Leadership development has long been valued as a key outcome of undergraduate student experiences (Dugan, 2017; Komives et al., 2009, 2011, 2013). To this end, student organizations can be catalysts for student leadership opportunities (Kim & Holyoke, 2022; Mainella, 2017) and development (Komives et al., 2011; Riutta & Teodorescu, 2014; Rosch & Collins, 2017). One organization that scholars have closely examined for its ability to leverage social groups and organizational structures (Martin et al., 2012; Sessa et al., 2017) to promote leadership development is the college sorority (e.g., Dugan, 2008). Yet a question persists, one that frames the title for an essay by Bureau (2010): “Fraternities and sororities support leadership development! How do we know?” (p. 1). Addressing this question becomes especially important when moving beyond positional leadership (i.e., formal leadership roles such as president or treasurer; see Dugan, 2017; Pearlman et al., 2023) toward a nuanced understanding of relational forms of leadership indicated by emerging literature (i.e., leadership as an aspect of interpersonal and intergroup relationships; Kezar et al., 2017).

The rationale for this study encompasses two general currents. First, despite considerable attention to the questions posed by Bureau (2010) and reinforced through more recent considerations of women, leadership, and/or sororities in whole or part (Sessa et al., 2017; Workman et al., 2020), further work remains to be done at this intersection. For example, organizations and scholars have illuminated sorority’s relationship to race and racism, and specifically the ways student leaders have continued to grapple with organizational accountability around racist histories and practices (Beaird et al., 2021; Dodge & Patterson, 2019; Roland & Matthews, 2023). Second, we are acutely aware that sororities operate in a higher education climate that expresses wariness toward their existence. This includes critiques (e.g., Schwartz, 2022) that “sororities impose a kind of conformity that stifles growth…in the form of shared social, ethical and political attitudes and behavior, members are expected to adhere to the accepted mores of their Greek houses” (para. 4). To the extent these attitudes animate current thinking, it is important for researchers to leverage insights from data to test such claims.

The purpose of this study was to understand the extent to which personal, environmental, and experiential aspects of undergraduate sorority engagement promoted relational leadership development. We asked: What organizational perceptions and sisterhood experiences influence relational leadership development among NPC sorority members? We now detail our literature review and theoretical foundation.

Literature Review

We begin our review acknowledging the gendered and/or sexist dynamics often found in leadership experiences and leadership development (e.g., Hardaway et al., 2021; Workman et al., 2020), and that in college sororities, the role gender and leadership play in tandem (e.g., Jones, 2018). Here, we explore college women’s experiences with student leadership as well as sorority women’s experiences in college more broadly.

College Women in Leadership

Clubs and student organizations have historically been viewed as a positive experience for college students, filled with growth and socialization (Mayhew et al., 2016). Women engage in leadership in these spaces, despite experiences with gender bias (Rhode, 2019) and an environment that was otherwise not set up for them to thrive (Nash, 2018). Previously, scholars have compared women and men students’ experiences with leadership and social change, finding that women had an advantage over their peers in six of eight values (congruence, commitment, collaboration, common purpose, controversy with civility, and citizenship; Shim, 2013). Though relational leadership has been considered as a feature of leadership practice within sorority-specific studies (e.g., Pearlman et al., 2023) and deemed to merit greater attention (Barber et al., 2020), it is notable that studies focused on positional leaders and role-based approaches to leadership still hold primacy in the discourse. Emerging scholarly directions echo this understanding, with recent work by McCarron et al. (2023) highlighting that “social identities have become increasingly important in conceptualizing leadership as a relational process of understanding self and others in organizations and larger society” (p. 36) and Owen (2020) explicitly focusing on women and leadership.

Still, several studies have importantly examined women in student government and their experiences with positional leadership and relationships. For example, Workman et al. (2020) found that college student government was largely still a space for men, and that there was a chilly climate for elected women from elections through to experiences. Keating Polson et al. (2022) studied women student body presidents, finding that women felt an extreme pressure as president, while navigating complex gender expectations while making positive changes for their campus and community. Scholars have also called on administrators to create space for more women in elected involvement, as well as to offer resources such as developing relationships with administrators (Hardaway et al., 2021; McCready et al., 2023). In university-wide endeavors, collegiate student government is a form of positional leadership often undertaken by women; as such, sorority leadership structures can also be viewed as a microcosm of these larger structures.

Sorority Women’s College Experiences

Sorority women are subject to patriarchal constraints in higher education, and specifically power relations with men (Ispa-Landa & Risman, 2021). In looking at relationships between women’s and gender centers and sororities, Jones (2018) posited that sororities share a goal of empowering women, despite many not defining themselves as feminist or “overtly participating in the struggle against patriarchal oppression” (p. 18). Jones (2018) continued that many women turn to sororities “that emphasize women’s leadership and power as sources of comfort and support,” and that some even join sororities to achieve a space of “feminist-informed empowerment” (p. 18). Continuing this theme, Ispla-Landa and Risman (2021) wrote about gender equality, as well as the racial inequality within historically white sororities, something also illuminated in studies by Beaird et al. (2021) and Roland and Matthews (2023).

Key to the sorority experience are leadership and sisterhood (Tull et al., 2020), core values espoused by the NPC, the umbrella organization for the 26 historically and predominately white women’s sororities (NPC Member Organizations, n.d.). Sorority women have described relationships as core to the experience and through the lens of role modeling (Reynolds, 2020). Specifically, sororities create an environment between new and older members (Pearlman et al., 2023) where “women look up to older members and create informal mentor relationships” (Reynolds, 2020, p. 34). These relationships are also at play with connection to students’ experiences with leadership, programming, and current events. As Wessel and Salisbury (2017) found, opportunities for “development, relationships, and leadership” (p. 28) are important criteria for joining and engaging in sorority life.

Yet, limited efforts in the literature have sought to explicitly examine the connections between hallmarks of sorority life and leadership outcomes through quantitative examination that accounts for experiences at the student- and group-level across a wide diversity of campuses. Specifically, there have been calls to model experiential variables such as sisterhood and belonging through studies employing multilevel models with longitudinal data (Schutts et al., 2017) as well as to include quantitative measures that more fully capture distinctive sorority experiences (Barber et al., 2020). As sororities and fraternities are often lumped together in research, there is real value in disaggregating these communities and to focus on women’s leadership experiences and specifically around relational leadership.

Theoretical Foundations

Our study reaffirms a theoretical claim with deep roots in leadership research and practice: relational leadership can be developed. Synthesizing views on relational leadership incorporating perspectives from leadership studies (Fletcher, 2010; Uhl-Bien & Ospina, 2012) and higher education (Komives et al., 2006), Dugan (2017) introduced relational leadership using four principles:

(1) leadership exists both in and outside of formal leader roles, (2) leadership is enacted through relationship across the organization, not just through pre-established hierarchies, (3) these relationships play a role in creating and/or influencing both social order and social action, and (4) relationships are influenced by the contexts in which they are nested. (p. 229)

As Endress (2000) comments, promoting leadership development through these dimensions is important for college students, especially women, as it also encourages a shift away from a mental model in which ‘leadership’ represents a function of roles and titles (i.e., positional leadership) rather than a more expansive and shared framework of leadership being a shared and inclusive practice. Collectively, these features – which encompass the reality that leadership is fluid, co-constructed, and contextual – have the potential to be advanced through sorority experiences and merit further scholarly attention (Barber et al., 2020).

Often referred to as “women’s ways of leading,” approaches to leadership that include collaboration, relationships, and sharing are participatory and democratic shifts in leadership (Owen, 2020, p. 154). Providing additional theoretical guidance, relational leadership can be understood as being more in line with ecological-based models of student development (Renn & Arnold, 2003) than more discrete understandings of engagement; relational leadership as a process cultivated through a multitude of experiences (Dugan, 2017). It involves active listening, collaboration, and civil discourse as leadership actions across a variety of college settings (Owen, 2020).

Theory specifically informed our research question, methodology, and analysis in three important ways. First, we considered variables (e.g., sisterhood) that have been voiced as important to sorority members and reflect key principles of relational leadership. Second, we utilized multilevel modeling to explicitly account for the nested nature of both our data (i.e., students within institutions) and propositions of ‘nestedness’ inherent to relational leadership theory (Dugan, 2017). Third, we were able to incorporate a pre-test measure to understand and further extend considerations for leadership development as a process that unfolds over time and can be influenced by measurable aspects of collegiate engagement (Komives et al., 2013; see also Pascarella & Wolniak, 2004). We now reflect our positionality and our methods.

Positionality

It is most immediate to name that each of the authors of this manuscript identify as cisgender men. We come to this research from different vantage points, all as faculty in student affairs and higher education programs, and with varying experiences working with sorority and fraternity students on campus and inter/nationally. For example, the first author is a scholar and educator in a strategic leadership program and teaches courses on leadership theory and quantitative approaches to studying leadership development.

We also acknowledge that our relationships with Dyad staff afforded us access to the data for our study. While the leaders of Oracle Sorority allow Dyad to share data collected from its members for research purposes as long as the name of the organization is masked, we do not have relationships with Oracle Sorority or any participants. Though members consented to participating in research, they are likely unaware that we used their responses for our research.

We also come to this research understanding the evolving nature of leadership, subscribing to a belief that leadership can be learned and developed; outdated is the notion that leaders are born. Further, through our experiences in both research and practice, we are keenly aware of the ways women students, faculty, and staff are not fully represented in leadership in higher education (e.g., one third of presidents and less than half of all provosts are women; see Fuesting et al., 2022). Understanding and illuminating leadership perspectives, and particularly those of women students, is paramount to our work that centers equity and justice; and this study does just that (Núñez et al., 2023).

Methods

We used data collected by Dyad Strategies, LLC (Dyad) during two spring waves of an annual internet-based membership survey sent to undergraduate members of a single historically and predominantly white college NPC-affiliated women’s social sorority with chapters located at approximately 175 four-year higher educations throughout the contiguous United States. Though the sorority uses a Greek-letter name, we have decided to refer to it as Oracle Sorority. Dyad administered the survey to 19,238 students from February to March 2020 (Time 1) and 19,236 students from February to March 2021 (Time 2). The response rates were 84.9% during Time 1 and 85.5% during Time 2. Data for 8,974 participants from 172 institutions who were active undergraduate members of sorority during Time 1 and Time 2 and participated during both of data collection periods were retained for our study.Footnote 1 We purged 629 cases through listwise deletion to ensure that we only retained complete responses. See Table 1 for participant demographics.

Table 1 Descriptive Statistics for Participant Demographics
Full size table

The participants in the study each attended one of 172 colleges and universities. The mean institutional cluster size was 42.52 (SD = 36.59; median = 37), ranging from 2 to 192. Almost two-thirds (61.1%) of these institutions are classified as doctoral universities, 25.6% are master’s colleges or universities, and 13.4% are baccalaureate colleges. A majority are located in the South (57.0%), with the remaining 43.0% located in the Midwest (19.8%), Northeast (16.3%) and West (7.0%). With respect to context, we recognize that portions of this data were collected during the more restrictive periods of the COVID-19 pandemic and that experiences may have differed between chapters as a function of institutional policies, state laws, and personal comfort levels. This feature noted, we suggest that both the quantity and diversity of institutions engaged provide a robust quantitative picture of student experiences during this time period.

Measures

The outcome in our analysis was a 12-item measure of relational leadership reported in Time 2 (α = 0.935) developed by Dyad. Participants rated their agreement using a Likert format ratings scale from “strongly disagree” (1) to “strongly agree” (5) (e.g., “I guide others to be successful”). The two-factor scale measures students’ reported relational leadership behaviors. To validate the construct, we conducted principal axis exploratory factor analysis (EFA) and maximum likelihood confirmatory factor analysis (CFA) using the Time 2 data; the full EFA can be found in Table 2. We found the fit statistics for the CFA to be generally acceptable (Hu & Bentler, 1999), with CFI = 0.960, TLI = 0.937, and RMSEA = 0.081 CI [0.079, 0.082]. Given the relatively strong fit and high correlation path between the two constructs (r = 0.81), the factor was scored as an average of all twelve items, which allowed for all other variables to be regressed on a single dependent measure. To better understand our data, we additionally examined the correlation of the measure with other measures in our study (see Table 3). The descriptive statistics for the outcome and other continuous predictors at Time 1 and Time 2 are included in Table 4.

Table 2 Exploratory Factor Analyses Results for Relational Leadership at Time 2
Full size table
Table 3 Correlations Among Continuous Constructs at Time 2
Full size table
Table 4 Descriptive Statistics of Continuous Variables
Full size table

In addition to the identities and lived experience variables identified in Table 1 and students’ responses to relational leadership (Time 1; α = 0.916), our models included nine other individual-level predictors from Time 2. We group-mean centered all continuous variables, and effect coded our demographic predictors to avoid positioning any group as normative (Mayhew & Simonoff, 2015).

We included four measures of members’ perceptions of their chapters. We selected a six-item measure of affective commitment (α = 0.938) and a six-item measure of normative commitment (α = 0.928) as adapted by Dyad from Meyer and Allen (1991). The former measures the extent to which a student feels a sense of commitment to their sorority based on their emotional attachment to aspects of the experience (e.g., “I really feel as if my chapter's problems are my own”); the latter measures the extent to which a student feels a sense of commitment to their sorority based on a sense of duty or obligation (e.g., “I would feel guilty if I left my chapter right now”).

In addition, we added a six-item organizational identification scale (α = 0.898) adapted by Dyad based on the research of Edwards and Peccei (2017). The scale measures the extent to which a student makes the sorority an important part of their social identity (e.g., “When someone criticizes my chapter, it feels like a personal insult”). The final measure of chapter perceptions in our model was a six-item measure of members’ organizational conformity Measured using a six-item scale developed by Dyad (α = 0.925). The single factor scale measures students’ conformity to the behaviors of others within their sorority (e.g., “When it relates to my chapter, I usually go along with ‘the status quo’”).

Adding to the chapter perception measures, we included the five subscales from the 26-item Measure of Fraternal Sisterhood (Schutts et al., 2017). Schutts and colleagues argued the scale reflect five distinct schema of fraternal sisterhood: accountability (six-items; α = 0.905), belonging (five-items; α = 0.947), common purpose (five-items; α = 0.941), shared social experience (five-items; α = 0.825), and support and encouragement (five-items; α = 0.909). The researchers relied on exploratory and confirmatory factor analyses to develop and validate the scale. Schutts et al. (2017) found the scales had internal consistency reliabilities ranging from 0.61 for shared social experience to 0.94 for belonging. Scores were calculated based on the mean value of the items corresponding to the scale.

The group-level predictors included in analyses were the aggregated group means of students’ responses to relational leadership (Time 1), affective commitment (Time 2), normative commitment (Time 2), organizational identification (Time 2), organizational conformity (Time 2) and the five subscales from the Measure of Fraternal Sisterhood (Schutts et al., 2017).

Analysis

We relied on hierarchical linear modeling (HLM) to analyze the data. Prior to creating our models, we performed diagnostics to ensure the variables met all assumptions (Raudenbush & Bryk, 2002). After conducting this preliminary analysis, we created an initial unconditional model to identify the amount of the variance of the outcome explained by the variance between institutions. Individual-level predictors were then added across seven steps. Step 1included most demographic predictors. Step 2 included the student involvement predictors of number of volunteer hours per week and number of campus organizations. Step 3 included leadership demographics (e.g., highest leadership role in sorority). Step 4 included the pre-test measure of Time 1 relational leadership. Step 5 added the chapter perception predictors. We removed these predictors in step 6, and instead included the Measure of Fraternal Sisterhood schema to isolate these predictors. All predictors were included in the final step. During each step, models were specified. Dunnett’s test were performed to examine the statistical significance of the effect coded predictors. To specify the final student-level model, random coefficient models were constructed to explore the variability in the student-level random slopes and the random coefficient.

Upon the final specification of the individual-level model, intercepts and slopes-as-outcomes models were constructed to identify if any institution-level variables could explain the variability in the intercept and slopes. The aggregated group mean centered predictors were added to the model to identify if any variability in Time 2 relational leadership was due to the between-institution variability of these predictors.

Unconditional Model

The unconditional ICC for the model was 0.021, p < . 001. The reliability estimate was 0.451, indicating there was adequate stability across the parameter estimates for each institution (Raudenbush & Bryk, 2002). The deviance for this model with two parameters was 13,156.299. The design effect for the sample is 2.831. Therefore, the effective sample size is 2947.693, meaning there was sufficient statistical power to conduct HLM analyses.

Individual-level Models

The cross-sectional one-way ANCOVA with random effects models were constructed utilizing individual-level predictors that were not allowed to vary randomly along their slopes. The preliminary models will not be reviewed thoroughly because of space. In the final student-level specified model, the coefficient slopes of organizational identification and common purpose sisterhood varied randomly, ps < 0.001, and these random coefficients were included from the subsequent slopes-as-outcomes models. The reliability of the intercept in the final individual-level combined model was sufficiently reliable at 0.657.

Slopes-as-Outcomes Models

After specifying the student-level model, group-level predictors were added to the final three models at the intercept. The model was specified by comparing the deviance of the slopes-as-outcomes models with fixed effects and one with random coefficients. Because deviance of the model individual-level fixed effects (8529.853) was higher than the model with the two random coefficients (8488.092), the final model was specified to exclude the random effects. The reliability of the student-level intercept remained adequate at 0.173.

Limitations

Several limitations confront this study and should be considered when interpreting this analysis. To begin, sorority members in this dataset do not fully reflect the diversity of individuals and organizations which comprise sorority life in the United States. Second, we recognize that while our measures are psychometrically robust, they may not fully capture the highly dynamic nature of sisterhood experiences which can be contextual and specific. Finally, we share perspectives with the field of leadership studies that quantification of leadership and demonstration of its development via linear modeling reflects a helpful, albeit limited, window on growth trajectories for individuals and group members. These limitations noted, we now present our results.

Results

Preliminary Analysis

Descriptive statistics for relational leadership (Time 2) and all independent variables can be found in Table 1 and 2. The outcome was slightly negatively skewed (-0.552). We chose not to transform the variable because of the high response rates across both data collection periods and to retain the interpretability of our models. In addition, we examined the change in participants’ relational leadership scores from Time 1 to Time 2. The mean for change in relational leadership from Time 1 to Time 2 was -0.022 (SD = 0.549), and mode was 0.00 (16.2% of participants). While the difference was small, it was statistically significant, t(8344) = 3.732, p < 0.001. In other words, on average participants reported small, but statistically significant, declines in relational leadership from Time 1 to Time 2.

Specified Slopes-as-Outcomes Models

We present findings of the final three specified models in Table 5. Prior to the inclusion of the sisterhood subscales in the final combined model (Step 5), three of four of the individual-level chapter perceptions predictors were statistically significantly associated with the outcome (e.g., affective commitment, γ = 0.051, p < 0.001). The remaining individual-level chapter perceptions predictor, individual-level organizational conformity, was statistically significantly negatively associated with relationship leadership (γ = -0.079, p < 0.001). Two of the chapter-level chapter perceptions predictors were statistically significantly associated with the outcome, affective commitment (γ = 0.135, p = 0.010) and organizational identification (γ = 0.180, p = 0.005). Without controlling for members’ sisterhood orientations, a member’s perception of their chapter and the related climates of their chapter may affect relational leadership.

Table 5 Specified Coefficients for the Final Models
Full size table

Upon adding the sisterhood subscales and removing chapter perceptions (Step 6), four of the individual-level sisterhood predictors were statistically significantly associated with the outcome (e.g., supportive sisterhood, γ = 0.219, p < 0.001). The remaining individual-level predictor, belonging sisterhood, was negatively associated with relationship leadership (γ = -0.066, p < 0.001). Two of the chapter-level sisterhood predictors were associated with the outcome, common purpose sisterhood (γ = 0.126, p = 0.040) and supportive sisterhood (γ = 0.307, p = 0.004). Without controlling for their perceptions of their chapter, a member’s sisterhood orientation and the related climates of their chapter may affect their relational leadership orientations.

In the combined model (Step 7), at an individual-level, four sisterhood subscales were positively related to relational leadership (e.g., supportive sisterhood, γ = 0.190, p < 0.001), while belonging sisterhood was negatively associated (γ = -0.059, p < 0.001). Regarding organizational perceptions predictors, normative commitment (γ = 0.024, p = 0.009) and organizational identification (γ = 0.063, p < 0.001) were positively associated with the outcome, while affective commitment (γ = -0.035, p = 0.004) and organizational conformity (γ = -0.054, p < 0.001) were negatively associated with relational leadership. We found two group-level sisterhood subscales were statistically significantly associated with relational leadership as well: common purpose (γ = 0.166, p = 0.021) and supportiveness (γ = 0.292, p = 0.003).

Only two items were consistently statistically significantly related to a member’s relational leadership across the final models. Students who reported that they were only members of one other student organization besides their sorority had lower relational leadership orientations than peers involved in more organizations (e.g., γ = -0.031, p < 0.05 for final combined model). In addition, sorority members who identify as current sorority general members (i.e., not holding leadership roles) reported higher relational leadership than their peers (γ = 0.033, p < 0.01). Student involvement and positional leadership, at least in our model, are not strong predictors of a member’s relational leadership orientation. Additional demographic features found to be statistically significant are presented in Table 5 and are discussed in the next section.

The final model explained 77.5% of the between group variance in relational leadership, and 44.4% of the within group variance. The total amount of variance explained by this model reached 45.1%. Of note, there was no longer a statistically significant portion of between group variance explained by the inclusion of other predictors after adding all predictors in the final combined model (p = 0.109).

Discussion

We now discuss our findings relevant to the existing literature, demonstrating how our study uniquely adds to understandings of sorority women. We begin by considering significant findings from latent measures of sorority experiences (e.g., sisterhood) and comment on behavioral (e.g., engagement) and demographic (e.g., paying for college) indicators.

Sorority Experiences

The variable with the largest effect at the individual level on developing relational leadership is supportive sisterhood. Why? We return to our measurement of relational leadership, a more inclusive presentation of leadership being a group effort than positional leaders exerting power and authority over followers. Drawing on previous leadership and gender scholarship (e.g., McKenzie, 2018), we speculate that sisterhood support reflects a positive if complex space for engaging a leadership identity and practice predicated on “supporting a diversity of perspectives and building trust that fosters the exchange of knowledge, ideas and information” (Foss et al., 2022, p. 2). Engagement in this way allows for a (more) human-centered approach to students’ leadership abilities, approaches, and styles.

Accountability also stands out as having an effect at the individual and group level. Accountability is likely contributing to leadership development through its operation on multiple levels of the ecology (Renn & Arnold, 2003): individual (i.e., where students are responsible to themselves), interpersonal (i.e., where students are accountable to sisters), and organizational (i.e., where students are accountable to the chapter). Such a finding (re)affirms notions of sorority, in which community and sisterhood are core values (e.g., NPC member organizations’ missions; NPC, n.d.).

We note several additional findings that register as positive predictors, albeit with lesser effects. These include common purpose in sisterhood, shared social experience, organizational identification, and normative organizational commitment. In general, we interpret these findings as reflecting important contributors to the overall sorority experience and its ‘value-added’ effect on leadership. On the other hand, organizational conformity, affective commitment, and sisterhood belonging have negative effects. Such findings might reflect criticisms about sororities as locations that prioritize conformity (e.g., Schwartz, 2022) while reminding that these features are hardly the entire story.

Group-Level Findings

Supportive sisterhood and perceiving a common purpose in connection to sisterhood were significant at the group level. Building on our theoretical perspectives (Dugan, 2017), group-level findings reflect and model the recursive and nested nature of sisterhood experiences. That supportive sisterhood registers a stronger effect than the pre-test merits substantial attention for two intertwined reasons. First, it endorses perspectives from the literature (e.g., Pearlman et al., 2023; Reynolds, 2020) that sorority experiences demonstrate robust peer-effects; that supportive sisterhood can engender profound developmental gains both individually and, as theorized, collectively (Komives et al., 2013; Martin et al., 2012; Shim, 2013). Moreover, group-level pre-tests can further account for the many possible unmeasurable confounding factors introduced by various levels of self-selection, be these into an institution or chapter. As Núñez et al. (2023) remind, “HLM’s capacity to address multiple units and levels of analysis can provide a way of examining how structural and organizational factors affect educational possibilities” (p. 431).

Participation and Demographic Findings

Findings concerning general sorority membership (vs. positional leadership role) reflect on the roles of student engagement in modern college life. These findings emphasize themes including the importance of authenticity, agency, and context in student development overall (Tillapaugh, 2015) and leadership specifically. Although some early literature on sorority and fraternity life did not disaggregate gender in the exploration of students’ leadership (e.g., Harms et al., 2006), role modeling and authenticity were found as an advantage in the leadership space. We concur that authenticity is especially important in the relational setup of a college sorority.

We briefly comment on other variables with significant main effects at the individual level. Perhaps unsurprisingly, being a sophomore registers a negative effect. As observed by Astin (1993) and many others, the true gains of college must be considered as emerging over the longer span of the undergraduate timeline. This is especially true for leadership in connection with sorority life, and even more relevant as there is a trend of younger leaders taking executive roles in sororities and fraternities (Beach, 2021). Another, intertwining explanation could be the very real disruptions caused by the introduction of strict COVID-19 precautions on sophomore’s collegiate experiences. It is, of course, impossible to truly estimate the proximal and distal effects of such a major disruption during a college year of pronounced academic and social transitions (Chaffin et al., 2019).

The findings with respect to disability perhaps indicate that those with a physical disability may be somewhat more disadvantaged relative to leadership development, and possibly engagement in sorority life more generally, than those with a learning disability who appear to achieve marginally more robust gains in relational leadership. We remind with this finding the necessity of ensuring that physical disabilities are not limitations toward full engagement in sorority or campus life and that the needs of those with physical disabilities must be considered when planning and engaging in chapter activities (see Evans et al., 2023).

The finding concerning hours per week studying broadly suggests that one’s attention to academic work is associated with relational leadership, over and above other factors. We interpret these findings to be less about studying, per se, and more about how leadership and its various behavioral (e.g., team organizing) and affective (e.g., self-confidence) may connect in some important ways to academic engagement. Similarly, we found a continuum of effects with respect to co-curricular involvement (Tillapaugh, 2015). Our results suggest that a small benefit is demonstrated by being involved with the sorority and another organization while more demonstrable gains are realized in connection with being involved in four to five other organizations. Meanwhile, involvement in many groups beyond the sorority has a negative effect, suggesting that overinvolvement might thwart leadership development. We also observed a slightly negative effect of being minimally involved (1 to 5 h/week) in volunteering activities. This could suggest that any potential leadership development emerging in association with volunteering requires sustained participation.

Finally, we found that paying for college via ‘federal or state need-based grants’ was negatively associated with the outcome. Although it may be difficult to ascertain why this was the case from the available data, we turn to existing literature to offer potential explanations for this finding. Namely, those relying on these forms of aid may be less likely to have the ability to engage in positional leader roles given responsibilities for work; on a related note, they may also face classism from other members and additional institutional barriers (e.g., see Houze, 2021). In turn, these realities can inform the access students have to learning about relational leadership. Consequently, we see this finding as further interrogating the intersection of sorority engagement and funding or, more broadly, classism.

Considering the results collectively through the theoretical prism of relational leadership, we uncover support for how this theory accurately captures student experiences in a variety of ways. For example, we get a glimpse into how patterns of leadership come to be constructed overtime in close association with a holistic presentation of a student including their identities, outside-of-class activities, selection into co-curricular experiences, and patterns of engagement within organizations. In other words, understanding leadership – and, as Owen (2020) wonderfully considers, women’s leadership in particular – benefits substantially from the inclusion of a relational (vs. exclusively positional) paradigm. Here, we suggest that our study comprehensively supports a more expansive notion of leadership – one that is inherently nested and individuated in contrast to positional models that are inherently discrete and individual.

Implications for Practice and Future Directions

This study affirms the perspectives of many leadership scholars: leadership aspects, be they intrapersonal beliefs or active knowledge, skills and abilities, can be developed during college (Day & Thornton, 2017). Ultimately, it is important to illuminate how relational leadership has been found important when considering the intersection(s) of leadership and gender identity (e.g., Biddix, 2010; Carli & Eagly, 2017) and move toward improved practice and future research.

Recommendations for Practice

How can relational leadership be skills be developed in practice? One possibility is for campus-based and headquarters professionals to examine already-existing programs that could be transformed into leadership development experiences. For example, mentoring experiences (e.g., ‘big sister/little sister’), which already create a link between older and younger members, can become more intentional leadership development programs and/or spaces in which members practice mentoring skills that can be passed down from generation to generation. Modules and leadership learning objectives can be created rather than a ‘choose your own adventure’ approach to this mentoring space, which can home in on supportive sisterhood and general relationship-based engagement. Further opportunities exist for specifically engaging sorority members who live in residence halls toward ensuring social integration and leadership development (Wessel & Salisbury, 2017).

Additionally, considering the foundational premises of sororities in connection with our findings concerning accountability, we see substantial opportunities for justification and hopefully funding for relational leadership as concerns career development and progression. Learning how to create and sustain affirming, accountable, and active networks is essential for professional advancement and can be a key aspect of relational leadership skills honed during college and which can have outsized and long-term leadership effects.

We also draw attention to the finding that being a general member (i.e., member without a positional leadership role) is a significant predictor of leadership development. This finding perhaps (re)affirms campus based organizational programming that focuses on all members and recognizes that leadership development can and should be for everyone involved, not only those who hold positions and titles. Educators might now be emboldened to ask and answer: How can we make this experience a relational leadership development opportunity?

Recommendations for Theory and Research

With respect to theory, our study has integrated existing approaches to relational leadership both generally (Dugan, 2017) and within the undergraduate context (Komives et al., 2013) with considerations for women’s leadership development (e.g., Foss et al., 2022). Given the nature of our findings, we encourage future theoretical development in this area that considers the potential conceptual connections between expressions of sisterhood, relational leadership, and women’s leadership. We wonder, for example, whether expressions of sisterhood and relationality might be integral to helping sorority women accrue a “female leadership advantage” (Foss et al., 2022, p. 15) during college and in their future career endeavors. We remind here, as Dugan (2017) poignantly observes throughout his volume, that further engaging critical perspectives will only help to expand the circle regarding who, through what actions, and within what contexts ‘leadership’ is expressed and ‘leadership development’ occurs.

Regarding future directions for research, we speculate what this work might look like across different NPC sororities, and across sororities not associated with NPC (e.g., culturally based, multicultural, and historically Black sororities). Future quantitative research can examine relational leadership at the intersection of students’ affiliations outside of NPC. Work in this way may complement the qualitative work done by scholars who have illuminated experiences of leadership for women in historically Black organizations (Gross et al., 2014; Hardaway et al., 2021; Kimbrough, 1995), historically Native American sororities (Minthorn et al., 2023), and culturally-based and multicultural organizations (Garcia, 2020; Garcia & Duran, 2021).

Continued research on NPC sororities can build off this study, as well as the work of other scholars (e.g., Pearlman et al., 2023; Reynolds, 2020; Shim, 2013). Qualitatively, focusing on how relational leadership exists across age and year in school are paramount to understanding mentorship, role modeling, and experience in this leadership context. Future research can include qualitative and mixed-methods studies that engage individual members and leaders in their understanding of relationships within the sorority environment.

Bridging theory and practice, we see this work as a positive step toward addressing calls in the literature base to approach the study of leadership development within higher education from an explicitly relational approach. Provided our coverage of this literature, theory, measurement, and modeling, we suggest that future research might extend the study of relational leadership within and beyond the sorority space to other areas of the curriculum and co-curriculum in which leadership development is the explicit or implicit focus.

Conclusion

Although leadership development is an espoused and enacted priority of sorority life, there remains limited attention to understanding the specific environments, experiences, and incoming characteristics that bolster this vital collegiate outcome. Through use of our longitudinal dataset and analytical strategy, we were able to draw robust inferences by connecting theory to measurable indicators of sisterhood experiences and relational leadership. Our hope is that this study provides insight and empirical justification to those seeking to champion sorority experiences as distinctive locations within the campus ecology for bringing women together in the spirit of support, accountability, and belonging — attributes that not only animate effective relational leadership, but in many ways, fulfill the highest aspirations of the undergraduate experience held closely by administrators, faculty, parents, and certainly students.