Narrative of Difference: The Effects of Social Class on Belonging at Liberal Arts Colleges

Low-income students have historically struggled in applying to colleges, especially liberal arts colleges. However, presently low-income students have had significantly more access to university due to changes in admission policies; the largest challenges now facing low-income students are during their time at university, where they struggle to maintain similar retention rates and comparable social and academic integration when compared to wealthier peers. In this research, I examined Grinnell College as a case study of liberal arts colleges as it is an interesting example due to no significant differences in average GPA or retention rates between social classes. I used an explanatory, sequential mixed-methods design to explore the experiences of low-incomes students at liberal arts colleges, focusing on belonging. I found that low-income students, despite claiming to belong at similar rates to wealthier peers, express narratives of difference where they have distinctly different experiences from wealthier peers. Differences included but are not limited to distinctive interpersonal relationships with faculty, staff, and peers, and low-income students’ lower participation levels in social events. However, despite these differences, lowincome students often “work the system” to equalize their differences. Based on my findings, I propose a 7-point plan for improving the experiences of low-income students at liberal arts colleges.


INTRODUCTION
Facing financial, informational, and academic insufficiencies, low-income students (LISs) have historically had less access to higher education.This narrative occurs across higher education despite the preponderance and efforts of universities.Additionally, LISs are also less likely to consider liberal arts colleges (LACs) because they often lack familiarity with the institutions; LACs are perceived as abundantly wealthy (i.e., culture of money; Somers, Woodhouse, & Cofer 2004;Ostrove & Long 2007) and lacking in vocational training.The aim of LACs to develop the whole individual, to train students to think independently across a wide range of disciplines, cannot be simply explained.For LISs, specifically those without the social and cultural capital to understand the differences among institutions of higher education, LACs are a place of privilege.They are an education developed by and for wealthier students, an education that does not contribute to their individual career goals or future plans.Furthermore, LISs struggle to explain what a LAC education provides to their familiesfamilies who often face even larger social and cultural capital gaps than those of their peers.
Despite this perceived gap, many LACs are some of the most economically diverse colleges in the nation.Vassar College, Amherst College, and Pomona College are nationally ranked eighth through tenth respectively on the college access index, a combined measure of students who receive Pell grants, the graduation rates of these students, and the price tag of colleges for low-and middle-income students (New York Times 2015).Grinnell College also ranks as one of the most economically diverse elite colleges in the nation, with 5.5% of the students from the top 1% (incomes greater than $630K) and 24.6% of their students from the bottom 60% (incomes less than $65K) (New York Times 2015).Grinnell College prides itself on this diversity, claiming that 19% of its student body self-identifies as a LIS today (D. Shorb, personal communication, September 1, 2017).This ranking is used to demonstrate that the college supports access to elite higher education for low-income students.
The statistics presented by Grinnell College nor the New York Times (2015) do not reveal the level of support on campus after acceptance, however; access is only the first obstacle that LISs must overcome in higher education; once there, retention and student success are the next challenges.With these secondary challenges, LISs often struggle to maintain similar retention and success levels when compared to their wealthier peers at other universities across the nation (Lynch & O'Riordan 2006;Ostrove & Long 2007;Pascarella, Pierson, Wolniak, & Terenzini 2004;Thayer 2000).
Interestingly, not all universities find differences in retention and success levels between LISs and wealthier peers.In particular, Grinnell College LISs do not have significantly lower GPAs nor retention rates (Stern & Wilcox 2017).Nevertheless, LISs and first-generation college students (FGSs)1 have noted improvements should be made to better their experiences, such as strengthening discussions of class and having first-generation and low-income student programming (Lange 2014).These improvements are crucial as Grinnell College has struggled with discussions regarding this disparity as far back as institutional memory dates (Jones 2000; United States.International students whose parents attended university outside of the United States are also considered FGS if they request the title.Lange 2014).In all, these suggestions for improving LISs' and FGSs' experiences demonstrate that there are different experiences between LISs and wealthier peers at Grinnell College and LACs in general.
Prior to the year 2000, there is little information on social class at Grinnell College.Jones (2000) studied the lives of LISs at Grinnell, finding four major themes: (1) Grinnell College was largely romanticized among the lower class; (2) a common disdain of wealth and therefore a dress-down culture limited the visibility of social class; (3) a perception that there were no LISs at Grinnell College; and (4) that there is no framework for pride of lowerclass roots.This research initiated the conversations held on campus since 2000.Beginning with the class awareness week discussions hosted by now Grinnell College Art History Professor Alfredo Rivera in 2002 through 2006, dialogue on social class has become much more prominent (A. Rivera, personal communication, November 17, 2016).Based on this background, Grinnell College provides an interesting case study of how LISs belong at LACs.
Here, I examine Grinnell College as an illuminating case of the larger trends at LACs across social classes.First, I present a critical review of the literature of LISs and FGSs, discussing larger trends across the nation.Second, I then describe my sequential mixed methods study, explain the quantitative and qualitative results, and conclude by suggesting policy changes to better support LISs and FGSs.I found that LISs exhibited no superficial quantitative differences in belonging; however, I discovered that LISs express narratives of differences but also increased effort to offset said differences when questioned about their experiences at LACs.

Background
LISs and Access to Higher Education Historically, LISs have had considerably less access to higher education than wealthier peers.Research has demonstrated that applying to and obtaining acceptance into university was the most challenging feat for LISs (Engle & Tinto n.d.; Pallais, Turner, & National Bureau of Economic Research [NBER] 2006; Stinebrickner & Stinebrickner 2003;Tinto 2007).Pallais, Turner, and NBER (2006) further found that three factors led to this challenge: (1) precollegiate achievement and preparation, as financial limitations prevent attending preparatory classes; (2) credit constraints, as LISs may be unable to finance college thus creating the need to work or live at home with parents; and (3) information constraints, as LISs (especially those who are also FGSs) may not know about opportunities at top schools such as financial aid offers and scholarships.But since the late 1990s, a considerable amount of evidence demonstrates the increase in access to higher education that LISs and FGSs have had since the late 1990s (Hoxby & Turner 2014;Pallais, Turner, & National Bureau of Economic Research [NBER] 2006;Stinebrickner & Stinebrickner 2003;Tinto 2007).Therefore, as LISs have seen increased access to college, presently the primary challenge for LISs is no longer applying to university, but rather being successful during their time at university.

Retention: A Function of Social and Economic Integration
As LISs have seen considerable increases in access to college, the primary challenges LISs now face during their time at university are retention, academic success, and social integration.In recent years, a lack of resources and support are the most common explanation for LISs dropping out (Jones 2000;Lange 2014;Lynch & O'Riordan 2006;Malecki & Demaray 2006;Ostrove & Long 2007;Pascarella, Pierson, Wolniak, & Terenzini 2004;Tinto 2008;Thayer 2000;Willett 2002;Zinshteyn 2017).In particular, Ostrove and Long (2007) discovered a "culture of money," where meal plans, decorated dorm rooms, and computer access demonstrate inequalities that affect the college experience.Many LISs cite these factors and others in exit interviews from university, explicitly stating that "college never felt like home" because of this culture of money (Ostrove & Long 2007).LISs are often alienated from their peers and feel as if they live "on the margin of two cultures" for which they are not prepared (Ostrove & Long 2007;Pascarella et al. 2004;Thayer 2000).Students from a lower socioeconomic status (SES) within the culture of money often express a sense of difference as their social and cultural capital differences are perceived through their visible financial differences.This represents a key gap in our understanding of LISs' experiences at university: it is unclear if LISs should adopt a middle-class or elitist perspective to fit in or if they should rely on their experiences as lower-class to thrive at university.Furthermore, Rubin (2012) found that social class and social integration are tightly linked: lower-class students were found to participate in fewer social activities.Significantly, Rubin (2012) found that nine factors ranging from finances for socializing to ethnicity mediate and/or moderate the link between social class and social integration, all of which relate back to social and cultural capital theories.Importantly, this research demonstrates that increasing the integration of LISs is a challenging and ambiguous trial; various actions such as spreading awareness of socioeconomic diversity on campus, increasing the number of free events on campus, and/or requiring students to live on campus may better assimilate LISs into campus culture.
Assimilation into campus culture can be further supported by strengthening learning communities and increasing knowledge and availability of emergency aid and financial resources.Drawing on student retention theory, which emphasizes policies supporting student engagement as a factor stimulating student retention and increasing social integration, Tinto (2007) calls for the melding of student affairs with coursework and rewards for professors who support retention efforts.Another suggestion is stronger academic integration, which would then lead to more thorough social integration by way of classroom bonding.Tinto (2008) asserts that when students learn together in a community, they are exponentially more likely to be socially integrated at university.These findings are well supported by other studies (Engle & Tinto n.d.;Pascarella, Pierson, Wolniak, & Terenzini 2004;Stinebrickner & Stinebrickner 2003;Terenzini, Cabrera, & Bernal 2001;Willett 2002;Zinshteyn 2017).
Financial limitations are another major factor impacting academic achievement and social integration for lowincome students at university.Students are less likely to have savings for emergencies and often have less money for entertainment.Kruger, Parnell, and Wesaw (2016) described different types of emergency aid that colleges and universities can provide to maintain high student retention, confronting the problem that universities often have emergency aid in the form of discretionary funds, but knowledge of these funds is spread by word of mouth and thus rarely used.This marketing strategy, word-of-mouth, is insufficient and leads to case-by-case management of requests where bias and stereotyping can impact decisions.

First-Generation Students
Another significant dimension of LISs is the identity of first-generation.Being an FGS is defined as being a child of parents who did not receive a degree from a four-year university, and this definition is widely accepted across the literature (Engle & Tinto n.d.;Lange 2014;Pascarella, Pierson, Wolniak, & Terenzini 2004;Somers, Woodhouse, & Cofer 2004;Thayer 2000).Although this identity often fits students who are low-income, the two identities do not always occur together.For example, when both parents are high school teachers, the low-income identity is appropriate but not first-generation; on the other hand, when parents labor in high-demand and high-pay jobs that require no college education, the first-generation identity is apt but lowincome is not.Furthermore, it can be described as an issue of paid employment and being an employee; families who have their own business may have high incomes but no college degrees.Somers et al. (2004) found that FGSs also face challenges distinct from LISs such as deferring extracurricular involvement to ensure academic success and survivor guilt over leaving families back home, contributing to feeling that students live "on the margins" of two cultures, facing a split-identity between home and college.

Conclusions
In summation, research on the low-income student experience has largely emphasized access to higher education rather than retention of low-income students.A recent shift in the literature toward examinations of resources and student success aims to address the newly arising issue of student support.(Jones 2000;Lange 2014;Lynch and O'riordan 2006;Malecki and Demaray 2006;Ostrove and Long 2007;Pascarella, Pierson, Wolniak, and Terenzini 2004;Tinto 2008;Thayer 2000;Willett 2002;Zinshteyn 2017).Factors such as survivor guilt and the culture of money are culprits in the lack of belonging, in creating the sense of living "on the margins" of two cultures (Ostrove and Long 2007;Somers, Woodhouse, and Cofer 2004).Second, LACs will share in a culture of money.Jones (2000) noted that Grinnell College did not have a culture of money, but rather a disdain of wealth and a dresseddown culture where students of all classes wear clothes stereotypically purchased by the lower class.I do not expect these findings to be presently accurate, however, as admission and financial aid policies have changed significantly within LACs, thus leading to differences in student attitudes.Third, belonging will be further lessened in LISs who are also FGSs.FGSs are less likely to academically and social integrated (Engle and Tinto n.d.;Thayer 2000), and I anticipate a synergistic negative effect of the LIS identity with that of FGS.

Social Class
In contemporary American society, class is marked by differences in wealth, material possessions, power, authority, and prestige distinguish people by social class (Lehmann 2009;University of Delaware n.d.).Key to understanding social class is that a definition is illusive.There is no consensus within sociology as to a precise definition as theorists have constantly debated a definition (i.e., Bourdieu 1987;Goffman 1951;Marx, Engels, and Moore 1959;Weber 1946;Wright and Perrone 1977).
Two classical sociologists, Karl Marx and Max Weber, discussed social class extensively and laid the foundation of social class dialogues today.Marx centered his analysis around two key assumptions: (1) since emergence from a primitive state, human society has remained divided based on class interests; and (2) the relationships between men are shaped by their relative access to the means of production, by their access to scarce resources and power (Marx et al. 1959).Marx emphasized the power dynamics in his discussions of social class: the owners had power over workers by determining the wage for their labor, and the workers were ultimately powerless to argue due to their politically powerless position within society (Marx et al. 1959).On the other hand, Weber (1946) argued that class was a quantifiable economic position, with shared life-chances and circumstances, and he emphasized the importance of property ownership in determining class.This economic position could be examined through three types of capital: material, relational, and cultural.
Both of these classical sociologists have influenced the conversation on how economic status is perceived.Since their publications, other sociologists have built upon their initial definitions.Wright and Perrone (1977) emphasized the lack of research utilizing Marxist definitions of social class in quantitative research.Attempting to bridge that gap in research, Wright and Perrone (1977) utilized Marxist definitions of social class in their quantitative research and concluded that there is a significant interaction between class position and income returns via education.Bourdieu (1987) also built on the work of Weber and Marx, emphasizing the importance of capital, of which there are four forms: social (Weber's relational), cultural, economic (Weber's material), and symbolic.It is important to note that symbolic capital was a new form conceptualized by Bourdieu (1987) which emphasizes the honor, prestige, and recognition that one holds, and this development plays out in my conceptualization through the lack of symbolic capital associated with LISs and/or FGSs.Using these four types of capital, Bourdieu (1987) detailed a coordinate map, where each type of capital positions individuals within larger society.Furthermore, it is at these positions that individuals adjust and develop what Erving Goffman (1951) calls a "sense of one's place."Acquisition of this sense is crucial, as individuals learn to behave similarly (Bourdieu 1987) and influence others' perceptions of social class (Goffman 1951).
Within the sociology of education, social class is variably defined and assessed due to the lack of consensus on one definition of social class.One such obscurity becomes evident when comparing the terms LIS, working-class student, and FGS.While working-and lower-class are often used interchangeably, there is a certain stigma attached to working-class that follows from associations with blue-collar work.For the purposes of this study, lower-class and lowincome were used in order to allay this stigma.Ostrove and Long (2007) defined social class through two methods: subjectively, by asking their identification with the labels "poor," "lower-class," "lower middle class," "upper middle class," or "upper class"; and objectively, by asking participants to indicate their annual family income on a fourteen-category scale ranging from <$10,000 to >$1,000,000.Jones (2000) based low-income status solely on subjective identification, ignoring quantifiable measures such as income.Another study measured socioeconomic status (SES) rather than income alone, combining annual parent income, educational attainment, and prestige of parent job to create an SES variable (Walpole 2003.This definition and methodology was not replicated in other studies. In this study, I adopt a nominal definition of social class by combining the aforementioned research with classical definitions.Relying heavily on Goffman (1951) and Bourdieu (1987), I claim that social classes are about the location of a person or group within the class structure of society based on economic, social, cultural, and symbolic capital.Within the college atmosphere, social class manifests in a plethora of means such as by dis/allowing financial expenditures, affecting the participation of the student in the culture of money, and impacting students' ability to focus on studies.Through these means, among others, social class manifests a student hierarchy at university, reflective of the hierarchy present in larger American society (Ostrove and Long 2007).Furthermore, this study focuses on four specific dimensions of social class: (1) social capital, which are the resources, information, ideas, and support provided by social networks (Lareau 1987;Pallais, Turner, and entertainment (Ostrove and Long 2007); (3) cultural capital, which are non-economic resources such as knowledge, skills, and education; and (4) identity, which incorporates social characteristics such as race, gender, sexuality, and the firstgeneration status into the intersection with social class (Engle and Tinto n.d.;Hoxby and Turner 2014;Lange 2014;Pascarella, Pierson, Wolniak, and Terenzini 2004;Somers, Woodhouse, and Cofer 2004;Thayer 2000).
Based on similar methods to Ostrove and Long ( 2007), both subjective identification with a social class and objective identification based on five income categories were utilized.Questions surrounding the material aspects of social class, aiming to examine the students' participation in the culture of money, were asked.These questions, such as interviewing participants about the activities they participate in with friends and the amounts of money they spend on such events, provided evidence of the effects of being a low-income student on belonging.
I also asked questions relating to identity to determine any trends and effects of the level of parents' education on college students.Students were questioned about their relationship with their family to determine the presence of survivor guilt (Piorkowski 1983).I also asked about the familial relationships prior to Grinnell, as well as to examine how the relationship dynamics have changed since entering college.Lastly, I questioned participants of their extracurricular involvement to see if students deferred their involvement until a later time for academic purposes.

Belonging
Belonging is not a common, explicit concept within the sociology of education.Instead, the focus is on academic and social integration.Ostrove and Long (2007) completed one of the few studies to stress the importance of belonging for LISs, emphasizing belonging due to the importance of it on social, psychological, and physical well-being.Therefore, as educational institutions have class-based markers that explicitly or implicitly define those who belong and those who do not, belonging is crucial when discussing the low-income student experience.
As Ostrove and Long ( 2007) are representative of the field (e.g., Leonhardt 2005;Ostrove and Cole 2003;Rubin 2012), I adopt their definition of belonging.The definition, which describes belonging as systematically created through social backgrounds and social networks, emphasizes various aspects of the college experience such as the culture of money.Two dimensions of this concept are central to my study: (1) academic integration, important for curricular learning; and (2) social networks, where the density and depth of connections to others affect belonging in all spaces (Pascarella, Pierson, Wolniak, and Terenzini 2004;Tinto 2007).Additionally, when considering both of these dimensions, the transition to college is also pertinent as this time is crucial in social network development and comfort in the classroom.
Furthermore, key to operationalizing belonging at university are Durkheim's (1995) four dynamics of belonging, adapted from the study of religious life: (1) physical co-presence of people, as face-to-face encounters are stimulating in various physiological and psychological ways; (2) a shared focus of attention, as sharing struggles over homework, college football, or singing in a choir can bond people; (3) ritualized common activities revolving around number two, like cheering at a football game, allowing people to fall into fairly set forms and patterns; and (4) exclusivity, which limits the groups' boundaries and allows for more tightly bound kinships (Chambliss and Takacs 2014;Durkheim 1995).Therefore, to measure academic integration, I asked participants various questions about their comfort in the classroom, ranging from topics such as being open about your social class identity to talking with professors about your class and family.Social networks were examined by asking about friendships and the depths of the relationships, group membership and extracurricular involvement, and faculty and staff relationships.To examine the transition to LACs, I asked participants if they felt supported by various offices on campus, but also what experiences challenged them and the support they received to overcome the challenge.I also inquired about participation in a pre-orientation program, as Jack (2015) found that these programs provide social and cultural capital.Furthermore, related to social and cultural capital, I surveyed participants of their pre-college education and experiences as this background is crucial to the development of social networks and academic integration.

Research Design
To explore the experiences of LISs at LACs, I conducted explanatory research to determine what meanings individuals give to their experiences at college, and to examine the differences in meanings and experiences by social class.Differences by social class were examined through both qualitative and quantitative methods in a sequential mixed method design as past studies of liberal arts students (i.e., Jones 2000) lack content validity as they do not measure objective and subjective identification with social class.Previous studies also fail to provide a complete and comprehensive description of how LISs belong at LACs.Therefore, I qualitatively examined differences to subjectively identify how experiences varied by class, focusing on meanings that students attached to their experiences and how students make sense of their experiences, and quantitatively examined differences to objectively identify variances.Both methodologies were necessary as each provides distinct evidence of differences, thus ensuring stronger content validity.The research was conducted with individuals using a crosssectional design.I used this design as I was not interested in change over time, but the current experiences to date and associated meanings of LISs at LACs.

Case Selection and Sampling
I conducted this research between August and December of 2017.The independent variable in my study was social class, and the dependent variables were the experiences of LIS and the extent to which they belong.The population of my study was liberal arts college students; the distinction of liberal arts students is crucial as their education significantly differs from that of other universities.A liberal arts education is distinct due to the small class sizes, greater number of faculty interactions, and the emphasis on classroom discussion rather than large lectures.These assumptions were significant as they allow the visibility of students' social and cultural capital within the classroom.The target population was Grinnell College students currently enrolled and on campus, as many offices on campus such as the Registrar and the Office of Financial Aid have full sampling frames of the student body.Due to confidentiality, a sampling frame was not obtained.
I selected cases through nonprobability sampling methods.I attempted to have a representative sample by: emailing minority student groups such as Grinnell QuestBridge Scholars (a low-income student group), Grinnell College's Concerned Black Students, and the Asian American Student Association asking them to distribute my survey link to their email list; sharing my survey link with residence hall coordinators at Grinnell College to email to their charges; and using Grinnell College Facebook pages (see individuals attach meaning to their experiences based on their social class.I also used interviews to examine further quantitative trends.Both the survey and interviews included measures to increase construct validity related to my theoretical discussions aforementioned. To study a greater sample of LISs at LACs, I used a survey.The triangulation of methods attempted to eliminate bias in my conclusions.Using a survey increased the reliability of my study, and allowed for comparisons to other studies (e.g., Ostrove and Long 2007) through the use of similar questions (i.e., "I feel that I fit in well at Grinnell College.")As surveys do not examine the complexity of social life, I also interviewed students.I wanted to understand personal experiences and perspectives, to get a "portrait of the people" (Adler and Adler 2003).

Biases and Relationship to the Topic
As a low-income and first-generation student, I established rapport with many individuals from the lower class quickly, understanding their experiences all too well; similarly, I established rapport with FGSs, despite class due to shared experiences.This familiarity with their experiences is both a challenge and strength in design, as I am in danger of being too close to the participants, but I am also able to draw from my own lived experiences to converse with LISs and FGSs about specificities of their own experiences.To ensure my experiences and identity as a LIS and FGS did not affect my data, I avoided taking a stance on the experiences of participants and maintained my position as a critical sociologist.On the other hand, my identity as low-income and first-generation challenged me when interviewing individuals from more privileged social classes, as I did not relate to them as easily.When interviewing these individuals, I was less challenged to maintain my critical position, and able to inquire further into responses given that I had no preconception of their lives.
In the interview sample, 6.3% came from families earning <$20,000 annually, 69% $20,000-50,000, 12.5% $100,000-200,000, and 12.5% >$200,000; this sample was purposely higher in LISs due to my focus on their experiences.In terms of race in the survey sample, 66% identified as white, 11.5% Asian, 5% Black or African American, 7.9% Hispanic or Latino, and 9.6% as mixed.A similar racial-ethnic composition was present in the interview sample.
$1 00,000-200,000 $20,000-50,000 $50,000-1 00,000 <$20,000 >$200,000 I also found that LISs felt as if their social class limits their participation in social events at LACs (F=19.66,P<0.001, Df=4,215; Figure 3) despite no significant attendance differences at social events between classes (F=1.6,P=0.18, Df=4,108).This raises a unique question then as it is unclear if these findings are because (a) LISs actually fit in at Grinnell (as Figure 2 suggests) but are just limited financially (as Figure 3 suggests) or (b) the questions influenced responses (see Appendix A for survey questions).Similarly, LISs were also fearful of sharing their parents' educational This result is representative of similar findings at other colleges.For example, research has demonstrated that LISs who have developed a positive framework for considering their social class inequalities often acquire their motivations from the shame placed on their parents. 4nterestingly, their motivations largely culminate in a desire to be middle class, and this desire stems from recognition of the limits their parents face in everyday life (Lehmann 2009).

Category
On the other hand, I found support for the second hypothesis, that LACs would have a culture of money.LISs noted how they felt that they stick out because of their class (F=2.5, P=0.044, Df=4,215; Figure 4).Research has shown that LISs hold this sentiment as they are in an unfamiliar environment (Thayer 2000).Their own lack of familiarity, in combination with the lack of familiarity from their family, leads to these students viewing their world with uncertainty; those who have supported them for their entire lives are no longer knowledgeable about their current situation.The disconnect the students face with university, as well with their families, leads to finding themselves "on the margin of two cultures" (Thayer 2000).It is important to note that these findings are likely confounded by students also being FGSs, not solely LISs.
Similarly, I found support for hypothesis three, that LISs who are also FGSs belong significantly less (F=2.64,P=0.036, Df=4,215).This finding is not particularly surprising, as between 2003-2004 approximately 64% of students with one identity (LIS or FGS) graduated whereas only 43% of those with both identities graduated (Pascarella et al. 2004).These findings are further supported by research demonstrating that FGSs are less involved in extracurriculars (Pascarella et al. 2004).
Lastly, I found significant differences in every parameter I used to investigate pre-Grinnell experiences, which are important for the development of social networks and academic integration (see Table 3).When questioned over access to various items such as good schooling in elementary or high school, ACT/SAT preparatory courses, or music lessons LISs had significantly less access than their wealthier peers (for all: F>6.12, P<0.001, Df=4,215).These findings are particularly important considering that research suggests pre-collegiate experiences, especially within education, are highly predictive of success during the college years (Stinebrickner & Stinebrickner 2003).I also discovered parents of LISs were significantly less involved pre-Grinnell in class choices (F=5.22,P=0.001, Df=4,215), extracurricular activities (F=5.09,P=0.001, and realistic life-experiences that help these students to overcome structural disadvantages.Despite full support for two of my hypotheses and partial support for one hypothesis, my quantitative data still left many questions unanswered.To start, I was interested in the differences between social classes in pre-Grinnell experiences.If pre-Grinnell experiences are predictive of college success (as found by Stinebrickner & Stinebrickner 2003), why do LISs at LACs, as evidenced at Grinnell, perform similarly to wealthier peers despite these differences?Furthermore, how do LISs apply to Grinnell without the support of parents, and how do they persevere despite considerably less support from parents?
One of the most important questions was why students superficially claimed to belong at similar levels despite class, but when questioned further in the survey differences by social class became clear.5I also wanted to explore further the support networks LISs had built for themselves and examine the role these networks may play in supporting similar average GPAs and retention rates among classes.Lastly, why did students feel more comfortable around faculty, staff, and students who share their social class background?In addition to this, I wondered how FGSs felt at Grinnell in academic environments and in relationships with faculty, staff, and other students.FGSs' relationships with family were also a significant variable that I had not considered in my survey, despite research suggesting that parents are significant providers of social support and mediate academic success at university (Malecki et al. 2006).

Qualitative Results
In their pre-college experiences, LISs often fall into a binary when considering their academic history: the doubly disadvantaged vs. the privileged poor (Jack 2015).The doubly disadvantaged, the poor who went to public high schools, are less represented at LACs but still present.These students are more often those that will struggle at college.When questioned about high school experiences, Maggie, a FGS, LIS, and double social sciences major, said: I went to one of the worst public high schools in the state of Kansas.almost no work and got all A's… It was a joke, really.Like, I took a few college courses that were harder, but there were zero opportunities to actually learn anything; we didn't even have AP classes.These low-resourced high schools were often in lower-class neighborhoods, and students were generally expected to go as far as the state school; LACs were never considered in the college decision process.But, for those students who chose LACs, they were often tracked there; many high schools with lower resources track students with aptitude and provided them with superior resources.
On the other hand, the privileged poor are those who attended private, wellresourced high schools, and they compose the majority of LISs at LACs.These students receive comparable if not better resources than their high-achieving, tracked doubly disadvantaged students.At these high schools, college after graduation is the norm, and college prep was largely successful: I went to a private college prep high school.The goal was that all of the graduates would go to college.My high school experience was stressful and very valuable.Coming into Grinnell I found that I was over prepared and most of what I learned in tutorial was kind of a repeat.The majority of LISs had parents who preferred the natural growth parenting strategy, where students are more independent (Lareau 1987), while wealthier parents preferred concerted cultivation, playing a more active role in the lives of students, such as in class choices or extracurricular activities.Facing tougher odds than wealthier peers, LISs often feel immense pressure during the college admissions process.During the college selection process, this effort often manifests in the "college hoe" phenomena.Sophia, a FGS, LIS, and STEM major of color, when questioned about the presence of pressure, said: Yeah, there was definitely pressure.The whole thing is that everyone does well, that everyone, well 99%, goes to college.So, you're stuck in this thing where you don't want to be a college hoe, but you are.
Everyone's just trying to get leadership positions… trying to be the best for every school.Interestingly, this phenomenon occurs in wealthier peers as well; however, wealthier peers are generally less anxious due to higher faith in their success.This faith manifests in applying to more schools in some cases (e.g., when students apply to more competitive schools such as the Ivy League) but less schools in others (e.g., when students apply to less competitive schools such as elite liberal arts schools).Wealthier high school students are also less likely to have a job in addition to their extracurriculars.Furthermore, the "college hoe" phenomenon occurs in LISs due to their own personal motivation and drive as well as social pressure, while wealthier students are more likely to face considerable parental pressure to become a "college hoe." When interviewing students about their experiences at Grinnell College, I found an interesting trend across class years: group chats played a significant role in friend group development for current firstyears, but not other class years.Every first year I interviewed mentioned being a member of at least 2 group chats through their Facebook class page.Students often joined groups based on music taste, sexuality, race, and sports, to name a few things: "We had a ton of group chats, and all the first-years kind of splintered off.I have been in one for a while now centered around music tastes, and that's the biggest one for me in terms of friends now."These group chats started as early as May, when their respective high school careers ended.Building these friendships early is especially important, as some research documents that friends can be viewed as a prerequisite for success in college (Chambliss & Takacs 2014).Despite many "not really planning on keeping these friends," the friendships were concrete from the whole summer of fairly constant activity.Other class years at Grinnell College did not have this experience and it showed in their stories of delayed friendship building, some as late as their second semester at Grinnell.As firstyear students comprised approximately 30% of my sample, it is possible that these early friendships led to no clear differences in belonging as Chambliss and Takacs (2014) suggest.
Another possible explanation of no differences in belonging is closeness with a staff member.LISs often had a close relationship with a (or many) staff member(s) at the college.On many occasions, they are incredibly close, and it is a symbiotic relationship; both groups are listening-ears, and both respond to the concerns of the other.LISs have particularly strong relationships as staff members are often more like home, and not academic, life.For example, Marie, a LIS and double language major, mentioned of her dining hall boss that, "I like to hear about her life a lot.She works more than she should, though."When talking, while the student may share about school work and the chef understands the experience despite not attending university, conversations center on the staff member; for students, these conversations are an escape from the academic environment.
One of the closest relationships I discovered was of a boss and her student employee, Annie: We actually went out last night.All the employees I worked with last night, they went out to get drinks, and they invited me to go with them.It's more of like a friendship now.This person still sees me as a daughter, and I her a motherly figure.
But like a friendly mom figure.I can talk to her.Annie, a senior FGS, LIS, and STEM major, has a particularly strong relationship with her manager as Annie is venturing into the workforce, and sees the manager as a mentor and mother figure, not solely a boss.The boss understands the college experience despite not attending, and she performs many of the duties that parents would normally.In many ways, this relationship fulfilled Annie's need for social support and understanding of the college experience that her family could not provide.
Part of why LISs succeed at LACs and belong at comparable rates to wealthier peers is that they are "working the system."This phenomenon begins as early as high school, when students may call on resources such as guidance counselors to make up for their lack of college admissions process knowledge.For example: It was really just whatever you wanted help with.And since I didn't have anybody, I turned to him [a guidance counselor] for everything.You were supposed to make meetings with him and such, but I'd just drop by and he'd be like, 'I have a meeting in 20 minutes,' and I'd reply, 'that's okay!We can take 10.'He was really helpful.This student, Sophia, an FGS, LIS, and STEM major, went out of her way to get the resources she needed.Many LISs often felt that they could always succeed at LACs (or anywhere) due to the resources available: "If I'm looking for a connection, I know I can find it.And that was my biggest thing when I came to college.I knew that wherever I go, I could find someone and the resources to support me.I knew I could do it."These students, while recognizing their independence, also note that relying on others and exploiting available resources is the best path towards success.
Working the system can also take other forms.Winnie, a first year FGS, LIS, and intended STEM major, shared that her family never had extra money to help her purchase running shoes.Thus, her father suggested she earn her money, despite being too young to work: During the school year I had a candy bar business… So, we weren't allowed to sell anything on campus, but I sold candy bars on campus every day.It's how I'd buy running shoes and concert tickets.It'd also pay for my running camp.I'd make like $800 during the school year.This student is the epitome of working the system: when faced with a financial barrier, she worked to equalize herself and maintain similar opportunities to wealthier peers.
At LACs, students also often work the system by working multiple jobs on campus.Many LISs work beyond required work-study hours, and they use this money for spending money during the semester.LISs often feel alienated from peers when others expect them to miss work.Maria, an FGS, LIS, and double social science and humanities major, explained the stress associated with how her athletic team always asked her to call in to work so that she could attend team events: Sometimes I get frustrated, and this is why I call my mom all the time.
No one understands how hard it was for me to get from my home to here.Like, it's because people are always like, 'I totally get it.'…but they don't understand that I have to work.Maria's experience emphasizes how LIS and wealthier peers face a disconnect in why working is important.Wealthier peers do not fully understand the depth of the differences in experiences with LISs.
Many students, however, expressed disdain for having to work the system.While many elite LACs often have large amounts of resources (Hoxby & Turner 2014), finding these resources is not always so easy.Kruger et al. (2016) points out that many universities use word of mouth tactics to spread knowledge of resources, but this method is especially not helpful for LISs or FGSs as they often do not receive this information through the grapevine.LISs often mentioned that jumping through the bureaucratic hoops to reach resources can be especially draining.Maggie said that: I think we're very invisible.We have to jump through a lot of hoops to get the resources we need here… I think there are some [resources], but to find them you have to email like 5 people and jump through all of these hoops to reach any help.When faced with the LAC workload and dealing with the additional stress of jobs on campus, the stress of having to search for resources yourself becomes excessive and can further impair the student.This stress grows when faced with the time that working the system can take.It is not an immediate process, and at LACs the process is often especially slow due to limited staff.Waiting this long for help only adds stress to the already stressful workload, and for LISs this may lead to a breaking point (see culture of suffering below).
LISs at LACs often feel alienated from peers.Henrietta, a senior LIS and Sociology major, put it bluntly: "rich kids, they don't even know!"One such reason for this is the dress-down culture present at many elite LACs (Jones 2000).Students of all classes often wear thrift-store clothing because it is fashionable; in doing so, wealthier students appropriate clothing.
Jerome, a senior person of color, LIS, and interdisciplinary major, faced a particularly challenging situation where dress-down culture hid a peer's class: So, my best friend in second year was in the 1%, and I had no clue.Like I knew he was like well-off, but everyone dresses in the shitty thrift clothing.But I started getting upset.
Why is it that he gets to be depressed and struggling like me, when he's super rich and has access to all these resources?That was really frustrating, and that limited my ability to relate to him.Thus, the dressed-down culture can hide social class and prevent you from finding others with similar experiences, making other LISs "exceedingly difficult to find."Students mentioned that, "after complaining a lot, I found other poor kids," but that, "you don't want to, like, out yourself as poor of sh*t."This conundrum is largely because money is stigmatized on campus; students do not want to discuss family financial history, especially if they have large scholarships.But for the most part, students simply say that social class "is just not a thing that's talked about" at LACs.As Goleman (2013) points out, it is largely because wealthier people care less about their appearance and can afford to do so in all senses: socially, financially, emotionally, and psychologically.For example, Zweigenhaft and Domhoff (1991) note how wealthy white college students at elite colleges can be found to walk around with holes in the soles of their shoes.LISs do not have this luxury and feel the need to work the system to equalize themselves with wealthier peers.
FGSs at LACs often feel uncomfortable sharing their parents' jobs and educational backgrounds.Their stress does not end here, however; while parents remain proud and supportive, they are not helpful and do very little.Often low-income parents prefer natural growth care (as opposed to concerted cultivation; see Lareau, 1987).Winnie described it as, "my parents' method of taking care of me was to let me motivate myself, so that without them I'd be fine.So, if I were stressing about an assignment or whatever, they wouldn't tell me to do it.They're not going to pressure me.They'd let me do it."Students described that parents rationalized this behavior by thinking in the long-term; LISs' parents often claimed that students need to be prepared and internally motivated without parent support as parents will not always be there.
On the other hand, some low-income parents try to be helpful and end up leading to negative consequences.Students in these situations are likely to express disdain towards their family: "I kind of just want them to be like, 'yes, I'm proud.'I wish I could just send them a resume and they could sign off with their pride.They don't need to know how I did it.They just need to know I came out how they wanted me to."This sentiment reinforces a split-identity in many FGSs.
The split-identity materializes when FGSs feel as if they are different people at home than at college.The academic atmosphere is extremely different from that of their normal lives.Part of this is the "first-gen thing" where parents are, "extremely happy.They want me to do well.They want me to be the best that I can be.But sometimes, they don't understand what I'm going through…It's a different life."But when talking about college, students often lessen the work families have to do to understand; they bridge the gap in knowledge and "translate" experiences.On top of the Grinnell workload, this effort can be extremely stressful.Students claim they "need them [family] to understand" and provide social support to make it through college.
The split identity is further exacerbated by the political and social atmosphere of LACs.Most low-income students do not leave LACs the same socially or politically.Jerome shared that his mother called him out for the changes, noting that, "once I went there, I knew how to fix the world.Since then I've been more assertive and even critical of even their parenting techniques and decisions."Jerome further mentioned that: The biggest change is that my politics have become really radicalized.I've also come to think things through much deeper and I don't want to come across as a huge know it all.I had to rethink how I talk to them.But since I've gotten a better handle of that, it's been better.I had to realize how much I've changed, and how that might be startling to what they knew of me.This narrative displays how changes induced by LACs often serve to further distance FGSs from family, exacerbating the stress of "living on the margins" of two cultures (home and college; Somers et al. 2004).
With FGSs, a common theme is that of independence.FGSs rarely have significant amounts of aid from parents (Engle & Tinto n.d.) often due to information barriers and gaps.Many did not have parental guidance in the college selection process, and this situation often does not change once at college.For example, Mickey, an FGS, LIS, and undeclared first-year described that: "I very much enjoy [independence].It's funny because I've talked to some people who had parents who were so involved in everything, and I was like, 'Wow, not much has really changed for me!' It's just like the same as before, except living somewhere else."Therefore, in transitioning to Grinnell and dealing with the day-to-day decisions of faux-adulthood, FGSs often feel more prepared than peers-and this sentiment is a first for them.FGSs are used to this liminal state between home and fuller, sociallyrecognized independence.Independence is the norm-confusion and a lack of a clear future is typical.But, to have the upper hand over wealthier peers is a new phenomenon.
Other students, despite being middle or upper class and having parents who attended university, also claimed this narrative of independence and confusion.Phillip, a wealthier Classics and Economics double-major, noted that, "I was the last of 4 kids, and my mom kind of took the time off."In describing his college selection process story, the student often used similar words to that of FGSs, coopting their language and stories as his own.It is important to note that his experience is greatly different from first-generation students, despite both being independent of their family; the first-generation identity is more than not just having help in applying to college.Often, FGSs lack the cultural and social capital to understand what help to ask of from others such as guidance counselors.FGSs also lack family that understand life at college.Lastly, the importance of material wealth underlies FGSs' experiences, as in the end it determines the choices one has and the decisions one makes.
Many students, despite class, have struggled with mental health issues at college.Part of this process is competing with others for who is doing worse; for example, Marie described it as "a big culture of the struggling Olympics.A lot of people, when they are engaging in conversation, they talk about how much work do you have to do.Are you 'dying right now'?" Another student described the culture of suffering from a faculty and staff etiology: "The workload is too f*cking much.It's taught me to balance my workload, but there's just so many things you can do that I still take on too much."Furthermore, LISs are often more negatively impacted by this culture.The workload is only exacerbated by paid jobs and extracurriculars, often to the point of utter exhaustion.Janet, a senior LIS and social science major, described the suffering: I'm constantly exhausted and it's not good for my mental health.Janet and other LIS often hit a breaking point: there is only so much they can bear.The stress of the culture of suffering adds to the stresses of identities such as low-income and first-generation.When you suffer from "living on the margins" of two cultures (home and college) as FGSs and many LISs do (Somers et al. 2004), the culture of suffering only furthers this stress.
Students often cite professors as the best aspect of Grinnell, no matter their class.Students with personal connections to professors often have the closest bonds.For example, Marie bonded particularly well with an Education professor as they shared a similar health-related illness: She's a very empathetic person, and we would start talking, and you know, we started bonding with each other.And then she asked me to be the note-taker.And we had coffee, and she offered me a position for next summer, for a week, and I feel like I can talk to her about anything… I would talk to her about academics, my family life, some of my social experiences, like, if I were having like academic worries, or even social worries, I'd be comfortable talking to her about that.Marie relied on this bond as an outlet from the academic world.This professor became a friend, not just a mentor, and it was often because of the professor's ability to simply listen to Marie.
Students often also form personal connections with professors based on shared identities.Susan, an LIS, FGS, and STEM double-major that identifies as a person of color, shared that: "I would say the first professor I got close with was my chemistry professor first semester.I remember going to her office and being like, 'are you Puerto Rican?'And she was like, 'yes.'And I was like, 'let's get to know each other.'"These shared identities allow bonding over minority experiences, and often lead to much deeper relationships.Susan further described that: "I've gone into her office and cried about Organic chemistry, and I've talked with her about the recent tornado.I've also talked to her about her son and my dogs.The connections I do form here with professors are enough to talk with them about anything."These connections with faculty are vital to student belonging at LACs.
Two other findings with professor relationships are evident: (1) Grinnell Science Project (GSP) often leads to strong relationships, and (2) language courses at LACs foster deeper bonds than other courses.GSP introduces students to faculty in their intended fields of study prior to New Student Orientation, and they often meet future advisors and professors.For example, Erika, a freshman FGS, LIS, and intended STEM and social science double major, met a Psychology professor, and realized how interested she was in the Psychology of Humor.Erika was so enthused that she "sacrificed [her]  In many cases, language courses at LACs are another huge boon for personal relationships.The small class sizes and oral exams often lead to professors knowing students incredibly well.Students are also in language courses for an extended period, and so conversations "often stray."A language professor often "asks how I'm doing and what my weekend was like.He doesn't want us to spend the whole two hours talking about nineteenth-century German lit, so we spend up to like twenty minutes a class talking about what our weekend was like and such."In diverging from the academic, language professors know students on a personal level and foster tight bonds.Furthermore, the content of language courses often forces closeness with faculty.Practicing vocabulary in class often includes storytelling about, for example, family and pre-college experiences.

CONCLUSIONS
In sum, my results support research claiming that LISs are at a disadvantage at university (Stinebrickner & Stinebricker 2003).I found superficially conflicting results between quantitative and qualitative data.In general, my quantitative data, which uses broad, thematic questions, (e.g., agreement with the statement "I feel like I fit in at Grinnell College") suggests that LISs fit in at Grinnell College but narrow, deeper questions show how: money is a limiting factor for LISs, belonging is limited due to social class, and fear and anxiety builds over sharing their parents' educational backgrounds.The qualitative data supports this superficiality of belonging as all experiences shared with me in interviews were tinged with anxiety, fear, and a lack of belonging-at least at some point during their collegiate training-due to social class.
These issues were often mediated by bonds built with faculty, staff, and friends of similar social classes to provide social support.These relationships help to mediate success at university as Malecki et al. (2006) has found, and these relationships, especially with staff members who did not attend university, represent an escape from the collegiate atmosphere.Similar to previous research (Lynch and O'riordan 2006;Malecki and Demaray 2006;Ostrove and Long 2007;Pascarella, Pierson, Wolniak, and Terenzini 2004;Tinto 2008;Thayer 2000; Zinshteyn 2017), I found that LISs less often belong at university due to factors such as alienation from peers and reduced social event participation due to financial limitations, despite no superficial quantitative differences in fitting in at LACs.Furthermore, I found that LACs have a culture of money and that LISs who are also FGSs struggle to belong even more at LACs.
The trends in data suggest that, when LISs claim to belong, they do so through a narrative of difference but also increased effort to offset said difference.LISs often feel alienated from peers, but they offset impact of differences by working more hours and hiding their class identity.Quantitative data shows little differences in typical measures of belonging and success at university, but in interviewing LISs it is clear that experiences are vastly different.Furthermore, the challenges for LISs do not end with their peers, as relationships with faculty and staff are also affected, and familial relationships are also hugely different.
My research is limited in that it did not use a random sample and may not be representative of all LACs nor did it examine retention of LIS due to a crosssectional research design.Future research should examine these limitations, working with LAC administrations to garner funds and institutional support for such studies.Additionally, research should also consider: (1) the cooptation of the first-generation identity by wealthier students, as FGSs experiences are significantly deeper than not having parental support in the college selection and application process; (2) faculty-student relationship differences across academic divisions and between social classes, as the surge in STEM enrollments has limited tight bonds between science faculty and students, and LISs and FGSs face considerable familial pressure to have a marketable degree, often leading to a STEM major; (3) admission policies and the doubly disadvantaged, as doubly disadvantaged students face considerable hurdles in applying to colleges, and changes in admission policies may better support their endeavors; and lastly (4) the culture of stress across social classes and the synergistic effects within LISs, as LISs' experiences at LACs may be considerably more challenging due to this phenomenon.
Despite these limitations, my research supports key proposals for changes that can better support LISs and FGSs at LACs.To maintain these similar rates of success seen at Grinnell College and the marginally different success rates seen at other LACs, we must develop policies with LISs and FGSs in mind.Therefore, I propose a seven-point plan for policy changes at LACs (see Table 4).These policy changes would better support LISs and limit how much they have to work the system in order to thrive-rather than merely survive-at university.They fail to account for still unequal rates of access, but, as aforementioned, access is no longer the primary problem facing LISs.These changes would make colleges more consistent and successful in their outreach and support of LISs at university and my research, especially qualitative data, support these actions at Grinnell College and other LACs.Campus resources should have drop-in hours throughout the week to support students in need.For example, LISs often face considerable stress when their financial aid malfunctions, so having drop-availability in the Financial Aid Office (as Grinnell College currently does and LISs' noted as helpful) would limit the duration of this stress.Furthermore, these stresses may also be limited in duration if knowledge of resources was spread more effectively.Word of mouth techniques alone are not sufficient, nor is a campus wide email.LACs must use a widearray of outlets to disseminate information on resources.

Figure 2 .
Figure 2. No significant differences were found in how students claim to fit in at Grinnell based on class.Percent Total Within Income Level

Figure 4 .
Figure 4. LIS often feel as if they stick out because of their class.Percent Total Within Income Level the National Bureau of Economic Research [NBER] 2006); (2) material wealth, which are environmental aspects such as college living and

Table 1 .
Grinnell College Facebook groups I used for requesting survey participation.

Table 4 .
Suggested policy changes to better support LISs and FGSs at LACs. 4. Do you like being at Grinnell College?How do you feel about your family and their views of Grinnell?•Do you act differently around them? o If yes, what caused the shift?oIf no, why do you think this is?Appendix C. Table 1.Results of ANOVA for each item ranked on a 4-point parentinvolvement scale in a Pre-Grinnell activity.Overall, these results demonstrate that social class likely influenced parental involvement in students' lives.These results are also an example of concerted cultivation vs. natural growth; low-income parents (anyone $50,000 or less) had less to do with their children than their wealthier counterparts.

Table 2 .
Results of ANOVA for agreements to statements by social class.Overall, the students were not that different based on class.The most important differences arose in buying supplies and books for classes, as well as the limits on participation in social events at Grinnell (although the students attended the same number of events on average-see below).There were also differences in what students and faculty students want to be around; low-income students prefer friends, faculty and staff (currently or previously) from a lower social class.However, students feel as if they fit in overall despite class differences.Despite class, students did not different in their presence in student organizations (P=0.987,F=0.08,Df=4,188), in their number of close friends (P=0.995,F=0.05,Df=4,179), or attendance at social events (P=0.18,F=1.6, Df=4,108).