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

Navigating Disparate Pathways to College: Examining the Conditional Effects of Race on Enrollment Decisions


by Mark E. Engberg & Gregory Wolniak - 2009

Background/Context:This study draws from the growing body of research dedicated to understanding how students navigate disparate pathways to college. The extant research has demonstrated the various stages that undergird the college choice process, drawing largely from economic and sociological perspectives related to human capital formation, status attainment, and social and cultural resources. Despite this growing body of research, our understanding of the college choice process across racial groups remains inconsistent and unclear.

Purpose/Objective/Research Question/Focus: The purpose of the current study is to better understand the various factors and resources that influence the decision to attend a particular college or university, with an emphasis on how the choice process manifests across different race groups. Through the development of a series of statistical models, we offer a unique glimpse into whether a common model of college choice exists, while exploring the possibility that the process and underlying factors that influence matriculation may differ depending on one's racial group membership.

Research Design: Using admissions and financial aid data from eight private colleges and universities, we performed secondary data analyses on general and race-conditional models of college choice.

Analysis: The current study primarily used blocked logistic regression to understand the main and conditional effects of enrollment for the general population and across four race groups.

Conclusions: Our findings suggest that the college choice process unfolds differently for students depending on their racial group membership. Although our models captured a significant amount of the variance for each racial group, our study reveals important distinctions related to the college choice process based on students' academic preparation, the overall academic quality of their secondary milieu, and the strength of established feeder networks between a particular high school and college. Overall, our findings provide conceptual guidance for researchers interested in studying the college choice process and emphasize the importance of examining conditional effects to fully appreciate how the process unfolds for all students.

The American system of higher education is composed of approximately 15 million students who attend a vast collection of institutions that differ by cost, selectivity, social prestige, and programs of study. As the number of individuals entering the higher education system has increased, so too has the socioeconomic and demographic diversity of the college-going population (Baker & Velez, 1996; Goldrick-Rab, 2006). However, considerable research has demonstrated that increased higher education participation and attainment do not necessarily lessen inequalities at the societal level (Corcoran, 1995; Haveman & Smeeding, 2006; Lucas, 2001). Simply put, the American higher education system increasingly reflects the differentiated and stratified American society (Labaree,1990; Rhoades 1987), and there is a need to better understand the effects of different pathways of students into college, as well as how pathways may eradicate or reproduce inequalities among students.


Although college enrollment rates have consistently increased over the last two decades, both Blacks and Hispanics continue to lag behind White students. A recent report by the National Center for Education Statistics (NCES, 2004), for instance, estimates a 9% lag for Blacks and a 12% lag for Hispanics in 2001 college enrollment rates. Further, both Hispanics and Blacks are more highly concentrated in 2-year rather than 4-year institutions, whereas those who do enter 4-year institutions experience lower levels of 4-year baccalaureate completions rates and inferior labor market outcomes (NCES, 2003). Although there has been a concomitant increase in college choice research that pays attention to racial differences (Alon & Tienda, 2007; Hurtado, Inkelas, Briggs, & Rhee, 1997; Perna & Titus, 2005), most studies focus on the decision to attend or not attend college, leaving a gap in our empirical understanding of how students who decide to attend college choose among the various schools to which they were admitted. Given the emphasis that many schools place on diversity, an even smaller body of research exists that examines the final enrollment choices among students of varying racial group memberships. Within this small body of work, researchers have often reached very different conclusions about the likelihood of enrollment for different racial groups (cf. Perna, 2000; Plank & Jordan, 2001), and few studies have specifically addressed how models of college choice differentially predict enrollment at a particular college (e.g., Perna, 2000).


Thus, the central purpose of our study is to better understand the process by which students belonging to different racial groups choose to enroll in a particular college or university. Such a process involves a variety of sequential decisions, beginning with forming educational aspirations and ending with students’ decisions to enroll in college (Cruce, 2004; DesJardins, Ahlburg, & McCall, 2006; Flint, 1992; Hearn, 1992; St. John, Asker, & Hu, 2001). At every stage of this process, the success of any given student is confounded by a disparate array of economic, cultural, and social resources. In addressing the various factors and resources that influence the decision to attend postsecondary education, we propose a series of statistical models to understand how the college choice process manifests across different race groups. In doing so, we offer a unique glimpse into whether a common model of college choice exists, while exploring the possibility that the process—and underlying factors that influence matriculation—may differ depending on one’s racial group membership.


THEORY AND EVIDENCE


The college choice process has been aptly described as a tripartite model that includes an initial predisposition stage, in which students develop educational aspirations, a search stage, in which students search for information about colleges and determine their choice set, and a choice stage, in which students make decisions about enrolling at a particular college or university (Hossler & Gallagher, 1987; Terenzini, Cabrera, & Bernal, 2001). Although researchers have proposed a number of theories to understand these stages, our conceptualization of the college choice process is largely shaped by economic and sociological perspectives related to human capital formation, status attainment, and social and cultural resources (Hossler, Braxton, & Coopersmith, 1989; Paulsen, 1990; Perna, 2006).


Despite a growing body of research that has operationalized these perspectives and highlighted important variables related to the college choice process (e.g., students’ background characteristics, socioeconomic characteristics of families, and contextual characteristics of peers, neighborhoods, high schools, and institutions; see Ehrenberg & Sherman, 1984; Fuller, Manski, & Wise, 1982; Heller, 1997; Kane, 1999; Leslie & Brinkman, 1987; Paulsen, 1990; Weiler, 1994; St. John, et al., 2001), our understanding of the college choice process across racial groups remains unclear. Hurtado et al. (1997), for example, demonstrated that Black students were less likely to attend their first-choice institution compared with White students, whereas other studies uncovered that Black students were more likely to enroll in college than White students (Perna, 2000; Plank & Jordan, 2001); similar inconsistencies were also evident among studies examining Hispanic students (cf. Perna, 2000; Plank & Jordan).


In the section that follows, we provide a closer review of the extant literature related to the college choice process and the disparate pathways to college that exist for different students. We organize past research according to the perspectives that have dominated college choice research (i.e., economic and sociological perspectives) and focus on empirical evidence that highlights differences based on students’ racial group memberships.


ECONOMIC PERSPECTIVES


Human capital theory lends itself to our understanding of the college choice process by grounding the decision to attend college in the language of productivity-enhancement and investment returns (Becker, 1993; Paulsen, 2001). Within this theoretical framework, attending college is based on a rational decision in which the potential gains in productivity (and therefore improved earnings and other monetary or nonmonetary returns) are compared with the direct and indirect costs associated with acquiring a college education (Cohn & Geske, 1990).


Although human capital theory postulates that education increases productivity net of a variety of background characteristics, enrollment in postsecondary institutions is also a function of family income and other economic resources (Ellwood & Kane, 2000). Researchers, for instance, have shown that the enrollment decisions of students from lower income families are more sensitive to changes in college costs (Avery & Hoxby, 2004; Haveman & Smeeding, 2006). Research also demonstrates the clear link between financial aid offered to students and their likelihood of enrolling in postsecondary education (Avery & Hoxby; Catsiapis, 1987).


One of the strongest human capital predictors of college enrollment is academic preparation (Perna, 2004), which researchers have operationalized using a number of different constructs, ranging from enrollment in college preparatory tracks (e.g., Perna, 2000) to the highest level of mathematics coursework completed (e.g., Perna & Titus, 2005). More direct measures of academic preparation or achievement, such as standardized test scores (e.g., Perna, 2000; Perna & Titus, 2005) and high school grade point averages (e.g., Ellwood & Kane, 2000), demonstrate a strong relationship with enrollment in postsecondary education (Perna, 2000).


Although this framework accounts for some of the observed differences in college choice patterns, researchers interested in identifying and understanding group differences in college choice need to take into account additional factors, such as access to financial resources (e.g., student financial aid, current loan limits; see Paulsen, 2001) and the overall demand for human capital (Catsiapis, 1987). Heller’s (1997) review of state grant expenditures and tuition pricing, for instance, found that changes in both of these components had stronger impacts on enrollment outcomes for Asian, Black, and Hispanic students when compared with White students. Massey, Charles, Lundy, and Fischer’s (2003) research also demonstrates that minority students attending selective colleges have greater exposure to economic risks, requiring them to rely relatively more on financial aid to both access and persist in college.


Similar to financial resources, studies indicate that differential exposure to information resources affects college choice. Students’ decisions about college are predicated on access to information and an understanding of the college choice process. Research has shown that limited access to information and a lack of understanding of college choice may particularly disadvantage first-generation students (NCES, 2004) and that differences in information resources partially explain disparities in college enrollment among low-income, Black, and Hispanic students (Perna, 2006).




SOCIOLOGICAL PERSPECTIVES


Theories of class reproduction apply a sociological perspective rooted in the concepts of social and cultural capital and have been useful in explaining the influence of the social context (e.g., family, community, and social supports) on the likelihood of attending a postsecondary institution. For example, resources accessed through social networks enable students to gain access to other forms of capital and institutional supports that facilitate college enrollment (Morrow, 1999). The concept of habitus has additionally been used in college choice research to explain how an internalized system of thoughts, beliefs, and perceptions acquired through one’s parents or immediate community shapes the college choice process for students (McDonough, 1997; Perna & Titus, 2005).


Status attainment models have been employed to understand how ascribed characteristics (e.g., socioeconomic status, race, and gender) and achieved characteristics (e.g., academic preparation and performance) influence educational aspirations, which in turn predict educational attainment (Perna, 2006). For example, research on the relationship between gender and the likelihood of enrollment has been explored in a handful of studies and has produced mixed results. In some studies, women appear more likely to enroll (Perna & Titus, 2005), whereas other studies demonstrate equal propensities for both men and women (Perna, 2000).


In terms of high school contexts, McDonough (1997) demonstrated that in addition to academic and socioeconomic characteristics, secondary school characteristics are important factors to consider in the college choice process. McDonough’s study focused primarily on the role of guidance counseling, whereas Perna and Titus’s (2005) study examined the structural context of the high school by analyzing the amount of resources accessible to parents through social networks at the school. Operationalizing the volume of resources available to students (including the average levels of parental involvement, family income, parental education, and parental educational expectations), a strong relationship was found between resource availability and likelihood of attending a postsecondary institution. Noting several important differences across racial groups, Perna and Titus concluded that “Blacks and Hispanics not only possess fewer types of capital that promote college enrollment but also attend schools with fewer of the resources that promote college enrollment” (p. 509).


From a sociological perspective, it also appears that structural barriers may pose differential access to institutional resources for individuals belonging to different racial and ethnic groups (Dika & Singh, 2002; Lin, 2001). Recent research on structural barriers has emphasized the role of the quality of the high school and the historical relationships that have existed between a particular high school and college. Wolniak and Engberg (2007), for instance, found that students attending better quality high schools (as measured by average college-going rates, standardized test scores, and incidence of AP test taking) were less likely to attend a particular college. This particular study demonstrated the potential for students to have a larger set of college alternatives based on the overall quality of their secondary institution. Similarly, Espenshade, Hale, and Chung (2005) found that as two different indicators of high school quality increased (i.e., average SAT I scores and per capita senior AP test taking), the likelihood of gaining admissions at an elite college decreased.


In addition, Wolniak and Engberg (2007) uncovered the positive impact that historical relationships between high schools and colleges can have on the likelihood of matriculating at a particular college. This study raised questions of structure versus agency in the college choice process and revealed that established institutional networks were found primarily among White students and within high-quality high schools and wealthier communities. Similarly, Person and Rosenbaum (2006) introduced the concept of chain enrollment to help explain the college choice decisions among immigrant students, who often apply to and enroll at postsecondary institutions attended by other students within their social network. Their findings among Latino/Latina students suggest that established networks of social contacts are particularly important in obtaining information about college and may be important determinants of matriculating at any given college.




THE STUDY


Based on the mentioned literature, it appears that the college choice process unfolds according to a range of economic, social, and cultural resources that inform student decision making, determine opportunity, and may ultimately differ by racial groups. Researchers, for instance, have turned to the major tenets of human capital theory and theories of class reproduction in estimating empirical models of college choice. These models have operationalized the college choice process by focusing on the effects of students’ background characteristics (e.g., preferences and attitudes related toward educational attainment as well as a particular college), socioeconomic characteristics of families (e.g., household income and parent educational attainment), contextual characteristics (e.g., factors related to peers, neighborhoods, and high schools), and characteristics of institutions (e.g., cost of attendance, academic profile as a proxy for academic selectivity and social prestige, and programmatic offerings; see Ehrenberg & Sherman, 1984; Fuller et al., 1982; Heller, 1997; Kane, 1994; Leslie & Brinkman, 1987; Paulsen, 1990; Weiler, 1994; St. John et al., 2001).


Building on these previous models, we propose a model of college choice that employs a confluence of variables representing aspects of students’ social, cultural, academic, and financial resources, including measures of high school quality, the percentage of minority students within both a given high school and college, and the historical admission patterns that characterize the strength of social networks between high schools and colleges. These variables are conceptually linked to human, social, and cultural capital and are particularly relevant to understanding college choice decisions across different racial groups (Perna, 2000, 2006).


Specifically, we address questions related to college choice and the mechanisms by which different racial groups navigate pathways to college. Our primary aim is to identify if and to what extent different models of college choice exist for White, Black, Asian, and Hispanic students. Through our analysis, we address the following questions: (1) What are the net effects of students’ race and other background characteristics, high school characteristics, college search and recruitment variables, and college characteristics on matriculation? This question examines the extent to which prevailing college choice models explain matriculation patterns for our sample of students. (2) Is the model of college choice general, or is it conditional on race? This question examines if a common model of college choice exists across students of varying racial group memberships.


METHODS


DATA AND SAMPLE


Data for this study were drawn from a multi-institutional sample of eight private 4-year colleges located in the Northeast, Southeast, and Midwest regions of the United States. Our sample included all admitted first-time freshman students who entered college during the fall of 2006. The eight participating schools were selected for the study based on the availability of complete admissions and financial aid records obtained during the fall 2005 and spring 2006 recruitment and admissions cycle. Because the information collected was based solely on the population of students admitted to one of the participating institutions as of the final census date (typically the final date at which students are allowed to drop or add a class without penalty), our sample is fully representative and inclusive of the admitted and matriculated populations at each of the respective schools.


The sample was geographically and demographically diverse and consisted of 16,207 students, providing a statistically robust pool for analysis and generalizing results. White students and females represented the modal populations (77% and 59%, respectively), with Blacks, Latinos/Latinas, and Asians representing approximately 8%, 5%, and 10% of the population, respectively. The average academic composite score for the sample was approximately 5.5 based on a 10-point scale (described next). In terms of the participating institutions, their admitted populations ranged from a low of roughly 1,000 students to a high of 4,500 students, and total enrollments (based on Integrated Postsecondary Education Data System [IPEDS] data) ranged from a low of 1,245 students to a high of 12,134 students. Across all the participating schools, in-state enrollment represented 40% of total enrollment, with a range of 16% to 84%. In terms of institutional resources, the average percentage of financial need met was approximately 84%, and the average percent of students of color enrolled was around 24%. Table 1 provides a full descriptive review of all variables used in the analysis.


Table 1: Descriptive statistics of variables in model (N = 16,207)


 

 

 

 

 

 

 

 

 

 

M

SD

Minimum

Maximum

Label

Background Characteristics

     
 

White

0.766

0.423

0.000

1.000

White

 

Asian

0.103

0.304

0.000

1.000

Asian

 

Black

0.084

0.277

0.000

1.000

Black

 

Latino/a

0.047

0.212

0.000

1.000

Latino/a

 

Female

0.588

0.492

0.000

1.000

Female

 

Expected family contribution
(per $1,000)

58.104

41.662

0.000

99.999

EFC

 

Academic Performance Index

5.496

2.539

1.000

10.000

AcadIndex

High School Characteristics





 
 

Percent students of color

23.790

17.633

0.000

99.078

HSSOC

 

Academic strength

0.770

0.217

0.006

1.000

HSQuality

 

Feeder legacya

9.272

5.080

0.000

15.000

Feeder

College Search / Recruitment





 
 

Financial aid applicant

0.560

0.496

0.000

1.000

AidApp

 

Number of cross-applications

2.600

2.562

0.000

6.000

Crossapps

 

Total grant (actual) b

0.267

0.202

0.000

1.148

Tgrant

College Characteristics





 
 

Percent students of color

24.198

8.730

12.875

36.522

CollegeSOC

 

Academic strength

0.846

0.121

0.594

0.973

CollegeQuality

 

Average financial need met

84.401

14.280

57.000

100.000

AvgNeedMet

College Choice Indicator





 
 

Matriculation decision

0.274

0.446

0.000

1.000

Matric

 

 

 

 

 

 

 

a Feeder legacy is a three-item scale measuring the strength of ties between high schools and a given college or university. Constituent items include how many times, during the last 5 years, a

 high school produced an application to the college, an admitted student to the college, or a matriculate to the college. Αlpha reliability = .928


b Total grant (actual) measures students' total grant offered as a proportion of the institutions' cost of attendance.




KEY VARIABLES


The dependent variable in the model was dichotomous, with 1 indicating that a student matriculated at a particular college or university, and 0 indicating that the student did not matriculate. We organized our independent variables around four clusters that represented students’ background characteristics (both ascribed and achieved), high school characteristics, college search and recruitment variables, and college characteristics.


Within the first group of background characteristics, we included three ascribed demographic and socioeconomic measures: a dichotomous variable for gender; separate dummy variables that represented White, Black, Latino/Latina, and Asian racial group memberships; and a continuous variable that measured students’ expected family contribution (EFC) based on the Free Application for Federal Student Aid (FAFSA). In keeping with current federal methodology, which caps EFC at $99,999, we recoded all nonaid applicants to this value. The final EFC values were divided by 1,000 to create a scale with values ranging from 0 to 100. To measure students’ achieved characteristics and variations in human capital prior to college, we created an academic preparation variable. In operationalizing academic preparation, we used a continuous variable representing students’ overall academic profile based on a composite of their high school grade point average, high school percentile rank, and standardized test scores (i.e., SAT or converted ACT). We segmented raw scores of each component into deciles and assigned equal weight, giving our final academic profile variable a range of 1–10 (see Table 2).


Table 2: Mean component and overall academic performance index by race (N = 16,207)



 

 

 

 

 

 

 

White

Asian

Black

Latino/a

 

(n = 12426)

(n = 1668)

(n = 1356)

(n = 767)

High School GPA

3.50

3.67

3.40

3.49

High School Rank

0.82

0.91

0.84

0.83

SAT (ACT Converted)

1199.51

1328.84

1121.08

1185.03

Academic Performance Index

5.43

6.88

4.65

5.37

 

 

 

 

 


To control for differences in students’ social and cultural capital based on observable characteristics of their high schools, we included three variables related to the structural context. First, we constructed a composite measure of the overall academic quality of a student’s high school that included each high school’s average standardized test scores, percentage of AP test takers, AP test takers scoring a 3 or higher, and percentage of college-bound seniors. Second, we created a variable that measured the percentage of students of color at any given high school. This variable was constructed using national demographic data that identified race composition at the ZIP code level (obtained from Claritas, a market research firm) in conjunction with a crosswalk defining the percentage of each ZIP code that feeds into a particular high school. Finally, we included a variable that represented the historical feeder patterns between a given high school and college in our sample. This three-item construct was based on a 5-year enrollment history (2001–2005) that included the number of years a particular college received (1) applicants, (2) admitted students, and (3) matriculants from a given high school. These components were based on the aggregation of College Entrance Examination Board (CEEB) codes included in the recruitment and admissions data from each of the participating colleges. We then created a summated scale with values ranging from 0 to 15 and an inter-item reliability of 0.93.


Our next set of variables was related to the college search and recruitment process. We included a dummy variable representing whether a student filed a FAFSA, and an additional continuous variable representing the number of cross-applications that a student included in his or her choice set (based on the six available Title IV codes within the FAFSA). Our past research has shown that students who submit the FAFSA to a particular school typically yield at higher rates and demonstrate a stronger affinity toward that school than nonaid applicants. Similarly, the number of Title IV codes included on the FAFSA is an important proxy for understanding the size of a student’s choice set, where the size of the choice set is inversely related to the likelihood of matriculating at any one college. We also included a variable representing the total amount of institutional, state, and federal grants a student received from a particular college. To control for variance in college costs, we expressed total grants in terms of the actual cost of attendance at a given college.


The final set of variables included a number of controls for differences across our sample of eight colleges based on data obtained from the Common Data Set (CDS). We created two continuous variables that represented the percentage of freshmen students of color on a particular campus and average percent of financial need met for freshmen. In addition, we developed a composite measure of institutional quality based on the following variables: average SAT scores, admit rates, freshman-to-sophomore retention rates, and 4-year graduation rates.


ANALYTICAL DESIGN


Our analytical design consisted of three stages. First, we began by generating means and intercorrelations for all the variables in our study, paying particular attention to issues related to multicollinearity. In assessing mean values, we performed t tests to understand whether there were any significant differences across race groups for a selected group of variables in our model and to what extent racial differences exist in students’ access to resources important for entering college.


For the second stage of analysis, we applied multivariate techniques to understand which of our independent variables had a significant influence predicting matriculation. Given that matriculation was a dichotomous variable, we ran logistic regression models on four sets of independent variables (i.e., background characteristics, high school characteristics, college search and recruitment variables, and college characteristics). To establish the main effects of the model on students’ likelihood of matriculation, we ran our logistic model on the entire sample to assess the effectiveness of one prevailing (or general) model of college choice.


For the third and final stage of analysis, we proceeded to run separate models for White, Black, Asian, and Hispanic students. This last stage enabled us to assess the presence of conditional effects. In other words, this approach provided evidence of whether different models explain college choice for students of different racial groups.


More formally, our model is represented in Equation (1), where p is the probability of matriculating, x1 is a vector of student background characteristics, x2 is a vector of high school characteristics, x3 is a vector of variables representing college search and recruitment, x4 is a vector of variables that controls for characteristics of the colleges in our sample, and ε represents random error (see Table 1 for the variables within each vector). Equation (2) represents the race-conditional model where x’1 excludes the race variable, while all variables are uniquely identified according to students’ race.


Log (odds) = ln ﴾p /)1 – p)﴿ = B0 + B1x1 + B2x2 + B3x3 + B4x4 + ε

(1)

  

Log (odds)i = ln ﴾p /)1 – p)﴿i = B0i + B1i x’1i + B2ix2i + B3ix3i + B4ix4i + εi

(2)

 

where i = White, Asian, Black, or Latino/a


LIMITATIONS


With this study, our ability to address the research questions was ultimately limited by our data resources. In particular, missing data introduced the threat of response bias, and our sample of private institutions potentially compromised the study’s external validity. In terms of missing data, we were limited by the partial information available for students who did not file for financial aid at a particular school. For modeling purposes, we made the assumption that nonfilers had no demonstrated need. Our experience, however, suggests that nonfilers constitute a proportion of students who truly have no demonstrated need and a proportion of students for whom a particular college did not make their top six in terms of the available positions on the FAFSA. Thus, for this group of nonfilers, we have no real understanding of their choice set, their actual financial resources, and whether they actually filed the FAFSA with other schools. We did include an aid applicant dummy variable in our models as a means of controlling for this lack of information.


In terms of our institutional sample, the eight colleges and universities contained within our data represent only a small proportion of the American postsecondary system. The institutions within our sample are all private and vary in terms of their academic profile, enrollment, and locations, as well as institutional financial resources and aid offered to students. Although our sample is representative of the schools in our population and does not suffer from response or sampling biases found in survey-based research, we recognize that our findings may not be generalizable to all private colleges and are certainly limited in application to the public sector. Compared with public institutions, the admissions practices at private colleges incorporate a more general and inclusive approach to determine admissibility and in awarding financial aid (Alon & Tienda, 2007; National Association for College Admission Counseling, 2006). Therefore, the effects we found related to college search and recruitment, as well as background and high school characteristics, may only apply to the college choice process at private institutions.  


We were also limited methodologically by the data sample used in the study. Our sample, for instance, was based on the availability of institutional data, which limited our ability to apply more advanced techniques such as hierarchical linear modeling (i.e., students were not randomly selected within high schools). Future studies, however, are planned that will incorporate more advanced sampling procedures in which students nested within particular high schools will be randomly selected, allowing us to better isolate the impact of secondary school environments on the college choice process and to determine whether student preferences are partially mediated by the high school milieu.


RESULTS


GROUP MEAN COMPARISONS


We began our analysis by conducting t tests to understand whether there were significant mean differences across White students and students of color for selected variables in our model. As presented in Table 3, we noted several differences across variables related to financial need and financial aid awards. Asian students, for instance, were associated with the highest average EFCs (which accompanied the lowest average need) compared with White students, whereas Black students demonstrated the lowest EFCs (and highest average need) compared with White students. It is not surprising, therefore, that Asian students were also associated with the lowest average grants offered at the participating schools and the lowest average percent of aid applicants compared with White students, whereas the opposite trend was found among Black students.


In examining, the average academic profile of the students in our sample, we found that Asian students scored significantly higher than the White students in our sample. Black students, however, were associated with significantly lower academic profile scores compared with White students, and no significant mean differences were found among Latino/Latina and White students.


In terms of the high school structural variables in our model, White students attended high schools with the lowest average percent of students of color compared with the other race groups in our sample; Black students attended high schools with the highest average percent of students of color. In relation to the academic quality of high schools, Asian students attended significantly higher quality high schools than White students, whereas Black students attended significantly lower quality high schools. Furthermore, White students demonstrated access to more established feeder legacies, suggesting that the colleges in our sample have cultivated stronger relationships with high schools that provide a larger pool of White applicants compared with all other racial groups in our sample. This also suggests that White students have access to more established networks of institutions and the benefits that accompany high school-to-college linkages.


Table 3: Sample means by race (N = 16,207)


 

 

 

White

Asian

Black

Latino/a

 

(n = 12,416)

(n = 1,668)

(n = 1,356)

(n = 767)

EFC

$59,624

 

$64,288

**(+)

$37,643

**(-)

$56,227

 

AcadIndex

5.426

 

6.821

**(+)

4.624

**(-)

5.294

 

HSSOC

20.683

 

27.694

**(+)

42.293

**(+)

32.883

**(+)

HSQuality

0.778

 

0.818

**(+)

0.624

**(-)

0.778

 

Tgrant

0.268

 

0.221

**(-)

0.323

**(+)

0.257

 

Feeder

9.461

 

8.535

**(-)

8.720

**(-)

8.782

**(-)

AidApp

0.553

 

0.472

**(-)

0.729

**(+)

0.559

 

 

 

 

 

 

 

 

 

 

(+) Mean value is significantly greater than the combined mean of the other groups.

(-) Mean value is significantly less than the combined mean of the other groups.

*  p < 0.01. ** p < 0.001.


GENERAL MODEL OF COLLEGE CHOICE


We began our multivariate analyses by running a logistic regression model on our entire sample to understand the overall effects of students’ background characteristics, high school characteristics, college search and recruitment variables, and college characteristics on students’ likelihood of enrollment. We ran the model using a three-stage blocked logistic modeling procedure and found that each block was associated with a significant R2 that increased with each additional group of variables (Nagelkerke, 1991; see Table 4).


In examining students’ background characteristics, we found that Asian, Black, and Latino/Latina students were all significantly less likely to attend one of the participating schools in our sample compared with their White counterparts; Black students were associated with the lowest odds ratio (nearly 50% less likely to attend). Students who were less needy were associated with greater odds of attending one of the colleges in our sample, as reflected in the estimated effect of EFC. However, the direction of this effect changed as college-level variables were introduced to the model, suggesting that financial aid applicant status and total grant offered may be suppressing the unique effect of EFC. We also uncovered a highly significant inverse relationship between academic profile and likelihood of matriculation, with stronger academic profiles associated with lower odds of matriculation. This result is consistent with our expectation that a more academically competitive student has more educational options and is therefore much less likely to enroll at any one institution, irrespective of the characteristics of his or her high school, search activities, or college characteristics.




Table 4: Parameter estimates, B, and odds ratios, Exp(B), predicting student matriculation (N = 16,207)


 

 

  

Model 1

Model 2

Model 3

  

B

Exp(B)

 

B

Exp(B)

 

B

Exp(B)

 

[1] Background Characteristics

 

Asian (vs. White)

-0.279

0.756

**

-0.235

0.790

**

-0.459

0.632

**

 

Black (vs. White)

-0.555

0.574

**

-0.443

0.642

**

-0.700

0.497

**

 

Latino/a (vs. White)

-0.362

0.696

**

-0.289

0.749

*

-0.470

0.625

**

 

Female

0.041

1.042

 

0.014

1.014

 

0.081

1.084

 
 

EFC

-0.007

0.993

**

0.002

1.002

 

0.004

1.004

*

 

AcadIndex

-0.114

0.892

**

-0.113

0.894

**

-0.204

0.816

**

High School Characteristics


 



 



 
 

HSSOC

0.002

1.002

 

0.005

1.005

**

0.003

1.003

 
 

HSQuality

-0.198

0.820

 

0.107

1.113

 

-0.321

0.726

*

 

Feeder

0.027

1.027

**

0.022

1.022

**

0.025

1.025

**

[2] College Search / Recruitment

 

AidApp

   

 2.232

9.323

**

2.362

10.613

**

 

Crossapps

   

 -0.370

0.691

**

-0.393

0.675

**

 

Tgrant

   

 0.884

2.422

**

1.737

5.681

**

[3] College Characteristics

 

CollegeSOC

      

0.038

1.039

**

 

CollegeQuality

      

2.166

8.727

**

 

AvgNeedMet

      

-0.004

0.996

 
 

 

 

 

 

 

 

 

 

 

 

Model R2

 

0.062

**

 

0.127

**

 

0.149

**

 

Model R2 represents the Nagelkerke R2 statistic based on the likelihood ratio test with a range of 0–1.


* p < .01 ** p < .001



Of the three high school-level variables in our model, we only uncovered significant effects for the high school quality and high school feeder legacy variables in the presence of the full set of variables. High school quality was associated with a significant negative effect on matriculation, which suggests that students from higher quality high schools are less likely to matriculate at any one given college. As with our interpretation of the earlier finding related to students’ academic profile, students from higher profile high schools are likely afforded more opportunities to access a range of postsecondary institutions. In addition, the high school feeder composite variable was a highly significant, positive predictor of students’ likelihood of matriculation, suggesting that students attending high schools with more established historical relationships with a particular college are more likely to attend that college.

We also uncovered highly significant findings for all the college search and recruitment variables in our model. Aid applicants, for instance, were much more likely to matriculate when compared with nonaid applicants. Given that nonaid applicants were students with no demonstrated need, as well as students for whom one of the participating schools was not part of their choice set listed on the FAFSA, our finding that aid applicants are more than 10 times more likely to matriculate seems reasonable. Higher total grant awarded was also a strong positive determinant of enrollment, indicating that financial aid leveraging does work in providing additional incentives for students to attend a particular college. Finally, as students’ choice set increases (i.e., more schools are listed on the FAFSA), there is a concomitant decrease in their likelihood of enrollment.


Finally, in relation to college-level effects, we uncovered two significant relationships: Students who were admitted to colleges with a higher level of structural diversity (i.e., percentage of students of color on campus) were associated with a greater likelihood of enrollment, and students who were admitted to colleges with higher institutional profiles were more likely to enroll than students at lower profile colleges.


RACE-CONDITIONAL MODEL OF COLLEGE CHOICE


Our final set of multivariate analyses centered on understanding if and how our general model of college decision making differed across race groups. All of our models were associated with significant R2 statistics, whereas the Black and Latino/Latina models were the strongest fitting models regardless of specification, explaining over 20% of the variance in students’ matriculation decision when all variables were included (Nagelkerke, 1991; see Table 5). Among students’ background characteristics, higher academic profiles were negatively associated with the likelihood of matriculation, which was consistent across race groups, although the magnitude of the effect was smallest among White students and greatest among Latino/Latina students.


Among the structural characteristics of high schools, the percent of students of color was a significant and positive influence on matriculation among White students only—the group who attended secondary schools with, on average, the smallest proportion of students of color. Unlike our individual measure of academic profile, higher academic quality high schools were associated with a significant negative effect on the likelihood of matriculation among White and Black students only, although there were no significant differences in estimated effects across the different race groups. In addition, coming from a high school with an established feeder legacy had a positive and significant effect for all groups except Black students, although the estimated effect was only significantly different among Black and Latino/Latina students.


Among all the variables in our model, the effects of college search and recruitment variables proved to be the most consistent across race groups in terms of direction and significance. All the effects across groups mirrored those found in the general model (shown in Table 4), whereas the magnitude of the effects differed. For example, applying for financial aid had the greatest effect among Asian and Black students, although the effect on Asians was significantly larger in magnitude compared with Black students. The number of schools a student identifies on his or her FAFSA appears to have the greatest negative impact among White students and the smallest effect among Black students. Alternatively, total grant offered (as a percentage of college cost of attendance) had the greatest positive impact among Black students and the least positive effect among White and Latino/Latina students. Thus, the percentage of aid offered seems to be influencing the enrollment decisions of Black students in our sample to a much greater extent than all other race groups.


 

Table 5: Parameter estimates (B) and odds ratios (Exp(B)) predicting student matriculation by race.




  

White

Asian

Black

Latino/a

  

(n = 12,416)

(n = 1,668)

(n = 1,356)

(n = 767)

 

 

B

Exp(B)

B

Exp(B)

B

Exp(B)

B

Exp(B)

[1] Background Characteristics

 

Female

0.106

1.112*

-0.062

0.940

0.002

1.002

-0.020

0.981

 

EFC

0.002

1.002A

0.011

1.011**

0.001

1.001

0.007

1.007

 

AcadIndex

-0.185

0.831***AL

-0.292

0.747***

-0.210

0.810***L

-0.362

0.696***B

High School Characteristics

 

HSSOC

0.004

1.004**B

-0.002

0.998

-0.004

0.996

0.000

1.000

 

HSQuality

-0.278

0.758*

-0.560

0.571

-0.621

0.538*

-0.353

0.702

 

Feeder

0.022

1.022***

0.038

1.038**

0.009

1.009L

0.059

1.061**B

[2] College Search / Recruitment

 

AidApp

2.360

10.588***

2.808

16.584***B

1.522

4.583**A

2.585

13.269***

 

Crossapps

-0.413

0.661***B

-0.322

0.725***

-0.293

0.746***

-0.400

0.671***

 

Tgrant

1.368

3.929***AB

2.229

9.287***

3.057

21.263***L

1.344

3.834*B

[3] College Characteristics

 

CollegeSOC

0.038

1.039***

0.044

1.045***

0.042

1.043***

0.037

1.038*

 

CollegeQuality

2.313

10.105***B

3.510

33.448***B

-0.037

0.964A

1.958

7.083

 

AvgNeedMet

-0.004

0.996*

-0.011

0.989

-0.006

0.994

-0.001

0.999

 

Model R2 [1]

 

0.050***


0.082***


0.078***


0.125***

Model R2 [1, 2]

 

0.119***


0.138***


0.194***


0.185***

Model R2 [1, 2, 3]

 

0.141***


0.171***


0.207***


0.201***

 

Model R2 represents the Nagelkerke statistic based on the likelihood ratio test with a range of 0 - 1 (Nagelkerke, 1991).

* p < .05  ** p < .01  *** p < .001

A Estimated effect is significantly (p < .05) different from Asians.

B Estimated effect is significantly (p < .05) different from Blacks

L Estimated effect is significantly (p < .05) different from Latino/as



Finally, among the three different college characteristics we examined, only the average institutional profile differed significantly across race groups. For instance, both White and Asian students were more likely to enroll at higher profile colleges compared with Black students, who were less likely to attend higher profile and more selective institutions. The structural diversity on a college campus had a very similar positive and significant effect on enrollment across all groups, controlling for both the institutional profile and average amount of need met. Thus, for the sample of students and institutions in this study, structural diversity appears to be an important factor in the college decision-making process for all students.

DISCUSSION


As researchers interested in improving access and opportunity among underserved populations, we are constantly pushing ourselves to understand more clearly the factors that influence the college choice process for all groups of students. To date, the literature on college choice has identified clear patterns among those who choose to attend postsecondary education and the associated disparities that continue to reflect the differentiation and stratification in American society (Haveman & Smeeding, 2006; Labaree, 1990; NCES, 2003, 2004; Rhoades, 1987). Little has been documented, however, on how the decisions to enroll in college differ across race groups (St. John, Paulsen, & Carter, 2005), especially among a group of students who have been admitted to a similar set of institutions. Thus, with the present study, we sought to understand whether different factors related to human capital formation, status attainment, and social and cultural resources might differentially affect the enrollment decisions of White, Black, Asian, and Latino/Latina students.


GENERAL EFFECTS ON COLLEGE CHOICE

Our first research question addressed whether our conceptual model of college choice was relevant to the population under investigation. Our model proved to explain a significant portion of the variance in students’ enrollment decisions, and the results were generally consonant with previous studies investigating college choice. We uncovered significant racial differences among students of color and White students in our general model, which further reinforces the importance of disaggregating effects by racial group memberships. Although our racial effects run counter to some studies (Perna, 2000; Plank & Jordan, 2001), differences in our sample, enrollment model, and dependent variable limit our ability in making any true comparisons. In a similar vein as race, few studies have focused on gender as a variable of consideration in understanding enrollment decisions (Perna, 2006), and our results uncovered no significant effects for gender. Previous studies have noted mixed results, with some studies suggesting that men and women have equal propensities to enroll (Perna, 2000) and others finding women more likely to enroll in specific types of institutions (Perna & Titus, 2005).


We also uncovered important effects based on students’ academic preparation and level of financial need, which highlights the importance of human capital indicators in understanding college decision-making behavior (Ellwood & Kane, 2000; Perna, 2004). Further, both the academic quality of the high school and the historical feeder relationships established between a particular high school and college were significantly related to matriculation. These findings reinforce sociological perspectives that emphasize the structural context of the high school (McDonough, 1997; Perna & Titus, 2005; Wolniak & Engberg, 2007) and the importance of social networks in influencing college choice decisions (Person & Rosenbaum, 2006; Wolniak & Engberg).


The largest effects in our general model were found among college search and recruitment variables, which resonate with the previous literature on the importance of financial aid in predicting enrollment behavior (e.g., Avery & Hoxby, 2004; Catsiapis, 1987). Aid applicant status, in particular, was a strong signal of the likelihood of enrollment and resonates with earlier research demonstrating that nonaid applicants are generally associated with lower propensities to yield than their aid applicant counterparts (Wolniak & Engberg, 2007). Nonaid applicants, for example, typically constitute a smaller percentage of students who truly have no financial need and a larger proportion of students for whom a particular college did not make their top six in terms of choice (based on available positions on the FASFA).


Interestingly, our general model found no effects of structural diversity at the secondary level but strong positive effects at the postsecondary level. This suggests that for many students in our sample, final enrollment decisions are influenced by the diversity of their postsecondary choices irrespective of the level of structural representation in their secondary school environments.




RACE-CONDITIONAL EFFECTS ON COLLEGE CHOICE


Our second research question was formulated to understand more specifically how our general enrollment model explained enrollment behaviors across different race groups—a topic that has only recently been addressed in college choice research. Although all our models explained a significant portion of the variance in enrollment decisions, the model was strongest for Black and Latino/Latina students. This suggests that the prevailing theories grounding our study may be more relevant to understanding behavioral patterns among underserved populations, although we recognize that our conceptual framework only addressed a finite set of variables within each theoretical paradigm.


In relation to background characteristics, we uncovered few effects for Black and Latino/Latina students, with the exception of academic profile. Although gender was insignificant in our general model, we did find a small but significant effect for White students, which resonates with previous research demonstrating stronger enrollment effects for females (Perna & Titus, 2004, 2005). Although our present study supports this finding, it also suggests that gender effects may be more conditional on racial status and less relevant in understanding the enrollment decisions of underserved populations. Additionally, financial need (as measured through levels of expected family contribution) was only relevant for Asian students, who were also associated with the lowest overall need. This suggests that for Asian students in our sample, financial need is an important consideration in understanding their college choice decisions, even when controlling for the amount of financial aid offered.  


Academic preparation was a particularly important human capital predictor in our study across race groups, although the effects were most pronounced among Latino/Latina students. Although several researchers have documented the important effects of academic preparation on enrollment (Ellwood & Kane, 2000; Perna, 2000; Perna & Titus, 2005), this effect has not been examined across separate race group models. Although better academic preparation seems to increase access for all students, our study indicates that for Latino/Latina students, higher academic preparation constitutes a much lower propensity toward enrollment. It may be that these students are highly desirable and heavily sought out in the academic marketplace, effectively reducing their likelihood of attending any one particular institution. Alternatively, Person and Rosenbaum’s (2006) theory of chain migration suggests that Latino/Latina students may be more heavily influenced by the enrollment decisions of other students within their social network regardless of their academic profile and admissibility options at a range of postsecondary institutions.


In addressing our college search and recruitment dimension, one of the largest effects we uncovered across all models was whether a student filed the FAFSA at a particular college. Black students, in particular, were least affected by aid applicant status, although this is partially attributable to the small level of variance among Black students, with 73% of the sample filing for financial aid. Total grant offered, however, was the largest predictor of enrollment for Black students, and the estimated effects were significantly higher than Latino/Latina students. This may be based partially on the higher average awards offered to Black students, coupled with their higher overall need.


In examining the sociological dimensions of college choice related to the structural context of students’ secondary school environments (McDonough, 1997; Perna & Titus, 2005; Wolniak & Engberg, 2007), we found that the overall academic quality of the high school was only slightly significant for White and Black students. Thus, the structural context, as explicated by the average college-going rates and scores on standardized tests, is less predictive of enrollment decisions than other factors in our models, particularly among Asian and Latino/Latina students. Similarly, in examining structural diversity at the secondary level, we only uncovered significant effects for White students, which further suggests that the structural components of our model have limited utility in explaining the enrollment decisions among students of color.


The estimated effects of our high school feeder variable, a sociological construct that highlights the importance of social networks that exist between high schools and colleges (Wolniak & Engberg, 2007), revealed estimated effects that were significantly higher for Latino/Latina students compared with Black students. Person and Rosenbaum’s (2006) theory of chain enrollment may partially explain these differences because their results suggest that Latino/Latina students are guided in their college choice decisions by the enrollment patterns of previous cohorts within their social network. Given that access to feeder networks was quite similar for both Black and Latinos/Latinas, this may be one plausible explanation for why the effects were so different for these two groups.


Overall, the present study has demonstrated that one model does not fit all and that the factors that influence matriculation are quite different across students of differing racial group memberships. In operationalizing constructs related to human, social, and cultural capital, our results indicated that human capital variables were more consistent across race groups, and more important overall in explaining enrollment decisions. The high school and racial composition variables, which reflect the underlying social and cultural context, seem to matter most in explaining the enrollment decisions of White students. More research is needed, however, in understanding the choice process for all groups, especially at the initial predisposition and search stages and through a larger selection of college and universities that cut across the full range of Carnegie classifications.


Acknowledgments


Support for this work was provided by the Human Capital Research Corporation, Evanston, Illinois. A previous version of this article was presented at the 2007 annual meeting of the American Educational Research Association in Chicago.


References


Alon, S., & Tienda, M. (2007). Diversity, opportunity, and the shifting meritocracy in higher education. American Sociological Review, 72, 487–511.


Avery, C., & Hoxby, C. M. (2004). Do and should financial aid packages affect students’ college choices? In C. M. Hoxby (Ed.), College choices: The economics of where to go, when to go, and how to pay for it (pp. 239–302). Chicago: University of Chicago Press.


Baker, T. L., & Velez, W. (1996). Access to and opportunity in postsecondary education in the United States: A review. Sociology of Education, 69(Extra Issue), 82–101.


Becker, G. S. (1993). Human Capital: A theoretical and empirical analysis with special reference to education (3rd ed.). Chicago: University of Chicago Press.


Brinton, M. C., & Nee, V. (1998). The new institutionalism in sociology. Stanford, CA: Stanford University Press.


Catsiapis, G. (1987). A model of educational investment decisions. Review of Economics and Statistics 69, 33–41.


Cohn, E., & Geske, T. G. (1990). The economics of education (3rd ed.). New York: Pergamon Press.


Corcoran, M. (1995). Rags to rags: Poverty and mobility in the United States.” Annual Review of Sociology, 21, 237–267.


Cruce, T. M. (2004). The effects of institutional attributes on the initial choice of college. Unpublished doctoral dissertation, University of Iowa, Iowa City.


DesJardins, S. L, Ahlburg, D. A., & McCall, B. P. (2006). An integrated model of application, admission, enrollment, and financial aid. Journal of Higher Education, 77, 381–429.


Dika, S. L., & Singh, K. (2002). Applications of social capital in educational literature: A critical synthesis. Review of Educational Research, 72, 31–60.


Ehrenberg, R. G., & Sherman, D. R. (1984). Optimal financial aid policies for a selective university. Journal of Human Resources, 19, 202–230.


Ellwood, D. T., & Kane, T. J. (2000). Who is getting a college education? Family background and the growing gaps in enrollment. In S. Danziger & J. Waldfogel (Eds.), Securing the future: Investing in children from birth to college (pp. 283–324). New York: Russell Sage Foundation.


Espenshade, T. J., Hale, L. E., & Chung, C. Y. (2005). The frog pond revisited: High school academic context, class rank, and elite college admissions. Sociology of Education, 78, 269–293.


Flint, T. A. (1992). Parental and planning influences on the formation of student college choice sets. Research in Higher Education, 33, 689–708.


Fuller, W. C., Manski, C. F., & Wise, D. A., (1982). New evidence on the economic determinants of postsecondary school choice. Journal of Human Resources, 17, 477–498.


Goldrick-Rab, S. (2006). Following their every move: An investigation of social-class differences in college pathways. Sociology of Education, 79, 61–79.


Haveman, R., & Smeeding, T. (2006). The role of higher education in social mobility. The Future of Children, 16, 125–150.


Hearn, J. C. (1992). Emerging variations in postsecondary attendance patterns: An investigation of part-time, delayed, and nondegree enrollment. Research in Higher Education, 33, 657–687.


Heller, D. E. (1997). Student price response in higher education: An update to Leslie and Brinkman. The Journal of Higher Education, 68, 624–659.


Hossler, D., Braxton, J., & Coopersmith, G. (1989). Understanding student college choice. In J. C. Smart (Ed.), Higher education: Handbook of theory and research (Vol. 5, pp. 231–288). New York: Agathon Press.


Hossler, D., & Gallagher, K. S. (1987). Studying college choice: A three-phase model and the implications for policy-makers. College and University, 2, 207–221.


Hurtado, S., Inkelas, K. K., Briggs, C., & Rhee, B. S. (1997). Differences in college access and choice among racial/ethnic groups: Identifying continuing barriers. Research in Higher Education, 38, 43–75.


Kane, T. J. (1999). The price of admission: Rethinking how Americans pay for college. Washington, DC: Brookings Institution Press.


Labaree, D. F. (1990). From comprehensive high schools to community colleges: Politics, markets, and the evolution of opportunity. Research in Sociology of Education and Socialization, 9, 203–140.


Leslie, L. L., & Brinkman, P. T. (1987). Student price response in higher education: The student demand studies. Journal of Higher Education, 58, 181–204.


Lin, N. (2001). Social capital: A theory of social structure and action. New York: Cambridge University Press.


Lucas, S. R. (2001). Effectively maintained inequality: Education transitions, track mobility, and social background effects. American Journal of Sociology, 106, 1642–1690.


Massey, D. S., Charles, C. Z., Lundy, G. F., & Fischer, M. J. (2003). The source of the river: The social origins of freshman at America’s selective colleges and universities. Princeton, NJ: Princeton University Press.


McDonough, P. M. (1997). Choosing colleges: How social class and schools structure opportunity. Albany: State University of New York Press.


Morrow, V. (1999). Conceptualising social capital in relation to the well-being of children and young people: A critical review. Sociological Review, 47, 744–765.


Nagelkerke, N. J. D. (1991). A note on a general definition of the coefficient of determination. Biometrika, 78, 691–692.


National Association for College Admission Counseling. (2006). State of college admission. Retrieved September 14, 2007, from http://www.nacacnet.org/NR/rdonlyres/4B4C0DF4-BF0A-4B10-89F4-A3D631061305/0/06StateofCollegeAdmission.pdf


National Center for Education Statistics. (2003). Enrollment in postsecondary institutions, fall 2001 and financial statistics, fiscal year 2001. Washington, DC: Author. (NCES 2004-155)


National Center for Education Statistics. (2004). Digest of education statistics 2003. Washington, DC: Author. (NCES 2005-025)


Paulsen, M. B. (1990). College choice: Understanding student enrollment behavior. Washington DC: The George Washington University. (ASHE-ERIC Higher Education Report No. 90-6)


Paulsen, M. B. (2001). The economics of human capital and investment in higher education. In M. B.

Paulsen & J. C. Smart (Eds.), The finance of higher education: Theory, research, policy, and practice (pp. 55–94). New York: Agathon Press.


Perna, L. W. (2000). Differences in the decision to enroll in college among African Americans, Hispanics, and Whites. Journal of Higher Education 71, 117–141.


Perna, L. W. (2004). The key to college access: A college preparatory curriculum. In W. G. Tierney, Z. B. Corwin, & J. E. Colyar (Eds.), Preparing for college: Nine elements of effective outreach (pp. 113–134). Albany: State University of New York Press.


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


Perna, L. W., & Titus, M. (2004). Understanding differences in the choice of college attended: The role of state public policies. Review of Higher Education, 27, 501–525.


Perna, L. W., & Titus, M. A. (2005). The relationship between parental involvement as social capital and college enrollment: An examination of racial/ethnic group differences. Journal of Higher Education, 76, 486–518.


Person, A. E., & Rosenbaum, J. E. (2006). Chain enrollment and college enclaves: Benefits and drawbacks of Latino college students’ enrollment decisions. In C. L. Horn, S. Flores, & G. Orfield (Eds.), New directions for community colleges (pp. 51–60). San Francisco: Jossey-Bass.


Plank, S. B., & Jordan, W. J. (2001). Effects of information, guidance, and actions on postsecondary destinations: A study of talent loss. American Educational Research Journal 38, 947–979.

Rhoades, G. (1987). Higher education in a consumer society. Journal of Higher Education, 58, 1–24.


St. John, E., Asker, E., & Hu, S. (2001). The role of finances in student choice: A review of theory and research. In M. Paulsen & J. Smart (Eds.), The finance of higher education: Theory, research, policy, and practice (pp. 419–438). New York: Agathon Press.


St. John, E. P., Paulsen, M. B., & Carter, D. F. (2005). Diversity, college costs, and postsecondary opportunity: An examination of the financial nexus between college choice and persistence. The Journal of Higher Education 76, 545–569.


Terenzini, P. T., Cabrera, A. F., & Bernal, E. M. (2001). Swimming against the tide: The poor in American Higher Education (Report No. 2001-1). New York: College Entrance Examination Board.


Weiler, W. C. (1994). Transition from consideration of a college to the decision to apply. Research in Higher Education, 35, 631–646.


Wolniak, G. C., & Engberg, M. E. (2007). The effects of high school feeder networks on college enrollment. Review of Higher Education, 31, 27–53.






Cite This Article as: Teachers College Record Volume 111 Number 9, 2009, p. 2255-2279
https://www.tcrecord.org ID Number: 15395, Date Accessed: 1/25/2022 5:38:49 PM

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

Related Discussion
 
Post a Comment | Read All

About the Author
  • Mark Engberg
    Loyola University Chicago
    E-mail Author
    MARK E. ENGBERG is an assistant professor of higher education at Loyola University Chicago. Dr. Engberg's current research examines the secondary and postsecondary school nexus, with particular attention to how the college choice process unfolds for underserved populations. His research also explores the role of educational interventions in reducing intergroup bias and improving intergroup relations on college campuses. He is actively involved in a number of educational associations and has recently published in the Review of Educational Research, Journal of Higher Education, Review of Higher Education, Research in Higher Education , and the Journal of College Student Development.
  • Gregory Wolniak
    NORC at the University of Chicago
    E-mail Author
    GREGORY C. WOLNIAK is a research scientist with the National Opinion Research Center (NORC) at the University of Chicago. Dr. Wolniak's research focuses on pathways to college and the relationship between postsecondary education and socioeconomic outcomes. Current research projects include a study of high school contexts and institutional networks in relation to college enrollment, as well as an analysis of early socioeconomic outcomes among participants of the Gates Millennium Scholars program. Recent publications have appeared in the Review of Higher Education, Research in Higher Education, Journal of Higher Education, and Journal of Vocational Behavior.
 
Member Center
In Print
This Month's Issue

Submit
EMAIL

Twitter

RSS