Attachment and Student Success During the Transition to College

Academic advising is intended to help students be successful in college and to prepare for their next phase in life (National Academic Advising Association, 2003). Therefore, academic advising professionals need to better understand the various factors that affect student success, especially at the critical time in which students enter college and experience a major life transition into adulthood. Because many models of academic advising are rooted in psychological theories (Frost, 1993; Hemwall & Trachte, 2005), we employ the psychological theory of attachment as a framework to better understand student success during the transition to college.

The impact of attachment affects many facets of one's life experiences; however, attachment takes on added significance during various life transitions, including when one begins formal schooling (Matas, Arend, & Sroufe, 1978), leaves the parental home (Bucx & Van Wel, 2008), marries (Davila, Karney, & Bradbury, 1999), and becomes a parent (Wilson, Rholes, Simpson, & Tran, 2007). While many of these transitions share similarities (e.g., psychological and social challenges), the move from high school to college comprises a unique experience as individuals approach adulthood, become more independent, experience changes in social structures, and move out of the parental home. How successfully students handle the transition to college can potentially affect their college success as well as the next transition, which begins at graduation. In this study, we examine college success using an attachment framework.

Attachment theory states that children develop either secure or insecure attachment styles based upon their early interactions with their caregiver (Bowlby, 1969). According to Bowlby (as noted in Bretherton, 1995), children develop specific mental models or schemas for the world and for themselves within the world. These internal working models encourage children to feel that their attachment figure(s) will (or will not) be available if needed, and if available, then they make them feel loved. The strength of the attachment styles was so strong and the effects so enduring that Bowlby believed that an infant's attachment to her or his mother started early in life and exerted lasting effects over the individual's life span (as cited in Waters, Merrick, Treboux, Crowell, & Albersheim, 2000) or as Bowlby stated “from cradle to grave” (Shear & Shair, 2005, p. 254).

Based on mother–infant interactions in the home during the first year of the child's life and in the laboratory procedure called “the strange situation,” Ainsworth, Blehar, Waters, and Wall (1978) distinguished several styles of infant behavior and labeled them secure, insecure avoidant, and insecure ambivalent. Later, other researchers (Hazan & Shaver, 1987) transformed these styles into adult-appropriate self-report patterns of attachment styles that respondents could use to characterize their own relationships with romantic partners. However, soon after these models were developed, Bartholomew and Horowitz (1991) reformulated the three styles of attachment into four, more closely matching Bowlby's perspective of positive and negative views of self and others: Secure attachment is based on positive views of self and others, fearful individuals reflect negative views of self and others, dismissives have a positive view of self and a negative view of others, and preoccupieds have a negative view of self and a positive view of others. More recently, Brennan, Clark, and Shaver (1998) identified two dimensions of attachment: (a) anxiety related to worry about being abandoned or rejected and (b) avoidance related to one's comfort with emotional closeness.

Study 1: Attachment and Transition to College

According to Bucx and Van Wel (2008), life course transitions give rise to fluctuations or changes in parent–child relationships, which are affected by attachment bonds. During major life transitions, individuals meet with new social and psychological challenges as well as stressors that must be negotiated. Larose and Bernier (2001) discussed three steps for dealing with life occurrences: primary appraisal, defined as “the process of categorizing an encounter and its various facets with respect to its significance for well-being” (p. 97); secondary appraisal, which refers to “the individual's evaluation of his or her resources to deal with a stressful situation (e.g., personal and environmental resources)” (p. 97); and coping, which refers to “the person's cognitive and behavioral efforts to manage demands that are appraised as taxing or exceeding his or her resources” (p. 98). As an individual experiences new life transitions, she or he must undertake further exploration that results in a need for reassurance that the secure base will be available. Additionally, depending upon attachment style, individuals deal with new stressors emerging from life transitions in differing ways; that is, a person may react by dismissing, ignoring, withdrawing from potential threats, or suppressing threat-related thoughts.

Over the course of life, an individual negotiates a number of transitional experiences. The first time a child leaves home for extended socialization experiences is considered a transitional event. Matas et al. (1978) found that, when compared to insecurely attached infants, securely attached children at the age of 18 months were more enthusiastic, cooperative, and better prepared for future social interactions as they transitioned into daycare; avoidant and ambivalent toddlers explored less and thus were less involved in the classroom setting. As they continue through school, students rely less on parental support and more on teacher support, showing an additional shift in their attachment system as they cope with further independence from their parents (Harter, 1996). Additionally, evidence shows that as children grow and develop, they express individually unique attachment styles that affect their development through puberty (Mikulincer & Shaver, 2007).

Adults face life transitions as they leave the parental home, marry, start a family, and so forth (Bucx & Van Wel, 2008). Davila et al. (1999) found that, during the onset of long-term romantic relationships, individuals become more secure as they move toward marriage and as their marriage develops. In terms of the transition to parenthood, Wilson et al. (2007) found that more anxious individuals tend to cope less effectively than do secure individuals. Additionally, as a new family emerges, secure spouses score higher on clinical ratings and self-reported measures of family functioning (Mikulincer & Shaver, 2007). As one's aging parents experience deteriorating health, an adult often must reverse roles from care receiver to caregiver (Bradley & Cafferty, 2001). Furthermore, with advancing age, secure individuals, characterized by positive models of self and others, report less anxiety about death (Besser & Priel, 2008).

While a variety of life transitions affect children and adults, some critical events in the lives of adolescents create especially challenging situations. For example, college students must adjust to a new social and academic environment while simultaneously growing more autonomous in managing their academics and navigating their personal lives (Larose, Bernier, & Tarabulsy, 2005, p. 282). The college years also may bring many unique challenges and stressors (Compas, Wagner, Slavin, & Vannatta, 1986), and attachment style can influence how students react to and are affected by the pressures of college. In some ways, the transition to college resembles the strange situation developed by Ainsworth et al. (1978) because college challenges an adolescent's adaptive strategies and coping mechanisms (Rice, Fitzgerald, Whaley, & Gibbs, 1995).

Higher education administrators face challenges in their charge to produce successful college graduates. Before the role of attachment can be examined in the postsecondary context, success measures must be defined. Most research on college student success focuses on academics. Svanum and Bigatti (2009) stated that college success has been defined as performance in a specific class or a collection of courses, college retention, or (less commonly) degree attainment. Lounsbury, Fisher, Levy, and Welsh (2009) stated that a student's cumulative grade-point average (GPA) is often viewed as the most important measure of college student performance. Other researchers suggest that college success encompasses more than just a high GPA. Wolf-Wendel, Ward, and Kinzie (2009) stressed the importance of student involvement, engagement, and integration into the campus as a whole. As discussed by Reich and Siegel (2002), life at a university offers students a social environment in which to experience intellectual, moral, and social-emotional exploration such that they will face new academic, psychological, and ethical experiences.

Secure individuals are better prepared than less secure peers for the academic (Aspelmeier & Kerns, 2003; Larose et al., 2005; Mikulincer & Shaver, 2007), psychological (Frey, Beesley, & Miller, 2006; Muris, Mayer, & Meesters, 2000; Wei, Russell, & Zakalik, 2005), ethical (Albert & Horowitz, 2009; Mikulincer & Goodman, 2006; Van Ijzendoorn & Zwart-Woudstra, 1995), and social challenges and rigors (Laible, 2007; Larose & Bernier, 2001) that they will encounter in college. Individuals unable to form or maintain secure attachment bonds face a more difficult transition, putting them at risk for attrition. To understand the effects of attachment, we recruited a group of incoming freshmen to participate in our study of the relationship between attachment and college success as measured by academic, psychological, ethical, and social indices.

Results

One of the major limitations of this study revolves around the sample size. Although 800 students were sent the request to participate in the survey, 85 responded for a return rate of approximately 10.5%. A further examination of the 800 potential student participants showed that, except in terms of gender, the sample of respondents may adequately represent the population as indicated by their similar demographics (see Table 1). Although the gender information does not closely follow that of the larger population, research in attachment indicates that gender is not a good predictor of attachment styles (Mikulincer & Shaver, 2007).

Table 1. Demographic comparison of sample and populations

View Fullscreen

To provide a well-rounded view of attachment theory influence on student success, we used both categorical (four paragraphs) and continuous (RS questionnaire) attachment assessments. When discussing levels of attachment anxiety and avoidance, we refer to continuous variables, but when discussing specific attachment styles (secure vs. insecure), we refer to categorical variables.

Attachment and academic success

We calculated correlations among attachment anxiety and attachment avoidance (as measured with the RS questionnaire) with various academic measurements, including four high school measures: final high school GPA as well as math SAT, verbal SAT, and total SAT scores (see Table 2). We found a significant relationship between high school GPA and attachment avoidance (r = .267, p = .021). In addition, we found a significant negative correlation between students' total SAT scores and attachment anxiety (r = −.261, p = .024). We also looked at correlations of attachment anxiety and avoidance with academic indices following the students' first semester at college: credits attempted, credits completed, GPA, and GPA credits (completed credits that count toward a student's GPA). These analyses did not result in any significant correlations. However, we conducted additional examinations using students' high school GPAs and their first-semester college GPAs. Students with relatively high scores of attachment anxiety performed worse academically in college than they had in high school (r = .312, p = .007). In addition, students showing high indications of attachment avoidance attempted more credits in their first semester compared to their peers with low scores of attachment avoidance: t(1,72) = 2.626, p = .011.

Table 2. Correlational measurements of attachment avoidance and anxiety and academic measurements, N = 74

View Fullscreen

Attachment and academic cheating

Students responded to the following questions: “How willing would you be to cheat to get a good grade?” and “How willing would you be to cheat to help a friend?” Students who scored high on attachment anxiety were more likely to agree to both statements: willing to cheat to get a good grade (r = .406, p = .000) and willing to cheat to help a friend (r = .298, p = .008).

Attachment and psychological health

We used four psychological measurements to analyze locus of control, anxiety, depression, and self-esteem (see Table 3). Results indicate that securely attached individuals were less anxious (t[1,64] = 2.383, p = .020) and less depressed (t[1,60] = −1.693, p = .096) than insecure individuals (see Figure 1).

Table 3. Correlational analysis of attachment avoidance and anxiety and psychological measurements, N = 67

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

Students who reported higher levels of attachment avoidance or attachment anxiety also scored higher on scales of depression: r = .552, p = .000 and r = .510, p = .000, respectively. Students who scored higher in attachment avoidance or in attachment anxiety scored lower on measures of academic locus of control: r = −.397, p = .001 and r = −.310, p = .008, respectively. Students who scored higher in attachment avoidance scored higher on items of overall general anxiety (r = .287, p = .020) as measured by the Beck Anxiety scale. Similarly, those who scored higher in attachment anxiety scored lower on measures of self-esteem: r = −.427, p = .000.

In addition, as previous research has shown (Mikulincer & Shaver, 2007), we found a strong relationship between attachment and psychological health as measured by drug and alcohol use. According to our results, students who reported higher attachment anxiety stated that they were more likely to use hard drugs (r = .460, p = .000) and light drugs (r = .324, p = .008) than those with lower attachment anxiety. Students with higher attachment anxiety also reported that they were more likely to “drink to get drunk” (r = .358, p = .003).

Locus of control was also highly correlated with a number of academic, psychological, and ethical variables. Academically, students with high scores on locus of control measures performed better academically in college compared to high school (as per GPA) (r = .372, p = .001), and they earned higher GPAs in their first semester at college (r = .396, p = .001) and completed more credits (r = .278, p = .017) than those who scored low on locus of control measures. Psychologically, students who scored high on locus of control also scored higher on levels of self-esteem (r = .482, p = .000) and lower on measurements of anxiety (r = −.289, p = .013) and depression (r = −.400, p = .001). Results of the study of ethical variables show that students who scored low on locus of control items also indicated that they would be more willing to cheat to get a good grade (r = .400, p = .000) and cheat to help a friend (r = .304, p = .009) than participants who scored high on locus of control statements.

We conducted a regression analysis to examine the influence of attachment anxiety and attachment avoidance on first-semester college GPA while accounting for locus of control, high school GPA, and gender. The results indicate that attachment anxiety exerts a significant effect on students' academic success during the transition from high school to college. This model accounts for 31% of the variance (see Figure 2).

View Figure

• Download as Powerpoint
• Download Figure

Attachment and socialization

Several questions asked if the participants were currently members of or were planning to join a fraternity or sorority or any other student group. Using correlational analysis, we found no significant correlations between student responses to involvement questions and measures of attachment anxiety or avoidance (see Table 4).

Table 4. Correlational analysis of attachment avoidance and anxiety and student involvement

View Fullscreen

Study 2: Attachment and Ethics

Study 1 shows that attachment affects various aspects of student success during transition. In Study 2, we further examined the relationship between attachment and ethical behavior at college. Through Study 1, we found a relationship between attachment and students' willingness to cheat. We designed Study 2 to look at unethical behavior within the college setting.

Ethical reasoning has been related to a number of factors, including age, gender, religion, value and belief systems, and moral character (Albert & Horowitz, 2009). Attachment theory provides a framework through which one looks at self and others and as such offers a unique perspective on ethical reasoning. Prior to formulating attachment theory, Bowlby (1944) reported on affectionless thieves, a group of adolescents who had grown up with one or more separations from their attachment figures and who subsequently developed into delinquents who lacked concern for others. In fact, so closely related is attachment theory to ethical behavior that Levy and Orlans (2000) suggested that children with attachment disorders show a propensity for typically defined unethical behaviors including lying, cheating, and stealing. It is theorized that individuals who are more insecurely attached will engage in more unethical reasoning as displayed through a variety of unethical behaviors involving moral reasoning, business practices, romantic fidelity, and academic integrity.

Although the literature on the relationship between attachment and moral reasoning is not extensive, it shows evidence supporting the role of attachment style. Moral reasoning dilemmas depend on many factors that parallel attachment theory. According to Kohlberg (1984) and Van Ijzendoorn and Zwart-Woudstra (1995), moral dilemmas depend on empathy and perspective taking as well as autonomy and trust in others. Reimer (2005) found that motivation to act morally lies within one's willingness to support and have confidence in one's self. In addition, Sims (2002) found that individuals who did not care much about others' wishes and expectations also tend to engage in unethical decision making. According to attachment theory, role playing, empathy, trust, and positive view of and confidence in oneself are all traits of security.

Van Ijzendoorn and Zwart-Woudstra (1995) surveyed college students with measures of attachment and moral reasoning. They found no correlation between moral reasoning and the three separate attachment styles (secure, ambivalent, avoidant) in terms of level of moral development. They theorized that their college sample was too homogeneous (i.e., most college students were within the conventional level of moral development). However, they looked at the differences between attachment styles as they relate to Kohlberg's (1984) framework of Type A (the need to act morally based on authoritarian mandates) and Type B (the need to act morally based on fairness and universality) sublevels, and they found that secure individuals were more likely to engage in Type B than Type A moral reasoning. They argued that Type B moral reasoning allows individuals to properly balance individual rights with societal demands; secure individuals balance individuals' rights between partners as well as consider the partnership itself.

According to Study 1, students with high attachment anxiety indicated that they would be more willing to cheat to get a good grade and to help a friend. Because students responded to a hypothetical question, we used Study 2 to examine behavior within a university setting in which all students who commit plagiarism are required, as part of their sanction, to enroll in a workshop on plagiarism. Attendees serve as an appropriate population because they were being held responsible for unethical behaviors.

Results

Using the ECR-R survey (a continuous measurement), we found that students from the plagiarism group were significantly more anxious (t[1,112] = 2.446, p = .016) but not more avoidant than the group analyzed in Study 1 (see Figure 3).

View Figure

• Download as Powerpoint
• Download Figure

General Discussion

Postsecondary academic advisors continually focus on assisting at-risk students to be more successful while transitioning to and through the college experience. The results of Studies 1 and 2 show that attachment security is related to the academic, psychological, and ethical success of college students. Although several different models of advising (Walsh, 2003) offer appropriate means for working with at-risk students, intrusive advising proves the best for insecurely attached students (Heisserer & Parette, 2002).

According to Varney (2007, p. 11), “Intrusive advising involves intentional contact with students with the goal of developing a caring and beneficial relationship that leads to increased academic motivation and persistence.” This relationship could not only serve as a one-stop resource for students, but it can dually serve as a secure base for students with attachment needs.

As previously stated, insecure and anxious individuals are troubled by rejection and abandonment and tend to be unsure about whether or not their attachment figure will be attentive and responsive. Therefore, an academic advisor's primary goal involves establishing a sense of availability. The ability to decipher a student's needs, whether through academic support, psychological assistance, or as a sounding board, comprise crucial skills for assisting students with an anxious and insecure attachment style. Advisors must reciprocate this student's needs and make him or her feel good about collegiate pursuits.

In times of need, insecure avoidant individuals tend to pull back and disengage from others. They have developed schemas that indicate feelings of mistrust such that they only comfortably rely on themselves. To work with these students, an academic advisor must establish trust through consistent communications and maintained availability. An avoidant student benefits from being able to contact the advisor via different means and modes (e-mail, social media, and in-person, etc.) when needed as well as from fulfilled expectations for a reasonable response to queries. In addition, an advisor may need to undertake intrusive and persistent means when working with an insecurely avoidant student, who will not quickly reach out or share concerns. However, in this case, the advisor needs to determine when to provide greater psychological space for the comfort of the avoidant advisee.

In addition to giving advisors a better idea about ways to work with students, attachment theory also supports the administrative advising model in which advisors work with a specific group of advisees from enrollment through graduation. This delivery method allows advisors and advisees to build bonds from day one and better ensures that advisors offer consistent communication in line with their personal knowledge of students' needs.

As a result of the formation of a caring and nurturing relationship, an advisor serves as a secure base that a student feels comfortable accessing in a time of need. By having this relationship with advisees, the advisor demonstrates care about them and their success, which engenders in students a more positive view of self, and by proving a positive support and source of encouragement, advisors help advisees develop a more positive view of others. With the implementation of an intrusive advising model, advisors can develop a close, trusting relationship in which students feel the kind of care that moves them toward a more secure attachment style (Mikulincer & Shaver, 2007) and thus a more successful academic career (Heisserer & Parette, 2002).