2

Competencies for College Success

This chapter addresses the committee’s charge to identify a range of intra- and interpersonal competencies that may be related to persistence and success in undergraduate education and that evidence indicates can be enhanced through intervention. It opens with an overview of intra- and interpersonal competencies in general; the committee’s perspective on these competencies within home, community, and academic contexts; and the committee’s approach to addressing this part of its charge. The second section presents a developmental framework of intra- and interpersonal competencies that may be relevant to college experiences: (1) broad dispositions, (2) beliefs, (3) specific motivations, and (4) future identity. The following four sections examine each of these concepts and identify eight specific competencies within them, which appear, based on the limited evidence available, to be related to persistence and success in undergraduate education:

1. behaviors related to conscientiousness,
2. sense of belonging,
3. academic self-efficacy,
4. growth mindset,
5. utility goals and values,
6. intrinsic goals and interest,
7. prosocial goals and values, and
8. positive future self.

As noted in Chapter 1, the committee defines “competencies” broadly to include malleable attitudes, behaviors, beliefs, and dispositions. Although such competencies may be influenced by college environments or contexts, as they exist and are perceived, they ultimately reside within the individual student. As discussed further below, for example, controlled experiments have shown that interventions can develop students’ sense of belonging, that this competency persists over time, and that it is related to higher grade point average (GPA) relative to control groups of students in the same college context that did not receive the intervention. For each of these competencies, the committee reviews the evidence for a relationship to college persistence and success and its malleability in response to interventions. The chapter ends with conclusions and recommendations.

OVERVIEW

What student competencies (skills, abilities, beliefs, and attitudes) produce success in college? It is well documented that individual differences in general cognitive ability, as well as in many specific intellectual skills, such as critical thinking, writing ability, and computational expertise, are associated with individual differences in academic success. Simply put, students with a stronger and wider range of cognitive skills tend to do better in college. But other factors beyond cognitive abilities also appear to be predictive.

The most frequently used standardized measures of cognitive indicators, such as the SAT and ACT, account for a portion of the variance in college success and degree completion, but other factors, such as high school GPA, also account for some of this variance (Bowen et al., 2011; Massey et al., 2003; Vars and Bowen, 1998). There is evidence as well that traditional cognitive assessments may differentially predict performance across demographic groups. Across 110 U.S. colleges and institutions (151,316 students), the SAT was found to underpredict GPA for women relative to men (i.e., women tend to perform better than predicted by the SAT) and to overpredict GPA for minorities relative to the white majority (i.e., minorities tend to perform less well than predicted by the SAT) (Mattern et al., 2008).

Subgroup performance differences on tests and in college have been attributed to a wide range of societal and historical inequalities and messages (Massey et al., 2003; Steele, 1997). Likewise, at the level of the individual student, it is almost certain that a wide range of factors extending well beyond measured cognitive skills play important roles in determining the extent to which a student will succeed in college. Although many of these factors reside in the student’s environment (e.g., social class disparities, family resources, differential experiences of various kinds), others—including

intra- and interpersonal competencies—reside more clearly within individual students and therefore are more within their control. This chapter introduces the intra- and interpersonal competencies identified by the committee as those that appear to be most important for college success, based on its review of the research evidence.

The committee reviewed research on intra- and interpersonal qualities that speak to the social and emotional lives of human beings and to human motivation. This research examines intrapersonal competencies such as self-control and motivation to achieve and interpersonal competencies such as teamwork, social communication, and community involvement. In some studies, the distinction between “intra” and “inter” appears to be blurred because the competencies were found to be interrelated or to operate both within the individual and in interpersonal relationships. For example, teamwork is viewed as an interpersonal skill promoting positive involvement with other people, teams, and communities. However, effective teamwork also involves intrapersonal skills such as self-regulation and decision making. Overall, the committee found little rigorous research examining how interpersonal competencies may be related to undergraduate persistence and success.

The committee’s conceptualization considers the critical interplay between the characteristics of students and their academic, social, and cultural contexts. For instance, contextual factors such as family socialization and resources, campus climate and culture, quality of instruction, and other conditions of the student’s environment may influence—either independently or jointly—the development of competencies that affect how students respond to academic and social challenges. Students from diverse backgrounds may experience these contextual factors, and therefore college, in very different ways. For example, underrepresented groups¹ may find certain campus climates and disciplinary cultures and practices to be uncongenial or even aversive, yet find other campus climates supportive. For students to be successful in college, their unique experiences in these contexts may require that they draw on special strengths they have developed over the course of their lives, such as their capacity for adaptation and resilience.

With regard to success in science, technology, engineering, and mathematics (STEM), the joint consideration of competencies and the college context is especially relevant. For instance, a recent National Academies of Sciences, Engineering, and Medicine (2016a) report characterizes today’s overall culture of STEM and STEM college education as tending to emphasize inherent, or “natural,” cognitive ability as necessary for success. But

___________________

¹ As discussed in Chapter 1, the committee’s definition of “underrepresented groups” encompasses low-income and first-generation college students; women in certain STEM disciplines; and black, Hispanic, and American Indian student populations.

the report notes that STEM skills can be improved, and that many studies indicate that student attributes and beliefs are in fact malleable and can be improved through intervention. Furthermore, the report highlights how racial and gender stereotypes around ability are salient in STEM contexts, and can manifest themselves in how women and students of color are treated and responded to. When prejudice, discrimination, and perceptions thereof are not prevented at the group and individual levels, even the most successful high school students can be discouraged from pursuing a college major in STEM, with the ultimate effect of pushing many highly competent women and minority college students out of these fields (e.g., Ohland et al., 2008; Seymour and Hewitt, 1997).

Thus, this chapter identifies competencies that appear to be causally related to engagement and success in college and completion to degree, above and beyond the demonstrated effects of cognitive skills. The focus is on competencies that help explain persistence in college across students from diverse backgrounds. As noted, the committee sought to identify competencies that research suggests are malleable, and therefore can be improved in college through programs or interventions designed to enhance student development and thereby boost academic engagement, academic success, and college completion rates.

As discussed in Chapter 1, to identify such competencies, the committee searched for literature across a range of fields—including personality, development, educational, and social psychology—examining the relationship between intra- and/or interpersonal competencies and students’ college persistence and success. The committee also invited presentations by outside experts and commissioned the National Academies library staff to conduct four searches of literature databases (see Appendix A). These literature searches used various terms for intra- and interpersonal competencies that had been identified and organized in the prior report referenced earlier (National Research Council, 2012b). For example, one early search included the terms “interpersonal competence,” “teamwork,” “communication,” “intrapersonal skills,” “motivation,” “self-regulation,” “metacognition,” and “college persistence.” Although all four searches included various terms for interpersonal competencies, they yielded no rigorous research suggesting that these competencies are related to persistence and success in undergraduate education. Therefore, the committee began to focus on intrapersonal competencies showing evidence of a relationship to undergraduate success. Next, to identify competencies that are malleable in response to interventions and to find strong, causal evidence of a relationship with college success, the committee conducted a search of the intervention research.

In this process, the committee considered the potential role of students’ mental health or illness in their college success. The committee commissioned National Academies library staff to conduct a search of the litera-

ture using terms related to mental health derived from the above report (National Research Council, 2012b; see Appendix A). Some studies found that students who enjoy high levels of mental health are more likely than those suffering from depression, chronic anxiety, and related mental illnesses to make friends at school, join campus organizations, and perform up to their full academic potential (Megivern et al., 2003; Salzer, 2012). In addition, the committee considered research on how the personality trait of neuroticism² may influence students’ college success. As discussed in the following section, however, traits such as neuroticism are not considered malleable in response to the kinds of interventions that colleges can implement, and there is weak evidence that neuroticism is related to college persistence. Whereas a few studies suggest that neuroticism has a modest negative association with class grades (Chamorro-Premuzik and Furnham, 2003; Rigdell and Lounsbury, 2004), others show no such association (McAbee and Oswald, 2013).

After deliberating on this research, the committee determined that the broad domain of mental health lays outside its charge. In the committee’s view, improving mental health is a very different goal from enhancing college success and completion. Moreover, the kinds of interventions that are required for addressing problems in mental health include therapy, counseling, and clinical work. In contrast, interventions designed to develop competencies aimed at college persistence are more academic by nature, sometimes implemented directly within courses.

Through this process, the committee arrived at the list of eight intrapersonal competencies presented above. The identification of these competencies was based on both correlational and experimental research. The committee, however, made judgments about the strength of the evidence related to each competency based exclusively on experimental research, for two reasons. First, as noted in Chapter 1, the committee was charged to identify competencies that are malleable in response to interventions. Because experiments attempt to isolate an intervention as the cause for changes in competencies and outcomes (e.g., through random assignment and the use of control groups), the experimental evidence from intervention research can provide strong evidence on a competency’s malleability. Second, experimental manipulation can be stronger than nonexperimental approaches in reducing biases from omitted variables.³ The committee’s

___________________

² As discussed further below, neuroticism is one of the “Big Five” (Goldberg, 1993; McCrae and Costa, 2008) personality traits, or broad dispositions.

³ The committee acknowledges that both experimental and correlational published studies can suffer from a host of problems, including post hoc “fishing expeditions” that involve multiple measures, subgrouping categories, and so on (Gelman and Loken, 2013; Simmons et al., 2011). As one countermeasure, researchers encourage the preregistration of one’s study design, hypotheses, and analysis plan (e.g., on AsPredicted at http://aspredicted.org [January

search for relevant intervention studies is described in Box 2-1; the effects on academic outcomes found in these studies are summarized in Table 2-1. Each study is described in more detail in the Intervention Table,

___________________

2017]; and through the Open Science Framework at http://osf.io [January 2017]). Exploration is still encouraged, but preregistration helps distinguish the researchers’ a priori plan from their post hoc analyses. Also, the committee acknowledges that even under experimental control, there are many potential reasons for a treatment’s effectiveness that can be explored further in future research (e.g., further experiments that isolate aspects of the intervention, use of different higher education institutions and samples).

* The evidence is described in greater detail in the Intervention Table, which is available at https://www.nap.edu/catalog/24697/supporting-students-college-success-the-role-of-assessment-of-intrapersonal. Click on the Resources tab and then the link for Intervention Table.

which is available at https://www.nap.edu/catalog/24697/supporting-students-college-success-the-role-of-assessment-of-intrapersonal. Click on the Resources tab and then the link for Intervention Table.

A DEVELOPMENTAL FRAMEWORK

Every college student’s experience is unique. Gender, race/ethnicity, age, and social class—and their unique configurations—shape students’ col-

lege experiences in profound and innumerable ways. Students who attend elite private colleges tend to have different characteristics, enjoy different opportunities, and face different challenges relative to those who attend 4-year state institutions or community colleges, or those who take online courses, or those enrolled in technical schools aimed at training students for a particular niche in the economy. Graduation rates are higher at highly selective institutions and private nonprofit institutions than at less selective public colleges and universities (Kena et al., 2016). A 35-year-old minority commuter student who simultaneously holds down a full-time job would typically perceive the nature and value of her education very differently from an 18-year-old Caucasian freshman whose parents just drove him 500 miles to campus and helped him move into his dorm room. Likewise, two students with the same levels of achievement and aspirations in STEM fields in high school may be differentially challenged by a college environment if one is a minority who is reminded that few members of his race/ethnicity enter or succeed as STEM majors, while the other is a Caucasian student who typically experiences an environment of inclusion and high expectations (National Academies of Sciences, Engineering, and Medicine, 2016a). Thus, no single developmental course fully explains how all students move through their college years.

At the same time that these wide differences in situational contexts and individual characteristics influence college success, many college students face certain challenges in common, and in roughly the same order. Even before they take their first class, for example, students bring with them a set of competencies that reflect, in part, prior experiences in education and in life. These competencies speak to a basic question that many students may ask themselves even before entering college: What are my strengths? Such strengths may include broad dispositional resources, such as the ability to sustain academic effort even in the face of boredom or the abilities to recruit positive emotion and avoid or redirect negative emotion under conditions of stress. These basic dispositions may continue to develop through the college years, and the student may continue to draw on dispositional strengths in meeting the many challenges, both academic and personal, that arise in college.

Upon entering college, one of the first challenges a student tends to face concerns feelings or questions about fit or comfort in the new college environment. Whether the new student is trying to adjust to dormitory life or commuting from home to take courses in the evening, a central question that arises very early on is this: Do I belong here? As the student begins to attend classes and attempts to do the required academic work, another question is likely to follow: Can I succeed? Over the course of college—from the initial formative years until graduation—students will develop and refine beliefs, explicit or implicit, about the extent to which they belong in

their college and the extent to which they are likely to be successful. These beliefs themselves constitute competencies in their own right and may also inform the development and expression of other competencies instrumental to college success.

As students continue to take classes, develop new relationships, participate in extracurricular activities, and become involved in college activities, their specific motivations regarding college may become clearer to them, as well as to others. Most students eventually ask more pointed questions about their college experience, such as: What are my goals? What most interests me? What do I value? Of course, some students have goals and values regarding college long before they take their first college class. High school counselors, teachers, parents, and peers may ask prospective college students to explain why they want to go to college in the first place or what they plan to get out of college in the long run. For instance, some students have developed interest in and established academic goals for a STEM discipline. For these students, the departmental and classroom climates, interactions with faculty, the perceived relevance of the curriculum, and methods of instruction may be the most important factors in their commitment to completing a major (Carlone and Johnson, 2007; Chang et al., 2011; Seymour and Hewitt, 1997). Other entering students, however, may not initially self-identify as scientists (Chang et al., 2011) and may therefore tend to undervalue STEM course content and the particular instructional methods used to present it (e.g., Darling et al., 2008).

College success may hinge not only on goals and values but also on the student’s identity, or understanding of self, in the context of the college environment. Over the course of college, the following questions may become increasingly salient: Who am I? Whom do I want to become? Developing a positive identity regarding the current and future self and linking that identity in meaningful ways to one’s experiences in college is a daunting challenge for any college student. Indeed, the ability to do so may itself be considered an important intrapersonal competency, one that may be relevant at any point in a college student’s career. It is likely to be especially important toward the end of college, as students who anticipate graduation look ahead to their postcollege future. It also may be the case that students who develop positive identities tied to their early college experiences are more likely to persist and graduate compared with students who are unable to do so.

BROAD DISPOSITIONS: WHAT ARE MY STRENGTHS?

Before prospective college students ever set foot on campus or enroll in their first class, they bring with them a unique life history and profile of general competencies. The intra- and interpersonal competencies that

travel with the student into college, and evolve over time, include broad dispositional tendencies that pertain to emotional well-being and mental health, social and instrumental effectiveness, and attitudes and orientations regarding the self and the world. As an initial step in developing a list of competencies that appear to be related to persistence and success in undergraduate education, the committee drew on 50 years of research in personality psychology. This research has resulted in identification of the “Big Five” framework (Goldberg, 1993; McCrae and Costa, 2008) as a useful and generally comprehensive accounting of broad dispositions relevant to intra- and interpersonal competencies.

These five broad dispositions are typically labeled as conscientiousness, neuroticism (the opposite of which is emotional stability), extraversion, openness to experience, and agreeableness. Each of the five describes a set of intra- and interpersonal strengths or weaknesses students may bring with them upon entrance to college. Whereas all five dispositions can, in principle, have bearing on student success in college, the empirical evidence suggests that conscientiousness is the one trait among the five that is correlated most consistently with college persistence and success. Indeed, the correlational literature suggests that conscientiousness is the strongest predictor of college persistence and success among all intra- and interpersonal competencies.

Conscientiousness and Related Behaviors

Conscientiousness refers to a spectrum of dispositional tendencies that describe individual differences in the propensity to be self-controlled, responsible to others, hardworking, orderly, and rule-abiding (Roberts et al., 2014). The broad rubric of conscientiousness encompasses a large number of overlapping dispositional constructs that have, over the decades, been variously described in different theories of personality with such labels as self-control, ego control, effort control, delay of gratification, self-discipline, future orientation, constraint, planfulness, socialization, and grit. Individuals who score high on conscientiousness are typically described (by themselves and others) as especially well-organized, hardworking, dependable, disciplined, dutiful, and persevering. By contrast, those scoring low on the trait are seen as disorganized, impulsive, lazy, undependable, undisciplined, irresponsible, and lacking in self-control. At the temperamental heart of conscientiousness lies a capacity to set aside impulses and other short-term distractions to pursue long-term ends. In a useful conceptualization, Hill and Jackson (2016) describe conscientiousness as a general tendency for individuals to invest time and energy with a focus on long-term, socially valued payouts.

Within the Big Five taxonomy of dispositional traits, individual differences in conscientiousness are the strongest and most consistent predictor

of success in various life domains, including marital stability, health, and occupational attainment (Roberts et al., 2007). High levels of conscientiousness are prospectively associated with success in nearly every type of occupational field, but especially in jobs and careers that demand high levels of autonomy (Barrick and Mount, 1993).

The correlational research indicates that conscientiousness predicts academic success long before college. For example, a number of studies have found that high scores on self-reported assessments of conscientiousness are associated with better high school grades (e.g., Freudenthaler et al., 2008; Poropat, 2009). In examining the possible reasons for the link between conscientiousness and GPA in high school, researchers have pointed to the role of academic effort: teenagers with greater conscientiousness tend to work harder in school. One recent study of nearly 5,000 German students in grades 5-8 showed that conscientiousness was a strong predictor of academic effort, especially when students found a school subject to be uninteresting (Trautwein et al., 2015). The results of this study suggest that in the face of boredom, many students may curtail their effort, but those high in conscientiousness may continue to work hard, which pays off in higher grades.

At the postsecondary level, empirical studies (based on correlational research) consistently demonstrate that high levels of conscientiousness predict better academic performance and higher levels of college completion (e.g., Chamorro-Premuzic and Furnham, 2003; Trapmann et al., 2007), even in research controlling for general cognitive ability (e.g., Okun and Finch, 1998; Poropat, 2009). Examining multiple measures of the Big Five dispositions, McAbee and Oswald (2013) found that measures of conscientiousness showed consistently strong predictive correlations with college GPA, whereas the other four dispositions (i.e., neuroticism, extraversion, openness to experience, and agreeableness) did not. In their authoritative review of psychological correlates of academic performance, Richardson and colleagues (2012) report that high effort regulation and low procrastination were statistically significantly associated with college GPA. Effort regulation is virtually synonymous with the self-control and perseverance themes of conscientiousness. Procrastination would appear to track the low end of conscientiousness, as depicted in facets of conscientiousness that tap into punctuality, orderliness, and achievement striving (De Young et al., 2007; McCrae and Costa, 2008).

In an effort to tease apart the relationship between conscientiousness and academic success in college, Corker and colleagues (2012) examined the strategies, goals, and academic behaviors of nearly 350 college students over two semesters. They found that the link between high conscientiousness and high grades was explained by students with high conscientiousness engaging in such concrete behaviors as completing homework assignments

on time, studying hard for tests, and persevering even when the course material was boring. Highly conscientious students put more time into their studies, and they completed projects in a timely manner. They also managed to suppress distractions that might divert them from completing their assignments, even when the assignments were tedious and difficult.

Although conscientiousness generally is a powerful predictor of academic achievement, some evidence suggests that the strength of this correlation varies across different student populations. In their study of Austrian adolescents, for example, Freudenthaler and colleagues (2008) found that the correlation between conscientiousness and GPA was statistically significant for girls but not for boys, whereas school-related intrinsic motivation, school anxiety, and performance-avoidance goals explained additional variance in GPA only for boys. As noted above, researchers (e.g., Corker et al., 2012) suggest that conscientiousness influences high grades through effort. However, DeLuca and Rosenbaum (2001) analyzed data from the High School and Beyond study and found that the rewards to effort varied with socioeconomic status. Although high school effort had a statistically significant relationship with later educational attainment net of socioeconomic status, the attainment benefits of effort varied by socioeconomic status. De Luca and Rosenbaum (2001) suggest that, even if students of low socioeconomic status strive very hard, their educational outcomes may not be improved or rewarded as much as those of other students, so they may have less incentive for school effort.

Lundberg (2013) conducted a study that identified substantial differences across types of families in the personality traits that predicted successful completion of college, particularly for men. Using data from the National Longitudinal Study of Adolescent Health to model the associations between personality traits and college graduation for different groups, she found that these relationships varied by gender, socioeconomic status (based on mother’s level of education), and race. Among men, conscientiousness was statistically significantly related to college completion for those from families of high socioeconomic status but not for those from less advantaged families. Among women, however, conscientiousness was statistically significantly related to college completion regardless of socioeconomic status. For both men and women, openness to experience was significantly correlated with college completion for those from families of low socioeconomic status, but not for those from advantaged families. Estimating the model separately for black and white subsamples, Lundberg (2013) found that, regardless of socioeconomic status, the positive effect of conscientiousness on educational attainment was weaker for black men and women, and the marginal effect of openness to experience was stronger. As noted earlier, the committee believes further research is needed to understand how family and academic contexts influence the development of

intra- and interpersonal competencies and their relationships with academic achievement.

Most personality psychologists see all of the Big Five traits, including conscientiousness, as deeply ingrained and relatively stable tendencies that are difficult to alter. As with all dispositional personality traits, individual differences in conscientiousness reveal, in part, genetic variability among people. As measured by self-report tests and ratings of observers, individual differences in conscientiousness are at least 50 percent heritable (Bouchard, 2004; Jang et al., 1996; Power and Pluess, 2015). This means that at least half of the variation in conscientiousness scores within a population is attributed to genetic differences among people. Moreover, individual differences in conscientiousness tend to be relatively stable, and their stability increases over the adult life course (Roberts and DelVecchio, 2000). Put simply, the most conscientious person in a high school graduating class will probably score somewhere near the top of a distribution of conscientiousness scores at the time of his or her 10-year reunion.

Nevertheless, longitudinal studies also demonstrate that people tend to show a gradual increase in conscientiousness from late adolescence through midlife (Roberts et al., 2006). Underlying this tendency, different people change in different ways. A growing body of life-span personality studies demonstrates that major changes in social roles (e.g., getting married, becoming employed, having a child) are associated with changes in conscientiousness, although one study found that the changes in conscientiousness may have preceded these events (Specht et al., 2014). Nonetheless, major changes in social roles also constitute extremely intensive “interventions”—much more intensive than what might take place in a college setting.

Magidson and colleagues (2014) suggest that if an intervention aims to target conscientiousness directly, it should focus on the discrete behavioral and motivational factors that inform the broad trait of conscientiousness, ideally providing “detailed structure, a focus on values, guided action, goal-setting, immediate feedback on progress and challenges, clear accountability, and an opportunity for remediation” (p. 1445). The logic behind this approach suggests that conscientiousness is supported by, and articulated through, a collection of more specific intra- and interpersonal constructs that entail particular behaviors, goals, values, identities, and the like. These constructs and processes are perhaps not technically part of conscientiousness itself, but may work to support it and/or work with it to produce positive life outcomes related to academic success.

Moving beyond correlational research designs, the committee identified seven random-assignment evaluations of interventions that, as suggested by Magidson and colleagues (2014), target discrete motivations and behaviors related to conscientiousness. Most of these interventions consisted of setting, monitoring, and (in some cases) coaching various study goals and

work habits. Four of the seven interventions assessed impacts on academic outcomes. The largest of these studies is that of Bettinger and Baker (2014), who evaluated InsideTrack, a for-profit student coaching service that is now the largest provider of student coaching in the country. Most of the more than 13,000 participating students were nontraditional college students. The exact nature of the intervention activities is proprietary information, but one of the study authors reports that InsideTrack coaches reached out to assigned students (via phone, text, email, and social media). The goals of the meetings varied according to student need, but the coaches generally assessed potential risk factors, worked to develop both cognitive and noncognitive skills (e.g., time management, self-advocacy) and guided students in seeking and leveraging campus supports. Coaches usually worked with their students over the course of two semesters.

Bettinger and Baker (2014) were forced to rely on data that the company supplied, although extensive checks showed broad-based comparability between treatment and control groups. They found that this goal-setting and support intervention increased the fraction of students persisting in the university 6 months after the end of the intervention by 5.1 percentage points, the fraction persisting 12 months after the intervention by 5.3 percentage points, and the fraction persisting 24 months after the intervention by 3.3 percentage points. Rates of college graduation were 4.0 percentage points higher for the treatment relative to the control group. All of these differences were statistically significant.

Morisano and colleagues (2010) report that their web-based goal-setting intervention increased retention rates significantly but do not report the sizes of those effects. Liu and colleagues (2012) sought to motivate students with information that their performance on a test battery might be reported to employers, and found impacts of around 0.40 standard deviation (SD) on a proficiency test, scores on an essay, and a self-reported motivational assessment. Duckworth and colleagues (2016) found (in Study 3) that psychology students randomly assigned to a condition that set study goals and attempted to remove temptations that might interfere with meeting those goals had greater success at meeting the goals 1 week after the intervention relative to controls, although they did not test for achievement or attainment impacts.

Taken together, the evidence that conscientiousness-based interventions improve college persistence is modest (see Table 2-1). Most random-assignment studies failed to measure academic outcomes, and the one that did (Bettinger and Baker, 2014) has proprietary restrictions on revealing the exact nature of the intervention. The Bettinger and Baker (2014) intervention showed positive effects on the 12-month college persistence rate of 5.3 percent, and the goal-setting intervention (Morisano et al., 2010) showed significant improvement in students’ college semester GPA (effect size = 0.65 SD.

BELIEFS: DO I BELONG HERE? CAN I SUCCEED?

The transition to college presents an exciting opportunity for many new students, but it can also present daunting personal and interpersonal challenges. Many new students, just out of high school, are leaving home to live on their own for the first time. Some are leaving their families to live on a distant campus they may have seen only once or twice before. Many are breaking away from established friendship networks while establishing new ones. Others may be trying to juggle part-time jobs or existing family commitments to take on a demanding new role. Those who are the first in their family to attend college may have few identifiable role models in their immediate college environment to help them navigate the college experience. At the same time, first-generation students may also feel the pressure of high expectations, as if the hopes and dreams of the family are now riding on their efforts and success. Students from homogeneous communities may be experiencing racial and socioeconomic diversity for the first time in dorms and classrooms. Academically and socially, many new college students may feel as if they are now strangers in a strange land.

Indeed, models of voluntary student departure and persistence suggest that the transition from high school to college is a critically important period, as the early development of goals/intentions and institutional commitment is critical to engagement and eventual completion (DeAngelo et al., 2011; Nora and Crisp, 2009; Tinto, 1994). It is not surprising, therefore, that researchers and institutions have devoted significant attention to the transition experience and to interventions that can help new students navigate college.

The question Do I belong here? may arise in a multitude of contexts for students beginning their college experience. New students want to know: Can I make new friends in this new place? Are there groups on campus that will accept me? Do I feel comfortable here? Do I feel that I fit in? After placement tests and by midterm of the classroom experience, another question is certain to arise for many students: Can I succeed here? Students wonder, Can I meet expectations and perform well in these classes? If the academic strategies I have used in the past do not seem to work here, what should I do?

These questions are very personal, yet the ways in which a student answers them depend on many factors that go well beyond the student’s own dispositional strengths. Institutional factors include the general ethos of the college environment itself—its demographic makeup and peer climate, the level of resources devoted to student life and instruction, the opportunities made available for meaningful involvement, the quality of instruction, the selectivity of the institution, and the financial aid awarded (Bowen et al., 2011; Oseguera and Rhee, 2009; Singell, 2004; Titus, 2004). Experiences

in high school and financial concerns also are likely to have a broad impact on both initial college choice and retention (Cabrera et al., 1993; Paulsen and St. John, 2002).

Relevant as well are many individual factors that go well beyond the broad dispositional trait of conscientiousness. An important internal factor may be the particular beliefs about college that a student develops early in his or her career. Indeed, what a student believes or perceives about the environment may be nearly as important as the objective realities of the environment itself. For example, perceptions of the campus climate (e.g., hostility, competition) affect the successful transition to college for students from many different racial/ethnic groups and majors (STEM and non-STEM) in terms of managing academic adjustment and integration in college (Cabrera et al., 1999; Hurtado and Carter, 1997; Hurtado et al., 2007; Locks et al., 2008). Similarly, what the student believes about his or her own abilities to succeed may be nearly as important as the strength of those abilities.

When it comes to beliefs about college, then, the research literature suggests that at least three intrapersonal competencies may be involved in determining how well students do in college: a sense of belonging, academic self-efficacy, and a growth mindset.

Sense of Belonging

A student’s sense of belonging is an intrapersonal competency that is influenced by interpersonal/social relationships as well as academic/professional concerns. It exists as a complex and evolving set of beliefs in the mind of the student regarding the extent to which he or she fits into important niches or features of the college environment.

Motivational theory posits that individuals are most likely to engage and perform positively in settings in which they feel a sense of connectedness or relatedness to others (Deci and Ryan, 1991). At the heart of human nature lies a basic need to be part of a group, also referred to as a psychological sense of integration (Tinto, 1994) or sense of belonging (Hurtado and Carter, 1997; Strayhorn, 2012). The transition to college marks a critical period in a person’s life when the sense of belonging is likely to be strongly activated. As a complex social ecology, college presents a panoply of groups with which students may affiliate, including formal organizations with rules for membership (e.g., sororities and fraternities), informal friendship networks, religious and political student organizations, and groups organized around academic concentrations (e.g., major and professional career clubs) and other personal interests (e.g., hobbies and sports).

Research shows that students with a higher sense of belonging in STEM are “more likely to report having friends in the major, socializing with peers

and faculty in the field, and feeling like their ‘friends would miss [them]’ if they left” college (Strayhorn, 2012, p. 68). In the minds of some students, moreover, the college itself may resemble an extended group, as participation in these social niches enables them to feel that they are an integral part of the college writ large (Hurtado and Carter, 1997; Hurtado et al., 2007). A sense of belonging also is associated with whether students feel that the “climate” for learning is welcoming or unwelcoming in their first year of college (e.g., instructional practices that promote cooperation and inclusiveness rather than competition and stereotyping).

The research literature using correlational methods suggests that feeling a sense of belonging, bonding, or solidarity with others in college tends to be positively associated with college retention (Robbins et al., 2009). Although most students’ intention to persist in college tends to decline over the first year, a random-assignment intervention designed to cultivate a higher sense of belonging among students was found to stem this decline (Hausmann et al., 2009). The intervention, which entailed providing community-oriented letters from university administrators and gifts with university logos versus general letters from faculty and generic gifts, was found to have similar effects for both white and black students, on average.

Sense of belonging is positively associated with the ability to manage academic adjustment, with grades, with self-rated change in the ability to conduct research, and with perceptions of the relevance of coursework in the first year of college (Hurtado et al., 2007). Although performance in the first year of college is a strong determinant of degree completion within 6 years, where one goes to college (level of selectivity) and how students feel about belonging in their college during this transition year are associated with 6-year degree attainment across all racial/ethnic groups (Hurtado and Ruiz Alvarado, 2016). Pittman and Richmond (2008) found that higher scores on sense of belonging predicted higher grades, academic competence, and positive self-worth. On the other hand, Richardson and colleagues (2012) found little evidence in their meta-analysis that such variables as social integration and social support were associated with college GPA. Their review, however, focused not on specific measures of sense of belonging but on general social integration measures that often include a variety of social activities, nor was it specific to the first year of college.

Although belonging and connectedness to an educational context are important for all students, they may be especially important to college achievement and persistence for student groups that are underrepresented in higher education (Strayhorn, 2012). Some studies have shown that underrepresented minority groups report a lower sense of belonging relative to their white counterparts (Johnson et al., 2007), in part because of perceptions and experiences of discrimination (Hurtado and Carter, 1997). The former groups are especially likely to experience challenges to their identity

(in the form of negative climate, discrimination, and perceived microaggression) that can undermine their sense of belonging and connectedness to the academic context (National Academies of Sciences, Engineering, and Medicine, 2016a). Students from these backgrounds tend to be particularly underrepresented in some STEM contexts (e.g., being the only Latina in a science class), which can make it even more challenging to develop or maintain a strong sense of belonging.

The committee identified 10 intervention studies addressing students’ uncertainty about “belonging” in college settings. Of these 10 studies, 7 report academic outcomes, including GPA in 5 of the studies, and 8 report competency outcomes. In a laboratory intervention, Walton and Cohen (2007) attempted to mitigate doubts about social belonging in college. Participants in the treatment condition received survey information from upperclassmen indicating that most students worry about belonging during their first year, but that these concerns lessen over time. Control students received information unrelated to the issue of belonging. Relative to their control group counterparts, black students (but not others) in the treatment group experienced a substantial increase in their GPAs. The intervention also improved students’ achievement behavior, again for black participants only. Black treatment group students reported studying an average of almost 1.5 hours more per day compared with black control group students, and also sent more email queries to their professors. The black treatment group students also felt a greater sense of academic fit, both immediately after the intervention and 7 days later. The treatment, however, actually had adverse effects for white students, whose sense of academic fit had decreased immediately after the intervention and 7 days later.

In a subsequent study, Walton and Cohen (2011) implemented a similar social belonging intervention, in this case lasting 1 hour in a laboratory setting, involving students in their second semester of their first college year. As in the Walton and Cohen (2007) study, black students in the treatment group experienced a significantly higher increase in their GPA relative to black students in the control group, with the GPA effect persisting through their senior year. GPA trajectories of white students in the treatment condition were not significantly different from those of their control group counterparts. Overall, the intervention closed the minority achievement gap by 52 percent.

Walton and colleagues (2015) report on two interventions targeted at women in engineering. The first entailed communicating to 228 female and male students in their first college semester that uncertainty about belonging in engineering is common during the first year of college, but it is temporary. The second, a values-affirmation intervention, encouraged students to incorporate personally important self-identities and values in their daily lives as an explicit strategy for managing stress and threat. This

intervention was designed to empower the female students to maintain their own well-being. Students’ transcript data for their first year and survey data from questionnaires administered immediately after the intervention both showed that the first intervention was generally effective for women in male-dominated but not in gender-diverse engineering majors. Both interventions raised GPAs meaningfully (by about 11 points on a 100-point GPA scale) among women in male-dominated majors relative to the control condition, but no statistically significant differences emerged for women in the gender-diverse majors. Walton and colleagues (2015) also report positive competency outcomes. The two interventions combined generally improved women’s felt experience in engineering—as measured by a sense of belonging, self-efficacy, and enjoyment—relative to the control condition (a moderately large effect size equal to a standardized mean difference of 0.67). This outcome persisted to the second semester. The interventions also improved women’s confidence in their prospects for succeeding in engineering.

An intervention study conducted by Stephens and colleagues (2014) targeting first-year, first-generation college students provides further support for the efficacy of interventions targeting sense of belonging. The intervention consisted of students in both treatment and control conditions attending a panel discussion. Both sets of panelists highlighted how they had adjusted to and succeeded in college. The key difference between the treatment and control panels was whether the panelists’ stories highlighted how their social class backgrounds mattered for their college experience. First-generation students’ end-of-year GPAs were substantially higher for the treatment than for the control group, and not statistically significantly different from the GPAs of non-first-generation students. A number of psychosocial outcomes also were measured. Treatment participants (regardless of generational status) scored significantly higher than controls on the following measures: psychological well-being, social fit, academic identification, and maintaining relationships. They did not differ from control participants on psychological distress, social identity threat, and social support.

A study by Folger and colleagues (2004) represents the last one in this category to include GPA among its achievement outcomes. In this case, three different GPA outcomes were measured: first-semester, second-semester, and cumulative GPA. The study participants were academically at-risk first-generation college students who participated in the Freshmen Empowerment Program (FEP). Unlike the other interventions in this category, the FEP was intensive, not brief. As part of this program, treatment students met for 6 weeks to discuss a range of topics (e.g., academics, college resources, adjustment to college). At each time point, the mean GPA of the intervention group was statistically significantly higher than that of the control group, but effect sizes are not reported.

A belonging intervention study conducted by Walton and colleagues (2012) examined the relationship between social connections and achievement motivation. In this study, students were randomly assigned to a skill-promotive context condition or to a relational context condition. In both conditions, students read a fictional report written by a recent graduate of the math department. The reports were identical; they varied only in their characterization of the math department. In the skill-promotive condition, the department was portrayed as providing students with opportunities to develop their personal ability and interest in math, whereas in the relational context condition, the report portrayed opportunities for positive, collaborative social interactions. Students in the latter condition reported greater motivation for and sense of social connectedness with math.

Taken together, the evidence from interventions targeting sense of belonging is promising. Among the eight intrapersonal competencies identified in this report, interventions targeting this competency showed the most consistent impacts on college performance. Six of seven studies testing the relationships between these interventions and GPA showed significant positive effects (effect size range = 0.25-1.10 SD), although their effects were often selective—significant for some groups but not others. Importantly, the selective impacts tended to be stronger for underrepresented student groups—women in male-dominated engineering majors in one study, first-generation students in another, and black students in two others.

Research using correlational methods has shown that a strong sense of belonging early in college is associated with engagement in specific college activities and with success in college, and a growing body of intervention research demonstrates that it is possible to foster such a sense of belonging. More research is clearly needed on how to develop a sense of belonging among college students and how this competency promotes student success. It is especially important to disaggregate group data on this construct in order to examine differences and similarities across gender and other demographic groups. Sense of belonging may pose an especially vexing problem for underrepresented student groups (e.g., first-generation college students, underrepresented minority groups), who may feel disenfranchised or on the periphery in a given college context. Programs and policies aimed at promoting sense of belonging need to focus on the specific needs of and challenges faced by particular student groups, identified by particular demographic profiles and life history experiences.

Academic Self-Efficacy

For nearly four decades, psychologists have conceived of self-efficacy as a person’s belief that he or she can successfully carry out “courses of action required to deal with prospective [future] situations containing many am-

biguous, unpredictable, and often stressful elements” (Bandura and Schunk, 1981, p. 587). According to proponents of this construct, self-efficacy is specific to particular tasks or realms of life. Therefore, a person might have a different level of self-efficacy in, say, the realm of sports compared with the realms of work, friendships, health and diet, and so on.

Academic self-efficacy, then, is a student’s belief that he or she can carry out actions that will lead to success in school. For college students, academic self-efficacy is a subjective belief in one’s own competence to meet the academic demands that arise in college. The belief is surely influenced by prior experiences of success and failure in high school. Not surprisingly, one’s belief in one’s competence is a function of one’s actual competence in a given area. Nonetheless, the subjective belief itself is viewed as having strong motivational power, above and beyond objective competence itself. Indeed, the subjective belief can boost (or hinder) objective competence.

Self-efficacy beliefs are construed as empowering the individual to become a more effective agent in a given goal domain. Psychologists typically assume that self-efficacy beliefs derive from past events in which the person experienced mastery or success in the given domain. In expectancy-value theories of motivation (e.g., Eccles, 2009; Eccles et al., 1983; see also Wigfield and Cambria, 2010), expectancy refers to an individual’s subjective expectation of success for a given task—not unlike self-efficacy. According to expectancy-value theories, people are motivated to pursue goals or ends (1) wherein they expect success and (2) to which they assign importance or value.

A considerable body of research has shown that high levels of academic self-efficacy are positively associated with college students’ academic performance and college retention (e.g., Chemers et al., 2001; Matten and Shaw, 2010). It is important to note that this research provides no causal evidence. In a meta-analysis of psychosocial and study skills in college, for example, Robbins and colleagues (2004) found that academic self-efficacy was the strongest predictor of GPA. In their meta-analysis of a broad set of psychological factors implicated in college success, Richardson and colleagues (2012) identified performance self-efficacy as a significant predictor of college GPA, even after controlling for cognitive ability as measured by high school GPA and SAT or ACT scores. Lent and colleagues (2005) found that academic self-efficacy enhanced motivation for minority students pursuing engineering careers. Similarly, Nordstrom (2012) found that mathematics self-efficacy predicted achievement in basic-skills math courses for Hispanic students in a community college.

Developing and maintaining strong academic self-efficacy can be uniquely challenging for underrepresented student groups. A robust body of scholarship in the area of stereotype threat (e.g., Steele, 1997) has demonstrated how stigmatizing experiences based on race/ethnicity, gender,

or social class serve as “identity threats” that may undermine students’ academic task performance. These threat experiences are represented by academic settings that make stereotypes salient or by students’ encounters of stereotype-based treatment in their academic settings. Such experiences are consistently documented among racial/ethnic minority students and women in STEM (National Academies of Sciences, Engineering, and Medicine, 2016a). Repeated threat experiences over time are said to lead to students’ disidentifying with, or disconnecting their personal identity from, the academic domain in order to protect their self-concept from the deleterious effects of threat experiences. Ironically, the experience of threat is most likely for those students who identify most strongly with the academic domain. These processes have been demonstrated in STEM specifically. Chang and colleagues (2011), for example, report that students who possess a high science identity (i.e., a strong connection to and self-efficacy in science) but encounter stereotype-based experiences in college are more likely than students who do not have such experiences to change majors in the first year of college.

Research on improving self-efficacy is often specific to a particular knowledge domain or context. Fitch and colleagues (2012), for example, designed a “solution focused, goal-setting, group work” intervention for college students that resulted in statistically significantly higher scores on measures of academic self-efficacy for the experimental relative to the control group.

Two intervention studies that met the committee’s criteria targeted academic self-efficacy, although only one measured impacts on academic outcomes. Luzzo and colleagues (1999) assessed the academic impacts of an intervention that consisted of a short video presentation from two graduates of the university who described their early indecision over choice of major. The two were undeclared majors early in college, but after successful encounters with STEM courses had gone on to graduate and begin successful STEM-related careers. Intervention subjects (94 in all) were college students who at the time of the intervention had not yet decided what career to pursue and possessed at least a moderate degree of math ability. The intervention generated quite large positive impacts on STEM-related academic self-efficacy, as well as large impacts (standardized mean difference of 0.90) on subsequent enrollment in STEM courses and moderate impacts (standardized mean difference of 0.40) on opting to major in a STEM field.

Overall, evidence from interventions designed to boost academic self-efficacy is modest.

A single study showed significant, moderate to large effect sizes for achievement-related outcomes (effect size range = 0.42-0.90 SD). Although correlational research identifies this competency as a powerful predictor of college success, the modest evidence from interventions highlights the

need for further research to determine whether, and to what extent, simply believing that one can do well will help one do well. Future studies need to examine how self-efficacy develops and how it can be promoted in different student groups. It may be that specific kinds of efforts are needed to boost and/or help maintain high academic self-efficacy among underrepresented minority groups, women, students from low-income families, and first-generation students in STEM fields. Such research would attend to cultural or structural features of the academic context that may represent threats with the potential to negatively impact academic self-concept and self-efficacy among these underrepresented groups.

Growth Mindset

Whereas academic self-efficacy refers to a student’s beliefs regarding his or her ability to be successful in a particular academic pursuit, growth mindset refers to the extent to which a person understands his or her own ability to be fixed or malleable (Dweck and Molden, 2005). The target ability most often studied is general cognitive ability (intelligence). An entity-based mindset applies when a person understands his or her own intelligence to be a fixed entity: The person may believe that he or she has high (“I am smart”) or low (“I am dumb”) intelligence, or somewhere in between, but whatever the level is, the person assumes that the level is more or less unchangeable. A fixed-entity mindset on intelligence becomes especially problematic when the person encounters challenges that seem to go beyond his or her self-ascribed level of intelligence: “Because I cannot accomplish this task, I must be dumb; and because my intelligence is fixed, I cannot make myself any smarter than I am to accomplish this task.” In the face of difficult cognitive challenges, then, a person with an entity-based mindset regarding intelligence may give up. By contrast, a growth mindset assumes that intelligence is malleable and, therefore, improvable through effort. When faced with difficult cognitive problems, the person with a growth mindset regarding his or her intelligence may show more resilience and may be better able to learn from mistakes.

Believing that one’s intelligence can grow through experience thus may constitute an intrapersonal competency that has implications for both identity and motivation. From the standpoint of identity, the growth mindset indicates that “I am the kind of person who can learn from mistakes and improve over time; I can get smarter and accomplish more through hard work.” This understanding of the self may motivate the person to work harder, which may ultimately result in deeper learning and greater success.

A recent meta-analysis shows that growth mindset tends to be associated with greater self-control and with positive expectations in the pursuit of goals (Burnette et al., 2013; see also Robins and Pals, 2002). A growth

mindset also tends to be positively associated with academic achievement, although the effect sizes here are relatively modest (Blackwell et al., 2007).

Most social psychologists who study mindset assume that an individual’s mindset is readily modified, an assumption that has led to many interventions in this area. Among all the intervention studies meeting the committee’s criteria, the largest number (17) focused on mindset. Specifically, these studies examined the way in which students responded to failure events, and were meant to increase the likelihood that students would attribute failure to controllable causes, such as effort, and decrease the likelihood that they would attribute it to uncontrollable causes, such as ability. Within this group of mindset interventions, 16 assessed impacts on academic outcomes, although not all of them report impacts in a readily interpretable form. Aronson and colleagues (2002) targeted stereotype threat among African American students with a pen-pal condition that encouraged students to view intelligence as malleable rather than fixed. The intervention boosted beliefs about the malleability of intelligence several months after it ended, as well as enjoyment of education and identification of academic achievement. A number of these impacts were statistically significant for both African American and white students. Relative to controls, later GPAs were boosted for African American students only.

A number of studies consisted of an attributional retraining (AR) intervention. Prior to completing an achievement task, students viewed a video depicting two students discussing the ways in which poor performance can improve. The video dialogue focused on controllable attributions; after performing poorly on an exam, for example, students discussed how putting more effort into studying helped improve their grades. The video generally was followed by an activity that reinforced the importance and functionality of effort attributions.

In their intervention study, Hamm and colleagues (2014) were unable to detect an effect of their AR treatment on cumulative first-year GPA. Similarly, Haynes and colleagues (2008) were unable to find a direct relationship between AR and cumulative GPA; however, they detected an indirect relationship such that AR predicted mastery motivation, which in turn predicted GPA. Perry and colleagues (2010) found a significant treatment effect, with AR participants having higher GPAs compared with their no-AR counterparts (a difference of 0.26 grade points). In this case, AR participants also performed better on a postintervention class exam and earned significantly higher course grades relative to the no-AR group (effect sizes were in the moderate range for low- and average-performing students compared with controls). On the other hand, Hall and colleagues (2006) were unable to detect a statistically significant treatment effect for the final course grade. Finally, in a study of AR training, Ruthig and colleagues (2004) found a marginally significant treatment effect on cumulative GPA

and a statistically significant impact on voluntary course withdrawal (AR-trained students dropped fewer credit hours relative to those in the control condition). This finding is somewhat consistent with the finding of the Hamm and colleagues (2014) intervention study that among students characterized as failure acceptors, those in the AR treatment group were 61 percent less likely to drop a class relative to their peers in the no-AR condition.

In their intervention study, Eskreis-Winkler and colleagues (2016) sought to improve math performance among college students. The intervention was similar to prior AR interventions in that participants were taught that talent and effort both contribute to success. In this study, however, the intervention was designed to motivate participants to engage in deliberate practice (an especially effortful activity) and improve their achievement. Moderate treatment effects were found for end-of-year GPAs.

Yeager and colleagues (2016) report on an intervention study that took place after students had been admitted to college but before they had arrived on campus. The participants were primarily African American and first-generation college students, and their achievement outcomes were measured during their college years. In this study (Study 1 in Yeager et al., 2016) the authors tested two interventions—one focused on social belonging and the other on growth mindset. Both interventions were designed to teach students a lay theory for the transition to college. In essence, they provided students with a framework that helped them understand that challenges are common in the transition to college and that students often question their academic success and sense of belonging. They also taught students how to face those challenges effectively. The main outcome measured was full-time enrollment during students’ first year in college. While the growth mindset intervention showed no treatment effects, the social belonging intervention increased full-time enrollment over the next academic year among the African American and first-generation treatment group participants by 34 percent relative to the control group.

Taken together, the evidence on the effectiveness of growth mindset/AR interventions for college performance and persistence is promising. Six of seven studies that examined GPA as an outcome showed significant effects (effect size range = 0.38-0.86 SD); two of four studies showed significant effects on final course grade (effect size range = 0.57-0.78 SD); and four of four studies showed significant effects on course test performance (effect size range = 0.73-1.56 SD). Many impacts were limited to sample subgroups, such as underrepresented minorities and students at risk for academic failure. A growth mindset appears to influence motivation in a positive way, encouraging students to work harder even in the face of failure.

SPECIFIC MOTIVATIONS: WHAT ARE MY GOALS? WHAT MOST INTERESTS ME? WHAT DO I VALUE?

By the time they enter college, most students have some sense of why they are there. Some are attending college because doing otherwise is nearly unthinkable. Their parents, and perhaps even their grandparents, attended college and have always expected that they will follow suit. Other students come from families in which they are the first to attend college, and they may pursue college for an education and a degree that will allow them to attain employment and social and economic mobility not available to others in their families and communities. Many students hope that college will prepare them for gainful employment down the road. Some may even value education for its own sake, expressing a love for learning that may motivate a lifelong quest to expand and refine their minds. Still others have already developed a specific interest in a field of study. They know what their major is going to be, and they may even have a plan for what job or postgraduate studies they will pursue later on.

For many students, however, such goals, values, and areas of interest are fluid early on, and these students have only a vague understanding of their goals and values when they begin college. Moreover, those whose motivational agendas are more solidified at first may later change their aspirations dramatically. Many develop major and career interests they never knew they had. Many also confront intellectual and motivational challenges in the discipline they expected to pursue. Over time, students change majors; switch schools; and try out many different plans, projects, and programs. They are looking to identify their goals, values, and areas of interest and then to fulfill them. Questions concerning motivation thus may become more salient as students move through their college years: What are my goals? What most interests me? What do I value?

The research literature in developmental and social psychology and in higher education suggests that the specific ways in which students approach, work with, and seek to realize their goals constitute critical skills for success in college. Beyond broad dispositional strengths and particular beliefs about college, students’ motivations also contribute to their intrapersonal competencies.

Utility Goals and Values

According to expectancy-value theories of human motivation (Eccles, 2009; Eccles et al., 1983; see also Wigfield and Cambria, 2010), students will work hard to succeed when they (1) expect to succeed in (2) a valued goal domain. Whereas academic self-efficacy speaks directly to a student’s expectations of success, the extent to which the student values the particu-

lar goal area within which success may be expected is an important factor to consider as well. After all, a person may expect to succeed in a given goal domain but may not value the goal, and therefore may not pursue it with the vigor or commitment that is necessary for success. Such a student may say to herself, “I could succeed in Task A if I wanted to, but I don’t value Task A, and therefore I see no point in striving to achieve Task A.” Expectancy-value theories propose the common-sense idea that people do their best when they are striving to achieve goals they value. Of course, extrinsic values can be more immediate (e.g., valuing a goal because it will produce a good grade) or more distant (e.g., valuing graduation because it will lead to paid and sustained employment).

Immediate and distant values are clearly related, and understanding the utility of a goal is essential to appreciating value in all its forms. A student’s perception of the utility of coursework, for example, can be an important extrinsic payoff. For instance, achieving a high grade in a chemistry class may enhance the prospects for getting into medical school. The student may not value chemistry per se, but excelling in the coursework is a means to an end if she clearly values the goal of becoming a physician in the future. Similarly, a student may work hard to achieve a good grade in a boring introductory survey course because passing that course opens a gateway to other courses that are expected to be much more interesting. Another student may join a campus organization focused on addressing a particular societal problem. The student may not be especially interested in that problem itself, at least initially, yet may know that accumulating such experiences may pay off later on when he applies to law school. This same student may develop a real intrinsic interest in the societal problem, and may even change his career aspirations as a result. But in each of these hypothetical cases, the student is initially motivated to pursue a goal not because it is valued for its own sake but because it may facilitate attainment of the student’s long-term aims or long-range anticipated rewards.

The committee located 15 intervention studies that targeted utility goals or values. Each assessed at least one academic outcome, and only 5 assessed at least one competency outcome.

Harackiewicz and colleagues (2014) conducted a values affirmation intervention intended to close the achievement gap for first-generation college students studying STEM. Targeting students taking an introductory biology class, this intervention was delivered twice during the semester through a brief writing exercise in which students wrote about values most important to them. Students in the experimental group wrote about why two or three values selected from a list were important to them. By contrast, the control group students wrote about why values least important to them might be important to someone else. The effect of this type of intervention on GPAs was significant for first-generation students but not for “continuing-

generation” students. Further, the intervention promoted continued enrollment in the biology course sequence; first-generation students in the values affirmation group were more likely to enroll in the next biology course compared with continuing-generation students, a difference in enrollment of roughly 10 percent (75 versus 86%). They also were more likely to enroll in the next course compared with their counterparts in the control group—a difference in enrollment of roughly 20 percent (66 versus 86%).

In a subsequent study, Harackiewicz and colleagues (2015) implemented both a values affirmation intervention and a utility value intervention in an introductory college biology course. In both interventions, students wrote brief essays at three different times during the course. Those in the values affirmation treatment group wrote essays similar to those in the Harackiewicz et al. (2014) study, whereas those in the utility value treatment group were asked to select a concept or question relevant to the course and answer it, drawing on course material and explaining why this information was directly relevant or useful to their own lives. Although there were no significant effects for the values affirmation intervention, the utility value intervention significantly improved student performance relative to the control group. The impacts were largest in the case of 32 underrepresented minority students who were also first-generation students assigned to the treatment condition; their grades rose more than half a letter grade on average compared with their 32 counterparts in the control condition (slightly over 2.5 versus slightly under 2.1 on the GPA scale).

Brady and colleagues (2016) conducted a values affirmation intervention study focused on both achievement and competency outcomes, but here the sample consisted of 183 first- and second-year Latino and white college students. As with the Harackiewicz et al. (2015) intervention, the students completed a brief writing exercise in which those in the affirmation condition wrote about their most important value, while those in the control group wrote about a low-ranked value (ranked 9th out of 11) and why this value might be important to someone else. This study also was similar to that of Harackiewicz et al. (2014) in that the intervention’s effect size was positive and moderate—but only for Latino students. For Latino participants, GPAs collected 2 years after the end of the intervention were significantly higher for students in the affirmation condition than for their counterparts in the control group—on par with those of the white students in the control group. It appeared that the initial affirmation intervention, during the sensitive first year of college, helped Latino students see subsequent adversities in a more optimistic light and presumably helped them overcome those adversities, which further increased their confidence. Among white students, on the other hand, GPAs for those in the treatment group were significantly lower than GPAs for those in the control group.

The affirmation intervention also had a large positive impact (a standardized mean difference of 0.94) for Latino students on “adaptive adequacy”—a measure of self-integrity, self-esteem, and hope. On the other hand, the difference in adaptive adequacy between white treatment and control group participants was quite small and not statistically significant. A similar pattern emerged for academic belonging: Latino students in the affirmation condition felt a stronger sense of belonging in school compared with control group participants, whereas the difference between whites in the two conditions was not statistically significant.

Miyake and colleagues (2010) conducted a brief (15-minute) values affirmation writing intervention targeting students in an introductory physics course. Student performance was measured in two ways: with an overall exam score in the course (an average score of the percent correct for four exams) and with a Force and Motion Conceptual Examination (FMCE). As in other values affirmation studies, this intervention proved to be effective in reducing the achievement gap, but in this case only for women. Final course grades shifted from the C to the B range for women in the affirmation condition but not for women in the control condition, and there was no such final grade improvement for men. A similar pattern emerged for the FMCE score.

Martens and colleagues (2006) evaluated two values affirmation interventions designed to alleviate the performance gap between men and women due to stereotype threat. In both studies, women who had been exposed to stereotype threat were randomly assigned to a self-affirmation group or a no-affirmation control group. In both studies, students in the treatment condition participated in a standard affirmation exercise in which they wrote about their most important value. Women in the treatment condition performed statistically significantly better on a difficult math test (in Study 1) and a spatial rotation test (in Study 2) relative to the control group in each study. The authors do not present effect sizes for the two studies.

Turning from academic outcomes to competencies, a small number of studies demonstrate how these types of brief writing interventions impact the target competency. In a laboratory study, for example, Hulleman and colleagues (2010) addressed utility goals and values directly, as measured by situational interest in a new method for solving two-digit multiplication problems that was taught in the intervention. Before using this new technique to solve a series of problems, students participated in a writing activity that resembled the Harackiewicz et al. (2015) utility value intervention. Those in the relevance condition wrote about how the math activity could relate to their life or to the lives of college students in general. Control participants wrote about a topic unrelated to the math activity. Competency measures were assessed following treatment. Compared with those in the control condition, students in the relevance condition became more inter-

ested in the new math technique, and they indicated that they were more inclined to use the technique in the future.

Similar interventions were implemented and studied by Durik and colleagues (2015) and Schechter and colleagues (2010). Instead of participating in a relevant writing task, however, students were simply told about the usefulness of the new technique for their performance in future classes, preparation for graduate school admissions tests, and future careers. Treatment outcomes for situational interest differed depending on the type of student. In the Durik et al. (2015) study, for example, treatment students high in their initial perceived ability in math were more situationally interested in the math technique relative to the control group. Treatment students low in their initial perceived ability in math were less situationally interested in the math technique, relative to the control group. A similar pattern emerged when students were grouped by initial interest in math; however, treatment effects were not statistically significant. In the Schechter and colleagues (2010) intervention study, the intervention appeared to be particularly effective for East Asian learners with initially low math interest. This group showed more interest in the new math technique relative to Westerners.

Taken together, the evidence from interventions targeting utility goals and values is promising. Five of six studies show effects on final course grade (effect size range = 0.06-0.55 SD) and two of two studies show effects on college GPA (effect size range = 0.33-0.52 SD). Although many of these impacts are limited to sample subgroups such as first-generation college students, underrepresented minority groups, and women, these underrepresented student groups are most at risk of leaving STEM or leaving college altogether. The ability to recognize the utility or value in a given college activity (e.g., classwork) can potentially motivate a student to work harder on that activity even if the activity itself is not especially interesting, and even if the student is not dispositionally inclined to work hard in the first place (i.e., the student is not especially high in trait conscientiousness). However, further research is needed to explore this possibility and demonstrate more conclusively that the development of utility goals and values increases college persistence and success.

Intrinsic Goals and Interest

Another way in which a goal may be infused with value is if pursuing it is viewed to be intrinsically enjoyable or worthwhile and is valued for its own sake. The pursuit of intrinsically motivated academic goals involves the development of content knowledge (deeper learning, or sense making) and the development of interest in the discipline (see Renninger and Hidi, 2016).

A substantial body of research on the concept of intrinsic motivation suggests that pursuing goals perceived as rewarding or valuable in and of themselves is associated with many positive outcomes in life, such as higher levels of well-being and greater success (Deci and Ryan, 1991). Findings from neuroscience have established the association between interest, or seeking behaviors, and the reward circuitry of the brain (e.g., Gruber et al., 2014; Panksepp, 1998). This research provides evidence that once interest is developed, it becomes its own reward. Accordingly, those who do not have an interest are likely to need support to find learning rewarding, and once they receive such support (e.g., through interactions aimed at identifying meaning in content to be learned), their interest can be sustained and developed. Once interest begins to develop, moreover, it may promote the development of academic self-efficacy (e.g., Bong et al., 2015) and self-regulation (e.g., Sansone et al., 2015).

The development of interest (and its accompanying value) can be supported at any age and in any disciplinary context (see reviews in Hidi and Renninger, 2006; Renninger and Hidi, 2016). For the development of a new interest in college—say, mathematics—support from other people, such as mentors and instructors, often is required, as well as the design of engaging text, software, exhibits, and so on. In early phases of interest development, students may need to discover the utility or relevance of the content for their own lives.

Research on college students shows that infusing goals with intrinsic value is associated with more successful pursuit of those goals. Turner and colleagues (2009), for example, showed that both intrinsic motivation and academic self-efficacy predicted academic performance. Some studies have found mastery goals (typically seen as intrinsically rewarding) to be positively correlated with sustained interest in college subjects, as measured by course taking and choice of major (Harackiewicz et al., 2000, 2002). Harackiewicz and colleagues (2002) also found that early interest and final grade, measured in an introductory psychology course, were positively correlated with the likelihood of majoring in psychology. In an examination of interventions designed to broaden participation of underrepresented minority students in STEM disciplines, a group of investigators found that mastery goals (associated with intrinsic motivation) were correlated with higher cumulative GPAs (Hernandez et al., 2013).

Three intervention studies addressing intrinsic goals and interest met the committee’s inclusion criteria. One of these—Hamm et al. (2014), discussed above in the section on growth mindset—was designed to investigate the effects of a traditional attribution retraining intervention on at-risk students who were low in perceived control of their studies and either low or high in preoccupation with failure. The first-year students who participated in the study were classified, based on questionnaire

responses, into four different performance orientations: (1) failure-acceptors, who were low on both preoccupation with failure and perceived control; (2) failure-ruminators, who were high on preoccupation with failure and low on perceived control; (3) achievement-oriented students, who were low on preoccupation with failure and high on perceived control; and (4) over-strivers, who were high on both preoccupation with failure and perceived control. Intrinsic motivation was measured as one of three competency outcomes. Students in the treatment condition reported higher motivation relative to their control group peers, but this was the case only for the subset of students characterized as failure-acceptors and failure-ruminators.

An intervention study by Vansteenkiste and colleagues (2004b Study 1) provides possible insights into the process through which intrinsic motivation is developed and influences academic performance. In this study, 200 Belgian college students studying to become preschool teachers were randomly assigned to intrinsic and extrinsic motivation conditions before participating in a target activity. The intervention manipulation was embedded within the activity instructions for each condition. Following the intervention, the intrinsic motivation participants’ self-reported superficial processing scores (measuring the degree to which students had engaged with the material in a superficial way) were on average lower than those of the extrinsic motivation participants. Further, the intrinsic motivation participants’ deep processing scores (measuring the degree to which students engaged with the material deeply by questioning its underlying meaning and relating it to other concepts) were on average statistically significantly higher than those of the extrinsic motivation participants. This intervention study also measured achievement outcomes. The performance of intrinsically motivated students, measured by a written test of comprehension and group discussion, was significantly higher than that of their extrinsic group peers; additionally, they visited the library more often following treatment to learn more about the activity topic.

In summary, the evidence from intervention studies designed to foster intrinsic motivation and interest is modest. Two of two studies show significant effects on academic outcomes (measured through a written test of comprehension) by framing academic exercises in terms of intrinsic (versus extrinsic) goals (effect size range = 0.39-1.25 SD).

Further research is needed to examine whether development of these competencies is causally related to college persistence and success. Interest is a powerful motivator (Renninger and Hidi, 2011, 2016), and future interventions could potentially trigger and support the development of intrinsically motivated goals infused with interest in at least two ways. First, certain interventions might lead students to identify connections between new and important academic ideas and opportunities on the one hand and

what they already know on the other, encouraging them to develop their understanding further. Second, these interventions might encourage students to continue to pursue intrinsic goals and to develop and deepen their understanding of already existing interests. Research is needed to test such future interventions, as is more basic research on the processes that underlie the development of intrinsic motivation and interest.

Prosocial Goals and Values

Many parents, educators, and citizens believe that college should encourage students to formulate goals and values that are aimed at promoting the common good. One of the rationales traditionally invoked to justify a liberal arts education is the idea that a well-educated and well-rounded student will be prepared to make a positive contribution to society. In a related vein, recent years have witnessed a proliferation of civic engagement initiatives, service-learning opportunities, and other programs designed to stimulate and organize students’ prosocial motivations (see Chapter 5 for further discussion). These efforts provide new ways for students to become engaged in rich and meaningful college life. Cultivating prosocial goals and values also may promote college success, as indexed by such traditional metrics as grades and completion to degree.

In this light, prosocial goals and values are focused explicitly on promoting the well-being or development of other people, or of domains that transcend the self. Included under this broad rubric are student goals and values aimed, for example, at helping others, benefiting society, promoting a prosocial religious or political agenda, and contributing something positive to the next generation. A closely related construct may be what Wolniak and colleagues (2012) call “socially responsible leadership”—the extent to which “students possess values such as equity, social justice, self-knowledge, citizenship, and commitment towards social change” (p. 814).

Research in life-span developmental psychology suggests that prosocial goals tend to increase in salience from young adulthood through late midlife (Freund and Riediger, 2006). Evidence also suggests that valuing prosocial outcomes tends to be associated with positive psychological health and well-being in adulthood. Many college teachers and counselors contend that nurturing prosocial goals and values, including values linked to citizenship and cultural awareness, promotes a civically engaged and enlightened identity among college students.

To date, research suggesting that the cultivation of prosocial goals and values predicts college success is sparse. Still, Wolniak and colleagues (2012), analyzing data from the Wabash National Study of Liberal Arts Education, found that socially responsible leadership predicted persistence in college. Likewise, Yeager and colleagues (2014) found that holding a

self-transcendent purpose for learning was associated with higher levels of academic self-regulation among adolescents and college students.

The committee identified two intervention studies focused on prosocial goals and values that met its inclusion criteria. Both are reported by Yeager and colleagues (2014). In the authors’ Study 3, students assigned to a transcendent purpose condition and a control group completed an online exam review activity, which consisted of answering 100 multiple-choice questions. Students in the transcendent purpose condition reviewed self-transcendent purpose materials prior to the review. Online software tracked the amount of time students spent on each problem; this measure of time spent served as the main dependent variable indicating students’ intention to learn from the material. The results show that a transcendent purpose intervention can promote deeper learning behavior: students in the transcendent purpose condition spent twice as much time on each question compared with the control group, a difference that represents a moderate effect size.

Yeager and colleagues (2014) Study 4 included 429 students enrolled in a college-level psychology course; the intervention procedure was identical to that of Study 3. The goal of this study was threefold. First, it tested whether a purpose condition fostered deeper learning behavior in the face of desirable alternatives. As in Study 3, students reviewed a series of math problems; however, they were told they could quit any time and entertain themselves online. Second, the study examined whether a purpose intervention benefited students during the most uninteresting portion of a task. Indeed, students in the purpose condition completed 26 percent fewer problems during the most tedious portion of the diligence task (i.e., a set of math problems) than during the less tedious portion, whereas this differential was 44 percent for control participants. Lastly, the study examined whether self-oriented beliefs alone promoted academic self-regulation. To address this question, the study included another intervention group, one that emphasized self-oriented benefits (i.e., intrinsic motives for learning) as opposed to purpose beliefs. Self-regulation did not improve among the self-oriented intervention group compared with the control group as measured by the number of math problems solved.

In summary, these two interventions provide only modest evidence that the development of prosocial/transcendent goals and values is related to college persistence and success. Although both studies show significant, positive effects on deeper learning behavior, neither reports achievement impacts. Fostering such goals during the college years could potentially produce long-term benefits in the realms of personal happiness, fulfillment, and citizenship, as well as promote meaningful interpersonal relationships and group affiliations. However, whether developing these kinds of goals and values directly promotes success in college, as measured by grades and graduation rates, remains an open question.

Research is needed to examine in much greater detail how students’ prosocial goals and values develop in college, how they relate to intrinsic interest and more utilitarian goals, and the extent to which pursuing a prosocial or self-transcendent agenda in college may influence college persistence and academic performance. Moreover, the relationships and interactions that may be revealed in such research need to be understood in terms of students’ unique profiles of gender, race/ethnicity, class, and other characteristics. First-generation and underrepresented minority students, for example, may find strong motivation in a desire to give back to their communities of origin. If colleges could capitalize on this desire by explicitly connecting coursework and college activities to these kinds of prosocial goals and values, underrepresented student groups might achieve higher levels of college success.

IDENTITY: WHO AM I? WHOM DO I WANT TO BECOME?

Many theories in developmental psychology contend that the college years are prime time for the development of identity (Erikson, 1963; McAdams, 2015). In its broadest sense, identity refers to a person’s overall conception of self and the self’s relationship with the world. The central questions in identity development are Who am I? and Whom do I want to become? Identity questions thus are directed both backward and forward in time. The person aims to link the remembered past with an anticipated future, forging links among conceptions of past, present, and future selves. Identities typically refer to “the kind of person I see myself as being or becoming” or “the kind of life I hope to live.” Among other things, identity functions to provide a person’s life with a sense of unity, purpose, and coherence.

Do certain forms of identity promote success in college? Over the span of their college experience, students may begin to develop images, pictures, imagined trajectories, and narratives about the kind of person they hope to become in the future. Many social psychologists argue that these imagined future selves have motivational power (Markus and Nurius, 1986; Smith and Oyserman, 2015). Future selves may help organize and give meaning to a person’s goals, strivings, and values. They may impel the person to work harder to succeed in the field of study to which the future self may be connected. If a student articulates a clear image or story in her mind about becoming an engineer or a scientist, for example, she may find greater meaning and interest in her engineering or science classes and may work harder and perform at a higher level in those classes. She also may work harder to do well in college more generally and to complete it successfully so she can take the next step toward actualizing her imagined future self. Indeed, a student’s identification as a scientist, or with the broader scientific

community, has been cited as a significant predictor of persistence in STEM fields (Graham et al., 2014).

Positive Future Self

Developing a positive future self may increase the utility value of certain college activities and imbue those activities with clearer purpose and meaning. Many educators believe that linking academic work to a student’s developing image or story of whom he or she hopes to become in the future presents a powerful opportunity to promote psychosocial development and college success (Hanson, 2014; Markus and Nurius, 1986). With this in mind, Landau and colleagues (2014) designed and tested several interventions intended to encourage students to think of their academic life as a “journey.” They argue that the identity metaphor of life as a journey enabled the students to think more clearly and productively about how their current activities might lead to the realization of positive future selves.

The committee identified five intervention studies meeting its criteria that address the competency of positive future self. Each of these studies reports outcomes related to the target competency; only two, however, report academic outcomes. Three of the five interventions are found in the Landau and colleagues (2014) article described above. Their intervention addressed identity exploration by helping students make connections between their current identity and their possible academic identity. In all three studies, students in the treatment condition were primed to frame their academic career as a journey toward becoming an academically accomplished college graduate. Students in the comparison groups were primed to frame their academic careers differently (e.g., to see each college year as a separate container) and/or to imagine different future identities. In Study 3, following the treatment, students in the journey-framed group indicated that they planned to dedicate 23 percent more time to coursework relative to the container-framed group. Further, in this study, the journey-framed group earned significantly higher scores than those in the container-framed group on a final exam administered 1 week after the intervention.

Landau and colleagues (2014) also report competency outcomes (in Studies 1 and 2). In Study 1, journey-framed participants reported being more academically engaged and having more academic intentions relative to students in the comparison groups. In Study 2, journey-framed participants showed significantly higher academic engagement than students in the comparison groups, as measured by effort on an academic task (solvable math problems). However, the authors do not report the specific sizes of these effects.

In another intervention study meeting the committee’s inclusion criteria, Harrison and colleagues (2006) showed how academic conditions can

impede identity development. In this study, students in both treatment and control conditions were assigned to complete an academic test assessing math and verbal ability. Prior to the exam, treatment participants were exposed to stereotype threat; in particular, the exam instructions stated that the exam would be used in research to understand why lower-income students generally perform worse than their higher-income counterparts. Lower-income students in the treatment condition performed significantly worse than the control group on both the math and verbal portions of the exam. Relative to the control group, lower-income students in the treatment condition also indicated that they identified less with English and math subjects.

Taken together, the evidence from intervention studies on the competency of positive future self is modest. A single study showed a significant increase in student engagement with academic identity and subsequent significant improvement in exam performance (0.56 SD). Although the intervention results suggest that promoting positive future self may be able to improve students’ academic engagement and at least their short-run test performance, long-term effects on college persistence are not known.

Identity achievement may represent both a predictor of success in college and a valued outcome of undergraduate education. The process of developing a positive identity may draw on a range of competencies enumerated above, from deeply ingrained dispositional traits such as conscientiousness, to particular beliefs about college, to the specific motivational agenda—the goals, values, and interests—that a student formulates and begins to pursue over the course of college. Further research is needed to explore this potential.

Identity also is shaped by the exigencies of the college environment and by broad cultural forces that establish life expectations that may be contingent on gender, race, and class categories. In a sense, identity development is the culmination of a student’s educational experiences in college. In the best-case scenarios, the student has learned something important about who he or she was, is, and may someday become, and has embarked on a directed journey to become the positive future self now envisioned. Further research is needed to explore how to support the development of a positive identity during the college years.

CONCLUSIONS AND RECOMMENDATIONS

This chapter has reviewed the available research, using a developmental framework, to articulate the competencies that may be implicated in students’ college success.

Research Needs

Over the course of an extensive search for research evidence of possible relationships between intra- and interpersonal competencies and persistence and success in college, the committee found that the research base is limited, especially in the area of interpersonal competencies and college success. There were also particular gaps related to other important topics.

Conclusion: Only limited research has been conducted to date on the potential relationships between various intra- and interpersonal competencies and students’ college success. There are major gaps in the research evidence:

• Little research is available on the possible relationship between interpersonal competencies and students’ college success.
• The available research has been conducted almost entirely in 4-year institutions; very little experimental evidence is available on the possible relationship between intra- and interpersonal competencies and students’ success in community colleges.
• There is a paucity of evidence on the possible relationships between intra- and interpersonal competencies and the success of students intending to major in science, technology, engineering, and mathematics fields.

RECOMMENDATION 1: Federal agencies and foundations should invest in research exploring the possible relationships between various intra- and interpersonal competencies and students’ college success. To address gaps in the research base, these investments should include support for research examining

• how interpersonal competencies may be related to student success in 4-year colleges;
• how intra- and interpersonal competencies may be related to student success in community colleges; and
• how intra- and interpersonal competencies may be related to students’ success in 2- and 4-year science, technology, engineering, and mathematics programs and majors.

Promising Competencies

Based on its review of the limited available research, the committee identified promising competencies that appear to be related to college success. Correlational research suggests that, among the competencies re-

viewed by the committee, the most robust predictor of college success is dispositional conscientiousness—the tendency to be self-controlled, responsible to others, hardworking, persevering, rule-abiding, and achievement oriented. Conscientiousness is closely related to other constructs such as self-control, self-discipline, persistence, and grit. Indeed, scores on measures of conscientiousness are nearly as predictive of college success as are measures of general cognitive ability. Yet evidence supports conscientiousness as a deeply ingrained dispositional trait that is difficult to change, at least in the short term. Nevertheless, a few interventions have targeted specific behaviors associated with conscientiousness, to date yielding only a little evidence of effects on college success.

Conclusion: Beyond cognitive factors, correlational research has shown that individual differences in intrapersonal competencies predict college success and completion. These competencies include the broad personality trait of conscientiousness. Although an individual’s relative standing on conscientiousness tends to be highly stable over time, some interventions have successfully targeted task management and other specific manifestations of this trait.

Moving beyond dispositions and traits, the committee reviewed the available research on the relationships between other intra- and interpersonal competencies and college success. Through this process, the committee identified eight intrapersonal competencies. (Reflecting a lack of research evidence, this list includes no purely interpersonal competencies.) These eight competencies have been studied using a range of methods, and the committee identified them based on correlational and experimental research. However, the committee judged the strength of the evidence related to each competency based exclusively on research that has developed and tested them through interventions using random assignment.

Conclusion: The limited intervention studies conducted to date have generated promising evidence that the competencies of sense of belonging, growth mindset, and utility goals and values are related to college success and are malleable in response to interventions. Available intervention studies provide more modest evidence that five other competencies are similarly related to college success and malleable, yielding a total of eight identified competencies:

1. Behaviors related to conscientiousness—behaviors related to self-control, hard work, persistence, and achievement orientation.
2. Sense of belonging—a student’s sense that he or she belongs at a college, fits in well, and is socially integrated.

3. Academic self-efficacy—a student’s belief that he or she can succeed in academic tasks.
4. Growth mindset—a student’s belief that his or her own intelligence is not a fixed entity, but a malleable quality that can grow and improve.
5. Utility goals and values—personal goals and values that a student perceives to be directly linked to the achievement of a future, desired end.
6. Intrinsic goals and interest—personal goals that a student experiences as rewarding in and of themselves, linked to strong interest.
7. Prosocial goals and values—the desire to promote the well-being or development of other people or of domains that transcend the self.
8. Positive future self—a positive image or personal narrative constructed by a student to represent what kind of person he or she will be in the future.

Interventions that often required very little time and money to implement have helped students develop these eight competencies. Some of these interventions have been particularly effective for underrepresented student groups that are most at risk for academic failure.

Conclusion: Notably, evidence shows that low-cost interventions aimed at developing a sense of belonging, growth mindset, and utility goals and values have sometimes generated the largest benefits for underrepresented student groups that are most at risk for academic failure. Although encouraging, this evidence is limited and recent, and further research is needed to replicate and extend it.

RECOMMENDATION 2: Federal agencies and foundations should invest in intervention research and research employing a range of other methods to understand better the competencies identified in this report, their relationship to college success, and the mechanisms through which they operate to improve college success. Research focused on supporting the college success of underrepresented student groups should be a priority.

RECOMMENDATION 3: Colleges and universities should support the intervention research proposed in Recommendation 2 by facilitating the implementation and evaluation of random-assignment interventions, thereby gaining valuable information about their students and building the knowledge base on effective interventions needed to increase student retention and success.

Success in boosting the eight identified competencies will also require more research on their development and their role in undergraduate success generally, especially in the STEM disciplines. The findings in this regard to date are promising, but very limited. Research is needed on the interrelationships of these competencies with each other and with student success, both generally and for different student groups. For example, further research on sense of belonging could potentially find that it plays a fundamental role in supporting the development of academic self-efficacy, utility goals and values, and other competencies. Developing a robust body of research on the identified competencies will in turn require that funding agencies, journals, professional societies, and promotion and tenure committees all value and even prioritize replication studies. Studies should be conducted with adequate statistical power, and if they find a lack of statistical significance (e.g., null results), those results will also be important contributors to the base of scientific knowledge about these competencies. This commitment to rigorous research will allow institutions to know when an investment in developing these competencies is warranted.

Sensitivity to Context and Subgroup Effects

The committee’s review of the research indicates that issues of race, ethnicity, gender, social class, and culture need to be carefully considered when educators, administrators, researchers, and policy makers think about competencies and their contribution to college success. Certain competencies, and the problems they might help address or solve, may be more salient or useful for certain groups of students than others. Underrepresented minority students, for example, may bring to college such competencies as a strong racial or cultural identity that may help them navigate potentially unwelcoming academic or social environments. More research also is needed to examine how particular educational and cultural contexts currently influence the development of motivation, intrinsic goals and interest, and other intra- and interpersonal competencies among underrepresented groups (minority students, first-generation college students, students from low-income families, and women in certain STEM disciplines). Such research would inform efforts to improve educational and cultural contexts to ensure that they do not hinder, but rather support, the positive development of promising competencies among these groups. The need for such research is especially salient in the STEM fields, in support of the aim of enhancing diversity and promoting opportunities in these fields.

Conclusion: Certain competencies develop and function differently for different groups and within different cultural and educational contexts. For example, although a strong sense of belonging in college

is an important factor for success among underrepresented student groups, members of these groups may find it difficult to develop this competency if they experience campus environments that are discriminatory, negative, or unwelcoming.

RECOMMENDATION 4: To help reduce disparities in college success among student groups, institutions of higher education should evaluate and improve their social and learning environments to support the development of the eight identified competencies, especially among underrepresented student groups.