% Black, 49 or 5.3% other, 808 or 88.6% heterosexual, 52 or 5.7% bisexual, 26 or 2.9% gay/lesbian, 26 or 2.8% other). Similar demographics were observed among the 728 students from University 1 (M age = 20.8 years, SD = 1.41, 521 or 71.5% female, 207 or 28.5% male, 600 or 82.4% Caucasian, 55 or 7.6% Asian/Pacific Islander, 25 or 3.4% Hispanic, 10 or 1.4% Black, 38 or 5.2% Other, 628 or 86.3% heterosexual, 50 or 6.9% bisexual, 25 or 3.4% gay/lesbian, 25 or 3.4% other) and the 184 students from University 2 (M age = 19.5 years, SD = 1.45, 96 or 52.2% female, 88 or 47.8% male, 124 or 67.4% Caucasian, 16 or 8.7% Asian/Pacific Islander, 16 or 8.7% Hispanic, 17 or 9.2% Black, 11 or 6.0% other, 180 or 97.8% heterosexual, 2 or 1.1% bisexual, 1 or 0.5% gay/lesbian, 1 or 0.5% other).
A total of 229 (25.1%) of all participants engaged in chronic NSSI. Rates for men (24.9%) and women (25.4%) were very similar. These results are outlined in Table 1. Demographic characteristics, beyond those that are known risk factors for NSSI (identifying as female, Caucasian, and/or bisexual), were not associated with chronic NSSI in this sample, including participants’ age, r = −0.03, p = .48, 95% CI = [−0.08, 0.05].
Gender Differences in Frequency and Method of Chronic Nonsuicidal Self-Injury (NSSI).
A point biserial correlation revealed a small positive correlation between scores on the CMNI and being a chronic self-injurer, r = 0.07, p < .05, 95% CI = [0.01, 0.13] (Table 2). As anticipated, men scored higher on the CMNI (M = 28.73, SD = 6.15) than did women (M = 23.59, SD = 5.11), t(487.12) = 12.35, p < .001, 95% CI = [4.32, 5.95]. Scores on the CMNI were correlated with several chronic methods of NSSI. For all participants, regardless of gender, scores on the CMNI were correlated with punching a wall or object, r = 0.11, p < .001, 95% CI = [0.05, 0.17]. For men, scores on the CMNI were positively correlated with burning oneself with a lighter or match, r = 0.14, p < .05, 95% CI = [0.03, 0.25]. For women, scores on the CMNI were correlated with preventing wounds from healing, r = 0.10, p < .05, 95% CI = [0.02, 0.18].
Correlations for Method of Chronic NSSI and Masculinity by Gender.
Note. NSSI = nonsuicidal self-injury; CMNI-22 = Conformity to Masculine Norms Inventory–22-Item Version. Numbers that appear in brackets represent 95% confidence intervals for bivariate correlations.
p < .05. **p < .01.
Associations Between Known Risk Factors and NSSI in the Current Sample
Participants identifying as Caucasian (relative to participants identifying as other; OR = 2.01, 95% CI = [1.08, 4.06]), or bisexual (relative to participants identifying as heterosexual; OR = 4.23, 95% CI = [2.39, 7.49]) were more likely to be chronic self-injurers. By and large, there were no sex differences in chronic NSSI among this sample. These results can be seen in Table 1. A point biserial correlation revealed a moderate-sized relation between the number of self-injurers participants reported knowing and chronic NSSI, r = 0.30, p < .001, 95% CI = [0.24, 0.36] (Table 3). Participants who did not engage in chronic NSSI reported knowing an average of about two self-injurers (M = 2.24, SD = 2.74) whereas those who did engage in chronic NSSI reported knowing an average of more than four (M = 4.64, SD = 4.49). An independent samples t-test revealed this difference to be significant, t(287.30) = 7.63, p < .001, 95% CI = [1.78, 3.02]. Women reported knowing significantly more self-injurers (M = 3.13, SD = 3.58) than did men (M = 2.22, SD = 2.99), t(906) = 3.79, p < .01, 95% CI = [0.44, 1.39].
Intercorrelations for Key Study Variables.
Note. NSSI = nonsuicidal self-injury; CMNI-22 = Conformity to Masculine Norms Inventory–22-Item Version; PANAS Negative Affect = total score for negative affect items on the Positive and Negative Affect Scale.
p < .05. **p < .01.
Does Masculinity Predict Chronic NSSI Above and Beyond Other Known Risk Factors?
Hierarchical logistic regression was conducted. To account for variance associated with known NSSI risk factors (negative affect, depression, identification as female, Caucasian, and/or bisexual, number of self-injurers known, negative affect, depressive symptoms), these variables were entered as the first step. Masculinity, as assessed by the CMNI-22, was entered as the final step. Results from the analysis indicate that the model provides a statistically significant improvement over the constant-only model, χ2(7, n = 909) = 138.16, p < .001. The model correctly predicted 78.7% of those cases identified as chronic self-injurers. Scores on the CMNI significantly predicted chronic NSSI, even after controlling for other known factors, AOR = 1.05, 95% CI = [1.02, 1.08]. Results from this analysis can be viewed in Table 4.
Masculinity Predicts Chronic NSSI.
Note. NSSI = nonsuicidal self-injury; CMNI-22 = Conformity to Masculine Norms Inventory–22-Item Version. PANAS Negative Affect = total score for negative affect items on the Positive and Negative Affect Scale.
Discussion
Building on extant research and theoretical models, this study examined a proposed relation between adherence to masculine norms and NSSI. Results support study hypothesis that masculinity may be associated with NSSI. Adherence to masculine gender norms predicted chronic NSSI above and beyond other well-established risk factors. Results lend support to the hypothesis that masculinity may influence the specific methods of NSSI some men use.
Although results from this study are in line with previously suggested relations between masculinity and other self-destructive behaviors, including suicide, to the best of the authors’ knowledge, this is the first study to identify a relation between masculinity and NSSI. The strength of this association may appear relatively weak at first glance (AOR = 1.05), but examining this finding in the context of additional results makes its meaning clearer. Scoring 1 SD (approximately 6 points) higher on the CMNI-22 would result in a participant being approximately 30% more likely to be a chronic self-injurer. Furthermore, results suggest that masculinity may play a role not only in one’s engagement in NSSI but also what specific methods men may use to injure themselves. Among men, adherence to masculine norms was positively associated with NSSI methods that appeared to be in line with these norms (e.g., burning oneself with a lighter or match). Many studies have reported that men are more likely to burn themselves than are women (Andover, Primack, Gibb, & Pepper, 2010; Claes et al., 2007; Laye-Gindhu & Schonert-Reichl, 2005; Sornberger et al., 2012). However, this appears to be the first study to report that this may be a function of adherence to masculine norms. In contrast, adherence to masculine norms among men was negatively associated with NSSI behaviors that did not appear to be in line with these norms (e.g., scratching oneself). Though this finding appears to be consistent with what is known about masculinity and other self-damaging or high-risk behaviors, to the best of the authors’ knowledge, this is the first study to examine the link between masculinity and specific methods of self-harm. Findings linking engagement in specific methods of NSSI to masculinity provide empirical support to the notion that masculine gender socialization likely influences the methods of self-harm men use (Adler & Adler, 2011; Green & Jakupcak, 2015).
Findings surrounding known risk factors for NSSI (identifying as female, Caucasian, bisexual, the number of self-injurers known, negative affect, and depressive symptoms) were largely consistent with extant research (Gratz, Conrad, & Roemer, 2002; Klonsky, Oltmanns, & Turkheimer, 2003; Muehlenkamp & Gutierrez, 2004). There were no sex differences in engagement in chronic NSSI in the present study. This may provide further evidence that the long-standing gender gap in NSSI is closing and, as a result, may mean that identifying as female may no longer be a risk factor for engaging in NSSI. Participants identifying as bisexual were more likely to have engaged in chronic NSSI and, although this finding has been well documented (King et al., 2008; Whitlock, Eckenrode, et al., 2006), there have been few substantive attempts to address why this is the case. More broadly, researchers and clinicians have discussed the greater prevalence of mental health problems occurring in those identifying as a sexual minority, but have offered few theories in this regard, and have not addressed self-harm specifically (Gollust et al., 2008; King et al., 2008; Skegg, Nada-Raja, Dickson, Paul, & Williams, 2003). This is an important area for future research given the continued finding that identifying as a sexual minority significantly increases the likelihood of engaging in self-harm. There is evidence to suggest that application of minority stress models may provide a useful context for furthering this research (House, Van Horn, Coppeans, & Stepleman, 2011).
Study results indicate that the number of self-injurers known, which may speak to the perceived normativeness of NSSI, increases the likelihood of engagement in chronic NSSI. Those who engaged in chronic NSSI reported knowing an average of approximately five self-injurers whereas those who did not engage in chronic NSSI reported knowing an average of approximately three. This finding is consistent with the well-documented phenomenon of social contagion—the influence of an individual’s self-harm to modify (increase the frequency or change the method) another’s self-harm behaviors (Heilbron & Prinstein, 2008). Similarly, social norms theory indicates that the attitudes and behaviors of others are strong predictors of an individual’s own behavior, with individuals tending to change their own behavior in regards to frequency and/or severity to match the perceived norm; Berkowitz, 2005).
Risk factors identified in this study have important clinical implications, particularly given the strong link between NSSI and suicide (Nock, Joiner, Gordon, Llyod-Richardson, & Prinstein, 2006) and men’s increased risk of death by suicide (Centers for Disease Control, 2011; Oquendo et al., 2001; U.S. Department of Justice, 1995-2006). Although demographic risk factors are unchangeable, adherence to masculine norms as well as the perceived normativeness of NSSI, are likely important and modifiable treatment targets. Research suggests that incorporating discussions about adherence to masculine norms has positive outcomes around other mental health problems often linked with masculinity (depression, help-seeking, anxiety; Syzdek, Addis, Green, Whorley, & Berger, 2014). Similarly, interventions that focus on changing social norms, particularly around masculinity, have been effective in reducing several problematic behaviors among men (Berkowitz, 2002; Kilmartin & Berkowitz, 2005), including substance use, which is often categorized as indirect self-harm behavior. Related efforts targeted at men’s self-harm and self-damaging behaviors may be equally effective.
The strengths of some of the relations in the study, particularly those between masculinity and specific methods of NSSI, were relatively weak. There were several behaviors that are generally characterized as traditionally masculine (e.g., punching walls or objects) with which masculinity was not related. These findings may be due to the way that the items in the DSHI were worded, specifically, using the phrase, “intentionally (on purpose).” It has been suggested that, as a result of masculine gender socialization, men may struggle to articulate their thoughts and feelings (Levant et al., 2003; Levant et al., 2006; Real, 1999), specifically around self-harm behaviors. As such, some men, as a result of masculine gender socialization, may engage in self-harm, but do so impulsively (not “on purpose”) and, as such, would not have reported this on the study’s survey. There is some evidence to suggest that men’s self-harm behaviors are more often supported by sensation seeking or a desire to “get a rush or surge of energy” (Klonsky & Glenn, 2009; Whitlock et al., 2011) which may be consistent with a more impulsive self-harm presentation.
There are several study limitations that must be acknowledged. First, the study is cross-sectional in design, limiting the ability to make causal inferences about the effects of masculinity on NSSI behaviors. Future research will need to examine these relations longitudinally. Additionally, the final sample was composed of emerging adults in the northeastern United States. Though focusing on an emerging adult population was important to ensure that the study captured an appropriate number of individuals engaging in NSSI, a behavior that is known to occur more often among emerging adults, it may limit the generalizability of the study’s findings. Similarly, the present study made use of a convenience sample of college students. As such, results may not generalize to other emerging adults, or to the general population.
Related to efforts to recruit a large sample, study authors were mindful of the length and complexity of the study’s battery, using the shortened 22-item version of the CMNI. The most in-depth examination of the psychometric properties of the CMNI-22 suggests that the measure is marginally sound and that longer versions of the measure should be used (Owen, 2010). This assertion appears to be significantly flawed as the aforementioned study used a mental health treatment-seeking sample of men to investigate the CMNI-22, which includes items related to self-reliance and emotional control. Further investigation of the CMNI-22 in representative samples is clearly needed. In the absence of such investigations, the measure demonstrated acceptable internal consistency in the present study and is highly correlated with the full 94-item version (Mahalik et al., 2007). Findings from the present study should be viewed with some caution. Future research investigating the relations between masculinity and NSSI may benefit from using longer forms of the CMNI, not only to improve measurement characteristics but also because longer forms of the CMNI allow for investigation into the specific masculine norms that may play a role in NSSI. Future research should focus on better understanding the underlying components of masculinity as they relate to men’s engagement in self-harm behaviors.
As masculinity was not more strongly related to methods of NSSI, this may point to a larger problem with the study of NSSI in men. Researchers have suggested that some men, as a result of masculine gender socialization, may struggle to articulate their intentions around self-harm. As such, some men’s self-harm behaviors may not be appropriately captured by measures of NSSI, which require that the self-damaging act be intentional. Future research should examine men’s indirect self-harm behaviors, a class of behaviors for which the intention behind the act is often unclear (Pattison & Kahan, 1983). There is some evidence to suggest that men engage in these behaviors more often than they do NSSI (Hooley & St. Germain, 2014). Additionally, many of the traditionally masculine self-damaging behaviors that men are thought to engage in (punching walls, picking fights, hurting oneself as part of a stunt or dare), may be better studied under the umbrella of indirect self-harm behaviors.
The present study draws attention to masculinity as a potential risk factor for NSSI while highlighting the importance of studying this association further. Though theoretical links between masculinity and self-damaging behaviors have been suggested, research has yet to examine whether and to what degree these proposed links hold true. In particular, future research will need to examine specific masculine norms such as self-reliance and emotional control to better understand self-harm in men. Self-harm behaviors men are more likely to engage in, specifically, traditionally masculine indirect self-harm behaviors (punching walls, picking fights, hurting oneself as part of a stunt or dare), are an important area for future investigation. Despite anecdotal evidence from the clinical literature of men’s engagement in these behaviors (Adler & Adler, 2011), there have been few formal investigations of indirect self-harm (Hooley & St. Germain, 2014; St Germain & Hooley, 2012). Those that exist have reported that men are more likely to engage in these behaviors than they are in NSSI. The impact of masculinity on these behaviors has yet to be examined and is an important next step given the results of the present study.
Footnotes
Authors note
Annie M. Ledoux is now at George Mason University.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.