College Success Among Students From Disadvantaged Backgrounds: “Poor” and “Rural” Do Not Spell Failure

Abstract

The popular view of students reared in poverty is that they fall short across a wide range of domains relative to their more advantaged peers. In this ongoing longitudinal study, we follow a cohort of college students who come from poverty and were awarded full financial support for four years at a large public research university. The results after two full academic years are striking for the lack of difference in dropout rate and grade point average between these economically disadvantaged students and their college peers. We suggest that it is not poverty per se that leads to poorer college academic performance in such students but rather the demand on their time and energy to meet ongoing financial needs.

Keywords

college success GPA poverty retention

Introduction

A college degree is widely accepted as a minimum academic credential for pursuing an economically successful life, but coming from a background of poverty can be an obstacle to success in higher education (Bjorklund-Young, 2016; Johnson & O’Hare, 2004; Khattri, Riley, & Kane, 1997; Lee, Daniels, Puig, Newgent, & Nam, 2008; Miller, 2014; Miller, Votruba-Drzal, & Setodji, 2013; Votruba-Drzal, Miller, & Coley, 2015). Even within poverty, there is considerable economic heterogeneity ranging from extreme to moderate that may differentially influence a student’s preparedness for college. Furthermore, poverty may have different effects depending on whether a student attended high school in an urban versus a rural area. In particular, rural schools are often assumed to fall short in preparing students for college relative to urban schools in economically similar communities (Corley, Goodjoin, & York, 1991; Cox, Tucker, Sharp, Van Fundy, & Rebellon, 2014; Downey, 1980; Fan & Chen, 1999). There remains a largely unexplored issue of a possible interaction between type of geographic community and socioeconomic status on academic performance.

It is often assumed that rural life is characterized by unemployment, little social capital, and serious difficulties for students intending to go to college. The literature partially supports this view. Rural secondary schools are smaller, provide fewer comprehensive classes, have less well-prepared teachers with fewer credentials, and have access to fewer resources, especially Advanced Placement courses, than urban schools (Cook & Van Cleaf, 2000; Corley et al., 1991; Fan & Chen, 1999; Grace et al., 2006; Khattri et al., 1997; Phillips, 2015; Provasnik et al., 2007; Reeves & Bylund, 2005; Yoder, 2008). In addition, strong family and community ties and an historical tie to manual labor may lower the perceived value of a college education and make moving from home difficult. However, the situation is far more complex than this. Smaller schools present a paradox. They can be problematic, in that they often have fewer resources; however, smaller class sizes can allow for more individualized attention. Similarly, strong family ties can make it hard to leave home, particularly if the family views do not support independence; however, if the family is supportive, having a strong social network could increase success (Azmitia, 2018; Radmacher & Azmitia, 2016; Wang & Nuru, 2017).

The conflation between rural residence and poverty is inherent in the distribution of wealth. Consider one small town in Kentucky: the estimated median household income in Adairville is US$33,523 and the estimated median income for Kentucky is US$46,659 (2014; city-data.com/city/Adairville-Kentucky.html). Even if restricted financial resources are adequate in a small town of 889 (2014 population of Adair), moving to a large city where costs are higher may create a financial burden. Clearly, there can be drastic differences in money and population size as students move from small, close-knit communities to large, diverse campuses. Indeed, the size of the incoming 2016 class at the university examined in our study is 4 times the entire population of Adairville. This points to the potential importance of environmental and sociocultural influences on economic status.

Attempts to address these issues have relied on between-group comparisons such as poor versus not poor; rural versus urban (supra vide). In this study, we examined within-group differences and their interactions. We asked whether rural versus urban home of origin makes a difference among students who all came from poor economic backgrounds. Does degree of poverty within the group of those living in federally defined poverty make a difference or is poverty relatively homogeneous? Are rural and urban poverty different in their association with academic success?

To address these questions, we are following a cohort of students admitted under a program designed to support students from federally designated poverty in attending college by providing full funding for 4 years (see “Methods” section). The combination of three characteristics make this a unique study. First, we examine differences among students from poverty. Second, we compare poverty effects between rural and urban students. Finally, this study is longitudinal (currently in its third year) and uses performance predictor variables including linguistic, cognitive and affective styles, stress/trauma history, anxiety, and depression, collected either before admission to college or at the beginning of the first semester. By following these students over the entire course of their college careers, we hope to illuminate what differentiates student outcomes among those students facing the disadvantages of economic hardship and rural environment.

Methods

Participants

This study has been approved by the local institutional review board. Fifty-four students were recruited from the Cardinal Covenant Program (CCP) at the University of Louisville in the fall of 2016. By Spring 2018, 47 students were still enrolled, a 2-year retention rate of 87% which is above the University’s first to second year overall retention rate of 80% and the national average of 72% (https://www.collegefactual.com/colleges/university-of-louisville/academic-life/graduation-and-retention/). As is true for many college-based student support programs, the CCP provides proactive and reactive academic help and peer/mentor social support as well as free psychological services for its students. What makes the CCP unique is that the program covers the full cost of tuition, books, and room and board, for 4 years. To be eligible, the student must be at or below 150% of the federal poverty level and be a resident of Kentucky. In 2017, those in a household of three at 150% of the poverty level had an annual income of US$30,630. For a household of four, the annual income was US$36,900. As all reported family incomes for the Cardinal Covenant Program students fell well below federally determined poverty levels, they were coded into four categories: less than US$9,999, US$10,000–US$19,999, US$20,000–US$39,999, and US$40,000–US$59,999. Specific family income data were missing for four students (three urban and one rural), yielding a total sample of 43 students for whom four semesters of grades were available. The students who apply for the Cardinal Covenant Program must have an ACT composite score of at least 20 and a high school grade point average (GPA) of at least 2.5. The mean ACT for the Cardinal Covenant

Program students in this study was 25.8; the mean for the entire college cohort from which the Cardinal CovenantPogram group was drawn was 25.5. In addition, students must complete an essay about how the CCP will impact their lives and how they will use it to reach their full potential as a student at the University of Louisville (Cardinal Covenant, 2017).

All students in this study were incoming freshman for the 2016 to 2017 school year. Of the original 54 students participating in this study, 61% were females (vs. 65.8% for entire student body), 35% were males, and 4% chose not to specify their gender. All participants were 18 years of age, with the exception of one 19-year old and one 20-year old. Fifty-seven percent of participants were self-reported as European American (vs. 72.3% for entire student body), 13% (vs. 10.2% for entire student body) as African American, 13% as multiracial (vs. 4.0% for entire student body), 9% as Asian/Pacific Islander descent (vs. 4.2% for entire student body), 6% as Hispanic/Latino descent (vs. 4.2% for entire student body), and 2% did not specify ethnicity.

Measures

Poverty

In light of the substantial difference in poverty levels among students’ families, in combination with the limited sample size, we created two poverty groups for within poverty comparisons: “extreme” poverty (EP; household income < US$9,999; n = 15) and “moderate poverty” (MP; household income equal to or greater than US$10,000; n = 30).

Rural–urban classification

The first step was to determine whether students came from an urban or rural area. ZIP codes were used to determine the county that a student resided in during high school. Because the rural–urban spectrum is so complex, there were three classifications used to create the criteria. The first and primary one was based on the 2010 U.S. Census Bureau classification. We combined urban clusters (2,500–50,000) and urban areas (50,000 and higher) and distinguished them from rural areas defined as everything not classified as urban cluster or urban area.

The second classification system that was used was the 2013 National Center for Health Statistics Urban–Rural Classification Scheme for Counties. This classification breaks down each county into six different categories. For the purpose of this study, the metropolitan categories were classified as urban, and the micropolitan and nonmetropolitan counties were classified as rural.

The first step in classifying students’ backgrounds utilized the U.S. Census Bureau classification. If at least 70% of the ZIP code population were considered rural or urban, then that designation was used without consulting the other two classifications. This applied to 24 ZIP codes and 36 students, which is over half of the database. If at least 70% of the population of the ZIP code was not considered rural or urban, the ZIP code was then compared across all three classifications. Only three ZIP codes (four students) did not align across all three classifications as either rural or urban. A decision about how to classify these areas was ultimately made after identifying the city that the ZIP code encompassed—a city with a population of less than 20,000 was classified as rural and a population over 20,000 was classified as urban. After using this method to classify students’ urbanicity, there were 38 urban students and 16 rural students.

Beginning College Survey of Student Engagement

Data from the Beginning College Survey of Student Engagement (BCSSE) 2016 were used to measure students’ high school preparation and their perception of college (see Cole & Dong, 2013 for psychometric data in support of BCSSE’s reliability and validity). The BCSSE has been used by over 450 institutions to measure students’ past school experiences and expected engagement in college. The BCSSE was administered to all incoming freshman at the University of Louisville prior to the start of the school year in the fall.

Three sections from the BCSSE were used for this study—expected academic perseverance, expected academic difficulty, and perceived academic preparedness. Expected academic perseverance includes questions such as “how certain are you that you will participate regularly in course discussions?” Expected academic difficulty includes questions such as “how difficult do you expect learning course material will be?” Perceived academic preparedness includes questions such as “how prepared are you to learn effectively on your own?” Each section was represented as the mean of items scored on a 6-point Likert-type scale from not at all difficult (1) to very difficult (6). This was used to measure students’ preparation from high school, along with number of Advanced Placement (AP) classes taken (question 6), high school GPA, and American College Testing scores.

Resilience

The Brief Resilience Scale (BRS) is a six-item, self-report questionnaire that assesses an individual’s ability to bounce back from stressful events (Smith et al., 2008) and is psychometrically one of the best measures available (Windle, Bennett, & Noyes, 2011). Students completed the BRS soon after classes started. Scores range from 1 (strongly disagree) to 5 (strongly agree) with the mean of the six items used to reflect resilience (the higher the number, the higher the self-reported resilience).

Academic performance

Student transcripts from the 2016 to 2018 school years were obtained and used to examine academic achievement in college. GPA was analyzed for the Fall 2016, Spring 2017, Fall 2017, and Spring 2018 semesters.

Analysis

The final sample in this report was restricted to students who had completed the first four semesters of college not including summer semesters. In combination with missing income data for four students, the final sample in this study is 43. Table 1 provides mean (standard deviation) high school GPA, ACT scores, and term GPA for four semesters for urbanicity by poverty groups. Table 2 provides comparison mean term GPAs for all Cardinal Covenant Program students and all Arts & Sciences (A&S) students (the cohort of which the Cardinal Covenant Program students were a subgroup) for the semesters examined in this study.

Table 1.

College Term GPAs by Urbanicity and Poverty.

	Urban		Rural
	Moderate poverty (n = 18)	Extreme poverty (n = 13)	Moderate poverty (n = 7)	Extreme poverty (n = 5)
HS GPA	3.7 (4.3)	3.7 (0.41)	3.9 (0.39)	3.9 (0.34)
ACT^a	24.6 (3.66)	26.3 (3.61)	28.1 (3.28)	24.4 (2.51)
College GPA^b
Fall 2016	3.3 (0.48)	2.8 (0.79)	3.1 (1.10)	2.6 (1.0)
Spring 2017	3.1 (0.89)	2.8 (0.78)	3.1 (1.41)	2.5 (1.01)
Fall 2017	3.1 (0.71)	2.9 (0.44)	3.4 (0.50)	3.1 (0.68)
Spring 2017	3.0 (1.13)	2.8 (0.83)	2.8 (1.50)	2.8 (0.90)

Note. GPA = grade point average; HS = high school.

^aSignificant Urbanicity × Poverty interaction, F(1, 39) = 5.110, p = .029.

^bNo significant main or interaction multivariate analysis of variance effects.

Table 2.

Mean GPA by Semester.

	A&S degree seeking Students (n = 6,809)*	All CCP Students (n = 128)*
College GPA
Fall 2016	2.92	3.03
Spring 2017	3.00	3.08
Fall 2017	2.95	3.16
Spring 2017	2.98	3.10

Note. A&S = Arts & Sciences; CCP = Cardinal Covenant Program; GPA = grade point average.

For the primary research question, we conducted a repeated measures multivariate analysis of variance over four semesters using GPA as outcome measure and poverty and urbanicity as fixed factors.

Results

Dropout rates for urbanicity and poverty levels were comparable. The percentages of dropout rates were 20% and 11% for rural and urban students, respectively, and 10% and 17% for extreme and moderate poverty students, respectively. The percentages of extreme poverty among students included in the study and among those who dropped out were 41.9% and 28.6%, respectively; percentages of rural students included in the study and among those who had dropped out were 38.7% and 42.9%, respectively. While urbanicity was not associated with dropping out during the first 2 years of college, extreme levels of poverty were associated with remaining in school.

Table 1 summarizes the results of multivariate analyses of variance with the GPAs of each of the four terms as dependent variables. There were no sex or ethnicity differences either in the distribution of students across poverty and urbanicity or in mean multivariate effects over the four terms.

High school GPAs, not surprisingly, were quite similar across all four groups. In contrast, there was a significant interaction of poverty and urbanicity on ACT scores with students from extremely poor, rural areas scoring significantly higher than the other three groups; F(1, 39) = 5.110, p = .029.

Three findings stand out in examining the results of the poverty by urbanicity analysis. First, over the four terms, there were no significant main or interaction effects. Second, between-subject analyses revealed a near significant main effect of poverty for the first term (Fall 2016), F(1, 39) = 3.294, p = .077. Neither urbanicity nor urbanicity by poverty approached statistical significance. A close examination of the mean GPAs over the four semesters reveals that the initial difference between students from extreme poverty and those from moderate poverty consistently decreased over each semester as reflected in decreasing statistical significance: p = .077, .170, .307, and .690 for Fall 2016, Spring 2017, Fall 2017, and Spring 2018, respectively (see Table 2).

We conducted a post hoc analysis to explore the role of resilience in academic performance. None of the group comparisons in mean scores yielded statistically significant results. Urban and rural students reported identical BRS scores (2.94); resilience was somewhat higher in students from extreme poverty (3.00) than from moderate poverty, 2.83; t(48) = 1.778, p = .082. With respect to the multiple regression across term GPAs, including resilience had a minor effect on the first semester difference in poverty groups increasing the statistical significance from .077 to .099. None of the other results approached statistical significance (all p values > .7).

Finally, we note that the Cardinal Covenant Program students in this study compare quite favorably with the entire undergraduate body in both college GPA and retention rate and that the entire Cardinal Covenant Program cohort (n = 128–107 over four semesters; 16.4% dropout) performed better than the A&S student body in each semester (n = 6,809–5,870; 14.7%); see Table 2 for details.

Discussion

The popular view of the academic performance of those from backgrounds of poverty is captured in a recent newspaper article. “In every measurable domain, young people reared in poverty experience poor outcomes. Compared to more affluent peers, they show poorer academic and cognitive performance, psychosocial well-being and physical health” (Galván, 2018). The overarching message from our study suggests another perspective. A background of poverty by itself does not condemn incoming college students to failure. Specifically, dropout rates were lower, and GPAs were higher for the poor students as a group compared with the performance across all A&S students. Perhaps this is because poorer students felt they had to work harder and overcome more to come to college, so they are more motivated and better prepared for the challenges of college than students who did not have to work as hard to get there (i.e., more affluent students). This is consistent with the curvilinear hypothesis of resilience, which essentially says if someone has never encountered obstacles, they will be more likely to have poor outcomes than those who encountered a moderate level of difficulties. Experiencing adversity (as long as there are also protective factors present) can be a protective factor in the long term due to giving people an opportunity to practice coping skills (McGee, Hóltage, Maercker, & Thoma, 2018). Another interpretation is that the cohort from poor backgrounds and rural areas that heard about and completed the paperwork for Cardinal Covenant Program, and college is a more restricted and higher achieving cohort in general than their peers. It might be argued that the better performance of the CCP students was a result of having selected students with better ACT scores and high school GPAs than the general student body. However, the mean ACTs for Cardinal Covenant Program and overall student body were almost identical. Finally, controlling for resilience, ACT score, and high school GPA did not significantly affect any of our findings, suggesting a more complex process.

Neither urbanicity nor extreme poverty in our study had a major impact on academic performance during the first four semesters of college. Indeed, the dropout rate of both groups was lower than the A&S rate overall. Extreme poverty may be associated with a slight risk (lower GPA) for the first college term, but this difference consistently declined over the four semesters of the study. This was not due to academically poorer students dropping out (and therefore increasing the group mean GPA) as the study only included students with data for all four semesters. It may be that having navigated the initial stress associated with starting college, any negative socioeconomic background effects diminish over time. We note as well that this group of students from extreme poverty reported, at the time of admission, a higher resilience (albeit not statistically significant) than students from less extreme poverty. This small difference may have had an impact on academic performance.

Two aspects of the Cardinal Covenant Program need to be considered in interpreting the success of these students. The program offers mentoring and regular contact with support staff, a common feature of such programs. Such services have been associated with improved student college success (e.g., Tinto, 2016). Second, and much less common, is the economic support given the Cardinal Covenant Program students. While expected to work part time, the students do not have to worry about the necessary costs of college, namely tuition, books, and room and board. (Paradoxically, this sometimes creates other problems related to money, such as families pressuring students to share some of “the wealth.”)

Clearly, there is no simple explanation for why the students have done so well. At this point, the important message is to acknowledge that poverty, whether rural or urban, does not necessarily condemn college students to poor outcomes.

Limits and Strengths

The sample size (n = 43) in the study was small and limited to a single institution attended largely by within state residents, restricting the strength of our findings and favoring the absence of statistically significant results. This was especially true in examining the interaction of poverty and urbanicity. Clearly, the study of different colleges (including community colleges) and universities with different student populations is necessary to gain a more thorough understanding of how students from disadvantaged economic backgrounds can be academically successful. The longitudinal design of the study and the unique sample, however, offer an unusual look at how students from poverty fare in college and suggest that college success may be more complicated than is usually understood.

Conclusion

Atypical for most research reports, our findings are largely characterized by the absence of statistically significant differences. In contrast to the expected view, our study at the 2-year point suggests that there are well-prepared (emotionally, cognitively, and socially) students from disadvantaged backgrounds who will do well in college if they do not have to worry about money while in school. The Cardinal Covenant Program, in its complete financial support of its students, removes everyday worries about money that would otherwise be expected to distract poor students and require major commitments of time and energy in remunerated work. In the absence of that financial support, it is not uncommon for students to work 20 to 40 hours a week at a paying job, live with family at some distance from campus to save rent, and spend the weekends working fulltime. To point to such students as being ill-prepared and therefore at risk for academic failure miscasts the role of finances. Without the worry about generating income, students from poverty chosen for the same competencies as nonpoor students are just as likely to do academically well as their economically advantaged counterparts. This is not to suggest that there are no initial difficulties, in cultural adjustment, for example, or that such students come with the same educational exposure/experience as more advantaged students. Indeed, providing support to ameliorate the initial stresses of college especially during the first semester is important especially in what may be a situation not primarily overcome by resilience. However, the initial results of our study suggest that if we remove the immediate, practical need for income among students from poverty, we should have every reason to believe that they will do well. That they come from poverty does not condemn them to experiencing “. … poorer academic and cognitive performance, psychosocial well-being and physical health.” Moving forward, we suggest two changes that must occur to make a significant difference: a change in our understanding of and attitudes toward students from poverty and the willingness to provide substantial economic support to remove constant, practical worry about money from the academic environment. Finally, we hope that the results of this study will encourage others in varied settings to identify how and why students who often expected to do poorly, in fact do well.

Footnotes

Acknowledgments

The authors would like to thank Tony Robinson and Chelsea McKendree who enthusiastically supported and facilitated the initiation of this project. This study literally would have been impossible without their support. Since we initiated this study in 2016, the Cardinal Covenant Program (begun in 2007) has been discontinued in favor of a program that distributes significantly less money to a larger number of students from poverty. This change in policy will allow us to examine in some detail the impact of total versus partial support of these students. Our findings would predict that with less financial support students with economic hardships will be less successful academically than those equally poor students who have been economically strongly supported. This will be answered empirically moving forward.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Rich Lewine is a professor in the Department of Psychological and Brain Sciences at the University of Louisville and a licensed clinical psychologist in Kentucky. He teaches undergraduate psychology courses and serves as a supervisor in the graduate clinical psychology training program.

Kayla Manley graduated with Honors with a major in psychology from the Univesity of Louisvile in 2018. Her honors thesis focused on urbanicity and constituted a significant contribution to the current manuscript. She is currently an administrative assistant at the Leadership Alliance in St. Louis, Missouri.

Geoff Bailey is the Executive Director of REACH and an Adjunct Faculty member in the College of Education and Human Development at the University of Louisville. He teaches graduate coursework for both the College Student Personnel and Higher Education Administration programs.

Ashlee Warnecke graduated from the Department of Psychological and Brain Sciences at the University of Louisville in 2018. She is currently completing a clinical psychology residency at the Michael E. DeBakey VA Medeical Center, with a focus on trauma and anxiety disorders, and is provisionally licensed in the state of Texas. Upon completion of residency, she plans to begin as a staff psychologist with the PTSD Clinical Team at the Battle Creek VA Medical Center.

Darlene Davis graduated from the Department of Psychological and Brain Sciences at the University of Louisville in 2018. She completed her internship at the Dingell VA medical Center in Detriot, MI and is continuing her work with homeless veterans with parenting needs.

Alison Sommers is currently the assistant director for Academic Services in Psychological and Brain Sciences at the University of Louisville. She has been a psychotherapist at a university counseling center and in private practice.

References

Azmitia

(2018). “ Dropping out is not an option”: How educationally resilient first generation students see the future. New Directions for Child & Adolescent Development, 2(160), 89–100.

Bjorklund-Young

(2016). Family income and the college completion gap [Johns Hopkins Institute for Education Policy Commentary]. Retrieved from Johns Hopkins School of Education: http://edpolicy.education.jhu.edu/family-income-and-the-college-completion-gap/

Cardinal Covenant. (2017). Student financial aid office. Retrieved from University of Louisville: http://louisville.edu/financialaid/cardinal-covenant

Cole

Dong

(2013). Confirmatory factor analysis of the BCSSE scales. Bloomington: Indiana University Center for Postsecondary Research. Retrieved from http://bcsse.indiana.edu/pdf/BCSSE%20Psychometric%20Report.pdf

Cook

D. W.

Van Cleaf

D. W.

(2000). Multicultural perceptions of 1st-year elementary teachers’ urban, suburban, and rural student teacher placements. Urban Education, 35(2), 165–174.

Corley

R. E.

Goodjoin

York

(1991). Differences in grades and SAT scores among minority college students from urban and rural environments. The High School Journal, 74(3), 173–177.

Cox

G. R.

Tucker

C. J.

Sharp

E. H.

Van Fundy

K. T.

Rebellon

C. J.

(2014). Practical considerations: Community context in a declining rural economy and emerging adults’ education and occupational aspirations. Emerging Adulthood, 2(3), 173–183.

Downey

R. G.

(1980). Higher education and rural youth. Paper presented at the annual Kansas State University Rural and Small School Conference, Manhattan, KS (ED 201459). Retrieved from https://files.eric.ed.gov/fulltext/ED201459.pdf

Fan

Chen

M. J.

(1999). Academic achievement of rural school students: A multi-year comparison with their peers in suburban and urban schools. Journal of Research in Rural Education, 15, 31–46.

10.

Galván

(2018, November 13). Neuroscientists identify a surprising low-tech fix to the problem of sleep-deprived teens. The Conversation.https://theconversation.com/neuroscientists-identify-a-surprising-low-tech-fix-to-the-problem-of-sleep-deprived-teens-106651

11.

Grace

Shores

E. F.

Martha

Brown

Aufsesser

Bell

(2006). Rural disparities in baseline data of the early childhood longitudinal study: A chartbook. Retrieved from National Center for Rural Early Childhood Learning Initiatives, Mississippi State University Early Childhood Institute: https://eric.ed.gov/?id=ED495855

12.

Johnson

K. M.

O’Hare

W. P.

(2004). Child poverty in rural America. Reports on America, 4(1), 1–20.

13.

Khattri

Riley

K. W.

Kane

M. B.

(1997). Students at risk in poor, rural areas: A review of the research. Journal of Research in Rural Education, 13, 79–100.

14.

Lee

S. M.

Daniels

M. H.

Puig

Newgent

R. A.

Nam

S. K.

(2008). A data-based model to predict postsecondary educational attainment of low-socioeconomic-status students. Professional School Counseling, 11(5), 306–316.

15.

McGee

S. L.

Hóltage

Maercker

Thoma

(2018). Sense of coherence and stress-related resilience: Investigating the mediating and moderating mechanisms in the development of resilience following stress or adversity. Frontiers in Psychiatry, 9, 378. doi:10.3389/fpsyt.2018.00378

16.

Miller

(2014). Urbanicity, income, and academic and behavioral functioning across childhood: Longitudinal associations and mediating mechanisms (Doctoral dissertation, University of Pittsburgh). Retrieved from http://d-scholarship.pitt.edu/23728/1/Miller_PL_ETD_final.pdf

17.

Miller

Votruba-Drzal

Setodji

C. M.

(2013). Family income and early achievement across the urban-rural continuum. Developmental Psychology, 49(8), 1452–1456.

18.

Phillips

(2015). I had to learn that on my own: Successful first-generation, low-income college students from rural areas at an urban institution (Doctoral dissertation, University of Louisville). doi:10.18297/etd/2003

19.

Provasnik

KewalRamani

Coleman

M. M.

Gilbertson

Herring

Xie

(2007). Status of education in rural America. Retrieved from https://nces.ed.gov/pubs2007/2007040.pdf

20.

Radmacher

K. A.

Azmitia

(2016). Family emotional support and the individuation process among Asian- and Latino-heritage college-going emerging adults. Journal of Research on Adolescence, 26(4), 979–990.

21.

Reeves

E. B.

Bylund

R. A.

(2005). Are rural schools inferior to urban schools? A multilevel analysis of school accountability trend in Kentucky. Rural Sociology, 70(3), 360–386.

22.

Smith

B. W.

Dalen

Wiggens

Tooley

Christopher

Bernard

(2008). The brief resilience scale: Assessing the ability to bounce back. International Journal of Behavioral Medicine, 15(3), 194–200.

23.

Tinto

(2016). Research and practice of student retention: What next? Journal of College Student Retention: Research, Theory & Practice, 8(1), 1–19. doi:10.2190/4YNU-4TMB-22DJ-AN4W

24.

U.S. Census Bureau. (2010). County classification lookup table [Table]. Geography: Urban and Rural. Retrieved from https://www.census.gov/geo/reference/urban-rural.html

25.

Votruba-Drzal

Miller

Coley

R. L.

(2015). Poverty, urbanicity, and children’s development of early academic skills. Child Development Perspectives, 10(1), 3–9.

26.

Wang

T. R.

Nuru

A. K.

(2017). “ He wanted me to achieve that for our family and I did, too”: Exploring first-generation students’ experiences of turning points during the transition to college. Journal of Family Communication, 17(2), 153–168.

27.

Windle

Bennett

K. M.

Noyes

(2011). A methodological review of resilience measurement scales. Health and Quality of Life Outcomes, 9, 8.

28.

Yoder

M. J.

(2008). Rural students’ access to and success in higher education: A case study. Dissertation Abstracts International Section A, 69, 1303.