Introduction

Skin cancer (melanoma, basal cell carcinoma, and squamous cell carcinoma combined) is the most common form of cancer in the USA [1], and there are several differences in incidence across gender and ethnicity. US men have greater lifetime risk for melanoma than women; however, women have greater risk of developing melanoma before age 50 [2]. Melanoma affects non-Hispanic Whites (NHW) at a higher rate than Hispanics [2], but Hispanic adults with melanoma are younger at diagnosis, have more advanced disease, and have lower survival rates than NHW [3]. National rates of nonmelanoma skin cancer are more difficult to estimate, as they are not reported to national cancer registries, but individual studies estimate Hispanics account for approximately 2.9% of cases [4]. Also, nonmelanoma cancers have been largely studied on NHW skin, so differences in physical presentation in Hispanic cases may result in poor detection and delayed diagnosis [5].

Individuals can greatly reduce their risk of skin cancer by decreasing risky sun exposure behavior (e.g., intentional tanning) and increasing sun protection behavior (e.g., using sunscreen) [1]. Researchers and government health agencies have expressed the need to develop skin cancer interventions that are tailored toward specific demographic characteristics such as ethnicity and gender [6, 7]. Young adults, including college students, engage in less sun protection behavior and experience more sunburns than any other age group [8], making young adulthood an important time to intervene. Before designing tailored interventions and education for young adults, whose health behavior is often motivated by social factors, it is important to first identify the psychosocial cognitions that most strongly predict sun safety behavior in unique demographic groups [9].

Gender Differences in Sun Protection and Risk Behaviors and Cognitions

Past research demonstrates that sun-related cognitions and behavior differ by gender. Although young women and adolescent girls tend to report more positive attitudes toward being tan and tanning behavior than their male counterparts [10], they also report more concern about photoaging and skin cancer [11]. Accordingly, research shows that women engage in certain sun protection behaviors (i.e., using sunscreen, seeking shade) [12, 13] and experience fewer sunburns [8] compared to men. These differences may be partially explained by beliefs that caring for one’s skin is feminine, as men who more strictly adhere to masculinity norms report using less sunscreen [14, 15]. Overall, the fact that women report more tanning and more sun protection behavior than men suggests that men and women experience distinct behavioral risk factors for skin cancer.

Ethnic Differences in Sun Protection and Risk Behaviors and Cognitions

Some literature has documented ethnic differences in sun-related cognitions and behavior. NHWs protect their skin more than Hispanics, using sunscreen at much higher rates [16, 17]. Although Hispanics tend to have less sun-sensitive skin, a recent study found that almost 30% of Hispanics had experienced a sunburn in the past year [6]. Both NHW and Hispanic students endorse the belief that tanning makes people look more attractive [18], but Hispanics engage in significantly less intentional tanning behavior than NHWs [19]. However, Hispanic adults believe they are at low risk for skin cancer due to their darker skin tone and worry little about developing the disease [20, 21].

The ethnic differences in sun protection cognitions and behaviors may be a result of different communication about tanning and sun safety from media, friends, and family. For example, parents’ own sun safety practices strongly influence their children’s [22], and Hispanic parents report less positive attitudes toward skin protection than White parents [23]. In focus groups, Hispanic parents indicate that intentional sun exposure is part of Hispanic culture, demonstrate misconceptions about sunscreen (e.g., sunscreen is not necessary), and seem fairly ambivalent about sun safety [24]. Hispanic children also demonstrate less knowledge of sun safety and skin cancer, suggesting ethnic disparities in sun protection education [18]. Hispanic adults who report more acculturation to the USA engage in both more protective behaviors and tanning [25], adopting the paradoxically healthier and riskier sun behaviors more characteristic of Whites. Taken together, research suggests that the ethnic differences in sun-related behaviors are due, in part, to cultural messaging and education in addition to differences in skin tone.

There has been limited research examining how the unique combinations of ethnic and gender identities impact sun-related cognitions and behaviors. Research on this topic is important to promote health equity. Further, although existing research can help predict women’s attitudes toward tanning and Hispanics’ attitudes toward tanning, this information is insufficient for predicting, for example, how a Hispanic woman feels about tanning.

Prototype Willingness Model

To create effective skin cancer education and interventions for individuals, researchers must identify the cognitions that most strongly predict this behavior. The prototype willingness model (PWM) is a commonly used and well-validated model for predicting young adult and adolescent health behavior [9]. The PWM posits that both a reasoned and social reactive path lead to health behavior. Through the reasoned path, attitudes and descriptive norms (i.e., perceptions of the extent to which others engage in the behavior) influence intentions, which then lead to behavior [9]. In the PWM, attitudes are often conceptualized as perceived vulnerability (PV) to the risk associated with the target behavior [9]. However, some studies informed by the PWM have measured attitudes as the extent to which individuals enjoy the activity and/or its outcome [26, 27]. In the social reactive path, descriptive norms, PV, and prototypes influence “willingness,” the most proximal antecedent to behavior. Willingness is defined as an openness to engaging in a behavior under specific circumstances [9]. Prototypes are social images of the typical person who engages in the behavior [9]. The favorability of these prototypes influences the extent to which young adults are motivated to avoid or more closely identify with that social image. The extent to which young adults perceive themselves as similar to these prototypes also influences their willingness to engage in this behavior [28]. Meta-analyses suggest that it is ideal to examine prototype similarity and favorability separately, as these constructs may individually and differentially impact willingness [28].

PWM and Sun-Related Behavior

Some studies have applied the PWM to sun protection and risk behavior [e.g., 29, 30]. For example, Stock et al. [31] found an index of cognitions from the PWM (including tanning attitudes, prototypes, and willingness) mediated the relationship between an ultraviolet (UV) photo intervention and increased sun protection among male outdoor workers. Howell and Ratliff [32] showed the PWM demonstrated good fit in the context of sun protection; tanning attitudes, norms, and prototypes predicted willingness and intentions for sun-related behavior among NHW women. However, this study included sun protection and risk cognitions in the same model [32]. The paradoxical tendency for women to simultaneously engage in both more risk behavior (i.e., tanning) and more protection behavior than men [10, 13] suggests that risk and protection behaviors in this domain are not mutually exclusive and should be examined separately.

Current Study

This study uses the PWM to determine the roles gender and ethnicity may play in the context of skin cancer prevention. This research is important as it helps to dispel health inequities and demonstrates the utility of the PWM as a predictive model for skin cancer prevention cognitions. Thus, the current study seeks to inform skin cancer interventions tailored to different groups of young adults using cross-sectional data from a survey of NHW and Hispanic college students. First, ethnic and gender differences in sun-related cognitions and behaviors were examined to understand how the unique intersections of identity affect these health outcomes. Second, the PWM was used to examine which cognitions were the strongest predictors of sun exposure and protection willingness and intentions for NHW women, Hispanic women, NHW men, and Hispanic men, respectively.

Method

Procedure

The study was approved by the George Washington University Institutional Review Board (IRB) and Texas A&M University Corpus Christi IRB. After providing written informed consent, NHW and Hispanic young adults completed an online survey in which they self-reported sun protection and risk cognitions and behaviors included in the PWM. They also reported their endorsement of normative beliefs regarding gender, skin tone, and sun protection. Students were recruited through the psychology department (either using the subject pool or in classes) and received course credit for their participation.

Participants

Participants were 262 undergraduate students recruited from two college campuses, an urban university in the mid-Atlantic region and a mid-sized university in the southwest. The sample included 130 NHW women (49.6%), 39 NHW men (14.9%), 68 Hispanic women (26.0%), and 24 Hispanic men (9.2%). Mean age was 19.76 (SD = 1.68). Among Hispanics, five participants (5.4%) identified as Black (three were women) and three (3.2%) identified as Asian (one woman). The remaining identified as White.

Measures

Skin Sensitivity

Consistent with prior sun protection literature, an assessment of skin (or sun) sensitivity was provided [6]. As is common in studies of sun risk and protection [33, 34], participants reported how their skin would react if they were to go out for an hour in the midday sun wearing a swimsuit on the first sunny day of the year (1, always burn, not tan, to 6, not burn at all, I am naturally dark skinned) using the Fitzpatrick’s scale [35].

Cultural Variables

Participants reported how frequently their mother and father used sunscreen and tried to get a suntan on a scale from 1 (never) to 7 (always) (mother’s sunscreen use, father’s sunscreen use, mother’s tanning, father’s tanning, respectively). Messages about tanned skin (3 items) were the sum of whether they had received the message that tanned people are more attractive through their friends (yes/no), their family (yes/no), and the media (yes/no). Scores ranged from 0 to 3, where 3 indicated an affirmative response on all three questions.

Sun Protection Beliefs and Knowledge

Participants reported their agreement that it is more important for women to protect their skin (gender beliefs), that it is not “manly” to protect one’s skin from the sun (masculinity beliefs), and that it is more important for people with lighter (vs. darker) skin to wear sunscreen (skin tone beliefs). Responses ranged from 1 (strongly disagree) to 7 (strongly agree). Participants answered 7 true/false questions about sun protection (e.g., “People with naturally darker skin are not at risk for getting skin cancer”) and their skin cancer knowledge score was calculated as the sum of the number correct.

PWM Protection Cognitions and Behavior

To assess sun protection behavior (4 items, α = .624), participants reported how often they used sunscreen on the face, used sunscreen on the body, wore a hat, and wore clothing to protect their skin from the sun in the past year on a scale from 1 (never) to 7 (always). Sun protection norms were assessed by a single item: “How many of your friends of your same age and gender protect their skin from the sun?” (1 (none of my friends) to 7 (all of my friends) [36]. Participants reported their perceived vulnerability (2 items, r = .734) by estimating the likelihood that they would develop skin cancer and experience wrinkling/age spots if they were to tan regularly (1 (not at all likely) to 7 (very likely)) [30]. Sun protection prototype favorability (5 items, α = .791) measured participants’ beliefs that the typical person their age and gender who protects their skin is smart, attractive, mature, self-confident, and fun on a scale from 1 (not at all) to 7 (very much) [32]. Sun protection prototype similarity was a single item assessing how similar they believed they were to this prototype (1, not at all similar to 7, very similar) [29].

Participants’ sun protection intentions (4 items, α = .694) were assessed by their plans to wear sunscreen on their face, sunscreen on their body, wear a hat, and wear protective clothing next summer on a scale from 1 (strongly disagree) to 7 (strongly agree) [37]. To assess sun protection willingness (4 items, α = .701), participants were asked to imagine two scenarios in which they had the opportunity to spend the day in the sun. Participants reported their willingness to use sunscreen and to protect their skin with clothing during each scenario on a scale from 1 (not at all willing) to 7 (very willing) [30].

PWM Risk Cognitions and Behavior

Past sunburns were the number of times in their life participants had developed a sunburn that blistered and peeled (1, none to 4, yes, many times). Past hours in the sun was the average number of hours participants spent outside in the sun during spring/summer on a scale from 1 (0–5 h) to 7 (more than 30 h). To report tanning norms, participants indicated how many of their same age and gender friends sunbathe and go to tanning salons (1, none of my friends to 7, all of my friends) [24]. They also reported their tanning attitudes (3 items, α = .862) by indicating their agreement that a suntan makes them feel more attractive, look healthier, and look thinner on a scale from 1 (strongly disagree) to 7 (strongly agree) [38]. Participants’ risk prototype favorability (5 items, α = .823) was their belief that the typical person their age and gender who does not protect their skin (e.g., spends time in the sun or tanning beds trying to get tan or without sun protection) is smart, attractive, mature, self-confident, and fun on a scale from 1 (not at all) to 7 (very much); adjectives were pulled from pre-existing literature [31, 32, 34, 39,40,41]. Risk prototype similarity was how similar they felt to this prototype (1, not at all similar to 7, very similar) [29].

Tanning intentions were measured through a single item in which participants indicated whether they plan to try to get tan by sunbathing outside next summer (1, strongly disagree to 7, strongly agree) [42]. To assess risky exposure willingness (4 items, α = .783), participants read the same two scenarios used to measure sun protection willingness but reported their willingness to go with their friends and try to get tan and to go with their friends without any sun protection (1, not at all willing to 7, very willing) [30].

Statistical Analyses

Data that support the findings of this study are openly available in Open Science Framework at http://doi.org/https://doi.org/10.17605/OSF.IO/SNKCY.

Racial and Gender Differences in Cognitions

Two by two ANOVAs tested the influence of ethnicity, gender, and their interaction on all cultural variables, skin protection beliefs and knowledge, and PWM risk and protection cognitions. Pairwise comparisons examined differences between each unique gender and ethnic group.

Modeling Prototype Willingness

Path models tested whether PWM cognitions related to sun protection and risky sun exposure would predict corresponding intentions and willingness. Exploratory models were run separately for all demographic groups to compare the relative strengths of each predictor. All models controlled for skin sensitivity and past behavior—which are typical to include for PWM and sun protections-related literatures [6, 43].

Results

Descriptive Statistics

NHW participants reported a mean of 2.69 on the skin sensitivity scale (higher = less sensitive to the sun), and Hispanic participants reported a mean of 3.53. Only 7.7% of NHW and 20.4% of Hispanics indicated they had “never” had a sunburn that peeled or blistered. See Table 1 for correlations between study variables.

Table 1 Correlations between study variables
Full size table

Ethnic and Gender Differences

Below, significant main effects and interactions are reported from the 2 × 2 ANOVAs examining gender and ethnicity differences in sun protection beliefs, protection-related PWM cognitions and behavior, and risky exposure-related PWM cognitions and behavior. See Table 2 for pairwise comparisons between demographic groups.

Table 2 Pairwise comparisons from two-way ANOVAs
Full size table

Cultural Variables

There were no main effects or interactions on father’s sunscreen use or father’s tanning behavior. Compared to White participants, Hispanic participants reported that their mothers used less sunscreen (F(1, 247) = 10.60, p = .001, ηp2 = .041) but also tanned less (F(1, 249) = 6.18, p = .014, ηp2 = .024). There were no effects of gender on either of these variables.

Sun Protection Beliefs and Knowledge

Compared to women, men held stronger masculinity beliefs, gender beliefs, and skin tone beliefs (F(1, 255) = 14.29, p < .001, ηp2 = .053). There were no main effects of ethnicity on these beliefs (Fs < 0.60, ps > .05) and no significant interactions (Fs < 0.50, ps > .10). Women scored higher on skin cancer knowledge (F(1, 255) = 8.29, p = .004, ηp2 = .030) than men, and Whites scored higher than Hispanics (F(1, 255) = 16.31, p < .001, ηp2 = .059).

PWM Protection Cognitions and Behavior

Compared to men, women reported greater PV (F(1, 258) = 5.52, p = .020, ηp2 = .021), more favorable sun protection prototypes (F(1, 258) = 5.43, p = .021, ηp2 = .021), greater similarity to sun protection prototypes (F(1, 258) = 4.29, p = .039, ηp2 = .016), and greater intentions to protect skin (F(1, 258) = 4.31, p = .039, ηp2 = .016). There were no main effects of gender on protection norms, past protection behavior, or protection willingness (Fs < 3.40, ps > .05).

Compared to Hispanics, NHW reported more favorable sun protection prototypes (F(1, 258) = 7.28, p = .007, ηp2 = .027) and greater perceived similarity (F(1, 258) = 8.62, p = .004, ηp2 = .032) to prototypes. There were no main effects of ethnicity on past protection behavior, PV, protection intentions, or protection willingness (Fs < 3.50, ps > .05). There were no significant gender × ethnicity interactions on PWM protection cognitions (Fs < 0.95, ps > .10).

PWM Risk Cognitions and Behavior

Compared to men, women reported higher tanning norms (F(1, 256) = 14.60, p < .001, ηp2 = .054), more favorable tanning attitudes (F(1, 258) = 8.04, p = .005, ηp2 = .030), and greater tanning intentions (F(1, 257) = 15.10 p < .001, ηp2 = .055). There were no main effects of gender on past sunburns, past hours in the sun, risk prototype favorability or similarity, or risky exposure willingness (Fs < 2.78, ps > .05).

Compared to Hispanic participants, NHW participants reported experiencing past burns more frequently (F(1, 258) = 14.32, p < .001, ηp2 = .053), more favorable tanning attitudes (F(1, 258) = 18.92, p < .001, ηp2 = .068), and greater tanning intentions (F(1, 257) = 7.01, p = .009, ηp2 = .027). There were no main effects of ethnicity on the past hours in the sun, descriptive norms of tanning, risk prototype favorability or similarity, or risky exposure willingness (Fs < 2.50, p > .05). There were no significant gender × ethnicity interactions on any PWM risk cognitions (Fs < 0.75, ps > .10).

Comparing PWM Among Demographic Groups

See Fig. 1 for path models predicting sun protection intentions and willingness, stratified by demographic group. See Fig. 2 for path models predicting tanning intentions and willingness for risky sun exposure, also stratified by demographic group. Dotted lines in the figures indicate relationships between control variables and outcomes; solid lines indicate relationships between primary PWM cognitions and outcomes. Gray lines are not significant; black thick lines are significant at p < 0.05.

Fig. 1
figure 1

Model of Sun Protection for a) Non-Hispanic White Women, b) Hispanic Women, c) Non-Hispanic White Men, and d) Hispanic Men

Full size image
Fig. 2
figure 2

Model of Sun Risk for a) Non-Hispanic White Women, b) Hispanic Women, c) Non-Hispanic White Men, and d) Hispanic Men

Full size image

Protection Models

For NHW women (Fig. 1a), PV predicted greater sun protection intentions and willingness, whereas descriptive norms predicted only willingness. Protection prototype favorability was associated with greater protection willingness, but prototype similarity was not. For Hispanic women (Fig. 1b), prototype similarity was the only significant predictor of protection willingness. Among NHW men (Fig. 1c), descriptive norms were the only significant predictor of sun protection intentions. PV and protection prototype favorability predicted protection willingness, whereas descriptive norms and prototype similarity did not. Finally, for Hispanic men (Fig. 1d), PV and descriptive norms predicted greater protection willingness, but no other paths were significant.

Risk Models

Among NHW women (Fig. 2a), PV and tanning attitudes predicted intentions, but descriptive norms did not. More favorable tanning attitudes, greater descriptive norms, and greater risk prototype similarity all predicted greater risk willingness; however, PV and risk prototype favorability did not. For Hispanic women (Fig. 2b), tanning attitudes predicted both tanning intentions and risk willingness, and risk prototype similarity predicted risk willingness. No other paths were significant. For NHW men (Fig. 2c), greater PV and more favorable tanning attitudes were associated with greater tanning intentions and risk willingness. Descriptive norms predicted intentions but not willingness. Neither prototype favorability nor similarity predicted willingness. Among Hispanic men (Fig. 2d), PV and more favorable tanning attitudes predicted greater tanning intentions, but descriptive norms of tanning did not. However, greater tanning norms, along with more favorable tanning attitudes and greater risk prototype similarity, predicted risk willingness. Risk prototype favorability was not associated with willingness.

Discussion

Results revealed several differences in cultural factors, sun protection beliefs and knowledge, and protective and risk PWM cognitions among NHW women, Hispanic women, NHW men, and Hispanic men. Further, in testing the hypothesized pathways of the PWM in each demographic group, it became evident that specific cognitions influence behavioral willingness and intentions more strongly for certain groups. This data is cross-sectional and uses a relatively small sample, especially for Hispanic men. However, these results highlight the need to consider both the ethnicity and gender of the target audience and provide preliminary evidence that can be used to inform interventions tailored to these groups.

PWM Path Models

PWM path models of protection cognitions showed that, when controlling for all psychosocial cognitions, PV for skin cancer was unrelated to sun protection intentions for all groups other than White women. These null relationships reflect evidence that motivations unrelated to health risk are often the strongest predictors of health behavior for young adults, emphasizing the need for health models to consider the social context [9]. For example, in most of this study’s models, descriptive norms explained variance in intentions and willingness even after accounting for PV. Accordingly, interventions to improve sun-related behavior should address social factors—simply providing risk information may not be enough to motivate behavior change.

In risk models, PV was a more consistent predictor of intentions but not willingness. However, attitudes toward being tan were the strongest, most consistent predictor of risk intentions and willingness for all demographic groups. These results suggest that studies informed by the PWM should consider both PV and attitudes toward the behavior or behavioral outcome, as other researchers have done in the past [26, 27]. This may be especially important for behaviors like tanning, which young adults may perceive as highly desirable outcomes. Consistent with previous research [44, 45] prototype favorability was a less consistent predictor of willingness than prototype similarity. Thus, identifying with a health risk or protection image may influence personal behavior more strongly than simply liking that image [28, 45].

A recent study by Calderón and colleagues [16] found that all psychosocial variables they measured (namely, tanning beliefs, body image, and attitudes toward protection behavior) influenced self-reported sun protection behaviors in the same pattern for Hispanics and Whites. However, the present study considered several additional psychosocial cognitions and found that, except for tanning attitudes, the influence of these cognitions on behavioral willingness and intentions indeed varied by ethnic (and gender) identity.

Implications for Tailored Education and Interventions

In synthesizing results from ANOVAs and path models, several themes with important implications for tailoring skin cancer interventions emerged. First, results from the ANOVAs highlight the paradox that women are at greater risk for tanning than men yet are more inclined to use sun protection [10, 11]. The evidence of this paradox within a single sample emphasizes that men and women face unique behavioral risk factors for skin cancer, and interventions should be tailored accordingly.

White participants had more knowledge of skin cancer than Hispanic participants, highlighting ethnic disparities in skin cancer education and the need to include Hispanic children in this education. In addition to formal education, Hispanic children may learn different sun safety practices by modeling their parents. In this study, White mothers used sunscreen more but also tanned more than Hispanic mothers, demonstrating the tanning/sun protection paradox once again.

White Women

As expected, White women had the healthiest sun protection cognitions yet the unhealthiest tanning cognitions. A possible explanation for this paradox could be due to the role of sunscreen response efficacy—where these individuals believe sunscreen will reduce sunburn, allowing them to be outdoors for longer to obtain a better tan. Thus, rather than attempting to increase their already healthy protection cognitions, interventions should target White women’s attitudes toward themselves being tan, which were disproportionately high in this group and also strongly predicted risk intentions and willingness. White women also reported receiving the most messages (from media, friends, and family) that having tanned skin was attractive. Interventions that encourage women to see pale skin as beautiful have been shown to reduce tanning intentions [46, 47]. Interventions may also consider presenting White women with alternatives. Such interventions may also decrease tanning norms, which were a strong predictor of White women’s risk willingness, by increasing perceptions that most women choose safe sunless tanning products rather than UV exposure to develop a tan.

Hispanic Women

Prototype similarity was the only significant predictor of sun protection cognitions for Hispanic women. Further, only prototype similarity and tanning attitudes predicted Hispanic women’s risk willingness. Therefore, these variables’ influence may supersede that of descriptive norms, which has previously been shown to be a strong predictor of protection behavior in Hispanic adults [48].

The influence of prototype similarity is especially interesting because researchers have attributed rising skin cancer rates among Hispanics to their increasing assimilation to US culture and sun-related behavior [49]. Previous evidence demonstrates that greater US acculturation among Hispanics is associated with both riskier sun exposure and more protective behaviors, such as sunscreen use [25, 50]. Both risk and protection prototypes for sun-related behavior may be perceived as more American than Latin, and the extent to which Hispanic women identify with these American prototypes dictates their likelihood of engaging in such behavior. The culturally held beliefs of colorism can also impact sun protection behavior [51, 52]. Although colorism was not accounted for in the present study, prior work has demonstrated that cross-cultural perspectives can shift colorism beliefs (e.g., away from what is traditionally valued, if living in a different culture [52, 53]). Therefore, a lack of or identification with these colorism beliefs—which are exhibited in Hispanic culture [54]—could explain these differences of engaging in sun behaviors. Interventions may encourage Hispanic women to contrast with prototypical tanners and identify with protection prototypes. Such an intervention has demonstrated effectiveness in the context of young adult drinking; participants asked to contrast themselves with prototypical drinkers reported less willingness to drink [55].

Hispanic women may also benefit from interventions that target tanning attitudes, which strongly predicted tanning intentions and risky exposure willingness. Although ample resources have been devoted to developing interventions that reduce tanning risk for White women, these interventions may not be appropriate for Hispanic women. Hispanic women tend to have naturally darker skin than White women; therefore, messages that pale is also beautiful may be less relevant to them. Alternatively, because Hispanic women’s tanning is motivated by appearance concerns (i.e., more favorable attitudes toward being tan), interventions conveying UV damage harms appearance, such as age progression or UV photos may be effective [56, 57].

For Hispanic women and men, sun protection intentions were not explained by any of the PWM cognitions above and beyond past behavior. For Hispanics, who rarely receive messages about skin cancer [58], sun protection may be the subject of little contemplation and thus more predominantly determined by the social reactive pathway.

White Men

In general, White men had less risky tanning cognitions than women but had lower intentions to protect their skin. Unlike for White women, interventions for White men may be apt to focus on improving sun protection cognitions rather than tanning. Specifically, it may be most effective to target White men’s perceived norms of skin protection and their prototype favorability, which was a poor predictor of cognitions in most models, but strongly predicted men’s willingness to protect their skin. Men also had less favorable protection prototypes than White women and more strongly endorsed beliefs that skin protection is “feminine.” Previous interventions that create positive images of individuals who engage in healthy behavior have successfully reduced risk cognitions [e.g., 59]. Therefore, interventions that create more favorable images of the typical man who protects his skin may encourage greater sun protection behavior in this group.

Hispanic Men

Hispanic men had the lowest tanning intentions, attitudes, and descriptive norms. Therefore, skin cancer interventions for this population should devote their efforts toward increasing sun protection. Compared to other groups, Hispanic men most strongly believed that it is more important for people with light skin (vs. darker skin) and for women (vs. men) to protect their skin from the sun, suggesting that they do not see these behaviors as personally relevant. Likewise, Hispanic men reported the lowest PV of any group. Because PV strongly predicted protection willingness, interventions should focus on Hispanic men’s understanding of their personal risk for skin cancer, tailored to individuals’ skin sensitivity, and level of exposure and protection. UV photo interventions [57, 60], in which participants receive a photo of the existing sun damage on their face, may be effective for this purpose. Hispanic men may also benefit from feedback that increases normative perceptions of sun protection. For primarily White samples, conveying high descriptive norms of sun protection increases protection intentions and maximize the effectiveness of UV photo interventions [37, 61]. Furthermore, normative interventions, particularly for young adults, have been effective across health behaviors, such as sun protection, given that this age group (compared to older ages) tend to be more influenced by others [62,63,64].

Limitations and Future Directions

There are some limitations to the study. The data collected used a convenience sample—which puts limits on the generalizability of the data. For example, college students may not be fully representative of US young people. The sample was majority women, and there were only a few individuals who reported identifying as Hispanic and a racial minority—in the future, research should work to recruit a more even distribution of gender as well as take a more intersectional approach for those with multiple minority identities (e.g., do dual identities serve as a modifier for risk cognitions). Future research should explore how sun-related cognitions vary across other racial/ethnic groups. Additionally, skin tone beliefs or colorism was not included as a cultural variable and should be considered as such in future research. The skin type/sensitivity measure was self-report, which although research has demonstrated value in, future studies could also examine objective measures (e.g., phototest) for this variable [65,66,67].

This study used the PWM as the theoretical framing. Future research could investigate integrating additional theoretical frameworks, such as including sunscreen response efficacy (i.e., from the health belief model [68]) to possibly help explain the paradoxical beliefs held by White women. Although social factors were considered in our study (e.g., perceived norms, prototypes, behavioral willingness, and parental influence), future research may want to further explore specific peer influences (e.g., close friends, peer risk perceptions) relevant to college students, given that young adults can be influenced by them [62, 63, 69]. This study also did not examine future behavior. Although behavioral willingness and intentions are acceptable predictors of behavior [43], future research can examine how these cognitions predict future behaviors among diverse racial/ethnic groups. Finally, we acknowledge that for the exploratory path models, only the White women subgroup was sufficiently powered. Given this, the remaining subgroup models should be interpreted with caution and replicated in future research. We have included these subgroup models to be in alignment with NIH’s recent guidelines, which emphasize reporting analyses even if underpowered to promote and encourage health disparity research—such as with gender and racial/ethnic minority populations [70, 71].

Conclusion

This study examined how gender and ethnicity impact skin cancer prevention cognitions for young adult populations within the PWM. The results highlight the need for skin cancer prevention interventions tailored to specific audiences—as each gender and ethnic intersection demonstrated differences in sun-related cognitions. The results of this study provide preliminary evidence to inform interventions for NHW and Hispanic young men and women. This study provides an important foundation for these efforts by including data from Hispanic populations, who have been traditionally neglected in skin cancer prevention research, and by considering the intersection of gender and ethnicity in this context. Future research should explore other cultural variables that could interact with tailored intervention messaging.