The Education Pendulum

Disciplinary Foundations of the Conceptual Model

DRAWING ON THE RESEARCH REVIEWED for this monograph,this chapter describes the disciplinary foundations for each of the four layers of the conceptual model. Given the parameters of the review (such as attention only to articles published in top journals in four disciplines), the substantive consideration of each layer is designed to be illustrative rather than comprehensive. In other words, the purpose of this discussion is not to provide an exhaustive assessment of the forces at all layers of context that contribute to all indicators of student success. Instead, this presentation illustrates the per- spectives and emphases that each of the four disciplines contributes to each layer of the model. For each of the four layers of the model, we discuss research from the top journals in the following order: economics, sociology, psychology, and education.

Layer One: Internal Context At its core, student success is determined by the attitudes, motivations, and behaviors of individual students. Our review suggests that, of the four discipli- nary perspectives, psychology focuses greater attention than the other disciplines on understanding the cognitive and noncognitive processes that determine stu- dent success. Even the relatively small number of relevant articles in other dis- ciplines and fields are centrally informed by psychological theories and frames. In short, psychology differs from the other disciplines included in this review in its decided focus on the individual’s mental processes and behaviors—processes and behaviors defining Layer One of the model.

35Theoretical Perspectives on Student Success

Economics Aside from work only loosely coupled to cognitive dimensions of student success, the economics journals we reviewed yielded little insight into the contribution of Layer One to student success. An example of work loosely bearing on interests at this layer of the model is that of Jacob (2002), who models the influence of noncognitive skills on the gender gap in college participation. Jacob concludes that noncognitive skills influence college enrollment patterns even after controlling for high school performance and aptitude.

Sociology Relatively little work published in top sociology journals examines cognitive or affective processes defining the internal context of success indicators. One notable exception, related to the work in psychology by Perry, Hladkyj, Pekrun, and Pelletier (2001), identifies the components of an academic work ethic among college students and shows how an academic work ethic is related to student performance and to characteristics of institutions attended (Rau and Durand, 2000). Rau and Durand conclude that a strong relationship exists between disciplined study, as defined by their academic ethic measure, and academic performance.

Psychology Attention to the contribution of cognitive and noncognitive processes to stu- dent success is relatively common in psychology articles. More specifically, most of the articles published in top psychology journals inform understanding of the ways that such constructs as achievement motivation, self-efficacy, and stereotype threat contribute to student success.

A few articles published in top psychology journals focus on aspects of self- regulated learning, particularly perceived academic control and other strategies that regulate motivation. Research suggests that perceived academic control is positively related to final course grades and that students with high academic control and high preoccupation with failure receive the highest grades (Perry, Hladkyj, Pekrun, and Pelletier, 2001). Students with high academic control not only receive higher grades but also exert more effort, experience less anxiety, have greater motivation, tend to monitor progress in achieving goals, and

36

perceive greater control over course assignments (Perry, Hladkyj, Pekrun, and Pelletier, 2001). Other research shows that students’ strategies for regulating their motivation are related to their goal orientation (Wolters, 1998). The use of intrinsic regulation strategies is more common among those with mastery goal orientations, while use of extrinsic regulation strategies is more common among those with performance goal orientations (Wolters, 1998). High school students with autonomy orientations (that is, those who tend to participate in academic activities that they believe to be important to themselves) have more positive academic experiences, while students with control orientations (those who tend to participate in academic activities that they believe to be important to others) have lower academic performance and commitment (Wong, 2000).

A substantial number of articles published in top psychology journals examine the contribution of students’ goals to their academic performance. This research consistently supports a “multiple goals” perspective, whereby mastery goals promote interest (Harackiewicz and others, 2000; Harackiewicz, Barron, Tauer, and Elliot, 2002), performance goals promote grades (Elliot and Church, 1997; Harackiewicz, Barron, Carter, and Lehto, 1997; Harackiewicz and others, 2000; Harackiewicz, Barron, Tauer, and Elliot, 2002), and per- formance avoidance goals reduce academic performance (Elliot and Church, 1997). The positive relationship between performance goals and academic achievement appears to be mediated by such variables as persistence on task (Elliot, McGregor, and Gable, 1999); effort, self-efficacy, goal level (VandeWalle, Cron, and Slocum, 2001); and achievement motivation (Barron and Harackiewicz, 2001), while the negative relationship between performance avoidance goals and academic achievement appears to be mediated by test anxiety (Elliot and McGregor, 1999) and disorganization (Elliot, McGregor, and Gable, 1999).

Research published in top psychology journals consistently shows that academic self-efficacy, optimism, and hope are positively related to students’ academic performance (Brackney and Karabenick, 1995; Chemers, Hu, and Garcia, 2001; Gibbons and others, 2000; Snyder and others, 2002). The effects of such “trait-like” characteristics as general self-efficacy, goal orientation, and cognitive ability on academic achievement may be mediated by such “state- like” characteristics as task-specific self-efficacy (Chen, Gully, Whiteman, and

37Theoretical Perspectives on Student Success

Kilcullen, 2000). Psychopathology (psychological disorders, including anxiety and substance abuse disorders) is negatively related to students’ academic per- formance directly (Svanum and Zody, 2001) and indirectly through self-efficacy and resource management (Brackney and Karabenick, 1995).

With only a few exceptions (for example, Cullen, Hardison, and Sackett, 2004), research in top psychology journals consistently shows that stereotype threat contributes to gaps in academic performance between blacks and whites (Brown and others, 2000; Gonzales, Blanton, and Williams, 2002; Steele and Aronson, 1995), women and men (Brown and others, 2000; Brown and Josephs, 1999; Gonzales, Blanton, and Williams, 2002; O’Brien and Crandall, 2003), and students with and without mental illness (Quinn, Kahng, and Crocker, 2004). A self-evaluative stereotype threat is assumed to negatively influence performance when an individual’s focus is diverted from performing a particular task to worrying that low performance will confirm a negative stereotype about a group to which the individual belongs (Steele and Aronson, 1995). The negative effects of stereotype threat on performance may be reduced by other psychological characteristics, particularly a coping sense of humor (Ford, Ferguson, Brooks, and Hagadone, 2004).

Other research in top psychology journals shows the negative effects of par- ticular experiences for African Americans. A longitudinal study of African Americans at one predominantly white institution showed that grades declined over the period of the study for students who had high levels of race-related rejection sensitivity (those who “anxiously expect, readily perceive, and intensely react to rejection in situations where rejection is possible” (Mendoza-Denton and others, 2002, p. 896). Other research suggests that, compared with other African American high school students, those who had positive feelings about their racial group and viewed race as important to their identity had higher rates of college enrollment, while those who felt few connections to their racial group and had negative beliefs about their racial group had lower rates of college enrollment (Chavous and others, 2003).

Articles in psychology journals also include attention to differences in academic performance between women and men, describing gender differences in rates of Advanced Placement test taking and performance (Stumpf and Stanley, 1996) and explaining that gender differences in SAT math scores are attributable in part

38

to gender differences in mental rotation ability and math self-confidence (Casey, Nuttall, Pezaris, and Benbow, 1995; Casey, Nuttall, and Pezaris, 1997). Other research shows gender differences in the relationship between psychopathol- ogy and semester grades (Svanum and Zody, 2001), between control orientation and academic experiences (Wong, 2000), and between text anxiety and grade point average (Chapell and others, 2005).

A small number of articles published in top psychology journals suggest links between the internal layer of context and the school context (that is, Layer Two of the model) via students’ perceptions of this context. For example, one study suggests that increased perceptions of “situational constraints” (quantity and quality of resources available to support learning) indirectly reduce academic performance by reducing students’ performance goals (Villanova, 1996). A second study shows that undergraduates’ academic achievement is positively related to students’ beliefs about school, particularly their predisposition toward the learning context (Larose and Roy, 1995).

Education Only a small number of articles in top education journals examine the ways that students’ cognitive and noncognitive skills shape their success. Moreover, the education articles that include this focus tend to draw on psychological constructs. For example, one article shows that students’ academic perfor- mance in college is shaped by cognitive skills (as measured by test scores) as well as by noncognitive variables, including motivation and use of self- regulated learning strategies (Ruban and McCoach, 2005). Other work stresses the contribution of self-efficacy to students’ academic achievement. In a review and synthesis of prior research, Pajares (1996) concludes that self-efficacy beliefs shape student effort and perseverance, which in turn influence subse- quent academic performance. In other educational research, performance is viewed as a function of self-worth beliefs related to mathematics and gender (Stage and Kloosterman, 1995).

Summary The proposed conceptual model assumes that, at the “core” (Layer One of the model), student success is determined by an individual’s motivations and

39Theoretical Perspectives on Student Success

attitudes. Our examination of research published in the top journals in each of four disciplines shows the dominance of psychological perspectives for understanding these core attributes.

Layer Two: Family Context The second layer of the model, the family context, recognizes that both inside and outside the home, families may “manage” their children’s experiences to promote various indicators of student success (Furstenberg and others, 1999; McDonough, 1997; Perna and Titus, 2005). Although the aspect of families that is emphasized varies across disciplines, research using each of the four disciplinary perspectives includes attention to the influence of families on student success.

Economics Our research review identified several examples of the ways that economists view the contribution of families to student success. Some research in top econom- ics journals examines the role of parents’ occupation. One study shows that, compared with peers with traditionally employed parents, young people from families with family-owned businesses generally have lower academic perfor- mance during high school and are less likely to enroll in college (Davila and Mora, 2004). Ease of intergenerational transfer of these family-owned businesses is presumed to discourage academic engagement in high school and diminish college-going aspirations of children in these entrepreneurial families (Davila and Mora, 2004). In other work addressing the occupational background of families, Siegfried and Getz (2006) develop a novel analysis of college choice patterns of students from families with at least one parent who works on a uni- versity faculty. Siegfried and Getz were particularly interested in the degree to which these students may be advantaged by additional information about college quality that would be transmitted by their more knowledgeable parent(s). Although failing to provide a causal explanation, Siegfried and Getz note that students in their sample are more likely to attend research universities and selective liberal arts colleges than are their peers from nonacademic families.

Economists have also examined the contribution of family structure to student success. For example, Ver Ploeg (2002) isolated the effects of displaced

40

children on the likelihood of college enrollment and degree attainment. Although previous researchers have explained this disadvantage as a function of the typically diminished income of broken homes, Ver Ploeg controls for income and reveals a net negative effect of such circumstance.

Beyond the structural characteristics of families, economists have also devoted attention to dimensions such as the economic behavior of families with children in college. One example of this perspective is Bodvarsson and Walker (2004), who find that students whose parents pay for a substantial proportion of the costs associated with tuition and living expenses have lower grade point averages, are more likely to fail courses, and are less likely to persist to the baccalaureate than students who bear the lion’s share of these costs themselves through work or personal savings.

Sociology Sociologists have made a number of important contributions to our under- standing of the influence of family characteristics on student success. Cheng and Starks (2002) employ a symbolic interaction frame to examine the differ- ential role of significant others on the educational expectations of students from different racial groups. Symbolic interaction focuses on the ways that personal identity is developed through the interaction with others. The Cheng and Starks work points to processes through which the influences of significant others are conditioned by race. Central to their findings is the idea that the power of specific significant others (such as parents, teachers, or friends) to influence expectations about education varies across racial groups.

Some research shows the role that families play in determining the future paths of their children and ultimately the degree to which those future paths may reduce or magnify stratification in broader society. Conley’s account (2001) of the role of family wealth in college attendance and completion shows that tra- ditional models of attainment have ignored the role of family wealth, focusing instead on less useful measures of family income. This type of analysis taps a long-standing sociological interest in the long-term advantages conveyed through the intergenerational transfer of wealth. Other sociological research focuses on families and high schools and the ways that family background can determine students’ preparation for college and range of choices available. For example,

41Theoretical Perspectives on Student Success

Attewell (2001) argues that families seek to maximize the quality of schooling available to their children, often with the hope of improving their chances for college success. He shows that this pursuit on the part of the family may actually be counterproductive in terms of chances for admission to high-quality colleges. Other research shows that family background has an important influence on high school performance and college enrollment (see, for example, Muller and Schiller, 2000; Conley, 2001; Crosnoe, 2001; Cheng and Starks, 2002; Karen, 2002; Hofferth, Boisjoly, and Duncan, 1998). This influence is channeled through increased parental involvement (Crosnoe, 2001), noneconomic (cultural) by-products of family wealth (Conley, 2001), the influence of signif- icant others (Cheng and Starks, 2002), and the social networks and cultural connections of parents (Hofferth, Boisjoly, and Duncan, 1998).

Most of the sociological literature is dominated by either an exclusive cultural reproduction framing or some type of contrast between the repro- duction models and mobility models. Although reproduction models place primacy on the binding role of social origins, mobility models focus on the degree to which social status can change over the course of a lifetime. Educa- tion is a central feature in both models, serving as a reproductive mechanism in the former and a democratizing mechanism in the latter. A notable example of such a contrast is Aschaffenburg and Maas’s examination (1997) of the role of cultural capital in school success. In that work they test competing mobility (DiMaggio, 1982) and reproduction theories (Bourdieu and Passeron, 1977) and conclude that the mobility model is dominant but that the repro- duction model is the more important in terms of college enrollment.

Psychology A number of articles in top psychology journals examine the influence of the family context on student success. Among the relevant aspects of the family context, from the perspective of this discipline, are strength of ties to parents, parenting style, and parents’ job security. The sole qualitative psychological article in this review suggests that students’ plans are shaped, at least in part, by the tension between increasing autonomy and sustaining ties to parents and other loved ones (Shilkret and Nigrosh, 1997). Other research shows that students’ academic achievement is influenced by parenting style, although the

42

relationship appears to be weaker for college seniors than for other students (Glasgow and others, 1997; Strage and Brandt, 1999). An exploratory study suggests that, when students perceive job insecurity among their parents, the students experience cognitive problems, which reduce students’ academic per- formance (Barling, Zacharatos, and Hepburn, 1999).

Other research published in top psychology journals shows that the rela- tionship between aspects of the family context and student success varies based on parents’ educational attainment, ethnicity, and immigrant status. Research shows variations based on parents’ educational attainment and ethnicity in the effects of parental involvement on eleventh grade students’ educational and occupational aspirations (Hill and others, 2004) and students’ academic achievement (Hong and Ho, 2005). Other research shows that, compared with students from United States-born families, students from immigrant families have higher academic motivation (which promotes academic achievement) but greater family demands (which reduce academic achievement [Tseng, 2004]).

Education A substantial share of articles in education show that student success is related to students’ sociodemographic characteristics, particularly socioeconomic status, race, and ethnicity. Educational research consistently shows that, even after tak- ing into account other variables, socioeconomic status is positively associated with such measures of student success as choice of institution attended (Astin and Oseguera, 2004; Perna and Titus, 2004; Teranishi and others, 2004) and graduate school enrollment (Walpole, 2003; Zhang, 2005a). Educational research also shows that the predictors of such indicators as predisposition to college (Hamrick and Stage, 1998), college enrollment (Heller, 1999), college grade point average (Hoffman and Lowitzki, 2005), and plans for graduate school (Pascarella, Wolniak, Flowers, and Pierson, 2004) vary by racial/ethnic group. Although fewer studies examine variations in broad racial or ethnic cat- egories, the small amount of available research shows that such indicators as choice of institution attended vary by ethnicity in a particular group (for exam- ple, Asian Pacific American [Teranishi and others, 2004]).

Research in education also includes attention to the role of family or parental involvement in promoting student success. In particular, articles in

43Theoretical Perspectives on Student Success

top educational journals have examined the contribution of parental involvement to such indicators as college enrollment (Perna, 2000; Perna and Titus, 2005) and the role of family background in shaping impressions and realities of attendance and choice constraints (Paulsen and St. John, 2002).

Summary All four disciplinary perspectives inform our understanding of the ways that the family context (Layer Two of the model) influences student success. Econ- omists have examined the influence on student success of such characteristics as parents’ occupation and family structure as well as parents’ role in paying college prices. For sociologists, families are central to “status attainment” frame- works and perspectives examining the sources of continued stratification of educational opportunity and outcomes. Psychologists include attention to the strength of ties to parents, parenting style, and parents’ job security as well as variations in outcomes based on family characteristics. Educational researchers often examine the contribution to student success of such characteristics as socioeconomic status and parents’ involvement in their children’s education.

Layer Three: School Context Layer Three of the model, the school context, reflects the attention that various disciplines, particularly economics, sociology, and education, devote to under- standing the contribution of K-12 schools and higher education institutions to student success. Attention to the school context, from primary school through college, enables the identification and understanding of compounding effects that determine the educational resources, academic preparation, and educational orientations that subsequently determine success at the college level.

Economics School effects are an important domain of study in economics. Much of the eco- nomics literature in this layer addresses the role of years, type, and quality of education on subsequent indicators of student success. A review of literature published in top economics journals suggests three themes pertaining to the rela- tionship between institutional characteristics and practices and student success

44

indicators: (1) interplay between two-year and four-year institutions; (2) eco- nomic returns to institutional characteristics; and (3) the effects of institutional packaging of financial aid on retention and degree attainment.

Economists have devoted significant attention to understanding differences between and relationships among influences of two- and four-year institutions. The subbaccalaureate labor market is the focus of Grubb (2002a, 2002b), who, through reviews and his own empirical analysis, concludes that, although there is little effect of course taking by itself, there exists a significant return to completion of subbaccalaureate credentials in certain areas. Alfonso, Bailey, and Scott (2005), however, show that “occupationally” oriented students are less likely to complete their degree programs and call attention to mission ambi- guity in today’s community colleges. Other researchers focus on the difference in returns among two- and four-year graduates (see, for example, Kane and Rouse, 1995), concluding that, relative to high school graduates, an earnings premium exists at each level of attainment and to a lesser degree for those leaving college without a degree (Grubb, 2002a; U.S. Census Bureau, 2004).

The role of the two-year school in determining aspirations, transfer behavior, and completion has been the focus of a number of researchers in the economics literature. Ehrenberg and Smith (2004) develop a useful evaluation scheme for states to use in determining the degree to which four-year institutions graduate two-year transfer students. Their framework calls attention to the role the two- year schools play in terms of academic preparation and the responsiveness of four-year institutions to the needs of these transfers. Leigh and Gill (2003, 2004) show how two-year schools enhance educational aspirations of their graduates and improve their probabilities of baccalaureate attainment. Taken together, Sandy, Gonzalez, and Hilmer (2006) and Gonzalez and Hilmer (2006) show that two-year colleges democratize opportunity and improve the likelihood of bac- calaureate attainment for Hispanic students in particular. They explain the lower rates of baccalaureate attainment by two-year transfers as a function of their propensity to transfer to lower-quality four-year institutions rather than inade- quate preparation at the two-year level (Gonzales and Hilmer, 2006). Surette (2001) focuses on gender differences in two- to four-year transfer, failing to arrive at a plausible explanation for the persistently lower rate of transfer and comple- tion among women than men.

45Theoretical Perspectives on Student Success

46

While Kane and Rouse (1995) focus on two- and four-year rates of return, Arias and McMahon (2001) focus on an improved model for estimating private rates of return (which they claim have been seriously underestimated in most of the economics literature). Looking beyond direct wage benefits, Eide (Eide and Waehrer, 1998; Eide, Brewer, and Ehrenberg, 1998) offers considerations of the option value that more selective colleges confer on their graduates and argues that a sole focus on wage premiums obscures the larger picture of bene- fits accruing to graduates from more prestigious schools.

Pricing and the influence of financial constraints in attendance decisions and college choices have also received a great deal of attention in the economics literature. For example, financial constraint has been shown to play a lesser role in college choice at the application stage than students’ sense of institutional fit (Toutkoushian, 2001). Analyzing college application behavior, Toutkoushian concludes that a student’s sense of ability relative to that of the student body at colleges in a potential choice pool plays a larger role in deciding where to apply than does their sense of affordability.

Although substantial work has been devoted to understanding the effects of financial aid and subsidies on student attendance patterns more broadly, the literature at Layer Three reveals economists’ interest in institutional behav- iors relating to aid and subsidies that can influence student attendance. Good examples of research along this line include Singell (2004) and Kerkvliet and Nowell (2005). Both studies examine the role of aid in persistence and arrive at different but not conflicting conclusions: aid matters but it depends on the context of the institution and degree to which students are influenced by their perceptions of opportunity costs.

Sociology Sociologists also devote substantial attention to the role of the school context, with particular focus on structural antecedents to postsecondary student success indicators. Much of the research that is relevant to this layer focuses on theoret- ical tensions between cultural reproduction and social mobility. Karen (2002) paints a powerful picture of stratified opportunity, showing that although disadvantaged students begin with a lower chance of college continuation, those that do go on most often enroll in less selective institutions that provide fewer

academic support resources—institutions that have also been shown to confer lesser returns in the postgraduation labor market (Thomas 2000; Thomas and Zhang, 2005).

The process through which students are advantaged as a result of their K-12 experiences is an important interest of sociologists. Attewell (2001) and Espenshade, Hale, and Chung (2005) have scrutinized the effects of the ultra- competitive high school environments that many parents seek for their chil- dren. Attewell’s analysis (also cited in our consideration at Layer Two of the model) suggests that these schools penalize students because of the probability of an otherwise top-performing student’s being ranked lower in the class simply because of the intense academic competition. Espenshade, Hale, and Chung (2005) revised this “frog-pond” effect and reaffirmed Attewell’s findings with the important qualification that the overall performance of students in the schools generally outweighs any disadvantage resulting from lower ranking as a result of the intensity of competition at the school.

Alon and Tienda (2005) and Hurtado and Carter (1997) address issues of institutional “fit” among minority and Latino students, respectively. Alon and Tienda (2005) develop a more nuanced model to estimate the success rates of minority students attending highly selective institutions, concluding that, when modeled properly, the rates of minority and nonminority student suc- cess at these institutions is the same. In contrast, Hurtado and Carter (1997) find that Latino students’ sense of belonging on four-year campuses after trans- fer from a two-year school is not as strong as for their peers who enrolled as native freshmen on the four-year campus.

Although the majority of work in sociology addresses issues related to the process by which schools structure opportunities for success at the postsec- ondary level, a few studies examine program effectiveness in colleges. Two studies of interest have already been cited as examples of work addressing issues in Layer One of our model. The first of them is Rau and Durand’s examination (2000) of the academic work ethic and its relationship to academic perfor- mance. Although the idea of an ethic clearly relates to Layer One, the study also provides insight about how colleges could best foster this ethic on their campuses. Collier (2000) was concerned with the impact of capstone course experiences on the development of student identity, which is in turn linked

47Theoretical Perspectives on Student Success

to academic performance. Similar to Rau and Durand, Collier offers a con- sideration of what colleges might be able to do to make these programs effective on their campuses. Additional sociological work on program effects includes Deil-Amen and Rosenbaum’s examination (2002) of “stigma free” remedia- tion programs at two-year colleges. They use a qualitative approach and con- clude that the “stigma free” programs can work, sometimes too well. They note that some students failed to develop a realistic sense of their abilities and there- fore held unrealistic expectations about their future academic opportunities. Regina and James (2004) examine differences between two-year for-profit and not-for-profit schools’ programs for maximizing employment possibilities for their graduates. They conclude that community colleges could benefit from institutionalizing their employment processes in ways that are similar to their for-profit counterparts.

Psychology Only a small number of articles in top psychology journals examine the influ- ence of school context on student success, suggesting that psychology theories may be less relevant than other perspectives to understanding this layer. The available psychology articles generally illustrate the ways that particular inter- ventions or experiences contribute to various aspects of student success. For example, one study shows that the negative effects of stereotype threat on female students’ math performance are reduced when students learn of a “disconfirm- ing example,” for example, a female role model in math performance (Marx and Roman, 2002). Another study shows that, by building interpersonal skills and educational aspirations, participating in extracurricular activities contributes to higher levels of educational attainment at age 20 (Mahoney, Cairns, and Farmer, 2003). A third study shows that students who request and receive personal psy- chological counseling have higher retention rates than students who request but do not receive counseling (Wilson, Mason, and Ewing, 1997). Another study suggests that, among students with high levels of emotional and psychological stress, an intervention that requires students to write about the stress prevents a decline in grade point average (Lumley and Provenzano, 2003). Providing stu- dents with a decision-making aid that helps them to identify the reasons for their decision improves students’ satisfaction with their choice of college (Kmett, Arkes, and Jones, 1999). Other research suggests that students’ SAT verbal scores

48

(but not course grades) depend in part on their relative skill with multiple-choice rather than essay examinations (Bridgeman and Morgan, 1996).

Education Educational research also recognizes the role of the institutional context in deter- mining student success. Some studies demonstrate that particular indicators of student success are associated with the characteristics of the high school attended, while other indicators are related to the characteristics of the higher education institution. For example, studies show that high school context is directly related to college access and choice. Factors such as high school quality and ethnic mix shape students’ opportunity for college; these high school influences are known to vary by racial background (Perna, 2000). Other work shows that choosing to attend a historically black college or university rather than a predominantly white institution is related, at least in part, to the characteristics of the high school attended and students’ experiences at that school (Freeman, 1999). High schools with well-developed guidance and advising programs have also been shown to have important influences on subsequent college enrollment behaviors and insti- tutional choices (Plank and Jordan, 2001; Tierney and Jun, 2001).

In terms of the higher education context, research in top educational jour- nals shows that such institutional characteristics as single sex, race, and qual- ity contribute to student success. Based on his review of research, Mael (1998) concludes that academic achievement is higher for students who attend a single-sex institution (junior high school, high school, or college) rather than a coeducational institution. Other research suggests that success varies in part based on whether a student attends a historically black college or university or a predominantly white institution, with black students having more posi- tive experiences at historically black colleges or universities than at predomi- nantly white institutions (Fleming, 2002; Fries-Britt and Turner, 2002). The quality of the higher education institution attended is positively related to such indicators of student success as graduate school enrollment and degree completion as well as earnings (Zhang, 2005a, 2005b), and institutional expenditure patterns are related to students’ self-reported gains in various aspects of college performance (Toutkoushian and Smart, 2001).

Other educational research focuses on the contribution to student success indicators of particular programs or experiences at an institution. Using an

49Theoretical Perspectives on Student Success

experimental design, Nagda and others (1998) found that participation in an undergraduate research opportunity increased rates of persistence through graduation, especially for African Americans with academic achievement below the median. Other research suggests that participating in community service is positively related to such indicators as graduate school enrollment and degree attainment (Astin, Sax, and Avalos, 1999). Reviews of research suggest that academic achievement improves when a student participates in some form of peer assessment (Topping, 1998) or participates in an intervention designed to enhance study skills (Hattie, Biggs, and Purdie, 1996). For science, math- ematics, engineering, and technology courses and programs, academic achieve- ment and persistence increase when the student engages in some form of small-group learning (Springer, Stanne, and Donovan, 1999).

Summary The conceptual model also assumes that context of the “school” attended, including both K-12 and higher education institutions, influences student suc- cess. Attention to the school context is relatively more common in economics, sociology, and education than in psychology. Economic frameworks may be particularly useful for informing our understanding of differences in outcomes for students attending two-year and four-year colleges and universities, the eco- nomic returns to different characteristics of colleges and universities, and the effects of a college or university’s financial aid programs on student outcomes. Sociological perspectives shed light on the ways that K-12 schools and higher education institutions structure educational opportunities for students. Drawing on both economic and sociological perspectives, education research also stresses the ways that school and college characteristics influence various indicators of student success. Psychological perspectives may be especially useful for under- standing how the effects of a particular policy or program on a given outcome may be mediated by cognitive and affective processes.

Layer Four: Social, Economic, and Policy Context The fourth layer of the model, the social, economic, and policy context, rec- ognizes that numerous external forces also influence student college choice,

50

both directly and indirectly through other layers of context (Perna, 2006a). Although less commonly examined than the other layers of the model, at least some research in all four disciplines suggests the importance of considering how the social, economic, and policy context influences student success. Among the potentially influential forces are social conditions (societal norms), economic conditions (unemployment rate), and public policies (establishment of a new state-sponsored non-need-based grant program).

Economics Economists have contributed significantly to our understanding of issues connected to this layer of the model. Representative work addresses the role that state aid policies play in college choice (Niu, Tienda, and Cortes, 2006) and the ways that education-related debt influences employment decisions (Minicozzi, 2005) and educational attainment (Monks, 2001). Supply and demand issues related to regulation and state appropriations for higher education present challenges for state struggles to maintain enrollment levels (Berger and Kostal, 2002). As demand inevitably declines when tuition increases, Berger and Kostal (2002) argue that states must choose between reducing supply through further reductions in appropriations or increasing regulation of colleges and universities. Supply and demand are dominant themes in work related to this layer. Additional research in this layer of our proposed conceptual model considers effects of public financing of K-12 schools (Deke, 2003) and the role of financial aid (Dynarski, 2002; Keane, 2002; Ichimura and Taber, 2002).

Sociology Sociological inquiry draws attention to the ways that larger societal inequities may be related to higher education. Some of this work focuses on occupational gen- der segregation and its relationship to persistent gender bias in students’ choice of major (Bradley, 2000). Other work in sociology attempts to account for observed increases in women’s participation in science and engineering major fields. For example, Ramirez and Wotipka (2001) suggest that increased participation in science and engineering fields is simply a function of increased participation overall by women. More refined analyses of gender bias in occupations and earnings conclude that women’s major choices limit their occupational choices and that

51Theoretical Perspectives on Student Success

women are more likely to find themselves in lower-paying government and non- profit jobs (Roksa, 2005). Roksa’s work also shows that, despite lower pay in these jobs, women tend to enjoy more rapid elevation to management positions.

Psychology Very few articles published in top psychology journals inform understanding of the contribution of the social, economic, and policy context to student success. One exception shows the role of the media in shaping student success. This study suggests that, by activating stereotype threat, stereotypic commercials contribute to lower math performance among women than men (Davies, Spencer, Quinn, and Gerhardstein, 2002).

Education A small number of articles in educational journals illustrate the role of the pol- icy context in shaping student success. Some research shows that aspects of such state public policies as tuition, financial aid, appropriations, and K-12 academic preparation are related to college access and choice (Heller, 1997, 1999; Perna and Titus, 2004), while other research shows that racial or eth- nic stratification in college enrollment increased in one state despite the pres- ence of various higher education policies (Perna, Steele, Woda, and Hibbert, 2005). K-12 educational reform, challenges to affirmative action, and changed student demographics appear to shape college admissions processes (Sireci, Zanetti, and Berger, 2003). Although desegregation initiatives are typically associated with increased enrollment of black students in predominantly white colleges, one study shows that desegregation initiatives and demographic changes appear to have contributed to greater enrollment of white than black students at one historically black college (Brown, 2002).

Summary The outermost layer of the model explicitly recognizes that aspects of the social, economic, and policy context may also shape student success. Economic perspectives appear particularly useful for informing our understanding of the ways that particular public policies influence student success. Sociological perspectives draw attention to the role of broad social forces and social

52

inequities. Although few articles in psychology journals appear relevant to this layer (that is, Layer Four of the model), one potentially relevant perspective considers the role of the media in shaping student success. By including attention to federal and state public policies, demographic changes, and other forces, education research also suggests that the social, economic, and policy context influences student success.

Conclusion This chapter illustrates the perspectives that four disciplines (economics, soci- ology, psychology, and education) contribute to each of the four layers of the conceptual model. The chapter also offers examples of studies from each of the four disciplines for each layer of the model. Based on our review of research published in top journals, we conclude that psychological perspectives domi- nate the understanding of the “internal context” of student success (Layer One of the model). With their focus on an individual’s cognitive and affective processes, psychological perspectives seem ideally suited for understanding how such core attributes as an individual’s attitudes and motivations influence student success. Although stressing different facets, all four disciplinary per- spectives recognize the contribution of the family context (Layer Two of the model) to student success. Economics, sociology, and education perspectives often include attention to the ways that characteristics of the K-12 schools and higher education institutions that students attend (Layer Three of the model) influence student success, while psychological perspectives typically devote less attention to this layer of context. A review of research in these four disciplines also sheds light on the various ways that student success is influenced by the broader social, economic, and policy context (Layer Four). The final chapter builds on these understandings to offer implications of the model for policy- makers and practitioners as well as future researchers.

53Theoretical Perspectives on Student Success