Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13
Topics
Discussion
Announcements
 

Exploring Sources and Influences of Social Capital on Community College Students’ First-Year Success: Does Age Make a Difference?


by Xueli Wang, Kelly Wickersham, Yen Lee & Hsun-Yu Chan - 2018

Background/Context: Although numerous studies have emerged shedding light on community college student success, the enduring role of social capital is often overlooked. Furthermore, when conceptualizing social capital in the community college context and its diverse student population, age represents a unique form of diversity in these institutions that warrants further exploration.

Purpose: This research identifies the sources of social capital and the relationships between different sources of social capital and community college success, taking into account how the identified sources and relationships may vary based on age through the following questions: First, what are the major sources of social capital among first-year community college students, and how do sources of social capital vary based on age of students? Second, how do different sources of social capital influence first-year community college success? Third, how do influences of social capital on first-year community college success vary based on the age of students?

Research Design: Our study drew on Coleman’s conceptualization of social capital, along with survey, administrative, and transcript data from three 2-year colleges in a midwestern state. We performed factor analysis with invariance tests to investigate the sources of social capital among community college students and how the identified factor structure may vary by age. We further conducted a logistic regression to examine the relationship between social capital and community college student success across age.

Findings: Our findings indicate that social capital needs to be conceptualized differently for community college students across age because they indeed drew on multiple forms of social capital differently, and the sources of social capital that emerged in turn were related to student success in varied ways. Students under the age of 24 relied on institutional agents and academic interaction as dominant forms of social capital, whereas those over the age of 24 relied on significant other’s support. Students under the age of 24 were more likely to succeed if they frequently visited advisors for academic reasons. A low or high level of support for schoolwork was related to a higher chance of success for students between 24 and 29 years of age. For the students who were over 30 years old, a moderate level of engagement in their learning network and discussions with academic advisors was related to the lowest level of dropping out.

Conclusions: This study extends the social capital model by illuminating the varying types of social capital that students of different age groups engage with, particularly in the community college context, and pushes the boundaries of the knowledge base on how social capital functions in relation to student success in postsecondary education. The findings also elucidate new directions for research, policy, and practice in regard to cultivating and maximizing networks and information for community college students of all ages.



Community colleges have long served as an entryway to postsecondary education for a diverse population of students (Cohen, Brawer, & Kisker, 2014). At present, 1,108 community colleges enroll roughly 7.3 million students each year (American Association of Community Colleges, 2016), with approximately 54% between the ages of 18 and 24, 13% between 25 and 29 years of age, and 25% 30 years of age and above (National Center for Education Statistics, 2014). These institutions primarily award certificates, diplomas, and associate’s degrees that lead to job placement or prepare students for transferring to four-year colleges and universities, among other purposes (Cohen et al., 2014). Despite the diversity in student populations and degree offerings, these institutions have been scrutinized because of the gap between access and success: Although they afford many historically underserved individuals the opportunity to attend college, many community college students struggle to progress beyond the first year, let alone graduate (Martin, Galentino, & Townsend, 2014). Based on data from the National Center for Education Statistics (NCES), community colleges, both public and private, retain barely over half of their first-time students who enroll part- or full-time from the first year to the second (Snyder, de Brey, & Dillow, 2016). The vast potential of community colleges in democratizing higher education by providing open access to students from all backgrounds, academic preparation, and walks of life, coupled with the perennial challenge around students’ college completion after enrolling at these institutions, has gained remarkable attention from researchers, policy makers, and practitioners as they grapple with ways to facilitate student success.


Numerous studies dealing with community college student success have recently emerged to explore this issue. To date, theoretical frameworks and empirical evidence around community college student success underscore the importance of several key individual and institutional factors, including academic preparation (e.g., dual enrollment; An, 2013; Wang, 2015), enrollment behavior and intensity (e.g., academic momentum; Monaghan & Attewell, 2015), academic self-efficacy (Wang, 2013b), demographics (Calcagno, Bailey, Jenkins, Kienzl, & Leinbach, 2008), institution size (Calcagno et al., 2008), and institutional agents such as faculty (Calcagno et al., 2008; Eagan & Jaeger, 2009), among others. Although this prior work sheds light on the individual and institutional mechanisms underlying community college student success, the enduring role of social capital is often overlooked with merely a handful of recent exceptions (e.g., Kruse, Starobin, Chen, Baul, & Laanan, 2015; Starobin, Smith, & Laanan, 2016; Wells, 2008). Social capital can be viewed as a relational asset embodied in actions (e.g., social exchanges) that take place between actors. As social capital can facilitate individuals’ accomplishment of goals (Coleman, 1988), students can accumulate social capital through interpersonal interactions and support from both institutional agents (e.g., faculty, advisor, peers; Tinto, 1975) and family or friends (Chickering & Reisser, 1993), and utilize their social capital to achieve their educational goals.


Despite its theoretical appeal, much of the research on community college student success fails to capture social capital through relationships or networks, which serve as a significant contributor to achieving educational goals and success (Coleman, 1988; Tierney & Venegas, 2006). Limited existing empirical work has shown that social capital through academic and social integration has a positive impact on community college students’ educational expectations, plans, and success (Deil-Amen, 2011; Wang, 2013a). Yet, there exist little intentional work and robust measurement around specific sources of social capital and their connection to student success at community colleges. In addition, individuals from different demographic backgrounds do not build and utilize social capital equally (Lin, 2000), which would be especially applicable in community colleges given that many of the students who enter these institutions tend to possess a greater diversity of individual characteristics (Cohen et al., 2014). Thus, researchers should purposefully attend to the potentially varying sources of social capital among diverse community college students and which ones are most beneficial to students.


When conceptualizing the development and use of social capital in the community college context and its diverse student population, age represents a unique form of diversity. Unlike four-year institutions, community colleges enroll students from a much wider range of age and life stages, which include many biological, psychological, and sociological changes that transpire throughout their lifetime (Merriam & Caffarella, 1999). There exist a number of studies that focus on various aspects of nontraditional-age1 students enrolled at community colleges, often dealing with factors related to their college persistence, pathways, or completion (Bers & Smith, 1991; Calcagno, Crosta, Bailey, & Jenkins, 2007a, 2007b; Capps, 2012; Cox & Ebbers, 2010; Hazzard, 1993; Sorey & Duggan, 2008). Several theoretical models have also been advanced to better understand, support, and educate nontraditional community college students as a distinct group (Chaves, 2006; Levin, 2014; Montero-Hernandez & Cerven, 2013; Stahl & Pavel, 1992). Compared with their traditional-age counterparts, nontraditional students more often take on additional identities and responsibilities (Cox & Ebbers, 2010), such as employee, spouse, and parent, and thus form new or additional relationships through these roles, which in turn become deeply intertwined with their identity as a student.


These studies on nontraditional-age community college students offer a rich, complex portrait of the often changing and multiple identities of older students and the potentially unique impact on progression and outcomes. However, there is very limited empirical work that highlights age as a dynamic and varying influential factor in the context of community college student persistence and success. Age tends to be treated in a dichotomous manner (i.e., traditional vs. nontraditional), without considering the differentiation in persistence or completion across the spectrum of age. This is an important gap in the literature, in the sense that students may rely on and negotiate or renegotiate different information and social relations at different points in their lives. As a result, it is likely that uniform sources of social capital do not exist among community college students of different ages.  


A number of implications are associated with examining social capital in the community college. First, exploring different forms of social capital in this context and the potential variation across age will broaden knowledge on the conceptualization and function of social capital overall, in specific environments, and across the life span. Second, community colleges may be better informed of the various types of social capital that are most advantageous across the diverse student populations who enroll there, especially with respect to age. Third, and related, community college students may utilize social capital in different ways and to varying extents, meaning that their institutions may need to be more attuned to potential variation so that they may receive the necessary opportunities, supports, and environments that work toward each student’s success.


Therefore, we argue that there is great value in adopting the lens of social capital to examine community college student success, while teasing out the role of age in shaping the sources and influences of social capital. Accordingly, our research focuses on identifying the sources of social capital and the relationships between different sources of social capital and community college success, taking into account how the identified sources and relationships may vary based on age. Specifically, in this study, we ask the following questions:


1.

What are the major sources of social capital among first-year community college students, and how do sources of social capital vary based on age of students?

2.

How do different sources of social capital influence first-year community college success?

3.

How do influences of social capital on first-year community college success vary based on age of students?


Note that in this study, we adopt the term success to streamline the multiple outcomes that we investigated as an integrated measure in our study. Doing so takes into account the multiple forms of success that students may achieve in the community college with or without necessarily completing a credential. This variation in outcomes and success is often attributed to the open-access nature of these institutions, in addition to the diversity not only among the students who enroll at community colleges but also with respect to their achievement and reasons for attending (Belfield, Crosta, & Jenkins, 2014; Cohen et al., 2014). Carrying this over methodologically, we incorporated several outcomes (i.e., program completion, transfer to four-year universities, and continued enrollment in postsecondary education) and combined them into one measure during the analytical process. Thus, it is practical and consistent to reference one term throughout.


CONCEPTUAL FRAMEWORK AND RELEVANT LITERATURE


Social capital serves as the primary lens guiding this study. Social capital refers to the networks, formal or informal, that an individual has or develops over time through educational, professional, or social interactions (Bourdieu, 1986). These relationships serve as a benefit that permits both the access to and exchange of information (Coleman, 1988). In the present study, we adopt Coleman’s (1988) conceptualization of social capital and define the source of social capital by the structure (e.g., in school, at home) and the actions of social/institutional partners within the structure (e.g., interaction with these actors) of students’ social realm so that social capital functions as a ladder to greater human capital (i.e., academic success in the context of the present paper; Portes, 1998). Such conceptualization has been adopted in theoretical frameworks and empirical research, and researchers have documented the salience of social capital in college access and postsecondary academic achievement (e.g., D. H. Kim & Schneider, 2005; Perna, 2006). In the community college context, social capital may be exhibited through students’ knowledge of individuals, such as advisors, who could help guide course and program selection. Social capital in the community college could also take the form of relationships students have with faculty, staff, and student peers, which could influence the ways in which they seek and receive assistance. Social capital may be acquired in a formally structured manner (e.g., through classroom interactions) or informal social contexts that still allow for a flow of information and interaction to cultivate relationships (Deil-Amen, 2011), among other potential sources of social capital. Finally, support gained through relationships with family members and friends also becomes a form of social capital that community college students can draw on as they navigate postsecondary education.


SOCIAL CAPITAL IN THE COMMUNITY COLLEGE


There exists a very limited amount of research that directly addresses ways in which community college students build social capital. Within the community college, students may accumulate social capital through peers and institutional agents such as advisors and faculty (Deil-Amen, 2011; Trujillo & Diaz, 1999; Wang, 2013a). Outside and beyond the community college, students may cultivate social capital through internships, their fellow peers or support groups, as well as friends and family (Reyes, 2011). Yet, the benefits and use of social capital tend to be limited to the amount and types of resources and individuals who are available (Coleman, 1988; Eagan & Jaeger, 2009) to both the students and the institutions that they attend, which can have a lasting impact on students’ outcomes. For example, Eagan and Jaeger (2009) integrated social capital and human capital to examine the impact of part-time faculty on community college students’ likelihood of transfer. They found that, with an increasing exposure to part-time community college faculty, students were less likely to transfer to a four-year institution. This and other empirical work that focuses on social capital and community college students pays attention to longer term outcomes or educational expectations surrounding transfer and baccalaureate attainment (Reyes, 2011; Trujillo & Diaz, 1999; Wang, 2013a). It is arguable that to sustain and achieve long-term goals and success, research should also examine more incremental or short-term forms of success (i.e., persistence) in order to determine and create viable and accessible paths toward completion.


Research that centers on social capital and community college student persistence reveals mixed results. On the one hand, scholars contend that social capital in the form of academic and social integration has no impact on persistence (Voorhees, 1987). On the other hand, other studies have noted that this and other forms of social capital (e.g., parental education or expectations) are indeed relevant to persistence (Arbona & Nora, 2007; Deil-Amen & Rosenbaum, 2003; London, 1978; Neumánn & Riesman, 1980; Pascarella & Chapman, 1983; Pascarella, Duby, & Iverson, 1983; Rosenbaum, Deil-Amen, & Person, 2006; Weis, 1985; Wells, 2008). For example, Wells (2008) examined the influence of social and cultural capital on first- to second-year persistence among community college and four-year students. Although the author discovered an overall positive relationship between social and cultural capital and persistence, this relationship was less pronounced among community college students. In addition, the relationship between social capital and student outcomes is not always linear. For example, Pascarella and Terenzini (1977) found a quadratic (that is, nonlinear) relationship between student–faculty interaction regarding academic and career matters and college grade point average among four-year college students.


These mixed findings may have emerged because of the varying research populations and data sources across different studies. But, more important, there is a conspicuous lack of differentiation among the sources and forms of social capital. As an example, interaction with faculty has been viewed as a way to gain social capital toward academic integration (e.g., Voorhees, 1987); at the same time, this type of interaction could be disaggregated based on whether students interacted with faculty regarding academic matters or personal matters (e.g., Pascarella & Chapman, 1983). The latter approach would result in greater measures of social capital through different ways in which students interact with faculty, and subsequently how these different forms of social capital would influence students’ success differently. Accordingly, it is critical to dig deeper into the multiple forms of social capital to better understand those that are especially salient and play a role in student success, particularly in the community college setting.


Looking at social capital in the community college reveals that there is also a lack of differentiation based on subgroups of any type in general. Any disaggregation involving community college students remains broad in terms of accounting for the ways in which social capital may function differently between students, such as different institution types (i.e., two-year versus four-year; e.g., Wells, 2008). Considering the diverse population of students who enroll at community colleges and that social capital tends to differ across demographic characteristics (Lin, 2000), it is crucial to take a fine-grained exploration of social capital of students at these institutions. Careful attention to differences based on age is particularly important. Much of the scholarship previously noted that integrates social capital tends to concentrate on younger, traditional-age students who enroll in college directly from high school. However, as Wang (2013a) suggested, social capital and its relationship with educational outcomes may differ among younger and older community college students. As such, this leads us to explore existing empirical work on community college students that accounts for age.


COMMUNITY COLLEGE STUDENT SUCCESS: THE ROLE OF AGE


A great deal of research on community college student persistence and success pays attention to particular individual student characteristics—whether race/ethnicity, gender, or socioeconomic status, among others—revealing that demographics indeed play a role. Yet, much of this empirical work overlooks the role of age. The limited research that does account for age focuses on the experiences, persistence, and success of specific age groups, either traditional-age or adult students, essentially studying one group in isolation (e.g., Adelman, 2005; Capps, 2012; Chaves, 2006; Cox & Ebbers, 2010) or creating a dichotomy through which to examine younger and older community college students (e.g., Bers & Smith, 1991; Calcagno et al., 2007a, 2007b; Sorey & Duggan, 2008) with very few exceptions (see Feldman, 1993). Although integrating age to study community college student success is a critical first step, it is essential to look beyond a singular age group (such as traditional or nontraditional), break down the dichotomy, and think about age as a wide spectrum.


A finer categorization of age has its theoretical backing in psychology, particularly the Eriksonian identity development theories. In essence, Erikson (1950, 1968) argued that the way individuals interact with others and the environment (e.g., workplace, school) is subject to change as they develop identity over time. Taking community college students—the target population of the current study—for example, from late adolescence to middle adulthood, the key interaction partners that influence individuals at home are first parents and later romantic partners, spouse, and children. Their school life first centers on teachers and peers in high school and then faculty and advisors as individuals chart their career beyond college. As a result, individuals at different places along this spectrum of development would interact with family members and institutional agents very differently. This transition could take even longer for youths in industrialized countries when individuals navigate through emerging adulthood (Arnett, 2000, 2007), partly because of a prolonged educational journey (e.g., enrolling in postsecondary education) that often spans from late adolescence to young adulthood.


Furthermore, among nontraditional-age students, scholars recently argued that the 30s pose another milestone. In fact, college-educated individuals’ efforts toward excellence in schoolwork and career generally peak in their 30s (Whitbourne, Sneed, & Sayer, 2009). Also, major life events, such as being married, play a key role in nontraditional community college students’ academic performance (Peterson, 2016). For example, currently, the median age of first marriage is 29.3 (United States Census Bureau, 2016). Therefore, an in-depth understanding of salient interpersonal factors of academic achievement is needed for students in their 30s and beyond.2 Taken together, age is arguably an important factor to examine in nuanced ways because of the varying developmental aspects and academic and external obligations pertinent to individuals at different life stages, which in turn may prompt students to build and utilize social capital differently. Taking this into account would construct a more accurate picture of additional differences that may exist among students across age, especially in terms of the types of social capital they accumulate and use and how they relate to success.


Moreover, age tends to serve as a typical covariate when examining persistence but does not play a larger role, with mixed results of its significance. Generally, scholars have indicated that older students tend to be less likely to persist as compared with younger students (Bers & Smith, 1991; Capps, 2012; Sorey & Duggan, 2008), yet there is research that suggests that nontraditional-age students are just as likely, if not more so, to succeed as they navigate community colleges (e.g., Calcagno et al., 2007a). As another example of conflicting evidence, Fike and Fike (2008) investigated predictors of first-year persistence among community college students at one institution, controlling for age and other background characteristics. Using data spanning four years, the authors found age to be negatively related to student retention overall and positively associated with first fall-to-spring retention. As such, it is important to delve deeper into what is happening throughout community college students’ life spans and how various stages contribute toward their success.


As established in our literature review, age stage is an important factor that may intersect with how community college students accumulate social capital and how different sources of social capital influence success, because these stages could represent the different developmental needs that students face. For example, younger community college students may rely more on parents and peers as a source of information and support (Perna & Titus, 2005) given that those are the people with whom they most frequently interact at that point in their lives, in addition to institutional agents such as instructors and advisors whom they see on a regular basis. Community college students who are further along in their life stages may turn to their spouses and, perhaps later on, children as their major source of support (Bryant, 2001; Vaillant & Milofsky, 1980). Also, working adult community college students may tap into coworkers in addition to their peers and instructors, all of whom they may interact with more frequently versus other institutional agents, such as advisors, especially if they work full time and commute to school. Given that these sources are not equally available to community college students depending on where they are in their lives, students may turn to and utilize these sources of social capital differently depending on their age. Thus, it stands to reason that community college students of varying age are likely to rely on different kinds of social relations within and beyond college and family settings as their sources of social capital.


METHOD


DATA AND PARTICIPANTS


The data were collected in fall 2014 from the baseline survey of a longitudinal research project supported by the National Science Foundation. The overarching goal of the larger project is to examine factors contributing to the educational pathways and success of two-year college students beginning in programs or courses in science, technology, engineering, and mathematics (STEM) fields. The target study population includes 3,884 first-time students who matriculated in fall 2014 and were enrolled in STEM programs or courses across three large two-year institutions located in a midwestern state.


Of the target population, roughly 3,000 students were selected to participate in the study, with a stratified random sample of approximately 1,000 from each of the three research sites. We adopted a stratified sampling approach by oversampling underrepresented racial/ethnic minorities (URM, including American Indian, Black, Hispanic, multiracial, and those who did not provide related information), female students, and students in programs with fewer enrollments in order to ensure a sufficient number of students in each mentioned category. We calculated and applied a sampling weight across the analyses to restore the proportion of the strata in the target sample. The participants’ unweighted sociodemographic information is shown in Table 1.


Table 1. Summary of the Descriptive Statistics


Data Source

Variables

Overall

(N = 1,668)

Under 24
(n = 1,165)

24 to 29
(n = 216)

Over 30
(n = 286)

 

Outcome variables

    

Administrative Records

First-year success

68.23%

73.41%

63.13%

50.17%

Covariates

    

Female

41.55%

42.97%

40.55%

36.24%

Male

58.09%

56.60%

58.53%

63.07%

African American

7.61%

5.06%

10.60%

15.68%

Hispanic

10.37%

11.15%

11.06%

6.62%

Asian

7.31%

8.58%

5.99%

3.14%

Other race/ethnicity

9.47%

7.72%

14.29%

12.89%

White

65.23%

67.41%

57.60%

61.32%

Cumulative GPA in fall 2014

2.74 (1.09)

2.64 (1.06)

3.02 (1.10)

2.98 (1.13)

Survey Data

Using other language for daily life

16.67%

17.07%

15.67%

15.33%

Using English for daily life

    83.33%

82.85%

83.87%

84.32%

Low income (< $30,000)

35.07%

28.04%

63.59%

41.81%

Regular income (≥ $30,000)

63.61%

70.41%

36.48%

56.79%

Full-time employment

21.10%

12.52%

38.71%

42.51%

Part-time employment

53.66%

63.64%

37.33%

25.09%

No employment

25.12%

23.67%

23.50%

32.06%

Married

11.81%

1.37%

19.35%

48.43%

Not married

88.19%

98.54%

80.18%

51.22%

Single parent

5.22%

1.54%

13.82%

13.59%

Not single parent

94.72%

98.37%

85.71%

86.06%

First-generation student

31.41%

29.67%

25.35%

42.86%

Not first-generation student

68.47%

70.15%

74.19%

56.79%

Part-time enrollment

30.58%

19.90%

47.00%

61.32%

Full-time enrollment

69.42%

80.02%

52.53%

38.33%

Have declared a major

63.61%

61.32%

71.89%

65.85%

Have not declared a major

36.21%

38.42%

27.19%

33.80%

  Math self-efficacy

3.80 (0.87)

3.79 (0.85)

3.90 (0.90)

3.81 (0.90)

  Science self-efficacy

3.72 (0.86)

3.69 (0.84)

3.88 (0.82)

3.71 (0.92)

Note. Cell entries are percentage or mean (SD). The reference group is in bold and is the last category for each variable. Because of missing data, in some variables, the percentage does not add up to 1.



In fall 2014, we sent an email inviting the sampled students to participate in the research project. The students could opt in by clicking the link in the email and completing an online survey. After several reminders, a survey package was sent to the sampled students who had not completed the survey, and these students had the option to decline participation or complete the survey, either online using the web link included in the mail or in paper-and-pencil format, and mail it to the research team in the return envelope. A total of 1,668 students completed the survey, for a response rate of 56.6%. All of the survey participants consented to releasing their transcript and administrative records, which we were able to combine with the students’ survey responses.


MEASURES


Social Capital


To explore the potentially diverging forms of social capital across the students of varying ages, we drew upon eighteen 5-point Likert scale suvey items that capture relationships, networks, and interactions through which students navigated during their first term. These items measure the frequency of students’ engagement in schoolwork with others (e.g., instructors, researchers, or peers), their contact with institutional agents (e.g., peers, instructors, and academic advisors), and the support they receive from family members and friends for their education. See Table 2 for a detailed description of these survey questions, the Likert scales, and to which factor the survey questions belong for each age group based on our factor analysis.


Table 2. Standardized Factor Loadings by Age Group

 

Fit Index

Under 24 years
n = 1,161

24 to 29
n = 215

Over 30
n = 280

Factor Structure

RMSEA (95% Confidence Interval)

.084 (.080, .088)

.089 (.078, .101)

.071 (.060, .081)

CFI

.931

.933

.972

TLI

.915

.918

.964

Mean Difference

F statistic ([39_22249.htm_g/00002.jpg]-value)

F(5,10335) = 1430.97 (p < 0.001)

F(4,1540) = 420.24 (p < 0.001)

F(5,2635) = 453.67 (p < 0.001)

  

Factor Name (Cronbach’s [39_22249.htm_g/00004.jpg])

  

Learning Network (.85)

Learning Network (.86)

Learning network and academic advisor interaction (.85)

Mean (SD)

2.24 (0.82)

2.07 (0.82)

2.04 (0.74)

1. How often do courses require you to work in groups to research necessary background material in order to solve complex, realistic problems?

(Never, Rarely, Sometimes, Often, Very often)

.55

.65

.64

2. How often do you work with an instructor or a researcher at your college on a research project? (Never, Rarely, Sometimes, Often, Very often)

.73

.74

.73

3. How often do you work with an instructor or a researcher from a four-year college or university on a research project? (Never, Rarely, Sometimes, Often, Very often)

.85

.83

.80

4. How often do you participate in a learning community, which are classes that are linked or clustered, often around an interdisciplinary theme, and enroll a common group of students? (Never, Rarely, Sometimes, Often, Very often)

.78

.77

.76

5. How often do you participate in a community-based project as part of your course requirement, where community organizations like schools, neighborhood organizations, and civic organizations partner with your college to solve real-world problems? (Never, Rarely, Sometimes, Often, Very often)

.83

.84

.84

6. How often do you interact with the following individuals for academic purposes?  Academic advisors (Never, Rarely, Sometimes, Often, Very often)

  

.63

  

Academic interaction: Instructor and peer (.61)

Academic interaction: Instructor, peer, and advisor (.75)

Academic interaction: Instructor and peer (.81)

Mean (SD)

3.74 (0.84)

3.29 (0.81)

3.61 (0.94)

7. How often do you interact with the following individuals for academic purposes?  Instructors (Never, Rarely, Sometimes, Often, Very often)

.61

.65

.75

8. How often do you interact with the following individuals for academic purposes?  Student peers (Never, Rarely, Sometimes, Often, Very often)

.72

.79

.92

9. How often do you interact with the following individuals for academic purposes?  Academic advisors (Never, Rarely, Sometimes, Often, Very often)

 

.71

 
  

Significant other’s support for school and upward transfer (.85)

Significant other’s support for school (.83)

Significant other’s support for school (.87)

Mean (SD)

3.71 (0.95)

3.96 (0.93)

3.85 (1.06)

10. How supportive of your schoolwork are your family members? (None, A little, Some, A lot, A great deal)

.71

.89

.87

11. How supportive of your schoolwork are your friends? (None, A little, Some, A lot, A great deal)

.85

.79

.90

  

Transfer knowledge and significant other’s support for upward transfer (.80)

Significant other’s support for upward transfer (.94)

Mean (SD)

 

1.91 (0.93)

2.86 (1.35)

12. How much support do you have from your family for transfer to a four-year college or university? (None, A little, Some, A lot, A great deal)

.74

.92

.95

13. How much support do you have from your friends and peers for transfer to a four-year college or university? (None, A little, Some, A lot, A great deal)

.83

.92

.94

 

Transfer knowledge (.70)

 

Transfer knowledge

Mean (SD)

3.03 (0.99)

 

2.52 (1.21)

14. How much information do you have about how to transfer to a four-year college or university? (None, A little, Some, A lot, A great deal)

.79

.55

.89

15. How well do you understand which courses at your college are transferrable to a four-year college or university? (Not at all, A little, Somewhat, Very, Extremely)

.68

.57

.78

  

Transfer discussion network: classroom and home (.78)

Transfer discussion network: school and home (.90)

Transfer discussion network: school and home (.94)

Mean (SD)

2.53 (0.98)

1.91 (0.93)

1.68 (0.86)

16. How often do you contact each of the following individuals to discuss matters related to transfer to a four-year college or university? Instructors (Never, Rarely, Sometimes, Often, Very often)

.80

.87

.93

17. How often do you contact each of the following individuals to discuss matters related to transfer to a four-year college or university? Student peers (Never, Rarely, Sometimes, Often, Very often)

.73

.87

.94

18. How often do you contact each of the following individuals to discuss matters related to transfer to a four-year college or university? Family members or friends (Never, Rarely, Sometimes, Often, Very often)

.69

.77

.75

19. How often do you contact each of the following individuals to discuss matters related to transfer to a four-year college or university? Academic advisors or counselors (Never, Rarely, Sometimes, Often, Very often)

 

.82

.83

  

Academic interaction: Advisor (.68)

  

Mean (SD)

2.53 (0.96)

20. How often do you interact with the following individuals for academic purposes?  Academic advisors (Never, Rarely, Sometimes, Often, Very often)

.67

21. How often do you contact each of the following individuals to discuss matters related to transfer to a four-year college or university? Academic advisors or counselors (Never, Rarely, Sometimes, Often, Very often)

.78


OUTCOME VARIABLE: FIRST-YEAR SUCCESS


Using administrative records supplied by the participating institutions, along with data from the National Student Clearinghouse, we examined the students’ first-year success from fall 2014 to fall 2015, accounting for several scenarios along two-year college students’ educational path: program completion, transfer to four-year institutions, and continued enrollment in postsecondary education as of fall 2015. First-year success was coded 1 if a student achieved one or more of the mentioned indicators, and 0 otherwise. This is due to the analytical process chosen for the study, as well as to account for the multifaceted nature of success in the community college context as explained earlier.


Covariates


A number of sociodemographic background variables were included in the present study as covariates. These variables were either supplied by the institutions or shared by the participants in the survey. These covariates included age, gender, race/ethnicity, primary language, household annual income, employment status, marital status, self-efficacy in math and science, single-parent status, first-generation status, and whether students had declared their major. In addition, we controlled for students’ GPA from fall 2014. These covariates were integrated based on the long established literature demonstrating their correlation with college persistence and success more broadly or in the community college context (Adelman 1999, 2006; Cohen et al., 2014; Greene, Marti, & McClenney, 2008; Ishitani, 2006; Martin et al., 2014; Napoli & Wortman, 1998; Pascarella & Terenzini, 1991, 2005; Porchea, Allen, Robbins, & Phelps, 2010; Summers, 2003; Tinto 1993; Walpole, 2003). For instance, individuals from underrepresented ethnic/racial groups, including Hispanic and African American students, tend to be less likely to persist or succeed as compared with White students (Greene et al., 2008; Porchea et al., 2010). This trend is similar for first-generation students (Ishitani, 2006; Porchea et al., 2010) and students from low socioeconomic backgrounds (Walpole, 2003). As another example, community college students tend to be more likely to be employed full time and have family responsibilities (Cohen et al., 2014; Martin et al., 2014), often resulting in part-time enrollment, which in turn leads to these students taking longer to progress and being less likely to persist and succeed (Martin et al., 2014). The descriptive statistics of these covariates are shown in Table 1.


ANALYSIS


To answer our research questions, we ran a series of statistical analyses. In the first step, we executed an exploratory factor analysis (EFA) with the entire sample to investigate the common factor structure among the 18 survey items in the social capital scale. Next, we tested the invariance of the identified factor structure across three different age groups: under 24 (n = 1,165; 69.85%), 24 to 29 (n = 216, 12.95%), and over 30 years of age (n = 286, 17.15%). Based on this analysis, we decided the factor structure for social capital based on the age groups. As a final step, we conducted a logistic regression analysis to examine the relationship between social capital and first-year success separately for the students within each of the three age groups, while accounting for the set of covariates described earlier. We conducted all analyses using R 3.2.0 (R Core Team, 2015), except the invariance tests and modification, which we performed in Mplus 7.4 (Muthén & Muthén, 1998–2015).  


More specifically, we conducted the EFA with polychoric correlations, and we extracted the factors using principal axis factor (PAF) with Promax rotation. The use of polychoric correlations and PAF is appropriate for Likert-type scales (Costello & Osborne, 2005; Holgado-Tello, Chacón-Moscoso, Barbero-García, & Vila-Abad, 2010), as was the case in our study. Promax rotation, one of the oblique rotation techniques, allows factors to correlate (Costello & Osborne, 2005) and was suited for our study because different forms of social capital may well correlate with one another. We determined the number of factors mainly by the interpretability of the solutions within the range constructed by the minimal average partial test (MAP; Velicer, 1976), parallel analysis (PA; Horn, 1965), and Kaiser’s lower bound (Kaiser, 1960), methods widely applied and recommended in exploratory factor enumeration (Garrido, Abad, & Ponsoda, 2013; Ruscio & Roche, 2012).


Second, we conducted an invariance test of the identified factor structure across the three age groups to test whether they shared the same factor structure. In the invariance test, we fitted the identified factor structure from EFA via weighted least squares with mean and variance correction estimation (WLSMV; Muthén, 1984), which is recommended for Likert-type scales (DiStefano & Morgan, 2014; Flora & Curran, 2004). We evaluated four levels of invariance—configural invariance, metric invariance, scalar invariance, and error variance invariance—in order, as this process examines general to specific types of invariance (Cheung & Rensvold, 2002). The configural invariance holds given the root mean square error of approximation ≤ 0.08 (RMSEA; Steiger & Lind, 1980), and the Tucker-Lewis Index and Cumulative Fit Index ≥ 0.95 (TLI and CFI, respectively; Bentler, 1990; Browne & Cudeck, 1993; Hu & Bentler, 1999; Tucker & Lewis, 1973), for both the whole sample and each age group. The sample size of each age group exceeds 200, the recommended minimum for structural equation modeling to identify reasonable model fit index values (Kline, 2015; MacCallum, Browne, & Sugawara, 1996).


These fit indices evaluate the absolute and relative fit of the configural invariance model. In the rest of the invariance tests, the specific aspect of the proposed factor structure would be considered invariant upon ΔCFI ≤ .01 (the difference in CFI between the two sequential invariance tests). If the configural invariance does not hold, we halt the invariance tests and explore the factor structure of each age group according to the modification indices supplied by the confirmatory factor analysis. The modified factor structure would be considered acceptable if the 95% confidence interval (CI) of RMSEA covered 0.08, and TLI and CFI ≥ 0.90. Once we determined the factor structures, we evaluated the reliability of the factor scales with Cronbach’s α based on polychoric correlations. Then, we calculated factor scores using the means of the items loaded on the same factors, followed by repeated measure analysis of variance (ANOVA) and pairwise comparisons with Bonferroni correction to understand if the students in each age group reported varying degrees of social capital.


In the logistic regression analysis as the final step, we regressed the log odds of first-year success on the social capital factor scores along with the covariates. To enhance the interpretability of the coefficients, we standardized the factor scores when fitting the logistic regression with linear terms. If we recognized any nonlinear relationships between social capital and first-year success through examining the residual plots, which capture the unexplained relationships between the predicted value and social capital, we modified the model accordingly by specifying the nonlinear polynomial terms in the logistic regression model in order to mirror the nonlinearity (Seber & Lee, 2003). The influence and suitability of nonlinearity terms were further evaluated by the likelihood ratio test. The model fit was assessed by pseudo R2.


MISSING DATA


We used listwise deletion for the participants with missing values in our analysis. Approximately 0.66% and 3.72% of the students had missing values in the EFA and the logistic regression analysis, respectively. Also, about 0.10% and 0.17% of the data points were missing in the EFA and the logistic regression analysis, respectively. Given the small portion of missing data, listwise deletion would have a limited impact on the analyses (Graham, 2009). As a result, the analytical sample size was 1,657 (99.34% of the initial sample size) or 1,656 (99.28% of the initial sample size) when the participants’ ages were required for the EFA and invariance test, and 1,606 (96.28% of the initial sample size) for the logistic regression.


RESULTS


VARIATION IN THE SOURCES OF SOCIAL CAPITAL BASED ON AGE GROUPS


With regard to our first research question on sources of social capital and how they may vary based on students’ age, our findings indicate that although a five-factor structure appears to hold for the entire sample, the three specific age groups have different factor structures for social capital, as revealed by the configural invariance test. Therefore, we explored the factor structure separately for each age group. In the following, we offer key results that answer our first research question. All detailed statistical findings associated with these analyses, both for the overall sample and each age group, are provided in Table 2.


To summarize, for the students under 24 years of age, a six-factor structure was identified; for the students between the ages of 24 and 29, a five-factor solution fit the data well; and for the students who were over 30 years of age when enrolled, a six-factor solution was selected. Table 2 displays the specific forms of social capital for each age group, and Table 3 shows the psychometric properties of the scales. A follow-up repeated measure ANOVA showed that the students within each age group differed in the level of the different types of social capital (see Table 2 for the associated F statistics). A further review of these findings illuminates some interesting variations across age groups. For example, although significant other’s support and academic interaction both emerged to be salient forms of social capital, they exhibited nuanced differences based on students’ age. To illustrate, one of the most used forms of social capital for students under 24 was academic interaction: instructor and peer, which was not shown in other age groups. In comparison, students aged 24 to 29 identified significant other’s support for school as the dominant form of social capital; the identified support was around school in general and distinct from upward transfer, as was the case for students under 24. This is more aligned with students over 30 years of age, who also reported the highest level of significant other’s support for school. With regard to academic interaction, it was the most prevalent form of social capital for students under 24, or the second most used social capital for students over 30, and the institutional agents involved this form of social capital for both age groups centered on instructor and peer. For students between 24 and 29, in contrast, academic interaction, which was important as well, also involved advisor in addition to peer and instructor. For more detailed differences, see Table 2.  


Table 3. Reliability and Factor Correlations

 

Mean (SD)

      

Under 24 years of age

       

    (I) Learning Network

2.24 (0.83)

.85

     

    (II) Academic interaction: Instructor and peer

3.74 (0.83)

.45

.61

    

    (III) Significant other’s support for school and upward transfer

3.71 (0.96)

.25

.41

.85

   

    (IV) Transfer discussion network: Classroom and home

2.54 (0.98)

.43

.45

.43

.78

  

    (V) Academic interaction: Advisor

2.54 (0.97)

.50

.51

.31

.77

.68

 

    (VI) Transfer knowledge

3.04 (0.98)

.27

.27

.45

.55

.59

.70

25 to 29 years of age

       

   (I) Learning network

2.10 (0.82)

.86

     

   (II) Academic interaction: Instructor, peer, and advisor

3.31 (0.80)

.63

.75

    

   (III) Significant other’s support for school

3.94 (0.95)

-.07

.18

.83

   

   (IV) Transfer discussion network: School and home

1.94 (0.94)

.28

.47

.09

.90

  

   (V) Transfer knowledge and significant other’s support for  

          upward transfer

2.99 (0.99)

-.12

.10

.49

.50

.80

 

Over 30 years of age

       

   (I) Learning network and academic advisor interaction

2.06 (0.74)

.85

     

   (II) Academic interaction: Instructor and peer

3.62 (0.93)

.59

.81

    

   (III) Significant other’s support for school

3.82 (1.07)

.03

.36

.87

   

   (IV) Significant other’s support for upward transfer

2.85 (1.35)

.13

.20

.67

.94

  

   (V) Transfer discussion network: school and home

1.69 (0.87)

.48

.31

-.07

.31

.92

 

   (VI) Transfer knowledge

2.53 (1.19)

.27

.20

.09

.42

.57

.82

Note. The diagonal elements are the reliability of each factor; the off-diagonal elements are the factor correlations.


RELATIONSHIPS BETWEEN SOCIAL CAPITAL FACTORS AND SUCCESS


To examine how various sources of social capital are related to first-year success across different age groups, we first fitted the models with linear effects of social capital separately for all three age groups. We then examined the residual plots and detected potential nonlinearity. Accordingly, we included quadratic terms of the social capital variables as a straightforward approach to accommodate nonlinear relationships (James, Witten, Hastie, & Tibshirani, 2013). According to the results of likelihood ratio tests, including the quadratic terms significantly improved our model fit (under 24: [39_22249.htm_g/00006.jpg]=23.00, [39_22249.htm_g/00008.jpg]<.001; 25 to 29: [39_22249.htm_g/00010.jpg]=11.29, [39_22249.htm_g/00012.jpg]=.046; over 30: [39_22249.htm_g/00014.jpg]=42.99, [39_22249.htm_g/00016.jpg]<.001). Thus, our final models included the quadratic terms, with pseudo R2 values all over 0.25, indicating good model to data fit (see Table 4 for specific findings). To facilitate interpretation, we also include a set of figures (Figures 1–3) plotting the predicted probability of the outcome variable occurring based on sources of the social capital.  


Table 4. Summary of Logistic Regression Results by Age Groups


 

Age Group

 

Under 24

25 to 29

Over 30

    Intercept

-0.41 (0.38)

0.61 (0.88)

-1.29 (0.70)

Covariates

   

    Female

-0.01 (0.12)

 0.31 (0.28)

 0.13 (0.25)

    African American

-0.16 (0.25)

-0.41 (0.45)

-1.05 (0.33) **

    Hispanic

-0.23 (0.24)

 1.14 (0.59)

-1.54 (0.54) **

    Asian

 0.39 (0.30)

-0.73 (0.71)

-2.51 (0.73) ***

    Other race/ethnicity

-0.29 (0.21)

-0.64 (0.37)

-1.05 (0.35) **

    Using other language for daily life

-0.41 (0.23)

-0.32 (0.50)

 0.69 (0.40)

    Low income

 0.11 (0.14)

-0.08 (0.30)

 0.66 (0.29) *

    Full-time employment

-0.24 (0.20)

-0.64 (0.35)

 0.45 (0.31)

    Part-time employment

-0.17 (0.14)

-0.61 (0.33)

 0.22 (0.31)

    Married

-0.06 (0.48)

-0.49 (0.35)

 0.31 (0.27)

    Cumulative GPA in fall 2014

 0.48 (0.05) ***

 0.55 (0.14) ***

 0.50 (0.12) ***

    Single parent

-1.15 (0.44) **

-0.74 (0.43)

 1.61 (0.39) ***

    First-generation college student

 0.09 (0.13)

 0.73 (0.36) *

 0.44 (0.23)

    Part-time enrollment

-0.98 (0.14) ***

-1.15 (0.28) ***

-1.34 (0.29) ***

    Math self-efficacy

 0.27 (0.07) ***

 0.17 (0.16)

 0.52 (0.17) **

    Science self-efficacy

-0.06 (0.08)

-0.29 (0.19)

-0.14 (0.16)

    Have declared a major

-0.04 (0.12)

-0.27 (0.29)

-1.25 (0.25) ***

Independent Variable: Linear Term

   

    Factor I

-0.05 (0.07)

 0.09 (0.17)

 0.56 (0.16) ***

    Factor II

 0.00 (0.07)

-0.07 (0.16)

-0.24 (0.14)

    Factor III

 0.09 (0.07)

 0.40 (0.18) *

 0.17 (0.18)

    Factor IV

 0.08 (0.07)

 0.35 (0.19)

 0.29 (0.14)

    Factor V

 0.09 (0.07)

-0.00 (0.16)

-0.25 (0.20)

    Factor VI

-0.19 (0.06) **

 

 0.18 (0.14)

Quadratic Terms

   

    Factor I

 0.01 (0.04)

-0.05 (0.11)

-0.47 (0.11) ***

    Factor II

-0.05 (0.04)

-0.06 (0.11)

-0.09 (0.10)

    Factor III

 0.08 (0.05)

 0.35 (0.11) **

-0.13 (0.12)

    Factor IV

-0.14 (0.05) **

-0.08 (0.13)

-0.26 (0.14)

    Factor V

 0.12 (0.05) *

-0.15 (0.11)

 0.01 (0.11)

    Factor VI

-0.12 (0.05) **

 

-0.25 (0.12) *

Pseudo R2

 0.25

 0.34

 0.50

Note. Outcome variable is first-year success. Cell entries are B (SE). The factor name can be found in Table 3.

*p < .05. **p < .01. ***p < .001.



For the students under 24 years of age, probability of achieving first-year success peaked when they engaged in the transfer discussion network in classroom and home (factor IV) just above the mean level (see panel A of Figure 1). In contrast, the students in this group who visited advisors for academic matters (factor V) above the average tended to succeed (see panel B of Figure 1). Finally, the students in this group who reported a high level of transfer knowledge (factor VI) tended to drop out at an increasing rate by fall 2015 (see panel C of Figure 1).


Figure 1. The Predicted Probability of First-Year Success by Different Social Capital for Students Under 24 Years of Age

[39_22249.htm_g/00018.jpg]

(A)

The Predicted Probability of First-Year Success by Transfer Discussion Network: Classroom and Home

[39_22249.htm_g/00020.jpg]

(B)

The Predicted Probability of First-Year Success by Academic Interaction: Advisor

[39_22249.htm_g/00022.jpg]

(C)

The Predicted Probability of First-Year Success by Transfer Knowledge


Note: The solid line indicates the predicted probability, and the dashed lines indicate the 95% CI of the predicted probabiltiy. The probability was calculated when the categorical covariates were set at the referenced group, and the continuous covariates were set at the mean values of the age group.



The students between 24 and 29 years of age were more likely to drop out by fall 2015 when they reported a moderate level of support from significant others for school (factor III), meaning that the perception of a low or high level of support for schoolwork was related to a higher chance of success (see Figure 2). For the students who were over 30 years old, a moderate level of engagement in their learning network and discussions with academic advisors was related to the lowest level of dropping out (factor I; see panel A of Figure 3). A similar but less drastic trend was found in these students’ transfer knowledge: Those who had a high or low level of transfer knowledge were more likely to drop out of college (factor VI; see panel B of Figure 3).


Figure 2. The Predicted Probability of First-Year Success by Different Forms of Social Capital for the Students Between 24 and 29 Years of Age


[39_22249.htm_g/00024.jpg]

Note: The solid line indicates the predicted probability and the dashed line indicates the 95% CI of the predicted probabiltiy. The probabiltiy was caculated when the categorical covariates were set at the referenced group, and the continuous covariates were set at the mean values of the age group.


Figure 3. The 95% CI Predicted Probability of First-Year Success by Different Forms of Social Capital for Students Over 30 Years of Age

[39_22249.htm_g/00026.jpg]

(A)

The Predicted Probability of First-Year Success by Learning Network and Academic Advisor Interaction


[39_22249.htm_g/00028.jpg]

(B)

The Predicted Probability of First-Year Success by Transfer Knowledge


Note. The solid line indicates the predicted probability, and the dashed lines indicate the 95% CI of the predicted probability. The probability was caculated when the categorical covariates were set at the reference group, and the continuous covariates were set at the mean values of the age group.



Some covariates were found to be associated with students’ likelihood of success. First-semester GPA turned out to be a strong positive predictor, and being a part-time student was related to a higher probability of dropping out across all age groups. The positive relationship between math self-efficacy and first-year success was only salient for the students who were under 24 years of age or above 30 years old. Similarly, being a single parent had a different influence on students in different age groups. For the students under 24, it impeded their likelihood to achieve first-year success, but being a single parent over 30 years old presented a lower chance of dropping out.


Finally, other covariates were salient predictors only for certain age groups. Those who were first-generation college students between 24 and 29 years old had a higher chance of achieving first-year success. For the students above 30 years of age, coming from a low-income family was related to a higher chance to succeed, but being an underrepresented minority student or declaring a major within the first year of college was related to a lower probability of achieving first-year success.


LIMITATIONS


A few limitations should be taken into account when interpreting the results of our study. First, the data from our study span one year, and social capital was measured only once. This does not account for ways in which social capital may be shaped and reshaped over time among the different age groups. At the same time, our findings serve as a foundation and allow for future, longitudinal inquiry toward this end. Second, the data collected come from three 2-year institutions in one state, which limits the generalizability of the results. Yet, the participating institutions included in the study are diverse with regard to both their student populations and geographic locations, arguably permitting some transferability. Third, the target sample of the project used for our study either majored or was taking courses in STEM fields. It is possible that the students pursuing majors or courses in these disciplines may contain a set of background or motivational attributes that could be potentially different from the overall student population, which may lead them to use social capital differently from their peers in other disciplines. Fourth, there are a few constraints with our data sources. For example, academic interaction: instructor and peer and academic interaction: advisor for students under 24 years of age did not achieve high internal consistency. Although this is not uncommon among constructed factor scales with a limited number of items, it may still pose a possible threat to our findings because the potentially large measurement errors associated with a small number of items on a factor scale serving as a predictor variable may attenuate regression coefficients, which indicates the possibility of low power (Fox, 2008). On the other hand, there is support that this may not seriously jeopardize the reliability of the scale if the interitem correlations are over .40 (Robinson, Shaver, & Wrightsman, 1991), or, more precisely, .43 and .52, as was the case in our study. Finally, it is important to note that the survey data rely on self-report of behaviors instead of objectively observed ones. Finally, with the lack of randomization and the nature of our data being observational, we do not claim causality of the findings.


DISCUSSION


The findings from this study demonstrate that social capital needs to be conceptualized differently for traditional-age and nontraditional community college students. Our study further offers a deeper look at age groups that may bear theoretical and practical implications, given the varying sources of social capital for students in different age groups, and invites more studies to explore meaningful ways to account for age differences in the research on community college students. We are able to offer evidence that students of various age ranges engage with multiple forms of social capital differently and that these sources of social capital influence short-term success in varied ways.


ENGAGING WITH VARIOUS FORMS OF SOCIAL CAPITAL: LOOKING ACROSS AGE


Traditional-age students’—specifically those under the age of 24 in our study—higher factor scores related to transfer-specific forms of social capital as compared with older students may reveal varying educational goals. For instance, traditional-age students may cultivate transfer-related social capital with the obvious intent to transfer and obtain a bachelor’s degree, whereas older students who attend community colleges tend to come in expecting to earn a qualification there without moving on to another institution. This aligns with previous research indicating that traditional-age community college students tend to have higher educational aspirations as compared with older community college students (Laanan, 2003; Porchea et al., 2010).


While significant other’s support emerged as a prominent source of social capital across all age groups, for older students, particularly those aged 24 to 29 and 30 and older, social capital in the form of support from significant others for schoolwork ranked highest (factor III). This finding implies that, for nontraditional-age students, their lives do not center on courses, peers, and extracurricular activities. Rather, their priorities tend to include work, family, and other obligations (Cohen et al., 2014). As such, significant others such as family and friends play a large role in terms of the social capital that these students rely on as they navigate and progress through higher education.


Across all age groups, it is intriguing to note differences in the ways that the students engaged with certain institutional agents as social capital. For example, students of different ages interacted with their academic advisors differently. For the students under 24, interaction with advisors represented its own factor, whereas for the students aged 24 to 29, this interaction was combined with academic interaction with others at the community college, and for those over 30, it was integrated into their learning network. For students aged 24 to 29, it is possible that this group does not make a distinction among academic advisors, instructors, and peers as institutional agents that can assist with schoolwork in one way or another, which was one of the most important uses of social capital for them. For students over 30 years of age, however, academic interaction centers on instructors and peers—institutional agents more immediately accessible within the classroom context, while academic interaction with advisors alongside the learning network, often entailing time invested outside the classroom, falls under the same source of social capital that is less used. As pointed out earlier, older students tend to have more competing responsibilities and thus do not have as much time to spend on campus. These findings, taken together, pinpoint the value of tailoring advising services for students from different age groups, as well as the need to conceive creative ways to blend academic and social venues (Deil-Amen, 2011) through which capital is built, especially for older students.


SOCIAL CAPITAL AND FIRST-YEAR SUCCESS BY AGE GROUP


In general, our findings reveal a clear linkage between building multiple sources of social capital and first-year college success. However, this relationship was not straightforward when we factored in students’ age. For example, the students under 24 years of age had a higher probability of first-year success if they moderately tapped into individuals inside and outside of college to discuss transfer (factor IV). This means that engaging with friends, family, instructors, and peers can be helpful to an extent, but too little or too much of it could prove to be detrimental. These students need enough discussion to help guide them with regard to transfer, but engaging too much could potentially result in an information overload. Discussing transfer with fellow students, friends, and family can help guide students in terms of where they may wish to transfer; yet, students may struggle to figure out the best institution with respect to program fit and maximizing credit transfer based on such discussions (Wang, Wickersham, & Sun, 2016). Although students may persist, the extra noise may create additional uncertainty as they work toward their educational goals.


Also, these students had a higher probability of first-year success when they interacted with academic advisors about academic matters (factor V) very frequently (i.e., above the average level). This reinforces the critical role of academic advisors in the persistence and success of community college students (Bahr, 2008). Advisors can serve as a helpful source of information regarding program planning and academic resources to ensure that students succeed in college. Accordingly, our finding supports relationships with advisors as a critically important source of social capital for community college students under the age of 24 if they indeed engage in a high level of interaction with advisors.


What is even more interesting is that the younger students tended to drop out without completing a credential or transfer when they reported an above-average level of transfer knowledge (factor VI). This finding certainly warrants further inquiry, but one possibility would be that, as students gain more information and knowledge regarding transfer, they might have a better grasp of the transfer process. In turn, by gaining such knowledge, they may better gauge their chance of successfully transferring to a four-year institution, which could overwhelm them or make them realize that it is no longer a goal. This ultimately has the potential to shape and change students’ educational aspirations, including transfer intent (Wang et al., 2016), which can ultimately impact persistence, transfer, and graduation.


Across all three age groups, building certain sources of social capital at a moderate level was associated with either the highest or lowest chance to be retained, graduate, or transfer in one year. For example, for the students aged 24 to 29, a very low or high level of support from significant others for schoolwork (factor III) was linked to a better chance to achieve first-year college success. It is possible that those who reported low levels of support may already perform well academically. As such, if they are doing well when it comes to their studies, they do not have a great need for additional support from significant others. On the other hand, some students in this age group may need a substantial amount of support from significant others for schoolwork in order to succeed. These students may rely heavily on family or friends to help them with assignments or exams; these significant others essentially become study partners or personal coaches who provide the necessary academic and emotional support to guide them toward success more broadly.


For the students 30 and older, a balance needed to be struck in terms of their engagement with their learning network and interaction with advisors (factor I). The students who had higher factor scores in these areas may already be having trouble with school, causing them to tap into these networks for help (Chan & Wang, 2016). On the opposite end of the spectrum, if students take very little advantage of these sources, they may not obtain critical information and networks, such as referrals to academic resources or transfer guidelines that could facilitate their success. In a sense, it is not about the sheer quantity of the interactions but their quality, and students making the most of those sources of social capital, which can subsequently impact student outcomes.


ADDITIONAL TAKEAWAYS: BACKGROUND AND BIGGER PICTURE ISSUES


There are some other intriguing findings that warrant discussion. For example, being a single parent under the age of 24 had a significantly negative influence, whereas the association was significantly positive for those over 30 years old. The negative relationship for the 24 and under age group aligns with previous empirical work indicating single-parent status to be a risk factor impacting college success (Coley, 2000; Dowd & Coury, 2006; Greene et al., 2008; Horn & Premo, 1995). However, this research did not distinguish this risk factor based on age, making the finding related to the students aged 30 and older contradictory, pointing to a need for further investigation. At the same time, the conflicting results between these age groups may suggest that any impact of being a single parent with short-term success is much more complex than what we may have previously known. One possible explanation may be that the younger single-parent students have younger children who require more attention and care, while the single parents age 30 and older may have children who are older, perhaps of school age and more capable of caring for themselves to an extent, thus less distracting. Moreover, older students who are single parents may have more experience managing multiple responsibilities, such as work and family, than those who are not single parents. As a result, they may be better able to successfully budget and prioritize time and effort (Grimes, 1997) allotted to school and external responsibilities. In addition, while the disparities in achieving short-term college success between White and racial/ethnic minority students in STEM fields were consistent for students of all age groups, they were statistically manifest only for the students aged 30 or older. This troubling finding bears practical implications and highlights that further empirical evidence is needed to determine any issues or barriers that these student populations face, especially within this particular age group.


Coming back to the idea of balance with regard to community college student use of social capital brings up a broader issue of the extent to which students cultivate and utilize networks. Making connections and tapping into essential information is certainly helpful, but social capital alone may not be the sole contributor to student success. There may be a greater context in which persistence and success occurs, and social capital may likely represent a complex and elusive piece of that puzzle. At the same time, we argue that students should also cultivate their own motivation to keep progressing toward their educational goals. As such, community college students should carefully negotiate and reconcile their own agency along with the benefits of networks and information that are made available to them in paving their success.


IMPLICATIONS FOR FUTURE RESEARCH AND PRACTICE


A number of implications for research and practice emerge from our study. First, the transfer-specific and academic integration-related forms of social capital that resulted from our study present a fine-grained understanding of the specific contexts in which social capital is formed and for what purpose. This has implications for the flow of information and resources through such networks or social ties as related to students’ educational success in the larger sense. Broader implications of these findings rest with a deeper exploration of community college students’ forms of social capital and contexts to better understand the types of social capital that exist, how students use them, and in what ways they contribute to student persistence and longer term success such as completion. Conducting further research toward this end will not only expand our horizons on social capital and its function as a whole, but also has the potential to inform community colleges on the types of social capital that are most beneficial across the various student populations enrolled there.


Second, future research needs to continue conceptualizing and examining social capital as more authentically exhibited in relationships as opposed to static characteristics. Also, future quantitative research in this area should move beyond measuring the exposure to opportunities of gaining social capital to more intentionally measuring actual accumulation of social capital. Building on quantitative inquiry, mixed methods research is needed to lend deeper insight by exploring structural elements of social capital first, followed by how students make meaning within structural contexts, and then how all of it contributes to community college students’ aspirations and success. Finally, a longitudinal approach should be taken to studying social capital. Following community college students and measuring their accumulation of social capital over time not only allows researchers to reveal the complex nature of this construct but also pinpoints the developmental process of capital building that will shed light on both research and practice.


Third, the results of our study emphasize the importance of paying purposeful attention to human agency. Cultivating a sense of agency is important in developing the networks and structures that facilitate the process of building community college students’ social capital. The measures in our study speak to this given the initiation implied in many of the items (e.g., engaging in discussions with peers, instructors, advisors, and so on). That is not to say that community colleges should not help with providing critical information and networks, especially given that some of the student populations concentrated there (e.g., women, ethnic/racial minorities) tend to have unequal amounts of social capital (Lin, 2000). The point is that community colleges should move beyond the mere provision of resources and tap into student agency in the process of maximizing available resources to ensure optimal flow of connections and information.


Fourth, as evidenced by our study, community college students in various age groups engage with different forms of social capital to varying extents. This means that community colleges need to be mindful of such variation in the sources and uses of social capital in order to provide the best opportunities and environments for their students. For instance, for older students, this may mean equipping faculty to also act as advisors for students, which has proved to be an effective practice at community colleges (McArthur, 2005; Packard, Tuladhar, & Lee, 2013). That way, these students can still take advantage of this social capital, despite having a full schedule with competing obligations. At the same time, it is still important to offer traditional advising services with which younger students tend to engage.


CONCLUSION


Our study set out to examine the various types of social capital that community college students engaged with, as well as how different forms of social capital influenced first-year success. Taking this a step further, we investigated the sources and influence of social capital based on age. Our findings extend the social capital model by illuminating the varying types of social capital students of different age groups engage with, particularly in the community college context. Our study pushes the boundaries of the knowledge base on how social capital functions in relation to student success in postsecondary education and elucidates new directions for research, policy, and practice in regard to cultivating and maximizing networks and information for community college students of all ages.


Acknowledgment


This study is based on work supported by the National Science Foundation under Grant No. DUE-1430642.


Notes


1. Although the term nontraditional may refer to age, this term has also referred to background characteristics or risk factors, such as race/ethnicity, socioeconomic status, enrollment status, employment status, financial dependence, and single-parent status, among others (Clark, 2012; Kane & Rouse, 1999; K. A. Kim, 2002; K. A. Kim, Sax, Lee, & Hagedorn, 2010; Levin, 2014).


2. We conducted sensitivity analyses to empirically test the robustness of the age cutoff at 30. For all statistical procedures, we performed sensitivity analyses using alternative cutoff points at 31 and 33, respectively, between the middle and older age groups. Results from the invariance test, factor analysis, factor mean score distributions, and logistic regression generally held across the cutoff points. For the proposed factor structures, all were well fitted except the 90% CI of RMSEA of middle age group (.082, .103) when the cutoff point was 33. However, the deviation is quite trivial (.02). For the logistic regression, the signs of the regression coefficients were all consistent across the cutoff points, and the results were robust for all predictors, except that for older students (over 31), English language learners had a significantly higher probability to succeed. Given these large patterns and key findings from our study, our sensitivity analyses suggest that having 30 as an additional age cutoff beyond the traditional binary approach to defining nontraditional-age is both theoretically driven and technically robust in the context of our study’s data.


References


Adelman, C. (1999). Women and men of the engineering path: A model for analysis of undergraduate careers. Washington, DC: U.S. Department of Education.


Adelman, C. (2005). Moving into town—and moving on: The community college in the lives of traditional-age students. Washington, DC: U.S. Department of Education.


Adelman, C. (2006). The toolbox revisited: Paths to degree completion from high school through college. Washington, DC: U.S. Department of Education.


American Association of Community Colleges. (2016). 2016 fact sheet. Washington, DC: Author.


An, B. P. (2013). The influence of dual enrollment on academic performance and college readiness: Differences by socioeconomic status. Research in Higher Education, 54(4), 407–432. doi:10.1007/s11162-012-9278-z


Arbona, C., & Nora, A. (2007). The influence of academic and environmental factors on Hispanic college degree attainment. The Review of Higher Education, 30(3), 247–269.


Arnett, J. J. (2000). Emerging adulthood: A theory of development from the late teens through the twenties. American Psychologist, 55(5), 469–480.


Arnett, J. J. (2007). Emerging adulthood: What is it, and what is it good for? Child Development Perspectives, 1(2), 68–73.


Bahr, P. R. (2008). Cooling out in the community college: What is the effect of academic advising on students’ chances of success? Research in Higher Education, 49(8), 704–732.

Belfield, C. R., Crosta, P. M., & Jenkins, D. (2014). Can community colleges afford to improve completion? Measuring the cost and efficiency consequences of reform. Educational Evaluation and Policy Analysis, 36(3), 327–345.


Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107(2), 238–246.


Bers, T. H., & Smith, K. E. (1991). Persistence of community college students: The influence of student intent and academic and social integration. Research in Higher Education, 32(5), 539–556. doi:10.1007/BF00992627


Bourdieu, P. (1986). The forms of capital. In J. G. Richardson (Ed.), Handbook of theory and research for the sociology of education (pp. 241–258). Westport, CT: Greenwood Press.


Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen & J. S. Long (Eds.), Testing structural equation models (pp. 136–162). Newbury Park, CA: Sage.


Bryant, A. N. (2001). ERIC review: Community college students recent findings and trends. Community College Review, 29(3), 77–93.


Calcagno, J. C., Bailey, T. R., Jenkins, D., Kienzl, G., & Leinbach, T. (2008). Community college student success: What institutional characteristics make a difference? Economics of Education Review, 27(6), 632–645.


Calcagno, J. C., Crosta, P. M., Bailey, T. R., & Jenkins, D. (2007a). Does age of entrance affect community college completion probabilities? Evidence from a discrete-time hazard model. Education Evaluation and Policy Analysis, 29(3), 218–235.


Calcagno, J. C., Crosta, P. M., Bailey, T. R., & Jenkins, D. (2007b). Stepping stones to a degree: The impact of enrollment pathways and milestones on community college student outcomes. Research in Higher Education, 48(7), 775–801.


Capps, R. (2012). Supporting adult-student persistence in community colleges. Change: The Magazine of Higher Learning, 44(2), 38–44.


Chan, H.-Y., & Wang, X. (2016). Interact for what? The relationship between interpersonal interaction based on motivation and educational outcomes among students in manufacturing programs at two-year technical colleges. Community College Review, 44(1), 26–48.


Chaves, C. (2006). Involvement, development, and retention: Theoretical foundations and potential extensions for adult community college students. Community College Review, 34(2), 139–152.


Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling, 9(2), 233–255.


Chickering, A. W., & Reisser, L. (1993). Education and identity (2nd ed.). San Francisco, CA: Jossey-Bass.


Clark, L. (2012). When nontraditional is traditional: A faculty dialogue with graduating community college students about persistence. Community College Journal of Research and Practice, 36(7), 511–519.


Cohen, A. M., Brawer, F. B., & Kisker, C. B. (2014). The American community college (6th ed.). San Francisco, CA: Jossey-Bass.


Coleman, J. S. (1988). Social capital in the creation of human capital. American Journal of Sociology, 94(1), S95–S120.


Coley, R. J. (2000). The American community college turns 100: A look at its students, programs and prospects. Princeton, NJ: Educational Testing Service.


Costello, A. B., & Osborne, J. W. (2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical Assessment, Research and Evaluation, 10(7), 1–9.


Cox, E. M., & Ebbers, L. H. (2010). Exploring the persistence of adult women at a Midwest community college. Community College Journal of Research and Practice, 34(4), 337–359.


Deil-Amen, R. J. (2011). Socio-academic integrative moments: Rethinking academic and social integration among two-year college students in career-related programs. Journal of Higher Education, 82(1), 54–91.


Deil-Amen, R. J., & Rosenbaum, J. E. (2003). The social prerequisites of success: Can college structure reduce the need for social know-how? In J. Jacobs & K. Shaw (Eds.), Annals of the American Academy of Political and Social Science (Vol. 586, pp. 120–143). Newbury Park, CA: Sage.


DiStefano, C., & Morgan, G. B. (2014). A comparison of diagonal weighted least squares robust estimation techniques for ordinal data. Structural Equation Modeling, 21(3), 425–438.


Dowd, A. C., & Coury, T. (2006). The effect of loans on the persistence and attainment of community college students. Research in Higher Education, 47(1), 33–62.


Eagan, M. K., Jr., & Jaeger, A. J. (2009). Effects of exposure to part-time faculty on community college transfer. Research in Higher Education, 50(6), 168–188.


Erikson, E. H. (1950). Childhood and society. New York, NY: Norton.


Erikson, E. H. (1968). Identity: Youth and crisis. New York, NY: Norton.


Feldman, M. J. (1993). Factors associated with one-year retention in the community college. Research in Higher Education, 34(4), 503–512.


Fike, D. S., & Fike, R. (2008). Predictors of first-year student retention in the community college. Community College Review, 36(2), 68–88.


Flora, D. B., & Curran, P. J. (2004). An empirical evaluation of alternative methods of estimation for confirmatory factor analysis with ordinal data. Psychological Methods, 9(4), 466–491.


Fox, J. (2008). Applied regression analysis and generalized linear models. London, England: Sage.


Garrido, L. E., Abad, F. J., & Ponsoda, V. (2013). A new look at Horn’s parallel analysis with ordinal variables. Psychological Methods, 18(4), 454–474.


Graham, J. W. (2009). Missing data analysis: Making it work in the real world. Annual Review of Psychology, 60, 549–576.


Greene, T. G., Marti, C. N., & McClenney, K. (2008). The effort-outcome gap: Differences for African American and Hispanic community college students in student engagement and academic achievement. Journal of Higher Education, 79(5), 513–539.


Grimes, S. K. (1997). Underprepared community college students: Characteristics, persistence, and academic success. Community College Journal of Research and Practice, 21(1), 47–56.


Hazzard, T. (1993). Programs, issues, and concerns regarding nontraditional students with a focus on model orientation session (Unpublished paper). Tallahassee: Florida State University


Holgado-Tello, F. P., Chacón-Moscoso, S., Barbero-García, I., & Vila-Abad, E. (2010). Polychoric versus Pearson correlations in exploratory and confirmatory factor analysis of ordinal variables. Quality & Quantity44(1), 153–166.


Horn, J. L. (1965). A rationale and test for the number of factors in factor analysis. Psychometrika30(2), 179-185.


Horn, L. J., & Premo, M. D. (1995). Profile of undergraduates in U.S. postsecondary education institutions: 1992-93 (NCES Publication No. 96-237). Washington, DC: National Center for Education Statistics.


Hu, L. -T. & Bentler, P. M. (1999). Cutoff criteria for fit indices in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1–55.


Ishitani, T. T. (2006). Studying attrition and degree completion behavior among first-generation college students in the United States. Journal of Higher Education, 77(5), 861–885.


James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. New York, NY: Springer.


Kaiser, H. F. (1960). The application of electronic computers to factor analysis. Educational & Psychological Measurement, 20, 141–151.


Kane, T. J., & Rouse, C. E. (1999). The community college: Educating students at the margin between college and work. Journal of Economic Perspectives, 13(1), 63–84.


Kim, D. H., & Schneider, B. (2005). Social capital in action: Alignment of parental support in adolescents’ transition to postsecondary education. Social Forces, 84(2), 1181–1206.


Kim, K. A. (2002). ERIC review: Exploring the meaning of “nontraditional” at the community college. Community College Review, 30(1), 74–89.


Kim, K. A., Sax, L. J., Lee, J. J., & Hagedorn, L. S. (2010). Redefining nontraditional students: Exploring the self-perceptions of community college students. Community College Journal of Research and Practice, 34(5), 402–422.


Kline, R. B. (2015). Principles and practice of structural equation modeling (4th ed.). New York, NY: Guilford Press.


Kruse, T., Starobin, S. S., Chen, Y. A., Baul, T., & Laanan, F. S. (2015). Impacts of intersection between social capital and finances on community college students’ pursuit of STEM degrees. Community College Journal of Research and Practice, 39(4), 324–343.


Laanan, F. S. (2003). Degree aspirations of two-year college students. Community College Journal of Research and Practice, 27(6), 495–518.


Levin, J. S. (2014). Nontraditional students and community colleges: The conflict of justice and neoliberalism (Rev. ed.). New York, NY: Palgrave Macmillan.


Lin, N. (2000). Inequality in social capital. Contemporary Sociology, 29(6), 785–795.


London, H. B. (1978). The culture of a community college. New York, NY: Praeger.


MacCallum, R. C., Browne, M. W., & Sugawara, H. M. (1996). Power analysis and determination of sample size for covariance structure modeling. Psychological Methods1(2), 130–149.


Martin, K., Galentino, R., & Townsend, L. (2014). Community college student success: The role of motivation and self-empowerment. Community College Review, 42(3), 221–241.


McArthur, R. C. (2005). Faculty-based advising: An important factory in community college retention. Community College Review, 32(4), 1–19.


Merriam, S. B., & Caffarella, R. S. (1999). Learning in adulthood: A comprehensive guide (2nd ed.). San Francisco, CA: Jossey-Bass.


Monaghan, D. B., & Attewell, P. (2015). The community college route to the bachelor's degree. Educational Evaluation and Policy Analysis, 37(1), 70–91.


Montero-Hernandez, V., & Cerven, C. (2013). Adult student development: The agentic approach and its relationship to the community college context. In J. S. Levin & S. T. Kater (Eds.), Understanding community colleges (pp. 69–85). New York, NY: Routledge.


Muthén, B. (1984). A general structural equation model with dichotomous, ordered categorical and continuous latent variable indicators. Psychometrika49(1), 115–132.


Muthén, L. K., & Muthén, B. O. (1998–2015). Mplus user’s guide (7th ed.). Los Angeles, CA: Muthén & Muthén.


Napoli, A. R., & Wortman, P. M. (1998). Psychosocial factors related to retention and early departure of two-year community college students. Research in Higher Education, 39(4), 419–455.


National Center for Education Statistics. (2014). Digest of education statistics. Washington, DC: U.S. Department of Education.


Neumánn, R. W., & Riesman, D. (1980). The community college elite. New Directions for Community Colleges, 1980(32), 53–71.


Packard, B. W. -L., Tuladhar, C., & Lee, J. -S. (2013). Advising in the classroom: How community college STEM faculty support transfer-bound students. Journal of College Science Teaching, 42(4), 14–20.


Pascarella, E. T., & Chapman, D. W. (1983). A multiinstitutional, path analytic validation of Tinto’s model of college withdrawal. American Educational Research Journal, 20(1), 60–75.


Pascarella, E. T., Duby, P. B., & Iverson, B. K. (1983). A text and reconceptualization of a theoretical model of college withdrawal in a commuter institution setting. Sociology of Education, 56(2), 88–100.


Pascarella, E. T., & Terenzini, P. T. (1977). Patterns of student-faculty informal interaction beyond the classroom and voluntary freshman attrition. Journal of Higher Education, 48(5), 540–552.


Pascarella, E. T., & Terenzini, P. T. (1991). How college affects students: Findings and insights from twenty years of research. San Francisco, CA: Jossey-Bass.


Pascarella, E. T., & Terenzini, P. T. (2005). How college affects students: A third decade of research (Vol. 2). San Francisco, CA: Jossey-Bass.


Perna, L. W. (2006). Studying college access and choice: A proposed conceptual model. In J. C. Smart (Ed.), Higher education: Handbook of theory and research (Vol. 21, pp. 99–157). Dordrecht, The Netherlands: Springer.


Perna, L. W., & Titus, M. (2005). The relationship between parental involvement as social capital and college enrollment: An examination of racial/ethnic group differences. Journal of Higher Education, 76(5), 485–518.


Peterson, S. (2016). Community college student-parents: priorities for persistence. Community College Journal of Research and Practice, 40(5), 370–384.


Porchea, S. F., Allen, J., Robbins, S., & Phelps, R. P. (2010). Predictors of long-term enrollment and degree outcomes for community college students: Integrating academic, psychosocial, socio-demographic, and situational factors. Journal of Higher Education, 81(6), 750–779.


Portes, A. (1998). Social capital: Its origins and applications in modern sociology. Annual Review of Sociology, 24(1), 1–24.


R Core Team. (2015). R: A language and environment for statistical computing. Vienna, Austria: Author.


Reyes, M. -E. (2011). Unique challenges for women of color in STEM transferring from community colleges to universities. Harvard Educational Review, 81(2), 241–262.


Robinson, J. P., Shaver, P. R., & Wrightsman, L. S. (1991). Criteria for scale selection and evaluation. In J. P. Robinson, P. R. Shaver, & L. S. Wrightsman (Eds.), Measures of personality and social psychological attitudes (pp. 1–16). San Diego, CA: Academic.


Rosenbaum, J. E., Deil-Amen, R., & Person, A. E. (2006). After admission: From college access to college success. New York, NY: Russell Sage Foundation.


Ruscio, J., & Roche, B. (2012). Determining the number of factors to retain in an exploratory factor analysis using comparison data of a known factorial structure. Psychological Assessment, 24(2), 282-292.


Seber, G. A., & Lee, A. J. (2003). Linear regression analysis. New York, NY: Wiley.


Snyder, T. D., de Brey, C., & Dillow, S. A. (2016). Digest of Education Statistics 2014 (NCES 2016-006). Washington, DC: U. S. Department of Education.


Sorey, K. C., & Duggan, M. H. (2008). Differential predictors of persistence between community college adult and traditional-aged students. Community College Journal of Research and Practice, 32(2), 75–100.


Stahl, V. V., & Pavel, D. M. (1992, April). Assessing the Bean and Metzner model with community college student data. Paper presented at the annual meeting of the American Educational Research Association, San Francisco, CA.


Starobin, S. S., Smith, D. J., & Laanan, F. S. (2016). Deconstructing the transfer student capital: Intersect between cultural and social capital among female transfer students in STEM fields. Community College Journal of Research and Practice. 40(12), 1040-1057.


Steiger, J. H., & Lind, J. (1980, May). Statistically-based tests for the number of common factors. Paper presented at the annual meeting of the Psychometric Society, Iowa City, IA.


Summers, M. D. (2003). ERIC review: Attrition research at community colleges. Community College Review, 30(4), 64–84.


Tierney, W. G., & Venegas, K. M. (2006). Fictive kin and social capital: The role of peer groups in applying and paying for college. American Behavioral Scientist, 49(12), 1687–1702.


Tinto, V. (1975). Dropout from higher education: A theoretical synthesis of recent research. Review of Educational Research, 45(1), 89–125.


Tinto, V. (1993). Leaving college: Rethinking the causes and cures of student attrition (2nd ed.). Chicago, IL: University of Chicago Press.


Trujillo, A., & Diaz, E. (1999). “Be a name, not a number”: The role of cultural and social capital in the transfer process. In K. M. Shaw, J. R. Valadez, & R. A. Rhoads (Eds.), Community colleges as cultural texts: Qualitative explorations of organizational and student culture (pp. 125–151). Albany: State University of New York Press.


Tucker, L. R., & Lewis, C. (1973). A reliability coefficient for maximum likelihood factor analysis. Psychometrika, 38(1), 1–10.


United States Census Bureau. (2016). 2011–2015 American community survey 5-year estimates. Retrieved from https://factfinder.census.gov/faces/nav/jsf/pages/index.xhtml


Vaillant, G. E., & Milofsky, E. (1980). Natural history of male psychological health: IX. Empirical evidence for Erikson's model of the life cycle. The American Journal of Psychiatry, 137(11), 1348-1359.


Velicer, W. F. (1976). Determining the number of components from the matrix of partial correlations. Psychometrika, 41(3), 321–327.


Voorhees, R. A. (1987). Toward building models of community college persistence: A logit analysis. Research in Higher Education, 26(2), 115–129.


Walpole, M. (2003). Socioeconomic status and college: How SES affects college experiences and outcomes. Review of Higher Education, 27(1), 45–73.


Wang, X. (2013a). Baccalaureate expectations of community college students: Socio-demographic, motivational, and contextual Influences. Teachers College Record, 115(4), 1–39.


Wang, X. (2013b). Why students choose STEM majors: Motivation, high school learning, and postsecondary context of support. American Educational Research Journal, 50(5), 1081–1121.


Wang, X. (2015). Pathway to a baccalaureate in STEM fields: Are community colleges a viable route and does early STEM momentum matter? Educational Evaluation and Policy Analysis, 37(3), 376–393.


Wang, X., Wickersham, K., & Sun, N. (2016). The evolving landscape of transfer research: Reconciling what we know in preparation for a new era of heightened promise and complexity. New Directions for Institutional Research, 2016(170), 115–121.


Weis, L. (1985). Between two worlds: Black students in an urban community college. Boston, MA: Routledge & Kegan Paul.


Wells, R. (2008). The effects of social and cultural capital on student persistence: Are community colleges more meritocratic? Community College Review, 36(1), 25–46.


Whitbourne, S. K., Sneed, J. R., & Sayer, A. (2009). Psychosocial development from college through midlife: A 34-year sequential study. Developmental Psychology, 45(5), 1328–1340.




Cite This Article as: Teachers College Record Volume 120 Number 10, 2018, p. 1-46
https://www.tcrecord.org ID Number: 22249, Date Accessed: 4/18/2021 2:40:10 AM

Purchase Reprint Rights for this article or review
 
Article Tools
Related Articles

Related Discussion
 
Post a Comment | Read All

About the Author
  • Xueli Wang
    University of Wisconsin–Madison
    E-mail Author
    XUELI WANG is an associate professor in the Department of Educational Leadership and Policy Analysis at the University of Wisconsin–Madison. Her research interests center on college student pathways and success, with a focus on community colleges and undergraduate STEM education. Her recent work includes “Toward a Holistic Theoretical Model of Momentum for Community College Student Success,” published in Higher Education: Handbook of Theory and Research, and a coauthored piece, “Does Active Learning Contribute to Transfer Intent Among 2-Year College Students Beginning in STEM?” published in The Journal of Higher Education.
  • Kelly Wickersham
    University of Wisconsin–Madison
    E-mail Author
    KELLY WICKERSHAM is a doctoral candidate in the Department of Educational Leadership and Policy Analysis at the University of Wisconsin–Madison. Her research interests revolve around evolving community college student pathways and success, including student pathways in STEM. Her recent publications include coauthored pieces, “Turning Math Remediation Into ‘Homeroom’: Contextualization as a Motivational Environment for Community College Students in Remedial Math,” published in The Review of Higher Education, and “Math Requirement Fulfillment and Educational Success of Community College Students,” published in Community College Review.
  • Yen Lee
    University of Wisconsin–Madison
    E-mail Author
    YEN LEE is a doctoral student in the Department of Educational Psychology at the University of Wisconsin–Madison. Her research interests focus on nonnormal data generation, Bayesian statistics, and machine learning. A recent publication is a coauthored piece, “Construct Validity of the Activities of Daily Living Rating Scale III in Patients with Schizophrenia,” published in PLoS One.
  • Hsun-Yu Chan
    Texas A&M University–Commerce
    E-mail Author
    HSUN-YU CHAN is an assistant professor in the Department of Psychology, Counseling, and Special Education at Texas A&M University–Commerce. His research interests center on adolescent development, including peer influence and parental interaction, as well as community college student self-efficacy and success, particularly in STEM fields of study. His recent publications include coauthored articles “A Nuanced Look at Women in STEM Fields at Two-Year Colleges: Factors That Shape Female Students' Transfer Intent,” published in Frontiers in Psychology, and “Optimizing Technical Education Pathways: Does Dual-Credit Course Completion Predict Students’ College and Labor Market Success?” in the Journal of Career and Technical Education.
 
Member Center
In Print
This Month's Issue

Submit
EMAIL

Twitter

RSS