Introduction

Emerging adulthood (ages 18–25 years) is considered a developmental period of increased exploration of possible life directions, experimentation, and transitions (for most) while attending higher education and deferring adult roles (Arnett, 2015). During this period, individuals explore various facets of their identities such as occupation, romantic relationships, and sexuality; however, some scholars argued that these exploratory experiences may not be afforded to those who do not attend postsecondary education settings such as colleges or universities (Hendry & Kloep, 2007). Scholars (Hendry & Kloep, 2010; Reifman, Arnett, & Colwell, 2007; Zaluski, 2012) have compared college-attending and non-college populations (ages 18–25 years) and found differences in perceptions of identity establishment and deferment of adult roles. Studies that have compared these populations (college-attending vs. non-college-attending) have also identified differences in sexual engagement including intercourse experience, casual sex, and sexual risk behaviors (Bailey, Fleming, Henson, Catalano, & Haggerty, 2008; Olmstead, Norona, & Anders, 2018); however, to date, no studies were found that have compared the sexual expectations of college-attending and non-college-attending (termed non-college henceforth) individuals.

Sexual expectations (i.e., expectations of future sexuality) are important to address during this developmental period as these expectations likely vary based on individuals’ inclusion and/or opportunities for exploration and experimentation in areas of sex and sexuality. Sexual expectations in emerging adulthood may be further understood by examining and comparing college-attending versus non-college individuals’ possible selves. Possible selves are internalized expectations or fears for the upcoming year (Markus & Nurius, 1986) and can be influential on behaviors and goals during transitional periods such as emerging adulthood (Oyserman & James, 2009).

Despite its potential utility, previous research is limited in examining the content of emerging adults’ future perceptions of their sexual selves (see Anders & Olmstead, 2018; Anders, Olmstead, & Johnson, 2017) and has yet to examine sexual possible selves differences between college-attending and non-college individuals. The purpose of this study was to examine the sexual possible selves (SPS) of a sample of college-attending emerging adults and a sample of non-college emerging adults to determine whether differences exist in their reported SPS themes. Specifically, we were interested in identifying and examining: (1) prominent categories of SPS and fears within each group, (2) associated strategies for achieving and avoiding these possible selves and fears within each group, and (3) group differences in the emergence of categories based on gender and college status.

The Developmental Period of Emerging Adulthood

Emerging adulthood (ages 18–25 years) is a time of identity exploration and experimentation (Arnett, 2015). Individuals in this developmental period typically experience relative independence from normative social expectations, the deferment of “adult roles” (e.g., marriage), and an acceptance of an experimental nature of young individuals (Schwartz, Zamboanga, Luyckx, Meca, & Ritchie, 2013). This period is thought to be distinct from other life stages based on demographic factors, subjective perceptions, and the role of identity exploration (Arnett, 2015).

Although shifts in current societal and economic trends have allowed some individuals to take these “time outs” from adult roles to pursue identity explorations, these moratoriums might look different for some emerging adults or may not be afforded altogether for others (Hendry & Kloep, 2007). For example, studies have indicated that emerging adulthood may only be observed among those attending college (Hendry & Kloep, 2007), specific industrialized cultures such as the U.S. (Arnett, 2015; Cote, 2014), and those in the middle to upper class (Cote & Bynner, 2008). Although presented as a universal life course stage, demographic variability may influence how, or even if, young individuals experience development considered distinct for this period (Bynner, 2005). For example, non-college emerging adults will, on average, marry younger (Arnett, Kloep, Hendry, & Tanner, 2011; Goldstein & Kenney, 2001), start a family sooner (Kloep & Hendry, 2011), and indicate earlier subjective feelings of adulthood (Nelson & Barry, 2005). College-attending emerging adults typically experience greater freedom to explore various lifestyles, make mistakes, and are often given freedom to explore their identities, including sexuality, as compared to their non-college peers (Lam & Lefkowitz, 2013). Expectations for non-college individuals may differ as they are expected to assume and establish patterns of adult roles and behaviors (such as immediate entry into the workforce), which may influence expectations that individuals hold for themselves.

The majority of research on emerging adults’ sexuality has focused on college-attending populations; however, recent evidence suggested that differences between these two populations exist (Bailey et al., 2008; Goldstein & Kenney, 2001; Lyons, Manning, Giordano, & Longmore, 2013). For example, Bailey et al. (2008) found differences in sexual risk trajectories and sexual engagement based on college enrollment (i.e., emerging adults in 2- or 4-year universities versus non-college emerging adults). Specifically, non-college emerging adults were found to have higher rates of sexual intercourse, casual sex, and engage in higher risk sexual activity, including inconsistent condom use, compared to their college-attending peers. Further, a comparison by Lyons et al. extended this research to examine different levels of educational attainment (e.g., high school degree versus community college) in relation to casual sex behaviors. Results indicated that emerging adults currently enrolled or graduated from a 4-year college had the least number of casual sex partners, whereas those who had less than a high school degree were 220% more likely to have more casual sex partners. These findings, along with continued scrutiny by scholars (e.g., Claxton & van Dulmen, 2013), emphasize the need for continued research on casual sex relationships (Grello, Welsh, & Harper, 2006), romantic relationships (Goldstein & Kenney, 2001), and sexual expectations (Anders et al., 2017; Stinson, 2010) of non-college samples.

College-attending emerging adults’ expectations about sexuality may differ from their non-college peers. Sexual expectations encompass an individual’s perceptions of and expectations for normative sexual behaviors, including those of their peers (Lambert, Kahn, & Apple, 2003; Stinson, 2010). These expectations focus on either current external behaviors or social norms, or may include future expectations. For example, individuals who were not sexually active indicated that they expected college to be a time to engage in sexual behaviors, and a time for initiating sexual intercourse within relationships (Cohen & Shotland, 1996; Stinson, 2010). Studies have examined these expectations among college-attending emerging adults and found that individuals consistently overestimate their peers’ engagement in various sexual behaviors (e.g., hooking up) as well as peer acceptance of casual sex behaviors (Fielder & Carey, 2010; Lambert et al., 2003). In contrast, evidence suggests that non-college individuals may be less susceptible to these overestimations, as their perceptions of casual sex acceptance were lower than their college-attending peers (Twenge, Sherman, & Wells, 2015). Overall, these expectations are important as they often influence establishing patterns of behavior and identity formation during the period of emerging adulthood (Barriger & Velez-Blasini, 2013; Fielder & Carey, 2010); yet, few studies have examined expectations, particularly sexual expectations, among non-college populations.

The Theory of Possible Selves

One perspective for examining emerging adults’ expectations is through their possible selves. Possible selves are considered as an individual’s future-oriented selves he or she hopes to become (i.e., expectation) or hopes to avoid becoming (i.e., fear) in the proximal future (Markus & Nurius, 1986). Possible selves are thought to be representative of salient goals or fears that are important to future identity development (Oyserman & James, 2009), and can serve as pathways to attaining or avoiding future identities (Oyserman, Bybee, Terry, & Hart-Johnson, 2004). Examining the specific behavioral plans individuals use to obtain or avoid these future possible selves (i.e., strategies) can provide insight into barriers to goal attainment. For example, one study found that students who articulated goals for academic achievement, yet were unable to articulate specific strategies to attaining these goal were less likely to achieve their future-oriented academic self (Oyserman et al., 2004). Possible selves are also thought to be representative of developmental milestones (Oyserman & James, 2009) and social contexts (Oyserman & Fryberg, 2006). For example, a possible self reported by an adolescent may focus on academic goals, whereas a possible self for a middle-aged individual would likely be more family related. Further, possible selves are situated in various cultures (e.g., college cultures) and influenced by social contexts (e.g., social norms; Oyserman & Fryberg, 2006). Overall, possible selves tend to be representative of where individuals see themselves in the proximal future (i.e., typically 1 year out) and are influenced by their current life stage and developmental paths (e.g., emerging adulthood).

Sexual Possible Selves

Sexual possible selves (SPS) is a narrowing of the possible selves theory (see Anders et al., 2017) to focus on the sexual expectations and fears individuals hold for themselves in the proximal future. These sexual goals hold important implications for which aspects of sexuality are most salient and can serve as guidelines or paths to achieving individuals’ future idealized sexual selves. SPS may also be reflective of developmentally salient sexual goals or identities, particularly based on current life stage.

As the identification and examination of SPS is fairly recent phenomenon, a review of the literature indicated few studies have examined the SPS of emerging adults (for exceptions see Anders & Olmstead, 2018, 2019; Anders et al., 2017) and has highlighted gaps within our understanding of these future sexual selves. For example, of those that have examined SPS, all have focused on college-attending emerging adults and have not extended this research to examine developmental differences in non-college emerging adults. Specifically, the original study (Anders et al., 2017) used a mixed-methods approach to examine the sexual expectations, fears, and strategies of first-semester college students (N = 282; ages 18–25 years). Themes that emerged included abstinence, physical/sexual health, sexual exploration, and interpersonal relationships. This study also revealed a discrepancy between reporting of SPS and the articulation of specific strategies participants were using or planned to use for obtaining these future-oriented selves. For example, although some reported fears of negative physical/sexual health selves in the future, most were unable to articulate realistic and specific strategies they were using to avoid these feared selves. As sexually transmitted infections (STIs) are a prevalent health concern due to high rates among young adults (e.g., highest rates of Chlamydia are among 20–24-year-olds; Centers for Disease Control and Prevention 2017), this is a concern for the sexual health of emerging adults if they are unable to articulate ways in which they can decrease the likelihood of these outcomes. Anders et al. also found differences based on sample demographics, including participant sex, intercourse experience, being in a romantic relationship, and religiosity.

Across these SPS studies, it was found that students varied in their SPS and fears. For example, gender differences (i.e., men versus women) in expected SPS, feared SPS, and strategies were an important finding in all previous SPS studies. Specifically, gender differences for themes related to sexual engagement (e.g., explore/experiment), reputation, and rape/assault hold important implications for sexuality research. Men were found to report expectations around exploring or experimenting during their first semester in college, including expectations for non-committed sexual behaviors (e.g., hooking up). Although not inherently risky, sexual explorations outside of committed relationships can increase the potential for sexual risk, yet men from this study also reported less physical/sexual health strategies for avoiding deleterious sexual health outcomes (Anders et al., 2017). Further, women were found to report more feared SPS focused on getting a negative reputation in college and being sexually assaulted or raped during the first semester than did men. This fear is not surprising as 20–25% of women are sexually assaulted during their college experience (Exner & Cummings, 2011). Interestingly, Anders and Olmstead (2019) found that men also reported fears around sexual assault; however, these fears were in the context of committing “unintentional” assault. Continued research examining gender differences in expected and feared SPS is needed, as limited research has examined these gender differences using a non-college emerging adult sample (for exception see Zweig, Barber, & Eccles, 1997).

A recent qualitative study by Anders and Olmstead (2019) examined the developmental influences that shape first year college students’ SPS and strategies (N = 35; ages 18–19 years). This study highlighted how these emerging adults perceived college cultures, along with associated alcohol use and partying, as playing an important role in shaping their expected and feared future selves. As this study focused on college students only, developmental differences among those who do not attend college may highlight differences in sexual expectations, fears, and strategies of those who are not exposed to these college cultures.

The differences found across these studies highlight a need for further examinations of SPS among diverse populations. One step to extending this SPS research to a more diverse sample would be to focus on expectations, fears, and strategies among those who opt out of or unable to attend postsecondary education. Examining differences between college-attending and non-college emerging adults’ SPS would advance our understandings of salient sexual themes and gaps in behavioral pathways that may be crucial for more developmentally appropriate sexual education/interventions for each population, rather than a “one-size-fits-all emerging adults” approach. In addition, as this area of research grows, it is important to continue to examine gender differences with diverse samples to determine whether women and men differ in their expected and feared SPS and the strategies they use to achieve their expected sexual selves or avoid their feared sexual selves.

Current Study

This study extends the possible selves literature and previous SPS research by examining and comparing expected and feared SPS using a large sample of college-attending emerging adults and a large sample of non-college emerging adults. In addition, we examine and compare the strategies (or lack thereof) these groups articulate to achieving or avoiding their SPS. Our study will enhance current understandings of similarities or differences in how these groups experience and/or hold expectations for sexuality during the developmental period of emerging adulthood. We also attend to whether SPS differ based on gender generally, and within each group. Doing so increases our understanding regarding differences or similarities in men’s and women’s approaches to their sexual selves. Findings likely hold important implications for sexual education and/or intervention for those in non-postsecondary education settings, along with continued sexual health and well-being promotion for college-attending emerging adults. Our study was guided by five research questions developed from theory (emerging adulthood and possible selves) and the extant literature on SPS and sexuality in emerging adulthood:

  • RQ1 How do college-attending and non-college emerging adults describe their (a) expected and (b) feared sexual possible selves?

  • RQ2 What are college-attending and non-college emerging adults’ strategies associated with their (a) expected and (b) feared sexual possible selves?

  • RQ3 Do college-attending and non-college emerging adults vary in their (a) expected and (b) feared sexual possible selves? If so, how do they vary in these possible selves realms?

  • RQ4 Do emerging adult men and women vary in their (a) expected and (b) feared sexual possible selves? If so, how do they vary in these possible selves realms?

  • RQ5 Do these (a) expected and (b) feared sexual possible selves vary based on gender and college attendance status (i.e., gender by college attendance status)?

Method

Participants

Participants were 800 college-attending (n = 400) and non-college (n = 400) emerging adult men and women (age range = 18–25).

College-Attending Sample

The majority were women (67.1%). Most (68.3%) identified their race/ethnicity as White, non-Hispanic followed by Black/African-American (10.0%), Latino/a or Hispanic (8.8%), Asian-American/Pacific Islander (8.8%), Native American/American Indian (1.8%), or “Other” (2.5%; e.g., multiracial). The majority (70.3%) identified as heterosexual; however, 18.0% identified as bisexual, 5.8% as gay/lesbian, 3.3% as unsure, and 2.8% as “Other” (e.g., pansexual or asexual). A large proportion of participants (47.0%) were in an exclusive romantic relationship, followed by no romantic relationship (31.5%), non-exclusive romantic relationship (18.8%), and “Other” (2.8%; e.g., “It’s complicated”). The sample reported being moderately religious with an average religious intensity of 7.38 (SD = 6.79; range = 0–20). The largest proportion (31.3%) of college-attending participants were Sophomores, followed by Juniors (28.3%), Seniors (26.8%), Freshmen (11.0%), and “Other” (2.8%; e.g., 5th year).

Non-College Sample

The majority were women (58.9%). Most (71.4%) identified their race/ethnicity as White, non-Hispanic followed by Black/African-American (13.3%), Latino/a or Hispanic (8.8%), Asian-American/Pacific Islander (3.5%), Native American/American Indian (1.0%), or “Other” (2.0%). The majority (73.9%) identified as heterosexual; however, 16.0% identified as bisexual, 5.0% as gay/lesbian, 2.0% as unsure, or “Other” (3.0%). About half (50.4%) were in an exclusive romantic relationship, followed by non-exclusive romantic relationship (24.3%), no romantic relationship (23.3%), and “Other” (2.0%). The sample reported being moderately religious with an average religious intensity of 7.16 (SD = 6.95; range = 0–20). No significant demographic differences were found between the college-attending and non-college samples (analyses not shown).

Procedure

Upon approval from the university’s institutional review board (IRB), individuals were recruited through advertised “tasks” on Amazon Mechanical Turk (MTurk) during summer 2016. MTurk is an Amazon Inc.-based crowdsourcing system used for various “Human Intelligence Tasks” and data collection that has become prevalent among experimental and survey-based social science researchers, as it increases the likelihood of obtaining individuals from non-college settings (Ipeirotis, 2010). Two separate surveys were posted on the site to collect data for each of the two populations (i.e., college-attending versus non-college). Potential participants first completed a brief eligibility questionnaire to see whether or not they met the inclusion criteria for one of the two separate surveys. The inclusion criteria for non-college individuals included only those who were: (1) between the ages of 18 and 25 years (i.e., emerging adulthood), (2) not currently enrolled in postsecondary education (i.e., college or trade school), and (3) have never attended or received a degree from a postsecondary education setting, including 2- or 4-year institutions. Additional exclusion criteria included individuals who have taken some college/university courses but are no longer enrolled, as they have been introduced to the college cultures and may hold biased responses. Inclusion criteria for college-attending individuals included only those who were: (1) between the ages of 18 and 25, (2) currently enrolled in postsecondary education (i.e., college), and (3) have never received a degree from a postsecondary education setting, including 2- or 4-year institutions. Both populations were also required to currently live in the U.S. and their primary language be English.

Participants who met the inclusion criteria received access to a restricted-use online survey (i.e., Qualtrics) and were presented with an informed consent page that described the goal of the survey, the risks and benefits of the study, their ability to withdraw at any point, and described how their responses would remain anonymous. Participants who provided consent were given access to the survey. The survey included a series of open-ended questions and questionnaires, demographic items, and quantitative survey measures. Participants who completed the survey received a small monetary compensation (i.e., $1.01). After 400 surveys for each population were completed, the recruitment was removed from the site. This study focused on participants’ responses to the sexual possible selves questionnaire and various demographic measures, which occurred right after the demographic items to avoid a priming effect (Creswell, 2007) from the other measures in the study.

Measures

Sexual Possible Selves

Participants’ SPS and strategies were measured using an adaptation of the standard open-ended format of the possible self and strategy measure (Anders et al., 2017). This measure focused on participants’ (1) SPS expectations, (2) strategies for attaining the associated expectations, (3) SPS fears, and (4) strategies for avoiding the associated fears. The questionnaire was formatted to be neutral to participation or non-participation in postsecondary education and focused on the upcoming year (see Anders et al., 2017 for description of questionnaire).

College Attendance Status

Participants were grouped based on (1) college-attending and (2) non-college.

Gender

Participants reported their gender as (1) Male, (2) Female, or (3) Other, please specify. The majority of participants from our sample reported their gender as either male or female (n = 796); however, four participants responded as other (e.g., transgender male). Due to this small sample size of participants who identified as other than male and female, we did not include their responses in the quantitative comparisons for this measure.

Analyses

Open-Ended Response Coding

Participants’ open-ended responses of their expected SPS, feared SPS, and associated behavioral strategies were examined using qualitative content analysis (RQ1 and RQ2; Krippendorff, 2013). First, responses were carefully examined by three trained coders for recurring topics or themes across responses, and keywords (i.e., units; Krippendorff, 2013) were created as the basis for developing coding categories. Category definitions were developed to help emphasize differences in themes and strengthen coding of responses. Using these identified coding categories, coding of the SPS responses was then conducted to categorize each individual’s responses. To minimize bias, coders were blind to participants’ gender and college attendance status throughout coding (Creswell, 2007). Further, one coder was blind to previous SPS research (i.e., Anders et al., 2017), limiting priming bias and allowing categories to emerge naturally rather than referencing previous SPS categories. Coding was conducted as a group until acceptable inter-coder agreement was met (i.e., > 80%). Coding checks were then conducted throughout to continue calculating coder agreement, allowing us to discuss any discrepancies in coding until agreement was met. Inter-coder agreement was high at 94.2% for expected SPS, 92.5% for expected SPS strategies, 94.2% for feared SPS, and 94.4% for feared SPS strategies.

Quantitative Analyses

Comparisons between college-attending and non-college SPS categories were conducted in two ways. First, after open-ended responses were categorized, counts of the frequencies of each of the identified categories were examined. Second, variability in expected SPS and feared SPS was examined using a series of Pearson chi-square comparisons. As multiple expected and feared SPS categories were reported by participants (described in more detail below), the data contained a preponderance of zeros for each category leading to skewness. For example, a participant who reported one type of expectation may not also report another expectation in a category that opposes the first and therefore that category would remain at zero. To accommodate for skewness, categories were dichotomized to either (0) did not report SPS in category or (1) reported SPS in category and chi-square comparisons between who did or did not report each type of SPS category were conducted. Specifically, proportion comparisons of each SPS category were conducted based on college attendance status (RQ3), gender (RQ4), and gender by college attendance status (RQ5). Proportional comparisons allowed us to examine whether specific expectations or fears varied by group membership (e.g., men versus women). Between-group comparisons can indicate differences in experiences, particularly in their expectations and fears about sexuality.

Results

Sexual Possible Selves

Our qualitative content analysis indicated a great deal of variance in participants’ sexual expectations, fears, and associated behavioral strategies, regardless of college attendance status. Overall, 11 categories were found for expected SPS, ten categories for expected SPS strategies, 11 categories for feared SPS, and ten categories for feared SPS strategies. SPS categories were coded on a continuous scale, with zero indicating that no SPS was reported for that category and each one point increase representing a SPS for that category. Although participants were limited to four open-ended response boxes, multiple SPS could be reported and coded within each box.

Expected SPS

Emergent categories for participants’ expected SPS focused on who they hoped to be sexually in the proximal year (see Table 1). On average, participants reported 2.16 expected SPS (SD = 1.25; range = 0–5). Eleven categories emerged for expected SPS including abstinence (e.g., “to not have sex”), explore/experiment (e.g., “To experiment new thing and sensations”), quantity (e.g., “more partners”), quality (e.g., “To enjoy it more”), sex specific (e.g., “Anal”), interpersonal relationship (e.g., “To find a long-term partner), sexual health/well-being (e.g., “To have safe sex”; i.e., positive physical health engagement or outcomes), maintain (e.g., “Nothing to change”), family formation (e.g., “To have a baby”), decreasing/avoidance (e.g., “To not use condoms”), and no expectations (e.g., “Nothing”). Frequency comparisons were conducted for participants who reported at least one expected SPS. The most frequently reported category was quantity with 34.9% of the sample reporting at least one quantity-related expectation (range = 0–4). The least frequently reported category was decreasing/avoidance expectations (1.8%; range = 0–1).

Table 1 Expected SPS and strategies categories with descriptives (N = 800)
Full size table

Expected SPS Strategies

Associated behavioral strategies were examined for each reported SPS. Expected SPS strategies focused on behaviors in which participants were actively engaging to meet their expectations for the year (see Table 1). Participants reported an average of 2.17 strategies (SD = 1.55; range = 0–8). Ten categories emerged for expected SPS strategies, including abstinence (e.g., “Not engaging in it”), interpersonal relationship (e.g., “Communicating with my partner”), dating (e.g., I joined a few dating sites”), making opportunities (e.g., “Going to clubs and meeting guys”), self-improvement (e.g., “Exercising to improve physical appearance”), sexual health/well-being (e.g., “Always use protection”), expanding boundaries (e.g., “Trying new things in bed”), restricting boundaries (e.g., “Avoiding women”), sex specific (e.g., “We are having sex”), and no strategy (e.g., “No”). Frequency comparisons were conducted for participants who reported at least one expected SPS strategy. The most frequently reported strategy category was making opportunities with 30.8% of the sample reporting at least one strategy that increased opportunities to reach their expected SPS (range = 0–7). The least frequently reported strategy participants were actively utilizing to attain their expected SPS was abstinence strategies (2.4%; range = 0–2).

Feared SPS

Following the research of Anders et al. (2017), participants’ feared sexual selves were examined to understand sexual motivations or goal development (see Table 2). On average, participants reported 2.03 feared SPS (SD = 1.18; range = 0–5). Emergent categories included: abstinence (e.g., “Engaging in sexual activities”), interpersonal relationship (e.g., “Being cheated on”), quantity (e.g., “Having less sex”), quality (e.g., “Bad sexual experiences”), self-focus (e.g., “Feeling bad about my body”), partner focus (e.g., “Partners who are no good for me”), sexual health/well-being (e.g., “STIs and getting pregnant”), increased sexual risk (e.g., “Drunk sex”), sexual assault/coercion (e.g., “Being pressured into unwanted sexual situations”), sex specific (e.g., “Oral”), and no fear (e.g., “Not really anything”). In contrast to participants’ expected SPS, less variance was found in participants’ feared selves. Frequency comparisons indicated the most frequently reported feared category was sexual health/well-being with 47.0% of the sample reporting at least one feared SPS focused on their sexual health (i.e., negative physical health engagement or outcomes; range = 0–4). The least frequently reported category was no fear with only 4.4% of participants, indicating they did not have a feared SPS for the upcoming year (range = 0–1).

Table 2 Feared SPS and strategies categories with descriptives (N = 800)
Full size table

Feared SPS Strategies

The associated behavioral strategies in which participants indicated they were actively engaging to avoid their future feared selves were examined (see Table 2). Reported strategies ranged in specificity, with some strategies being detailed and well-articulated, whereas others were vague and were not effective behavioral plans for goal attainment. Feared SPS strategies were reported on average of 2.08 (SD = 1.53; range = 0–9). Ten themes within the strategies emerged and included: abstinence (e.g., “Dating others that share my belief of no sex before marriage”), interpersonal relationship (e.g., “Staying faithful”), sexual health/well-being (e.g., “Getting a better birth control option”), self-assertion/protection (e.g., “Being more assertive”), restricting boundaries (e.g., “Staying clear of certain types of people”), expanding boundaries (e.g., “Stepping out of my sexual comfort zone”), thoughtful/informed decisions (e.g., “By making smart choices”), making opportunities (e.g., “Socializing more”), self-improvement (e.g., “Allowing myself to think that I am beautiful, regardless of societal standards of beauty”), and no strategy (e.g., “No”). Frequency comparisons were conducted for participants who reported at least one feared SPS strategy. The most frequently reported strategy category was sexual health/well-being with 41.6% of the sample reporting at least one strategy to avoiding poor sexual health selves (range = 0–6). Similarly to expected SPS strategies, the least frequently reported feared strategy was abstinence strategies (6.4%; range = 0–3), indicating that participants who had goals around abstinent behaviors had the least strategies to attaining these abstinent future selves.

Quantitative Comparisons

Demographic comparisons were conducted to examine potential differences in expected SPS, feared SPS, and the associated behavioral strategies (see Table 3). Specifically, demographic differences based on college attendance status, gender, and gender differences based on college attendance status were examined through a series of Chi-square tests. These tests allowed us to examine group differences based on whether or not a SPS was reported for each category, along with examining differences in reported behavioral strategies for attaining or avoiding these future selves. Due to limited expected SPS responses within the categories of family formation (n = 15; 1.9%) and decreasing/avoidance (n = 14; 1.8%), these categories were not included in our analyses. Due to length of analyses, we only present significant results in this manuscript (contact first author for full results).

Table 3 Chi-square comparisons for SPS categories based on college attendance, gender, and gender by college attendance (N = 796)a
Full size table

College Attendance Status

We first examined how expected SPS, feared SPS, and behavioral strategies varied by college attendance status (i.e., college-attending versus non-college) to compare for differences in these facets of sexuality based on differences in developmental pathways. Overall, only one expected SPS category was found to differ by college attendance status. A greater proportion of college-attending emerging adults (26.2%) reported expectations around interpersonal relationship SPS than non-college emerging adults (16.9%). No differences were found for expected SPS strategies, feared SPS, or feared SPS strategies based on college attendance status.

Gender

Next, we examined how expected SPS, feared SPS, and the associated behavioral strategies varied by gender of the full sample (i.e., men versus women, n = 796). Five expected SPS categories were found to vary by gender. Specifically, a greater proportion of men (45.4%) reported quantity-related SPS than that of women (28.7%), and a greater proportion of men (15.0%) reported sex-specific-related SPS than women (7.3%). In contrast, a greater proportion of women (28.7%) reported quality-related SPS than that of men (16.1%), a greater proportion of women (24.0%) reported interpersonal relationship-specific SPS than that of men (17.9%), and a greater proportion of women (20.4%) reported sexual health-related SPS than that of men (11.8%).

Four expected SPS strategy categories were found to differ between men and women. A greater proportion of women (25.9%) reported strategies related to interpersonal relationship than that of men (14.6%), a greater proportion of women (20.2%) reported strategies related to expanding boundaries than that of men (10.2%), and a greater proportion of women (17.0%) reported strategies to attain positive sexual health/well-being expectations than that of men (8.2%). Lastly, a greater proportion of men (36.7%) reported strategies related to making opportunities to reach their expected SPS than that of women (27.5%).

Feared SPS and strategies were also examined for differences between men and women. Three feared SPS categories were found to differ proportionally. A greater proportion of men (10.7%) reported fears related to quantity SPS that of women (5.1%), and a greater proportion of men (7.1%) reported no fears than that of women (2.9%). In contrast, a greater proportion of women (14.3%) reported fears related to sexual assault/coercion than that of men (6.3%).

Three feared SPS strategy categories were found to differ between women and men. A greater proportion of women (22.3%) reported strategies related to self-assertion/protection SPS than that of men (9.7%) and a greater proportion of women (10.6%) reported strategies for expanding boundaries than that of men (4.8%). However, similarly to expected SPS strategies, a greater proportion of men (14.1%) reported making opportunities strategies than that of women (5.7%).

Gender by College Attendance Status

Lastly, we examined how gender differences by college attendance status varied for each of the SPS categories and behavioral strategy categories. Specifically, we compared college-attending women to non-college women and college-attending men to non-college men for each category. Two expected SPS categories were found to differ based on gender and college attendance status. A greater proportion of non-college men (15.4%) reported sexual health-related SPS than that of college-attending men (7.3%) and a greater proportion of college-attending women (28.2%) reported interpersonal relationship-specific SPS than that of non-college women (19.0%). One expected SPS strategy category was found to differ by gender and college attendance status. A greater proportion of non-college men (9.8%) reported sex-specific strategies than that of college-attending men (3.1%).

One feared SPS category was found to differ by gender and college attendance status. A greater proportion of non-college men (22.2%) reported partner focus-specific feared SPS than that of college-attending men (10.3%). Also, four feared SPS strategy categories were found to differ by gender and college attendance status. A greater proportion of college-attending women (26.0%) reported self-improvement strategies than that of non-college women (17.9%), a greater proportion of college-attending women (13.4%) reported thoughtful/informed decisions strategies than that of non-college women (6.3%), and a greater proportion of college-attending women (26.4%) reported no strategy than that of non-college women (18.4%). In contrast, a greater proportion of non-college women (21.7%) reported interpersonal relationship strategies than that of college-attending women (13.8%).

Discussion

This study examined and compared sexuality-related expectations and fears based on differences in developmental pathways during the life course period of emerging adulthood. Our study extended the limited literature examining sexual possible selves among college students to a non-college emerging adult sample. Findings from our study allowed us to determine whether differences in sexual expectations, fears, and behavioral strategies existed for those who have and have not attended college. Further, findings from this study, as we explain below, help to inform sexual education and intervention programs for developmental appropriateness based on differences in developmental pathways.

Sexual Possible Selves and Strategies

To address our first and second research questions, we replicated and extended previous SPS research (Anders & Olmstead, 2018, 2019; Anders et al., 2017) by examining expected and feared SPS themes using a large sample of college-attending and non-college emerging adults. Variations in SPS categories were found in the qualitative examination of and frequency counts within these categories. Whereas the original examination by Anders et al. (2017) found six expected SPS categories, 11 expected SPS categories emerged from our larger, more diverse sample. Consistent with previous research, the categories of abstinence, interpersonal relationship, sexual health/well-being, and explore/experiment were found; however, our findings also indicated expectations focused on quantity of sexual activity, quality of sexual experience, sex-specific behaviors, choosing to maintain their current sexual self in the upcoming year, family formation, and decreasing/avoidance sexual behaviors.

Additionally, differences in feared SPS were found between our study and Anders et al., 2017 study. Although several similar categories were found in both studies (i.e., abstinence, sexual health/well-being, sexual assault/coercion, interpersonal relationship), additional themes were identified in our study. Specifically, participants’ SPS revealed fears around decreased quantity of sexual activity (i.e., quantity), poor quality sexual experience (i.e., quality), internalized fears focused on their own sexuality (i.e., self-focused), externalized fears focused on partner characteristics (i.e., partner focused), fears of engaging in sex-specific behaviors (i.e., sex specific), and fears regarding behaviors that could increase their sexual risk (i.e., increased sexual risk).

Our study followed Anders and Olmstead (2018) in addressing a limitation in previous SPS research by coding behavioral strategies separately from their associated expected and feared SPS. For example, Anders et al. (2017) found if a strategy for attaining a future abstinent expectation was reported, then the strategy would also be coded as “abstinence.” However, Anders and Olmstead (2018) revised this protocol to code these strategies as unique categories serving as more representative of the strategy rather than requiring them to be coded the same category as the associated SPS. Our study allowed themes within the strategies to be identified naturally and to be coded separately from their associated SPS (e.g., making opportunities). Similar to Anders and Olmstead (2018), our study did find some strategies matched participants’ associated SPS, including abstinence, interpersonal relationship, sexual health/well-being, and sex specific; however, our analyses also indicated that individuals were engaging in strategies that increased opportunities for goal attainment (i.e., making opportunities), strategies of trying new activities or expanding their sexual boundaries (i.e., expanding boundaries), strategies focused on avoidant or restrictive behaviors (i.e., restricting boundaries), strategies focused on dating behaviors (i.e., dating), and strategies for improving facets of an individual’s internalized self (i.e., self-improvement). This approach to coding strategies allowed us to show that the emerging adults in our study used a variety of strategies for attaining and avoiding their future selves.

An unanticipated finding from our study was that participants frequently reported enacting the same type of behavioral strategies (e.g., restricting boundaries) regardless of whether they were articulating an expected or feared SPS. This indicates that strategies may serve the dual purpose of helping individuals to attain their future expected SPS and avoid their future feared SPS. These findings are consistent with the concept of balance (i.e., having both an expectation and fear in the same category) of possible selves (Oyserman et al., 2004). Previous possible selves studies have shown that having balanced expected and feared possible selves and strategies can increase the likelihood of goal attainment (Oyserman & Markus, 1990).

College Attendance

Our third and fifth research questions extended SPS research to examine the expectations and fears of emerging adults who do not attend postsecondary education (i.e., college). College-attending and non-college emerging adults have been found to differ in their perceptions and experiences of this developmental period (Hendry & Kloep, 2010). These divergent experiences may be reflected in the possible selves participants expect for themselves, as possible selves are thought to be reflective of one’s current developmental stage (Frazier, Hooker, Johnson, & Kaus, 2000) and social context (Oyserman & Fryberg, 2006). To date, no study has examined these developmental differences in expected and feared SPS. Our study indicated that similar sexual selves are salient for both groups regardless of college attendance, as the majority of their expected and feared SPS were reported equally. However, we did find differences in expected SPS for several categories, particularly when we made comparisons based on gender subgroups (e.g., college-attending men versus non-college men).

Differences found between these two groups included expectations for interpersonal relationship SPS for the upcoming year. College-attending participants reported more expectations for sexuality in interpersonal and romantic relationships than did non-college emerging adults. Many of these participants, particularly college-attending women, articulated expectations around entering a romantic relationship or engaging in sexuality within a romantic relationship. This interpersonal relationship focus may be reflective of a relationship trend among emerging adults, in which young adults engage in a series of brief, committed relationships (Arnett, 2015). As these relationships are more transient than long-term committed relationships (Arnett, 2015), access to potential partners may play a role in the likelihood of romantic relationship formation. College students may have access to more potential partners in the college environment and may be afforded more time for focusing on entering romantic relationships and navigating sexuality in romantic relationships than their non-college peers. This explanation may be particularly relevant for college-attending men as they articulated more strategies to attaining their future interpersonal relationship goals than non-college men.

In contrast, non-college emerging adults may be more focused on non-committed sexual encounters, such as casual sex (Bailey et al., 2008). Although college is often referenced as an environment for sexual exploration and engagement (e.g., high rates of hooking up; Fielder & Carey, 2010), casual sex experience has been found to be more prevalent among non-college individuals (Bailey et al., 2008). Additionally, our sample of non-college emerging adults reported being in more non-exclusive romantic relationships than their college-attending peers. Non-college emerging adults may be more focused on sexual experiences outside of committed relationships, thus leading to fewer expectations about interpersonal relationships. For example, our findings indicated that non-college men may be more focused on sexual partner characteristics, particularly those of casual sex partners, than college-attending men.

The increased pool of potential sexual partners characteristic of college campuses in conjunction with expectations to hookup may lead to increased relationship fears. Contextual factors, such as the college environment, can play an important role in sexual and relationship goals and behaviors (Olmstead et al., 2018). College environment factors such as access to a variety of romantic alternatives and a culture of alcohol and substance use can contribute to negative interpersonal relationships outcomes (Norona, Olmstead, & Welsh, 2018). Further, independence has been shown to be a developmental task of emerging adulthood (Lerner, Theokas, & Jelicic, 2005), particularly among college students who are afforded an increased moratorium to explore their sexual and relationship identities (Arnett, 2015); however, feeling that one’s independence needs (such as autonomy and identity) were not being met by a romantic partner was found to be an important reason for engaging in infidelity in a diverse sample of emerging adults (Norona et al., 2018). The college environment may not foster skills for maintaining healthy relationships and avoiding negative behaviors that would contribute to feared interpersonal relationship SPS, such as infidelity. These factors can also decrease feelings of efficacy or limit behavioral strategies for emerging adults, particularly those in college. Our findings support this perspective, as a greater proportion of non-college women reported behavioral strategies for avoiding their feared interpersonal relationship selves than college-attending women.

Gender Differences

To address our fourth research question, we examined differences in expected and feared SPS, and the associated behavioral strategies, by gender. Differences based on gender have been prevalent in sexuality research (Peterson & Hyde, 2010) and can be influential on a numerous facets of sexuality, including sexual identity development (Worthington, Savoy, Dillon, & Vernaglia, 2002), hooking up (Owen, Rhoades, Stanley, & Fincham, 2010), or SPS (Anders et al., 2017). Similar to previous studies (Anders & Olmstead, 2018, 2019; Anders et al., 2017), we found differences based on gender for multiple expected and feared SPS.

An interesting gender difference from our study focused on the categories of quantity and quality. Men reported more quantity SPS overall (both expected and feared) than women. Our findings may be explained by examining differences in motivations to have sex for emerging adult men and women, as these motivations may shape their sexual selves. Previous research has examined motivations for sexual intercourse (e.g., Leigh, 1989) and found differences between men and women on their reasons for having sex (Patrick, Maggs, & Abar, 2007). Men have been found to reference self-focused goals such as sexual opportunism and sexual goals as motivators for having sex (Eyre & Millstein, 1999; Patrick et al., 2007), whereas women indicate sexual intimacy and partner focus as motivations for sex (Patrick et al., 2007). Quantity focused selves may be indicative of placing importance on attaining a specific sexual goal for their sexual identity. In contrast, women may place more emphasis on quality or intimate experiences for their sexual identity.

Consistent with Anders et al. (2017), gender differences were also found for the category of sexual assault/coercion. In our study, a greater proportion of women were found to report feared SPS focused on avoiding rape or sexual pressure in the upcoming year. Fears within this category focused on both explicit fears around being sexually assaulted or raped, and fears of social or partner pressure to engage in acts in which they are not comfortable (e.g., “Having forced sex through pressure”). Of the sample, 14.3% of women indicated fears around these sexual assault or coercive behaviors, and these fears were reported similarly between college-attending and non-college women. Sexual assault is prevalent for both college-attending women (e.g., 25% of college women; Exner & Cummings, 2011) and non-college women (e.g., 1.2 times that of college women; Sinozich & Langton, 2014). Due to these high prevalence rates, women are often exposed to and educated about sexual assault and its prevalence in sexual education and media. Salient possible selves, including SPS, are often shaped by social norms or prevalent social scripts to which an individual is exposed (Oyserman & Markus, 1990). This emphasis on sexual assault among women may be socializing women to adopt these feared sexual selves. In contrast, discussion of sexual assault among men is less frequent and underemphasized in sexual education and media (Zweig et al., 1997). This lack of discussion may lead to a reduced focus on this form of sexual aggression for men’s sexual selves.

Limitations

Despite the importance of our findings, this study had several limitations. Our sample was not randomly selected, as we required several inclusion/exclusion criteria to be met for each sample. Although utilizing online crowdsourcing recruitment sites such as MTurk allows for increased access to wider samples as participants can be drawn from across the 50 United States, there are limitations with this form of data collection. Although the demographics of the MTurk sample pool have been shown to allow greater access to heterogeneous populations, particularly compared to college samples (Ipeirotis, 2010), evidence suggests that the majority are typically male, White non-Hispanic, and middle- to upper-class individuals (see Ipeirotis, 2010). Although closer to being a representative sample, our sample was fairly homogeneous in several ways demographically (e.g., race/ethnicity). Future SPS research would benefit from specifically focusing on an examination of expectations and fears among minority populations, including racial and ethnic minorities, as differences in emergent SPS and strategies may be indicative of variance in the impact of racial or ethnic context on sexual development of these populations. Although educational status (e.g., college attendance) can be used as an indicator of class, future research would benefit from directly focusing on the impact of socioeconomic status on sexual development through other more direct forms of class indicators (e.g., parental income) or subjective social status to identify the impact of these.

Participants’ responses were collected using an online cross-sectional survey on a crowdsourcing site. This form of data collection can limit exposure to a variety of participants as recruitment was limited to the MTurk site and individuals must be registered as a “worker” for MTurk to be able to participate in the survey. Further, multiple measure surveys can create priming effects due to survey/question order leading to primed responses (i.e., biased responses). To account for this priming effect, our survey first collected participants’ responses to the sexual possible selves measure before presenting them with other survey measures to reduce the potential for biased responses. Additionally, due to site error, data was only collected on participants’ age range and not collected on participants’ exact age; this error limits our ability to present findings on the mean age of the sample. Lastly, online surveys limit our ability to ask follow-up questions and clarification of participants’ responses. Mixed-method analyses, such as content analysis, benefit from utilization of member checking (Creswell, 2007) which allows researchers to ask for participants’ confirmation of themes that emerge from analyses.

Implications

Our study addressed several limitations in SPS research and our findings hold important implications for future sexuality research. As SPS research is relatively new, there is limited research examining expected and feared SPS among samples outside of higher education settings. Findings from our study addressed this limitation by showing a similarity between these two samples on the majority of sexual expectations and fears during the period of emerging adulthood. As SPS may be influenced by current developmental stage (Frazier et al., 2000), future research would benefit from extending beyond college attendance status as an indicator of developmental differences. Instead, examining SPS in relation to individuals’ experience or stage of sexual identity development (e.g., sexual identity commitment) may be indicative of differences in developmental life stage. As previous research has shown unique differences in sexual development based on sexual or gender orientation (e.g., Glover, Galliher, & Lamere, 2009), future research would benefit from focusing on the SPS of non-heterosexual or cisgender emerging adults. This research might explore potential differences in both expectations and fears, while also highlighting potential gaps in behavioral strategies for reaching these future selves. Examining these gaps may help to inform skill-building within sexual education programs, particularly among those who may have decreased exposure to or resources for improving their strategies (e.g., lower SES individuals). Further, such a study would provide greater insight into beneficial areas of education or interventions to be identified and aimed at sexual and gender minority students within sexual education settings (e.g., addressing gaps in STI prevention methods for sexual minority individuals).

Our findings also hold implications for sexual education among emerging adult populations as fears around sexual assault continue to be prevalent during this developmental period, including among non-college emerging adults. For example, one study found that around 11.5% of non-college emerging adults reported experiencing sexual aggression in the past year, as compared to 10.3% of their college-attending peers (Buddie & Testa, 2005); however, sexual education for emerging adults is often provided at postsecondary institutions, and thus might lead to deficits among those individuals who do not attend college. Continued emphasis on community level sexual education and violence prevention is important to reduce these rates among non-college emerging adults.

Our sample emphasized fears for both explicit acts of sexual assault (e.g., rape) and implicit acts (i.e., sexual coercion). Research has shown that sexual coercion education is often limited within education programs (Jozkowski, Peterson, Sanders, Dennis, & Reece, 2014). These findings emphasize the continued need for consent education focused on teaching protective and negotiation skills (e.g., sexual negotiations) that may improve emerging adults’ efficacy against sexual pressure. Research indicates limited consent education programs outside of higher education settings (Jozkowski et al., 2014); therefore, community settings would benefit from the development and implementation of sexual consent programs to reach emerging adults who opt out of college or continued education.