Measuring gender segregation in emerging adulthood: why it matters how we ask

Introduction: Gender segregated peer groups serve as an important context of socialization. The prevalence of gender segregation has typically been assessed via peer nominations, but this method has limitations.

Methods: The current study introduces two new measures of gender segregation—actual time spent with same-gender peers and ideal amount of time spent with same-gender peers—and compares the prevalence of gender differences in gender segregation on each measure to an established peer nomination measure. College students (N = 332; 58.4% women; 78.9% White; ages 18–29, M = 19.97) completed an online survey.

Results: The prevalence of gender segregation differed significantly depending on participant gender and measurement type, with men reporting more gender segregation than women on the actual time measure, whereas no gender differences were found for the peer nomination or ideal amount of time measures.

Discussion: These findings demonstrate that measurement choice can meaningfully shape researchers understanding of gender segregation. Relying solely on peer nominations may obscure important gender differences in peer interactions. By introducing new measures of actual and ideal time spent with same-gender peers, this study provides researchers with a behaviorally-based tool that can yield a more complete understanding of gendered social relationships and their developmental consequences across the lifespan.

1 Introduction

Gender segregation, the tendency to affiliate primarily with same-gender peers, is present in early childhood, persists into adulthood (Fabes et al., 2019; Mehta and Strough, 2009; Thorne and Luria, 1986), and is observed across cultures (Mehta and Smith, 2023). Same-gender peer groups serve as a powerful context for gender socialization. Peers shape social behaviors, communication styles, and gender-typed preferences through modeling, reinforcement, and negotiation of gender norms (Maccoby, 1998; Leaper, 2000; Furman and Rose, 2015). Yet, same-gender peer groups can also reinforce ingroup bias by shaping attitudes and stereotypes about both one's own gender and that of others (Bigler and Liben, 2007; Leaper, 2022). Persistent gender segregation may limit opportunities to hold meaningful connections with other-gender peers.

Developmental intergroup theory posits that gender segregated peer groups reflect the perceptual salience and social relevance of gender as an implicit and explicit category for organizing people and the environment (Bigler and Liben, 2007). Once formed, these categories provide a foundation for strengthening affective ties toward the ingroup and developing stereotypes and prejudice toward the outgroup. Positive interactions with outgroup members can help to reduce prejudice and promote more flexible attitudes, but a lack of contact perpetuates the development of rigid attitudes and stereotypes (Allport, 1954; Mikołajczak et al., 2025). To the extent that spending time in gender segregated groups limits contact with other-gender peers, gender segregation can be a mechanism through which gendered attitudes and stereotypes are reproduced. This underscores the importance of examining gender segregation in ways which capture individuals' behavioral and social realities.

Early work with younger children often relied on naturalistic observations of play to document the frequency and duration of interactions with same- vs. other-gender peers in classrooms, playgrounds, and other structured settings (Bianchi and Bakeman, 1978; Boyatzis et al., 1999; Fabes et al., 2003; Halim et al., 2021; Howes, 1988; Maccoby, 1998; Whiting and Edwards, 1973). For example, Howes (1988) and Fabes et al. (2003) observed preschool aged children's naturally occurring free-play behavior, recorded whether children interacted within same-gender or mixed-gender peer groups, and calculated proportion scores for same- and mixed-gender play. These types of observational methods are well-suited for studying children because children spend a large portion of their time in structured environments that are readily observable.

During adolescence, social networks broaden and peer interactions increasingly take place in informal or unsupervised settings which make observational methods less feasible (Brown, 2004; Rubin et al., 2007). Perhaps as a consequence, research on adolescent gender segregation has relied on peer nomination methods. In these studies, participants list the names and genders of a specified number of peers, typically five or more, with whom they would like to hang out or enjoy spending time with. Research based on peer nominations has consistently found that the majority of peers nominated by adolescents are of the same gender as the respondent (Keener et al., 2013; Mehta et al., 2017; Mehta and Strough, 2010; Mehta and Wilson, 2020; Poulin and Pedersen, 2007; Reeder, 2003; Strough and Covatto, 2002).

During the transition to adulthood, a developmental period referred to as emerging adulthood, individuals have considerable freedom to choose the peers with whom they interact. Serious romantic relationships become more prevalent (Arnett, 2000), but peers remain vital due to delays in the timing of marriage for current cohorts of emerging adults (DiDonato and Strough, 2013b). Even young adults who have other-gender romantic partners are more likely to seek support from their same-gender peers (Barbee et al., 1990; Monsour, 2001).

Attending college presents numerous opportunities for intergroup contact, yet aspects of campus culture may reinforce gender segregation (Leaper, 2022). For example, Greek life organizations, dormitories, and athletic teams are often gender segregated (Speer, 2017). Many academic majors are dominated by one gender as seen in women's greater representation in social sciences and humanities and men's greater representation in math, science, and business-related fields (Hawrami and Williams, 2025; National Center for Education Statistics, 2024). To the extent that these contexts shape daily peer interactions, they may limit opportunities for intergroup contact (DiDonato and Strough, 2013b; Speer, 2017). Indeed, similar to adolescents, college students' peer nominations typically indicate a greater proportion of same gender peers than other-gender peers (DiDonato and Strough, 2013a; Mehta and Wilson, 2020; Mehta et al., 2017; Wong et al., 2018). For example, Mehta et al. (2014) found that approximately 70% of nominated peers were of the same gender as the respondent. Although gender differences indicating greater gender segregation among men than women are sometimes found, the general pattern of nominating more same-gender peers than other-gender peers holds for both men and women (DiDonato and Strough, 2013a; Lemaster et al., 2015).

Nomination-based measures have been foundational in documenting the persistence of gender segregation into adulthood, but such measures have notable limitations. Nominations may not necessarily reflect actual time spent with peers but instead may represent desired or aspirational relationships (Strough and Covatto, 2002; DiDonato and Strough, 2013a; Mehta and Strough, 2010). Further, although some peer nomination measures specify contexts for nominations, such as home or school (Strough and Covatto, 2002), more typically participants are asked to nominate peers for informal social interactions. For example, Mehta and Strough (2010) asked participants to nominate peers that they would invite to 'hang out' such that nominations reflected desired affiliations rather than measuring time actually spent with same-gender peers. Affiliative preferences may or may not correspond to actual time spent. Time spent with peers may better represent opportunities for socialization than nomination measures. Indeed, time spent with same-gender peers across varying contexts has been proposed as one reason why peers are powerful socialization agents (Leaper, 2022; Mehta and Strough, 2009), making time spent with same-gender peers important to understand.

To address limitations of prior research that relied solely on peer nominations, the present study introduced two new self-report measures of gender segregation and examined their associations with a traditional nomination-based measure. First, the Actual Time Measure, which was adapted from the American Time Use Study (U.S. Bureau of Labor Statistics, 2021), captured how much of their leisure time individuals actually spent with same-vs. other-gender peers within 24 h. We focused on leisure time rather than total time because leisure contexts provide greater autonomy in choosing social partners and more accurately reflect individuals' voluntary peer interactions across different settings. Unlike obligatory activities, such as classes or work, leisure time captures the social environments in which individuals can actively shape their gendered peer experiences.

Second, the Ideal Time Measure assessed how individuals preferred to allocate their leisure time across same- and other-gender peers in that same 24-h period. This measure was meant to complement the Actual Time Measure by allowing individuals to specify their preferred time use. Together, these measures extend beyond peer nomination by assessing both behavioral (actual time) and preferential (ideal time) dimensions of gender segregation and grounding these measures in time-based perspective.

We designed the Actual and Ideal Time measures to provide detailed information about how individuals spend their time and the types of relationships involved. By correlating peer nominations with both actual time use and ideal time preferences, we aimed to clarify whether prior work based on nominations reflected enacted social behavior or preferred social behavior. Determining whether nominations capture preferences or actual time use is a key step in advancing the study of gender segregation. If nominations primarily reflect preferences, they may be better conceptualized as ingroup affiliation or a gender-typed attitude linked to how individuals approach relationships, work, and social interactions across contexts (Leaper, 2022; Mehta and Strough, 2009). If nominations primarily reflect behavioral tendencies, then they may be conceptualized as an indicator of opportunities for peer socialization. Because peer nomination measures of gender segregation have been linked to important social attitudes, including sexism (Jenkins et al., 2023; Keener et al., 2013) it is crucial to clarify whether peer nominations reflect actual behavior or merely preferences. If nominations correspond mostly to preferences, then this could suggest that associations between nominations and attitudes such as sexism may reflect desired affiliations rather than engagement in peer interactions.

In summary, the goal of this study was to clarify how measurement shapes our understanding of gender segregation and provide tools that better reflect real-world social interactions. To achieve this goal, we investigated whether the prevalence of gender segregation varied depending on the measure of gender segregation (nomination vs. actual time vs. ideal time) and an individual's gender. Our first research question asked, are peer nominations associated with Actual Time and Ideal Time measures of gender segregation? We hypothesized that patterns of gender segregation would be more similar between the Peer Nomination and Ideal Time measures than between the Peer Nomination and Actual Time measures. Our second research question asked, does the prevalence of gender segregation vary by individuals' genders and the measure of gender segregation (nomination, actual time, ideal time)? Based on prior research, we expected that there would be a main effect of gender, such that men would report higher levels of same-gender peer engagement than women (DiDonato and Strough, 2013a; Lemaster et al., 2015).

2 Methods

2.1 Participants

The analyses were based on 332 undergraduate students from a large, public, mid-Atlantic university in the United States (M_age = 19.97, SD = 1.90; 58.4% women, 41.6% men; 78.9% White; 4.5% African American/Black; 4.5% Latino; 7.2% Asian; 80.1% heterosexual). Gender identity was assessed by asking participants to report their current gender identity. Participants selected from a multiple-choice item that included woman, man, transgender woman, transgender man, non-binary or gender-fluid, and Two-Spirit. Participants also had the option to self-describe their gender identity in an open-ended field. This gender identity variable was used in all analyses. Sex assigned at birth was assessed in a separate item.

The final analytic sample was based on 58.4% individuals who identified as female and 41.6% who identified as male. The analytic sample reflected that of the 524 students who began the online survey (Mage = 20.28, SD = 3.29; 44.1% women, 33.2% men, 2.9% non-binary; 63.7% White; 3.6% African American/Black; 3.6% Latino; 5.7% Asian; 61.5% heterosexual), a total of n = 192 participants were excluded for various reasons (see Supplementary Figure 1 for a flow chart depicting selection of the analytic sample). Reasons for exclusion were as follows: not answering the demographic questions such that their data could not be used in the primary analyses (n = 33), falling outside the target age range of emerging adulthood due to being older than 29 years (n = 7), and concerns about data quality as indicated by extreme completion times of less than 10 min or more than 62 min (n = 136). Additionally, fifteen participants who identified outside of the gender binary were excluded from the analyses because the sample size was too small to allow adequately powered comparisons despite targeted recruitment attempts aimed at LGBTQ campus groups and other organizations. Descriptive information on key study variables for non-binary participants is reported in Supplementary Table 1.

Descriptive statistics for the final analytical sample are presented in Table 1 and for the full sample before exclusions, in Table 2. Descriptive statistics were consistent across the analytical and full samples suggesting that exclusions did not substantially alter the distributions of study variables. Specifically, exclusions based upon survey completion time affected a small number of non-white participants, but the overall racial composition of the sample was similar after exclusions were applied. Supplementary analyses indicated identical results when those with extreme completion times were included in the analytical sample, suggesting that the findings were robust to this criteria for exclusion.

Table 1

Table 1. Descriptive and frequency statistics for demographic variables.

Table 2

Table 2. Descriptive statistics for demographics and main study variables of full sample.

Participants were recruited using a combination of printed and digital flyers, classroom announcements, and outreach through university sources including academic advisors, instructors, and residence hall communications. Participants chose to receive either a $10 Amazon gift card or extra credit in a psychology course in exchange for participation. The vast majority selected the gift card (n = 310) over extra credit (n = 15), and a few (n = 7) did not provide a response regarding compensation. When analyses were restricted to those who selected the gift card, the significance of the results reported in the following sections was the same. The study protocol was approved by the university's Institutional Review Board (IRB# 2401905093), and all participants provided informed consent before participation.

2.2 Procedure

The survey was administered via Qualtrics and distributed through the university's SONA research system. After providing informed consent, participants completed the survey which was fully anonymous and de-identified. Survey measures were presented in the order listed below with demographic questions appearing at the end. The survey took approximately 20–30 min to complete. The Actual and Ideal Time Assessments were piloted with a small group of student volunteers before full data collection to ensure clarity and usability.

2.3 Measures

2.3.1 Gender segregation

This study included three measures of gender segregation: a traditional peer nomination measure and two newly developed time-based measures assessing participants' actual and ideal time spent with same-gender peers.

2.3.1.1 Peer nomination

The first measure of gender segregation was the established peer nomination measure commonly used in prior research on adolescence and emerging adulthood (e.g., Keener et al., 2013; Mehta and Wilson, 2020). Participants were asked to nominate up to five of their closest friends and to indicate each friend's gender (male, female, non-binary, or prefer not to say), see Supplementary Table 3 Appendix A for the full measure. A proportion score was calculated by dividing the number of nominated friends who shared the participant's gender by the total number of nominations (95.8% of participants nominated 5 friends). For example, if four out of five nominated friends were the same gender as the participant, the gender segregation score was 0.80 (see Table 3).

Table 3

Table 3. Descriptive statistics for main study variables.

2.3.1.2 Actual time assessment

Participants completed a structured 24-h recall measure designed to assess how they actually spent their time the previous day (Kotvas, 2025) (see Supplementary Table 3 Appendix B for the full measure). The measure mirrored the structure of the American Time Use Survey (U.S. Bureau of Labor Statistics, 2021). Participants were instructed: “Please reflect on yesterday and think of how you spent your time and who you were with. Splitting your time into hours, please describe how long you spent doing each activity, what kind of activity it was, and who you were with. You will be asked to categorize your activity as either leisure or obligatory.” Participants reported durations in minutes for each activity, and indicated whether the activity was leisure or obligatory, who was present during the activity and their gender. The total leisure minutes was calculated by summing the duration of all activities labeled as leisure.

Although participants were asked to report activities covering a full 24 h, the assessment did not restrict answers to total 24 h. Due to this limitation in the measure, the reported total hours of leisure time ranged from 1 to 65.5. On average, participants reported 1,431.79 min (23.85 h; SD = 558.10), suggesting that some participants under- or over-reported the intended period. Histograms of the Actual Time assessment are presented in Supplementary Figure 2. This variability was accounted for by using proportion scores, which capture the relative allocation of leisure time with same-gender peers to provide a meaningful behavioral index of gender segregation.

Participants reported the number of people present during the activity by selecting one of 13 categorical options (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11–20, 20+, or alone). Participants reported the gender composition of those present during the activity by selecting from a drop-down menu that included the following options: women; men; non-binary or gender-fluid; mixed-gender: mostly women; mixed-gender: mostly men; mixed-gender: mostly non-binary or gender fluid; equal women and men; equal women and non-binary or gender-fluid; equal men and non-binary or gender-fluid; equal women, men, and non-binary or gender-fluid; alone; and mixed-gender: unsure.

Gender segregation was operationalized by calculating the proportion of leisure time, in minutes, that participants spent with same-gender friends and roommates. To calculate the score, first, for each leisure activity, a binary “same-gender peer” variable was created, coded as 1 when the reported friend or roommate's gender matched the participant's self-identified gender and 0 when it did not. Only activities that the participants categorized as exclusively involving same-gender others (women or men) were coded as 1. That is, activities that were categorized by participants as mixed-gender or various combinations of same- and other gender peers were coded as zero. We focused on activities that were classified as exclusively same gender so that scores were directly comparable to the peer nomination measure. That is, the peer nomination measure yielded a score that represented the proportion of exclusive same-gender nominations, not mixed-gender nominations. Thus, aligning the time-use measure with this operationalization allowed for direct comparisons across measures.

After intervals with same-gender peers were identified, the total duration of leisure time with same-gender peers was calculated by summing the durations of all same-gender peer leisure activity intervals. Next, total leisure time with all friends and roommates, regardless of gender, was summed across all intervals. Finally, to calculate the same-gender Actual Time use proportion score, the total duration of actual time spent in leisure activity with same-gender friends or roommates was divided by the total duration of actual time spent in leisure activity. For example, if a participant spent a total of 480 min (8 h) in leisure activities and 300 min (5 h) of those hours were with same-gender peers, the proportion score was 0.63. Descriptive statistics for the Actual Time Assessment are reported in Table 3. Higher scores reflected more time actually spent with same-gender peers during leisure time. As a robustness check, we computed an alternative score that assessed time spent in activities where at least most of the others who were present were the same gender as the participant. This alternative score was computed by summing time spent across all intervals that participants reported were mixed gender, but mostly the same gender as the participant and adding this sum to the score for exclusive same-gender intervals. Analyses using this alternative score are reported in Supplementary Table 2. A MANOVA using the combined variable revealed a significant main effect of measurement type. Whereas the main effect of gender and the measure x gender interaction were non-significant. Overall, these findings indicate that the observed interaction observed in the primary analyses was driven by the proportion scores that included only exclusive same-gender interactions.

2.3.1.3 Ideal time assessment

Immediately after completing the actual time assessment, participants were asked to report how they ideally wished to have spent their day (Kotvas, 2025) (see Supplementary Table 3 Appendix C for the full measure). The format and instructions were identical to the Actual Time Assessment, except that participants were asked to reflect on their “ideal” day rather than their actual day. Reported ideal day totals ranged from 45 min to 47 h. On average, participants reported 1,338.88 min (22.30 h; SD = 401.97), indicating that some participants under- or over-reported the intended time period. Histograms of the Ideal Time assessment are presented in Supplementary Figure 2. All coding and scoring procedures for Ideal Time were the same as for the Actual Time scoring. The overall gender segregation score for the Ideal Time assessment was calculated by dividing the total duration of time spent in ideal same-gender leisure activity with friends or roommates by the total duration of ideal time spent in leisure activity. Descriptive statistics for the Ideal Time Assessment are reported in Table 3. Higher scores reflected more ideal time spent with same-gender peers during leisure time.

2.3.2 Demographics

Participants reported their age, gender identity, sexual orientation, race/ethnicity, and class rank. Descriptive statistics for demographic variables are presented in Table 1.

2.3.3 Analyses

We used IBM SPSS Statistics (IBM Corporation, 2022) to conduct all analyses. Before conducting analyses to address our research questions, we tested whether the day of the week participants completed the survey influenced their responses, examined normality of distributions of variables, and tested whether the assumptions of the statistical tests we used were met. Sphericity was assessed and when violated, the Huynh–Feldt correction was applied. To address our first research question, whether peer nominations were associated with the Actual Time and Ideal Time measures of gender segregation, bivariate Pearson correlations were computed among the three gender segregation measures: peer nomination, actual time, and ideal time (see Table 5). To address the second research question, whether the prevalence of gender segregation varied based on participant gender and type of measurement, we conducted a 2 (Participant Gender: women, men) × 3 (Measurement Type: peer nomination, actual time, ideal time) mixed-model multivariate analysis of variance (MANOVA). Participant gender served as the between-subjects factor, and measurement type served as the within-subjects factor. The dependent variables were the proportion scores for the three gender segregation measures, which were standardized (z-scored) to allow for direct comparison across measurement types in the inferential analyses (see Table 4).

Table 4

Table 4. Descriptive statistics for Z-scored main study variables.

3 Results

3.1 Initial analyses

Descriptive statistics for the three gender segregation measures are presented in Table 3. Before conducting the analyses that addressed our research questions, we assessed whether the day of reporting influenced responses on each measure. This analysis was conducted because patterns of social interactions may differ between weekdays and weekends if weekends allow for longer intervals of social activities. We tested the associations between day of the week (coded as 0 = weekend, 1 = weekday) and each gender segregation score. Day of the week was not significantly related to any of the three gender segregation measures; peer nomination (r = 0.05), Actual Time (r = −0.11), Ideal Time (r = −0.05). Participants' reports of gender segregation were not associated with the day of the week that they completed the survey (see Table 5).

Table 5

Table 5. Bivariate correlations for main study variables.

To test whether the variables were distributed normally, we examined skew and kurtosis. All three gender segregation variables were slightly skewed due to the bounded nature of the proportion measures. However, both skewness (< 2) and kurtosis (< 7) remained within acceptable thresholds (Kim, 2013). Scatterplots indicated linear relationships between variables with no extreme univariate outliers. A test for multivariate outliers using Mahalanobis distance did not indicate any multivariate outliers. The sample size (N = 332) was large enough for the Central Limit Theorem and related large-sample approximations to justify treating the sampling distributions of the means and correlation coefficients as approximately normal. For the MANOVA, homogeneity of covariance was confirmed by a non-significant Box's M test (Box's M = 7.67, p = 0.27).

3.2 Research question 1. Are peer nominations associated with the actual time assessment and ideal time assessment measures of gender segregation?

Pearson correlations were computed to examine the associations among the three measures of gender segregation: peer nomination, actual time assessment, and ideal time assessment (see Table 5). It was hypothesized that patterns of gender segregation would be more similar between the Peer Nomination and Ideal Time Assessment measures than between the Peer Nomination and Actual Time Assessment measures. The correlations among the three gender segregation measures were statistically significant: nomination and actual time, r(318) = 0.14, 95% CI [0.03, 0.25], p = 0.01; nomination and ideal time, r(318) = 0.23, 95% CI [0.13, 0.33], p < 0.001; and actual time and ideal time, r(329) = 0.37, 95% CI [0.27, 0.46], p < 0.001.

To test whether the strength of these associations differed from one another, Fisher's r-to-z comparisons were conducted. Contrary to the hypothesis, the correlation between nomination and ideal time was not significantly stronger than the correlation between nomination and actual time (z = −1.30, p = 0.19), although the correlations were in the predicted direction. Additional Fisher's r-to-z tests indicated that the correlation between actual and ideal time was significantly stronger than the correlation between nomination and ideal time (z = −3.13, p = 0.002), but not significantly stronger than the correlation between nomination and actual time (z = −1.82, p = 0.07).

3.3 Research question 2. Does the prevalence of gender segregation vary depending on participants' gender and the type of measurement (e.g., peer nomination vs. actual time assessment vs. ideal time assessment)?

It was hypothesized that there would be a main effect of gender, with men reporting more same-gender peers compared to women. A mixed-model MANOVA was conducted to examine the effect of measurement type (within-subjects factor) and gender (between-subjects factor) on the prevalence of gender segregation. Standardized z-scores for each of the three gender segregation measures were used to account for differences in measurement scaling.

The multivariate test revealed no significant main effect of measurement type, Wilks' Λ = 1.00, F(2, 314) = 0.07, p = 0.93, partial η² = 0.00, or gender, F(1, 315) = 0.03, p = 0.87, partial η² = 0.00. The interaction between measurement type and gender was significant, F(2, 314) = 3.87, p = 0.02, partial η² = 0.02, indicating that gender differences in gender segregation varied across measurement type. Mauchly's test of sphericity was significant (χ² = 12.75, p < 0.01), indicating a violation of the assumption of sphericity; the Huynh–Feldt correction was applied (ε = 0.97), as recommended when ε > 0.75 (Blanca et al., 2023). The results remained robust, all main analyses and the significant interaction were unchanged.

Planned comparisons were conducted to examined to isolate the significant measurement type by gender interaction by examining the univariate ANOVAs. For the Actual Time measure, there was a significant gender difference, F(1, 315) = 4.05, p = 0.045, partial η² = 0.013, with men reporting a greater proportion of leisure time with same-gender peers (M = 0.12, SE = 0.09) than women (M = −0.10, SE = 0.07). The mean difference between men and women on the Actual Time measure was 0.23, 95% CI [0.005, 0.45], Cohen's d = 0.23, indicating a small effect. No significant gender differences were observed for the Peer Nomination F(1, 315) = 2.13, p = 0.15, partial η² = 0.007, Cohen's d = 0.16, or Ideal Time measures, F(1, 315) = 0.03, p = 0.87, partial η² = 0.00, Cohen's d = 0.02. Overall, these comparisons indicated that the gender difference in gender segregation was localized to the Actual Time measure (see Figure 1).

Figure 1

Figure 1. Proportion of same-gender peers by measurement type and gender. Note The figure depicts the standardized (z) scores for the proportion of same gender peers by measurement type and gender. Positive values indicate more than average time spent with same-gender peers or a greater proportion than average of same-gender peer nominations, while negative values indicate less than average. Error bars depict the standard error of the mean. *p < 0.05.

To evaluate the robustness of the observed effects, additional analyses were conducted. First, a sensitivity analysis was conducted by excluding cases where reports of time use exceeded 24 hours. The pattern and significance of the results remained the same (see Supplementary Table 2). Second, we investigated the prevalence of same-gender interactions across all relationships. That is, instead of restricting the analysis solely to peer-only interactions, we computed proportion scores that reflected the total proportion of leisure time spent exclusively with those of the same-gender irrespective of relationship category (Peers, Family, Coworkers, Roommate, Classmate, and Other). When using this indicator of gender segregation in MANOVA, the measurement type by gender interaction was no longer significant, suggesting that this finding was specific to exclusive-same gender peer contexts and do not generalize across other types of exclusively same-gender relationships.

4 Discussion

The goal of this study was to clarify how measurement shapes our understanding of gender segregation and to provide new tools to assess the prevalence of same-gender peer interactions in daily life. Prior research on gender segregation in adolescents and young adults has been based exclusively on peer nominations (Keener et al., 2013; Mehta et al., 2017; Mehta and Strough, 2010; Mehta and Wilson, 2020; Poulin and Pedersen, 2007; Reeder, 2003; Strough and Covatto, 2002). A key contribution of our study is its measurement of gender segregation using multiple methods. In doing so, our findings provide new insights about what is assessed by the peer nomination measures of gender segregation commonly used in studies of adolescents and young adults. Our findings have implications for the use of self-report measures of gender segregation across diverse age groups.

In our study of emerging adult college students, the prevalence of gender segregation as assessed via peer nominations was similar to the prevalence reported in other studies of this age group (DiDonato and Strough, 2013a; Lemaster et al., 2015). That is, approximately 75% of nominated peers were the same gender as the participant. Our findings add to the literature by providing new insights as to what is assessed by nomination measures. Our findings indicate that peer nominations relate to, but do not fully capture, participants' actual behavior of time spent with peers. Time spent in same-gender peer groups is thought to be important for understanding the power of peers as agents of socialization (e.g., Leaper, 2022). Contact with ingroup members is posited to strengthen affiliative ties with the ingroup while increasing stereotypes and prejudice toward outgroup members (Bigler and Liben, 2007). Our findings indicate that peer nominations are an imperfect proxy for time spent in contact with same-gender peers. Instead, nominations appear to be more closely aligned, albeit not significantly so, with individuals' aspirations about the others with whom they would prefer to spend their time. That is, the association between peer nominations and ideal time suggests that nominations may correspond relatively more to perceptions, social ideals, or aspirational behavior, rather than actual time spent with same-gender others in real, daily social routines.

The relatively weaker correspondence between preferences as assessed via nominations and behavior as assessed via actual time use may reflect the influence of social desirability, perceived norms, or self-presentation motives when nominating peers. For instance, college students may report relatively more balanced peer preferences to align with heteronormative concerns about socializing with members of the other gender but gravitate toward same-gender peers in daily life, resulting in a low convergence between measures. Overall, our findings suggest that the three measures capture related but distinct dimensions of gender segregation, with peer nominations being related to but not fully overlapping with individuals' actual leisure behavior. This finding has important implications when testing theories of gender socialization that emphasize time spent in contact with same-gender peers as a key mechanism (e.g., Leaper, 2022; Bigler and Liben, 2007). Sole reliance on peer nominations in such studies may inadequately represent the potential for same-gender peers to model and reinforce gender socialization through day-to-day interactions.

4.1 Gender differences across measures of gender segregation

We considered gender differences in the prevalence of gender segregation across three measures: a behaviorally based measure of actual leisure time spent with same-gender peers and two preference-based measures assessing self-reported ideal time and peer nominations. Significant gender differences were observed only for the actual time measure, men spent more leisure time with same-gender peers than women did, whereas no gender differences emerged for peer nominations or ideal time preferences. This finding implies that men's real-world behavior reflects a greater tendency toward same-gender interaction compared to women even if this difference is not evident in their nominations or preferred time preferences.

The gender self-socialization model (Tobin et al., 2010) proposes that individuals are both shaped by and actively shape their gendered social environments. Men's relatively greater time spent in same-gender groups compared to women may reflect they men are more concerned than women in maintaining group boundaries to protect the status of their ingroup from intrusions by outgroup members (McDonald et al., 2012). Men's greater time spent with other men could shape their gendered behaviors, beliefs, and attitudes through building affiliation with other ingroup members and by providing opportunities for modeling, imitation and reinforcement of actual behavior. Over time, social engagement with other men in exclusively same-gender contexts could shape stereotypic thinking about women as outgroup members (Bigler and Liben, 2007), and could help to explain why gender segregation has been linked to sexist attitudes (Jenkins et al., 2023; Keener et al., 2013). Similar processes may unfold when women interact exclusively with other women. However, the relatively lower prevalence of gender segregation in women's actual time use compared to men suggests that such socialization could be a somewhat less powerful mechanism for women.

The significant interaction between gender and measurement type suggests that measurement matters when assessing gender differences. Some prior research using peer nominations with college student samples had found that men nominated more same-gender peers than women did (e.g., DiDonato and Strough, 2013a; Lemaster et al., 2015), but other studies found no such differences (Mehta et al., 2017). In our study, greater gender segregation among men than women was localized to self-reports of time actually spent with same-gender peers. For the other two measures, this pattern was reversed, although the differences between men and women were non-significant. The present study extends prior research by showing that reported actual behavior may be more likely to show gender differences in social relationships than peer nominations. Overall, the actual time recall measure provides a more behaviorally grounded representation of gender segregation, revealing patterns that peer nominations may not fully capture.

4.2 Implications

Understanding gender segregation requires distinguishing between preferences and actual behavior. Our findings suggest that prior research that used peer nominations to assess gender segregation in adolescents and emerging adults (e.g., Mehta and Strough, 2010; Keener et al., 2013), may have primarily captured adolescents' affiliative preferences for same-gender peers rather than the time they actually spent with them. As such, prior research that has linked gender segregated peer groups to gender-typed attitudes and behaviors may underestimate the strength of the associations if time spent in actual contact with same-gender peers is the underlying mechanism. This highlights that interventions should target not only opportunities for contact with other-gender peers but also underlying preferences and, critically, actual behavior.

These findings have practical significance for reducing gender segregation. Because actual behavior is where gender differences in gender segregation emerged in our study, interventions should focus on changing realistic behavioral routines and structuring social environments, rather than solely addressing perceptions or preferences. Evidence from early childhood programs, such as Meet-Up Buddy-Up, demonstrates that structured, low-stakes cross-gender contact can improve interpersonal skills, reduce gender segregation, and enhance social climates (Fabes et al., 2019; Xiao et al., 2023). Although most prior intervention work has focused on young children, these strategies could be adapted for adolescents and emerging adults through collaborative group activities, academic projects, or social events that encourage meaningful cross-gender interactions. Such interventions are particularly important during adolescence, as individuals are starting to encounter contexts, such as school, workplaces, and romantic partnerships, where effective cross-gender collaboration is essential, and in emerging adulthood when individuals have considerable freedom to choose social partners.

Importantly, these efforts remain relevant well into adulthood, as gender segregation can persist in professional and social contexts, influencing opportunities for collaboration and advancement. For example, adults who primarily associate with same-gender peers may have fewer chances to develop cross-gender communication skills that are essential in workplaces, parenting partnerships, and social engagement (Mehta and Wilson, 2020). Interventions that promote inclusive, mixed-gender teamwork and networking can foster more equitable professional relationships, challenge gender stereotypes, and contribute to greater gender equity in leadership and decision-making roles. Overall, reducing gender segregation can help mitigate stereotypes, enhance prosocial behavior, and promote positive attitudes toward the other gender (Howes, 1988; Halim et al., 2021; Fabes et al., 2003). Importantly, accurate measurement of both preferences and behavior will allow researchers to identify whether interventions are changing attitudes, behaviors, or both, critical information for designing programs that produce lasting change. Future research should continue to examine how actual behavior differs from preference-based measures and how interventions targeting both behavior and social preferences can support more equitable and adaptive gendered social experiences across developmental periods.

4.3 Limitations and future directions

The present study has limitations that should be considered when interpreting the findings. First, although the study used multiple measurement approaches, each method has inherent limitations. For example, peer nomination captures preferences but not actual behavior, while actual and ideal time-use measures rely on retrospective self-report, which may be subject to recall bias or inaccuracies. Although the newly developed time-use measures introduced here offer richer contextual detail than peer nomination alone, they should be refined and validated further. For example, although participants were instructed to report activities over 24 h, self-reported totals frequently deviated from the intended timeframe, with some participants under- or over-reporting the duration of their day. However, these deviations represent a limitation of the time-use measures, as self-reports did not always align with the intended 24-h reporting period. Proportion scores were calculated to account for this variability, although using proportion scores does work to mitigate this issue, it does not fully eliminate measurement error. The proportion scores used in analyses capture the relative distribution of time across social contexts and offer a meaningful behavioral index of gender segregation, deviations from the intended 24-h reporting period may introduce measurement error that was not fully addressed by our use of proportion scores. Future research could address this issue by providing more precise instructions or implementing automated checks to ensure reported daily totals approximate 24 h. This would enhance accuracy and reliability of the measure

Our results must be interpreted in light of the methods used to compute gender segregation scores. Only interactions with exclusively same-gender peers (i.e., women-only or men-only groups) were included. Interactions involving other-gender individuals or those with diverse gender identities were not investigated. This method was adopted to facilitate comparisons to the nomination measure, which was our main focus, but it may mask individual differences in subgroups such as men who affiliate only with other men vs. those who primary affiliations encompass women or non-binary individuals. Future work could be directed toward understanding and identifying such subgroups to provide a more comprehensive understanding of how gender social dynamics unfold in individuals' daily lives.

We excluded fifteen non-binary participants from the main analyses due to the small sample, which prevented statistical comparisons. As such, the generalizability of our findings is restricted to individuals identifying within the gender binary. Future research is needed to better understand peer relationships among non-binary individuals, who may experience different patterns of gender segregation and peer support compared to binary-identifying peers (Salinas-Quiroz and Sweder, 2023; Kurup and Underwood, 2021). Future research is also needed to investigate the extent to which gender segregation differs as a function of culture, sexual orientation, and age. In such research, it will also be important to consider the intersection of these individual difference characteristics and contextual demands.

Although not the primary focus of the current study, our Actual Time and Ideal Time Assessments capture richer contextual details than peer nominations, including the type of peers (same-gender, other-gender, family, romantic partners) and whether activities were leisure or obligatory. This detail allows researchers to examine how peer interactions are embedded within broader patterns of daily life. Future work could explore these contextual differences more fully, compare results across age groups, and examine how interventions can leverage this detailed information to reduce gender segregation and promote positive other-gender interactions.

Replication studies should investigate whether the gender differences observed in self-reports of time use and the associations among peer nomination, actual time assessment, and ideal time assessment are consistent in other samples. Our sample was comprised of emerging adult college students from one institution. As such, findings may not generalize to other samples of emerging adults or to younger or older individuals.

Finally, future research would benefit from longitudinal designs and alternative methodological approaches to better capture how gender segregation develops and changes across time. Future studies could strengthen this approach by incorporating real-time tracking (e.g., ecological momentary assessment) or digital logs of social behavior. Building on the multi-method framework used here, comparing nomination, actual time assessment, and ideal time assessment, will facilitate continued refinement and validation of gender segregation measures to improve reliability and deepen our understanding of how to best measure gender segregation across the lifespan.

5 Conclusion

By introducing two new measures, the Actual Time Assessment and Ideal Time Assessment, and comparing them to an existing Peer Nomination measure, this study highlights how the choice of methodology shapes conclusions about gender segregation in peer groups. Although men and women showed similar levels of gender segregation on peer nominations and ideal time preferences, actual behavior revealed greater gender segregation among men, underscoring the importance of including behaviorally based measures. These findings demonstrate that different measurement approaches capture related but distinct aspects of gendered peer interactions and provide a foundation for future research examining the social and developmental consequences of gender segregation. More broadly, this work offers guidance for researchers and practitioners on selecting measurement tools, while emphasizing the need for continued methodological innovation and diverse sampling to fully capture the complexity of gendered social experiences across contexts and over time.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement

The studies involving humans were approved by WVU Institutional Review Board (IRB). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

Author contributions

KK: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Resources, Visualization, Writing – original draft, Writing – review & editing. JS: Conceptualization, Formal analysis, Methodology, Project administration, Supervision, Writing – review & editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This research was supported in part by funding from the Department of Psychology at West Virginia University. The content of this paper is solely the responsibility of the authors.

Acknowledgments

This manuscript is based upon the author's master's thesis completed under the supervision of Dr. JoNell Strough, whose substantive guidance shaped the project's development and refinement. The authors further acknowledge the contributions of the remaining thesis committee members, Dr. Julie Patrick and Dr. Kathleen Morrison, whose feedback and support strengthened this work.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fdpys.2025.1711011/full#supplementary-material

References

Allport, G. W. (1954). The Nature of Prejudice. Cambridge: Addison-Wesley.

Google Scholar

Arnett, J. J. (2000). Emerging adulthood: a theory of development from the late teens through the twenties. Am. Psychol. 55, 469–480. doi: 10.1037/0003-066X.55.5.469

PubMed Abstract | Crossref Full Text | Google Scholar

Barbee, A. P., Gulley, M. R., and Cunningham, M. R. (1990). Support seeking in personal relationships. J. Soc. Pers. Relat. 7, 531–540. doi: 10.1177/0265407590074009

Crossref Full Text | Google Scholar

Bianchi, B. D., and Bakeman, R. (1978). Sex-typed affiliation preferences observed in preschoolers: traditional and open school differences. Child Dev. 49, 910–912. doi: 10.2307/1128270

Crossref Full Text | Google Scholar

Bigler, R. S., and Liben, L. S. (2007). Developmental intergroup theory: Explaining and reducing children's social stereotyping and prejudice. Curr. Dir. Psychol. Sci. 16, 162–166. doi: 10.1111/j.1467-8721.2007.00496.x

Crossref Full Text | Google Scholar

Blanca, M. J., Arnau, J., García-Castro, F. J., Alarcón, R., and Bono, R. (2023). Repeated measures ANOVA and adjusted F-tests when sphericity is violated: which procedure is best? Front. Psychol. 14:1192453. doi: 10.3389/fpsyg.2023.1192453

PubMed Abstract | Crossref Full Text | Google Scholar

Boyatzis, C.J., Mallis, M., and Leon, I. (1999). Effects of game type on children's gender-based peer preferences: a naturalistic observational study. Sex Roles 40, 93–105. doi: 10.1023/A:1018882417926

Crossref Full Text | Google Scholar

Brown, B. B. (2004). “Adolescents' relationships with peers,” in Handbook of Adolescent Psychology, 2nd Edn., Eds. R. M. Lerner, and L. Steinberg (John Wiley and Sons, Inc., New York), 363–394. doi: 10.1002/9780471726746.ch12

Crossref Full Text | Google Scholar

DiDonato, L., and Strough, J. (2013a). Contextual influences on gender segregation in emerging adulthood. Sex Roles 69, 632–643. doi: 10.1007/s11199-013-0312-1

Crossref Full Text | Google Scholar

DiDonato, L., and Strough, J. (2013b). Do college students' gender-typed attitudes about occupations predict their real-world decisions? Sex Roles 68, 536–549. doi: 10.1007/s11199-013-0275-2

Crossref Full Text | Google Scholar

Fabes, R. A., Martin, C. L., and Hanish, L. D. (2019). Gender integration and the promotion of inclusive classroom climates. Educ. Psychol. 54, 271–285. doi: 10.1080/00461520.2019.1631826

Crossref Full Text | Google Scholar

Fabes, R. A., Martin, C. L., Hanish, L. D., Anders, M. C., and Madden-Derdich, D. A. (2003). Early school competence: the roles of sex-segregated play and effortful control. Dev. Psychol. 39, 848–858. doi: 10.1037/0012-1649.39.5.848

PubMed Abstract | Crossref Full Text | Google Scholar

Furman, W., and Rose, A. J. (2015). “Friendships, romantic relationships, and peer relationships,” in Handbook of Child Psychology and Developmental Science, 1st Edn., eds. R. M. Lerner. New York: Wiley, 1–43. doi: 10.1002/9781118963418.childpsy322

Crossref Full Text | Google Scholar

Halim, M. L. D., Martin, C. L., Andrews, N. C. Z., Zosuls, K. M., and Ruble, D. N. (2021). Enjoying each other's company: gaining other-gender friendships promotes positive gender attitudes among ethnically diverse children. Pers. Soc. Psychol. Bull. 47, 1635–1653. doi: 10.1177/0146167220984407

PubMed Abstract | Crossref Full Text | Google Scholar

Hawrami, R., and Williams, A. (2025). Major Changes: Gender Shifts in Undergraduate Studies Over Time. American Institute for Boys and Men. Available online at: https://aibm.org/research/major-changes-gender-shifts-in-undergraduate-studies-over-time/ (Accessed January 6, 2026).

Google Scholar

Howes, C. (1988). Same-and cross-sex friends: implications for interaction and social skills. Early Child. Res. Q. 3, 21–37. doi: 10.1016/0885-2006(88)90027-0

Crossref Full Text | Google Scholar

IBM Corporation (2022). IBM SPSS Statistics for Windows (Version 29) [Computer software]. IBM Corporation.

Google Scholar

Jenkins, D. L., Xiao, S. X., and Martin, C. L. (2023). Does the gender of your friends matter for sexist attitudes about women? Emerg. Adulthood 11, 380–393. doi: 10.1177/21676968221121165

Crossref Full Text | Google Scholar

Keener, E., Mehta, C., and Strough, J. (2013). Should educators and parents encourage other gender interactions? Gender segregation and sexism. Gender Educ. 25, 818–833. doi: 10.1080/09540253.2013.845648

Crossref Full Text | Google Scholar

Kim, H. Y. (2013). Statistical notes for clinical researchers: assessing normal distribution (2) using skewness and kurtosis. Restor. Dent. Endod. 38, 52–54. doi: 10.5395/rde.2013.38.1.52

PubMed Abstract | Crossref Full Text | Google Scholar

Kotvas, K. L. (2025). Does the time spent with same-gender peers matter? Gender segregation and attitudes about coercive sex in college students (Publication No. 12788) (Master's thesis, West Virginia University). West Virginia University Research Repository. Available online at: https://researchrepository.wvu.edu/etd/12788 (Accessed January 6, 2026).

Google Scholar

Kurup, A. R., and Underwood, M. K. (2021). Gender diversity in peer relations: Best research practices and marshalling peer influence. J. Appl. Dev. Psychol. 76:101328. doi: 10.1016/j.appdev.2021.101328

PubMed Abstract | Crossref Full Text | Google Scholar

Leaper, C. (2000). Gender, affiliation, assertion, and the interactive context of parent–child play. Dev. Psychol. 36, 381–393. doi: 10.1037/0012-1649.36.3.381

PubMed Abstract | Crossref Full Text | Google Scholar

Leaper, C. (2022). “Origins and consequences of childhood gender segregation: toward an integrative developmental systems model,” in Gender and Sexuality Development, eds. D. P. VanderLaan, and W. I. Wong (Cham: Springer International Publishing), 159–205. doi: 10.1007/978-3-030-84273-4_6

Crossref Full Text | Google Scholar

Lemaster, P., Strough, J., Stoiko, R., and DiDonato, L. (2015). To have and to do: masculine facets of gender predict men's and women's attitudes about gender equality among college students. Psychol. Men Masculinity 16, 195–205. doi: 10.1037/a0036429

Crossref Full Text | Google Scholar

Maccoby, E. E. (1998). The Two Sexes: Growing Up Apart, Coming Together. Cambridge: Belknap Press/Harvard University Press.

Google Scholar

McDonald, M. M., Navarrete, C. D., and Van Vugt, M. (2012). Evolution and the psychology of intergroup conflict: the male warrior hypothesis. Philos. Trans. R. Soc. B Biol. Sci. 367, 670–679. doi: 10.1098/rstb.2011.0301

PubMed Abstract | Crossref Full Text | Google Scholar

Mehta, C. M., Alfonso, J., Delaney, R., and Ayotte, B. J. (2014). Associations between mixed gender friendships, gender reference group identity and substance use in college students. Sex Roles 70, 98–109. doi: 10.1007/s11199-013-0334-8

Crossref Full Text | Google Scholar

Mehta, C. M., Hojjat, M., Smith, K. R., and Ayotte, B. J. (2017). Associations between gender segregation and gender identity in college students. Sex Roles 76, 694–704. doi: 10.1007/s11199-016-0685-z

Crossref Full Text | Google Scholar

Mehta, C. M., and Smith, K. (2023). Childhood gender segregation in context: a cultural sociocontextual approach. Rev. Gen. Psychol. 27, 135–154. doi: 10.1177/10892680221121324

Crossref Full Text | Google Scholar

Mehta, C. M., and Strough, J. (2009). Sex segregation in friendships and normative contexts across the life span. Dev. Rev. 29, 201–220. doi: 10.1016/j.dr.2009.06.001

Crossref Full Text | Google Scholar

Mehta, C. M., and Strough, J. (2010). Gender segregation and gender-typing in adolescence. Sex Roles 63, 251–263. doi: 10.1007/s11199-010-9780-8

Crossref Full Text | Google Scholar

Mehta, C. M., and Wilson, J. (2020). Gender segregation and its correlates in established adulthood. Sex Roles 83, 240–253. doi: 10.1007/s11199-019-01099-9

Crossref Full Text | Google Scholar

Mikołajczak, G., Fisher, A. N., and Ryan, M. K. (2025). Intergroup contact in the context of gender. A critical review of the literature and opportunities for future research. Group Processes Intergroup Relat. 28, 723–751. doi: 10.1177/13684302241311544

Crossref Full Text | Google Scholar

Monsour, M. (2001). Women and Men As Friends: Relationships Across the Life Span in the 21st Century, 1st Edn. Mahwah, NJ: Psychology Press. doi: 10.4324/9781410604613

Crossref Full Text | Google Scholar

National Center for Education Statistics (2024). “Undergraduate degree fields,” in Condition of Education. U.S. Department of Education, Institute of Education Sciences. Available online at: https://nces.ed.gov/programs/coe/indicator/cta (Accessed January 6, 2026).

Google Scholar

Poulin, F., and Pedersen, S. (2007). Developmental changes in gender composition of friendship networks in adolescent girls and boys. Dev. Psychol. 43, 1484–1496. doi: 10.1037/0012-1649.43.6.1484

PubMed Abstract | Crossref Full Text | Google Scholar

Reeder, H. M. (2003). The effect of gender role orientation on same- and cross-sex friendship formation. Sex Roles 49, 143–152. doi: 10.1023/A:1024408913880

Crossref Full Text | Google Scholar

Rubin, K. H., Bukowski, W. M., and Parker, J. G. (2007). “Peer interactions, relationships, and groups,” in Handbook of Child Psychology, 1st Edn., eds. W. Damon, and R. M. Lerner (New York: Wiley). doi: 10.1002/9780470147658.chpsy0310

Crossref Full Text | Google Scholar

Salinas-Quiroz, F., and Sweder, N. (2023). Authentic gender development in non-binary children. Front. Sociol. 8:1177766. doi: 10.3389/fsoc.2023.1177766

PubMed Abstract | Crossref Full Text | Google Scholar

Speer, J. D. (2017). The gender gap in college major: revisiting the role of pre-college factors. Labour Econ. 44, 69–88. doi: 10.1016/j.labeco.2016.12.004

Crossref Full Text | Google Scholar

Strough, J., and Covatto, A. M. (2002). Context and age differences in same-and other-gender peer preferences. Soc. Dev. 11, 346–361. doi: 10.1111/1467-9507.00204

Crossref Full Text | Google Scholar

Thorne, B., and Luria, Z. (1986). Sexuality and gender in children's daily worlds. Soc. Probl. 33, 176–190. doi: 10.2307/800703

Crossref Full Text | Google Scholar

Tobin, D. D., Menon, M., Menon, M., Spatta, B. C., Hodges, E. V. E., and Perry, D. G. (2010). The intrapsychics of gender: a model of self-socialization. Psychol. Rev. 117, 601–622. doi: 10.1037/a0018936

PubMed Abstract | Crossref Full Text | Google Scholar

U.S. Bureau of Labor Statistics (2021). American Time Use Survey. U.S. Bureau of Labor Statistics. Available online at: https://www.bls.gov/tus/ (Accessed January 6, 2026).

Google Scholar

Whiting, B., and Edwards, C. P. (1973). A cross-cultural analysis of sex differences in the behavior of children aged three through 11. J. Soc. Psychol. 91, 171–188. doi: 10.1080/00224545.1973.9923040

Crossref Full Text | Google Scholar

Wong, W. I., Shi, S. Y., and Chen, Z. (2018). Students from single-sex schools are more gender-salient and more anxious in mixed-gender situations: results from high school and college samples. PLoS ONE 13:e0208707. doi: 10.1371/journal.pone.0208707

PubMed Abstract | Crossref Full Text | Google Scholar

Xiao, S. X., Ren, H., Hanish, L. D., Miller, C. F., Martin, C. L., and Fabes, R. A. (2023). Meet up buddy up: effective strategies to promote 4th grade students' intergroup prosocial behavior. Front. Dev. Psychol. 1:1177566. doi: 10.3389/fdpys.2023.1177566

Crossref Full Text | Google Scholar