Preliminary Validation of a Short-Form Version of the Attachment Style Questionnaire for Use in Clinical Rehabilitation Counseling Research and Practice

Abstract

The purpose of this study was to develop and validate a short-form version of the Attachment Style Questionnaire (ASQ) in people with disabilities. The construction sample consisted of 108 participants recruited from spinal cord injury (SCI) advocacy organizations. The cross-validation sample comprised 140 individuals with traumatic injuries recruited from a rehabilitation hospital. Measures administered were the ASQ, Trait Hope Scale, Sense of Coherence Scale, and Satisfaction With Life Scale. Results showed that the three subscales of secure, anxious, and avoidant attachment from the short-form ASQ had high correlations with the three subscales from the long-form ASQ. The reliability of the subscales for the short-form ASQ was adequate and similar to the long-form ASQ. Both the short- and long-form ASQ subscales were found to correlate with hope, sense of coherence, and subjective well-being in the expected theoretical directions. Confirmatory factor analysis also supported the three-factor measurement structure of the short-form ASQ. This study provides evidence to support the psychometric properties of the abbreviated ASQ in people with disabilities. The short-form version of the ASQ is a brief, reliable, and psychometrically sound measure of attachment that can be used in clinical rehabilitation counseling research and practice.

Keywords

attachment positive psychology rehabilitation disability spinal cord injury short-form measure

Attachment theory postulates that healthy relationships between infants and their early primary caregivers are essential for the development of adaptive psychosocial competencies and formation of healthy relationships during adulthood. Accordingly, attachment profiles formed during infancy and early childhood carry over into adulthood and across the life span to influence social skills and functioning, personality features, and well-being (Bowlby, 1988). Attachment theory is among the most thoroughly investigated psychological theories, with empirical findings indicating that attachment profiles are predictive of a broad range of outcomes (Barbaro, Boutwell, Barnes, & Shackelford, 2017). The literature documents a number of instruments for assessing attachment and related constructs, reflecting the depth of attachment-focused research (Ravitz, Maunder, Hunter, Sthankiya, & Lancee, 2010). The Attachment Style Questionnaire (ASQ; Feeney, Noller, & Hanrahan, 1994) has been used to investigate relationships between attachment and rehabilitation-relevant outcomes, including coping, social support, depression, anxiety, pain, and employment outcomes (Blake, Brooks, Greenbaum, & Chan, 2017; Ravitz et al., 2010). The ASQ is a 40-item instrument that can be used to assess adult attachment according to the three fundamental attachment domains (i.e., secure attachment, anxious attachment, and avoidant attachment) identified by Ainsworth, Blehar, Waters, and Wall (1978) and Hazan and Shaver (1987). While the psychometric properties of the ASQ have been supported by recent rehabilitation counseling research, consideration of respondent burden for individuals with disabilities, and time and resource restrictions inherent in clinical and research settings, suggest that the 40-item measure may be too lengthy (Blake et al., 2017; Widaman, Little, Preacher, & Sawalani, 2011).

The goal of the current study was to develop a short-form version of the ASQ comprising four exemplary items for each of the three ASQ subscales (i.e., secure attachment, anxious attachment, and avoidant attachment). To this end, we followed Stanton, Sinar, Balzer, and Smith’s (2002) recommendations for developing short-form scales from existing self-report instruments, which integrates the most common methods for constructing short-form measures identified by Widaman et al. (2011). In keeping with Stanton et al.’s (2002) guidance, short-form item selection was guided by four separate analyses: (a) principal components analysis was used to evaluate individual item loading on respective factors (i.e., secure attachment, anxious attachment, and avoidant attachment); (b) correlational analysis was used to evaluate bivariate relationships between individual item scores and respective ASQ subscale scores (i.e., item-total correlation), and relationships between individual item scores and external, attachment-related constructs (i.e., convergent validity); (c) results from reliability analysis were used to evaluate individual item effects on internal consistency of respective subscales (Cronbach’s alpha); and (d) analysis of content validity ratings—provided by a panel comprised of the researchers—was used to evaluate level of agreement among authors regarding items included in short-form instrument.

Attachment and Rehabilitation Counseling

In rehabilitation counseling, adult attachment profiles can be viewed as protective or vulnerability factors that influence overall adjustment for individuals with chronic illness and disability (Blake et al., 2017; Umucu et al., 2016; Wilson et al., 2013). Attachment profiles are relatively stable across the life span, and affect a range of psychosocial factors that are important for adaptation to illness and disability, including social functioning, coping skills, relationship patterns, social support, and employment status (Blake et al., 2017; Bucci, Roberts, Danquah, & Berry, 2005; Kumpfer, 1999; Schmidt, Nachtigall, Wuethrich-Martone, & Strauss, 2002; Snyder, 2000; Umucu et al., 2016; Wilson et al., 2013). Several authors have suggested that insecure adult attachment profiles may have the potential to magnify effects of chronic illness or disability through psychological traits such as perfectionism and self-criticism (Ciechanowski, Walker, Katon, & Russo, 2002; Wei, Mallinckrodt, Russell, & Abraham, 2004; Wilson et al., 2013). Conversely, secure adult attachment profiles may shield people with chronic illness and disability from co-occurring psychological distress. Likewise, secure attachment is also a strong predictor of subjective well-being (SWB; Wilson et al., 2013). It is crucial to note that while chronic illness and disability themselves do not cause insecure attachment, persons with congenital disabilities or disabilities acquired during infancy or early childhood may be uniquely vulnerable to developing an insecure attachment profile, as parent–child interactions may be affected by parental stress and disability-related difficulties in communication (Howe, 2006). Given the potential for early childhood attachment experiences to influence adult psychosocial functioning and high rates of insecure attachment among people with disabilities (Howe, 2006), rehabilitation counseling professionals should attend to attachment issues in both research and practice to inform the selection of appropriate psychosocial interventions for individuals with chronic illness and disability.

Related Constructs

Convergent validity and discriminant validity refer to the degree to which empirical relationships between theoretically related measures are consistent with theoretical relationships between the measures’ respective constructs. Relative to the development of short-form scales from existing instruments, an important goal for instrument developers is that relationships between short-form scales scores and external, theoretically related constructs, are quantitatively similar to relationships between original (long-form) scales scores and external, theoretically related constructs (Stanton et al., 2002). Therefore, for the current study, short-form item selection was guided in part by examination of bivariate relationships between participants’ scale scores on individual ASQ items, and participants’ scale scores on three instruments designed to assess constructs with known relationships to attachment: (a) sense of coherence (SOC; Antonovsky, 1979), (b) hope (Snyder, 1994), (c) SWB (Diener, Emmons, Larsen, & Griffin, 1985), and (d) depression and anxiety (Avagianou & Zafiropoulou, 2008).

Sense of Coherence

Sense of coherence refers to individuals’ self-perception of (a) support or obstruction inherent in social contexts, (b) availability of necessary material and social resources, and (c) meaning and comprehensibility in their surroundings (Antonovsky, 1987; Eriksson & Lindström, 2005; Klepp, Mastekaasa, Sørensen, Sandanger, & Kleiner, 2007). Empirical findings provide support for SOC as a social/environmental construct that accounts for factors such as family dynamics, social support, and other social resources (Al-yagon, 2012; Drageset et al., 2009; Han, 2005; Ying, Lee, & Tsai, 2007). Along these lines, results from several studies have indicated significant relationships between attachment and SOC. Al-yagon (2012) found that higher levels of SOC were significantly related to higher levels of attachment security for children with learning disabilities (LD) and typically developing peers. Drageset et al. (2009) found a significant relationship between SOC and attachment for nursing home residents. Han (2005) found that SOC was significantly related to attachment for Southeast Asian American college students who were born to refugee families. Ying et al. (2007) found that SOC was predicted by attachment, with low levels of SOC predicted by poor early attachment relationships.

Hope

Snyder’s hope theory describes a cognitive/affective model for understanding motivation and goal-achievement behavior. Hope theory is centrally informed by attachment theory and a number of studies provide empirical support for theoretical relationships between hope and attachment (Snyder, 1994). Hope was found to be significantly related to attachment for U.S. middle school students (Jiang, Huebner, & Hills, 2013) and secure, avoidant, and anxious attachment for U.S. college students (Shorey, Snyder, Yang, & Lewin, 2003). In addition, Simmons, Nelson, and Quick (2003) found that hope was positively related to secure attachment, and that hope was negatively related to anxious and avoidant attachment for home health care nurses in the United States. Simmons, Gooty, Nelson, and Little (2009) found that hope was positively correlated to secure attachment for assisted living center staff in the United States. Blake et al. (2017) found that hope was significantly related to secure attachment, anxious attachment, and avoidant attachment among a participant sample comprised of individuals with spinal cord injury (SCI).

Subjective Well-Being

SWB (Diener, 1984) is the affective and cognitive self-appraisal of emotional processes and states of being—or happiness—in the broad context of social relationships, interests, demographics, mental health, personality, temperament, and adaptation style (D. G. Myers, 2000; Peterson, 2000; Taylor, Kemeny, Reed, Bower, & Gruenewald, 2000). A number of studies have provided empirical support for a predictable relationship between SWB and attachment. Jiang et al. (2013) found a significant positive relationship between SWB and attachment among a sample of typically developing middle school students. Gnilka, Ashby, and Noble (2013) found negative relationships for SWB and anxious attachment, and SWB and avoidant attachment for U.S. undergraduate students. Lavy and Littman-Ovadia (2011) found that SWB was negatively related to avoidant attachment and anxious attachment among a community sample comprised of Jewish Israeli individuals.

Depression

Bowlby’s model of attachment indicates that insufficient bonding with caregivers during infancy and childhood contributes to development of psychopathology in adulthood. Likewise, a body of empirical findings provides support for a relationship between adult attachment profiles and depression (Avagianou & Zafiropoulou, 2008). Furthermore, within the rehabilitation literature, healthy attachment profiles have been identified as a protective factor against depression for individuals with disabilities (Wilson et al., 2013). In an investigation of adult attachment profiles and depression, Gnilka et al. (2013) found that higher levels of attachment anxiety and attachment avoidance predicted higher levels of depression in a nonclinical sample of 180 undergraduate college students. Kenny and Sirin (2006) found that levels of attachment health predicted significant variance in levels depression in a participant sample comprising 81 young adults and their mothers.

Avagianou and Zafiropoulou’s (2008) findings on the relationship between depression and parental bonding among a nonclinical participant sample of 103 adults and 78 students (N = 181) indicated that early caregiver relationships that are overprotective or lack sufficient parental are linked to increased levels of depressive symptoms across the life span, while optimal relationships with early caregivers are related to fewer depressive symptoms. In a longitudinal study focused on parental bonding as a predictor of severe depressive symptoms, Grotmol and colleagues (2010) found that parental bonding, specifically low levels of maternal care, predicted severe depression symptoms for a cohort of Norwegian physicians (N = 265). Results from Sim and Yow’s (2011) investigation of relationships between attachment and psychological adjustment in a nonclinical sample of 236 Singaporean adolescents included a significant inverse relationship between attachment health and individuals’ levels of depression. In a study investigating a predictive model of mental health for college students, Shorey et al. (2003) found that secure attachment was negatively related to depression, whereas avoidant and anxious attachment were positively related to depression.

Measurement Issues and Study Purpose

Early research on attachment structure conceptualized the three primary attachment domains (i.e., secure, anxious, avoidant) as mutually exclusive, categorical types, or attachment styles (e.g., Ainsworth et al., 1978; Hazan & Shaver, 1987). More recent findings support models that conceptualize attachment domains as nonmutually exclusive, continuous dimensions, with a preponderance of current attachment measurement research indicating that dimensional measures yield accounts of interpersonal and intrapersonal functioning that are more precise and have greater utility than categorical measures (Blake & Norton, 2014; Fraley, Hudson, Heffernan, & Segal, 2015). Feeney and colleagues (1994) developed the ASQ to provide a comprehensive measure that operationalized the strengths of a continuous, dimensional conceptualization of attachment domains (Feeney et al., 1994). Although ASQ validity is well supported, the instrument’s length may not be ideal for rehabilitation counseling and rehabilitation research (Blake et al., 2017). Lengthier scales indeed possess value in comprehensively evaluating multiple subconstructs through a single measure; however, some researchers have suggested that longer scales do not always guarantee the quality and accuracy of responses. Furthermore, researchers in educational and psychological measurements have suggested that it is possible to scientifically reduce the length of existing scales while maintaining their psychometric properties (Stanton et al., 2002; Widaman et al., 2011). Thus, this article describes the development and validation for an abbreviated version of the ASQ. The availability of a short-form version of the ASQ may increase the utility of the scale in both research and clinical rehabilitation counseling practice by lowering time- and cost-burden, as well as decreasing the chances of respondent fatigue.

Study 1: Construction of the Short-Form ASQ

Method

Participants (Sample A)

One hundred eighty-five surveys were attempted with 108 completed (N = 108). They were recruited from SCI advocacy organizations including the National Spinal Cord Injury Association (NSCIA), Paralyzed Veterans of America (PVA), Spinal Cord Injury Network–United Kingdom (SCIN), and the Canadian Paraplegic Association (CPA). Participant ages were between 21 and 69 years (M = 49.84; SD = 11.84). The majority of participants were male (68.5%) and White (87%). Most participants completed high school (93.5%), and the majority had at least some postsecondary education (80.5%). According to the National SCI Statistics Center (2016), 51.5% of persons with SCI are high school graduates and 10.6% with a college education or higher at time of injury. The majority of participants in our study had postsecondary education; therefore, their educational attainment may be higher than the average level of people with disabilities including people with SCI. Thirty participants were employed full-time (27.8%), 12 were employed part-time (11.1%), and 62 participants were unemployed (57.4%). Participants reported more partial injuries (56%) than complete injuries (49%). Forty-four participants reported cervical level injuries, 44 reported thoracic level injuries, and 19 reported lumbar injuries.

Measures

Attachment Style Questionnaire (ASQ)

The ASQ is a 40-item, self-report measure comprising three subscales: (a) secure attachment, (b) avoidant attachment, and (c) anxious attachment (Feeney et al., 1994). Scale scores are consistent with the attachment domains reported by Hazan and Shaver (1987). Respondents rate each item using a 6-point Likert-type agreement scale ranging from 1 (totally disagree) to 6 (totally agree). Validity is supported by intercorrelations among scales. Secure and avoidant were negatively correlated (r = −.49), secure and anxious were negatively correlated (r = −.29), and avoidant and anxious were positively correlated (r = .35). Negative relationships between secure attachment and the insecure domains (i.e., avoidant, anxious) provided additional support for validity. Secure attachment scores also correlated with family intimacy, democratic parenting styles, and fewer instances of within-family conflict. Anxious attachment correlated with neuroticism. High secure attachment and lower avoidant attachment were related to extroversion. Test–retest reliability coefficients over 10 weeks were between .74 and .80. Internal consistency reliability coefficients (Cronbach’s alpha) ranged from .83 to .85 (Feeney et al., 1994).

Trait Hope Scale (THS)

The THS was developed by Snyder et al. (1991) to measure dispositional hope. It is composed of eight items and two subscales: (a) agency thinking, four items (e.g., “I energetically pursue my goals”) and (b) pathways thinking, four items (e.g., “There are lots of ways around any problem”). Each item is rated on a 4-point Likert-type agreement scale, ranging from 1 (definitely false) to 4 (definitely true). Validity support for THS scores has been based on association between the THS and optimism (Scheier & Carver, 1985), expectancy for attaining goals (Fibel & Hale, 1978), and self-esteem (Rosenberg, 1965). Cronbach’s alpha ranged from .70 to .84 for the agency thinking scale and from .63 to .86 for the pathways thinking scale.

Sense of Coherence Scale (SOC-29)

The SOC-29 assesses the SOC construct defined as the ability to cope, solve problems, and engage in healthy behaviors (Antonovsky, 1987). It is composed of three components: (a) manageability, (b) comprehensibility, and (c) meaningfulness. Respondents rate each item using a 7-point Likert-type agreement scale ranging from 1 (never have this feeling) to 7 (always have this feeling). In the present study, an abbreviated version of the SOC-29 (SOC-13) was used to operationalize the SOC factors. Cronbach’s alphas ranged from .83 to .95 for the SOC-29, and from .74 to .91 for the SOC-13. Several studies supported criterion validity for the SOC-29 and the SOC-13 as a measure of perceived stress, global self-orientation, health, and behavior and attitudes (Feldt, Leskinen, Kinnunen, & Ruoppila, 2003; Feldt et al., 2007; Gana & Garnier, 2001; Lin et al., 2009; Zimprich, Allemand, & Hornung, 2006).

Satisfaction With Life Scale (SWLS)

The SWLS (Diener et al., 1985) was used to assess SWB. The SWLS is composed of five items (e.g., “In most ways my life is close to my ideal”); each item is rated on a 7-point Likert-type agreement scale ranging from 1 (strongly disagree) to 7 (strongly agree). The instrument yields an assessment of individuals’ comparative self-appraisal of well-being against a self-constructed standard, relative to life circumstances in which higher scores (ranging from 5 to 35) indicate higher levels of SWB. Cronbach’s alphas were reported as ranging from .61 to .81. Reported test–retest reliability for the SWLS was .82, and the scale is associated with SWB measures including the Fordyce’s (1988) Happiness Scale, Rosenberg’s (1965) Self-Esteem Scale, and Bradburn’s (1969) Affect Balance Scale (Park, Peterson, & Seligman, 2004; Pavot, Diener, Colvin, & Sandvik, 1991).

Procedures

Following Institutional Review Board (IRB) approval from the University of Wisconsin–Madison, study announcements were posted on NSCIA, PVA, SCIN, and CPA websites and social media pages. In addition, announcements were forwarded by email, and published in organization newsletters. Individuals who wished to participate accessed a web-link to complete the research packet online. No compensation or incentives were offered or provided to participants and no identifying information was collected.

Data analysis

The goal of this study was to develop a short-form version of the ASQ by selecting four exemplary items for each of the ASQ subscales. Stanton and colleagues (2002) recommended the consideration of three salient qualities for evaluating items to shorten scales, including (a) internal item qualities, (b) external item qualities, and (c) judgmental item qualities. Internal item qualities refer to properties of items that can be assessed in reference to other items on the scale. External item qualities refer to connections between the scale and other constructs or indicators. Judgmental item qualities refer to issues that require subjective judgment (e.g., content validity rating). Based on the recommendations of Stanton et al. (2002), we selected items for each ASQ subscale by evaluating (a) each item’s factor loading, (b) item-total correlation, (c) the effect of the item on the internal consistency reliability (Cronbach’s alpha) of the subscale, and (d) correlations with attachment-related constructs (i.e., hope, SOC, and SWB), and (e) average content validity rating. Stanton et al. (2002) also indicated that if there are conflicts among the criteria, researchers should choose the “best” subset of items to ensure the resulting shortened scales preserve the optimal characteristics of the full-length scale. Principal components analysis was used to compute the factor loadings for each ASQ subscale; item analysis was used to compute the item-total correlation coefficients and Cronbach’s alpha if the item was removed; Pearson product–moment correlation coefficients were computed for the relationship between ASQ items and the external correlates; and content validity indexes were computed by averaging the ratings of the five researchers in this study who are knowledgeable about Bowlby’s (1988) attachment theory, using a 3-point Likert-type rating scale (0 = poor fit with the construct, 1 = adequate fit with the construct, and 2 = excellent fit with the construct).

Results

Table 1 presents the means, standard deviations, and the correlations among the short- and long-form ASQ subscales for Sample A.

Table 1.

Correlations, Means, and Standard Deviations for the Short- and Long-Form ASQ Subscales in Sample A.

ASQ subscales	1	2	3	4	5	6
Short-form ASQ
1. Secure attachment	—
2. Anxious attachment	−.63**	—
3. Avoidant attachment	−.59**	50**	—
Long-form ASQ
4. Secure attachment	.93**	−.56**	−.62**	—
5. Anxious attachment	−.60**	.88**	.58**	−.53**	—
6. Avoidant attachment	−.63**	.49**	.86**	−.70**	.58**	—
Mean	4.33	3.03	3.08	4.35	2.97	3.12
Standard Deviation	0.92	1.05	0.99	0.72	0.75	0.69

Note. ASQ = Attachment Style Questionnaire.

p < .01.

The average rating for the short-form secure attachment subscale was 4.33 (SD = 0.92), which was close to the “slightly agree” rating. The average ratings for the anxious attachment and avoidant attachment were 3.03 (SD = 1.05) and 3.08 (SD = 0.99), respectively, and these ratings were close to the “slightly disagree” rating. The short-form secure attachment subscale was found to be negatively associated with the short-form anxious attachment subscale (r = −.63, p < .01) and the short-form avoidant attachment subscale (r = −.59, p < .01). The short-form anxious attachment subscale was positively associated with the short-form avoidant attachment subscale (r = .49, p < .01). These results were similar to the mean ratings and correlation coefficients for the long-form ASQ subscales.

Table 2 presents characteristics of the items based on factor analysis, item analysis, and correlational analysis. Table 3 presents internal consistency reliability coefficients and Pearson product–moment correlations between the short- and long-form ASQ and hope, SOC, SWB, and depression for both Samples A and B.

Table 2.

Statistical Indices for Selecting Items for the Short-Form Version of the Attachment Style Questionnaire.

Item	Factor loading	Item-total correlation	Alpha if item deleted	Hope	Sense of coherence	Subjective well-being	Content validity
Secure Attachment subscale
1. Overall I am a worthwhile person.	.59	.46	.75	.37	.37	.30	1.0
2. I am easier to get to know than most people.	.63	.44	.75	.31	.15	.19	1.0
3. I feel confident that other people will be there for me when I need them.^a	.64	.49	.74	.32	.45	.49	2.0
19. I find it relatively easy to get close to other people.^a	.74	.58	.72	.23	.26	.24	2.0
31. I feel confident about relating to others.^a	.71	.55	.73	.36	.43	.22	2.0
33. I often worry that I do not really fit in with other people. (R)	.65	.50	.74	.26	.54	.25	1.4
37. If something is bothering me, others are generally aware and concerned.	.21	.15	.80	−.03	.07	.23	0.6
38. I am confident that other people will like and respect me.^a	.79	.68	.71	.30	.48	.25	1.4
Anxious Attachment subscale
11. It’s important that others like me.	.17	.18	.83	−.03	.00	.03	1.2
12. It’s important to me to avoid doing things that others won’t like.	.35	.33	.82	−.18	−.10	−.03	0.6
13. I find it hard to make a decision unless I know what other people think.	.31	.28	.83	−22	−.15	−.01	0.8
15. Sometimes I think I am no good at all.	.55	.44	.82	−.41	−.43	−.35	1.2
18. I find that others are reluctant to get as close as I would like.^a	.75	.64	.80	−.35	−.55	−.36	1.8
22. I worry that others won’t care about me as much as I care about them.^a	.63	.50	.81	−.37	−.43	−.34	1.6
24. I worry that I won’t measure up to other people.	.67	.58	.81	−.19	−.40	.05	1.4
27. I wonder why people would want to be involved with me.	.62	.46	.82	−.31	−.50	−.34	1.0
28. It’s very important to me to have a close relationship.	.29	.26	.83	.01	−.04	.10	1.0
29. I worry a lot about my relationships.^a	.73	.67	.80	−.19	−.39	−.24	2.0
30. I wonder how I would cope without someone to love me.	.47	.38	.82	−.08	−.23	−.07	1.2
32. I often feel left out or alone.^a	.75	.62	.80	−.26	−.65	−.45	1.8
35. When I talk over my problems with others, I generally feel ashamed or foolish.	.65	.52	.81	−.39	−.66	−.36	1.4
39. I get frustrated when others are not available when I need them.	.56	.49	.81	−.24	−.41	−.16	1.2
40. Other people often disappoint me.	.44	.29	.83	−.26	−.46	−.38	1.0
Avoidant Attachment subscale
4. I prefer to depend on myself rather than other people.	.11	.08	.85	.28	.10	.11	1.6
5. I prefer to keep to myself.^a	.46	.38	.84	−.26	−.31	−.28	1.4
6. To ask for help is to admit that you’re a failure.	.59	.51	.83	−.44	−.36	−.27	1.4
7. People’s worth should be judged by what they achieve.	.30	.25	.85	−.10	−.06	.05	0.8
8. Achieving things is more important than building relationships.	.48	.43	.84	−.18	−.12	.00	1.2
9. Doing your best is more important than getting on with others	.48	.41	.84	−.12	−.20	−.09	1.2
10. If you’ve got a job to do, you should do it no matter who gets hurt.	.51	.45	.84	−.12	−.16	.16	1.0
14. My relationships with others are generally superficial.	.52	.44	.84	−.18	−.29	−.32	1.2
16. I find it hard to trust other people.^a	.70	.61	.83	−.31	−.50	−.19	1.4
17. I find it difficult to depend on others.	.64	.55	.83	−.21	−.43	−.24	1.8
20. I find it easy to trust others. (R)	.56	.46	.84	−.25	−.39	−.26	1.0
21. I feel comfortable depending on other people (R)	.54	.46	.84	−.07	−.24	−.09	1.6
23. I worry about people getting too close.	.60	.49	.83	−.23	−.34	−.17	1.4
25. I have mixed feelings about being close to others.^a	.70	.60	.83	−.15	−.44	−.05	1.2
26. While I want to get close to others, I feel uneasy about it.^a	.72	.63	.83	−.19	−.41	−.07	1.2
34. Other people have their own problems, so I don’t bother them with mine.	.47	.37	.84	−.13	−.33	−.15	0.6
36. I am too busy with other activities to put much time into relationships.	.61	.52	.83	−.10	−.23	.04	1.6

Note. Item values in bold face represent the top items, generally the highest four items, for each item selection index based on factor loadings, item-total correlations, effect on internal consistency reliability, external correlates, and content validity within each ASQ subscale. Certain subscales have more than four bold-faced items due to ties on these criteria.

Items selected for the short-form version of the Attachment Style Questionnaire.

Table 3.

Internal Consistency Reliability Coefficients and External Correlates for the Short- and Long-Form Attachment Style Questionnaire for Samples A and B.

Variable	Sample A	Sample B
Internal consistency reliability coefficient (Cronbach’s α)
Secure attachment	.74 (.76)	.69 (.68)
Anxious attachment	.79 (.84)	.76 (.76)
Avoidant attachment	.78 (.85)	.73 (.69)
External correlates
Secure attachment
Hope	.38 (.40)	—
Sense of coherence	.64 (.59)	—
Subjective well-being	.39 (.39)	—
Depression	— (—)	−.27 (−.27)
Anxiety	— (—)	−.22 (−.22)
Emotional distress	— (—)	−.28 (−.28)
Anxious attachment
Hope	−.35 (−.38)	—
Sense of coherence	−.63 (−.68)	—
Subjective well-being	−.38 (−.30)	—
Depression	— (—)	.28 (.28)
Anxiety	— (—)	.41 (.38)
Emotional distress	— (—)	.41 (.38)
Avoidant attachment
Hope	−.31 (−.32)	—
Sense of coherence	−.56 (−.53)	—
Subjective well-being	−.21* (−.21*)	—
Depression	— (—)	.26** (.17*)
Anxiety	— (—)	.31 (.33)
Emotional distress	— (—)	.33 (.30)

Note. Depression, anxiety, and emotional distress were assessed using the Hospital Anxiety and Depression Scale.

Results for the short-form are not in parentheses and those for the long-form are in parentheses.

p < .05. **p < .01.

Secure attachment

According to Bowlby’s theory, secure attachment represents primarily confidence in establishing relationship with others, as well as comfort in depending on others. The four items with the highest factor loadings in the long-form secure attachment subscale were 19, 31, 33, and 38. They also had the highest item-total correlation coefficients and would reduce the internal consistency reliability of the long-form secure attachment subscale if removed from the subscale. These four items also correlated moderately with the external correlates (hope, SOC, and SWB) in the expected theoretical directions. In terms of content validity, both Items 19 (“I find it relatively easy to get close to other people”) and 31 (“I feel confident about relating to others”) received an average rating of “2” (“excellent fit with the construct”), while Items 33 (“I often worry that I do not really fit in with other people”) and 38 (“I am confident that other people will like and respect me”) each received an average rating of 1.4 (between “adequate fit with the construct” and “excellent fit with the construct”).

Notably, a close examination of Items 33 and 3 (“I feel confident that other people will be there for me when I need them”) indicated that both items perform almost the same in factor loadings (.65 vs. .64), item-total correlations (.50 vs. .49), and effect on internal consistency reliability (.74 vs. 74). Importantly, Item 3 outperformed Item 33 in its relationship with hope (.32 vs. .26) and SWB (.49 vs. .25) as well as in content validity (2.0 vs. 1.4). Based on these indices, we decided to retain Items 3, 19, 31, and 38 for the short-form secure attachment subscale. The short-form secure attachment subscale correlated highly with the long-form secure attachment subscale (r = .93) and the internal consistency reliability of the short-form secure attachment subscale was computed to be .74 (vs. .76 for the long-form). Both the short- and long-form secure attachment subscale correlated moderately with hope and SWB and highly with SOC. The results provide strong evidence for the psychometric properties for both the short- and long-form secure attachment subscale.

Anxious attachment

Anxious attachment represents primarily preoccupation with relationship, or an anxious reaching out to others to fulfill dependency needs. The four items with the highest factor loadings were 18, 24, 29, and 32. The removal of any of these items would also reduce the internal consistency reliability coefficient of the long-form anxious attachment scale. Items 18 (“I find that others are reluctant to get as close as I would like”), 29 (“I worry a lot about my relationships”), and 32 (“I often feel left out or alone”) also had the highest item-total correlation. Items 18 and 32 correlated moderately well with hope, SOC, and SWB. Item 29 correlated moderately with the external correlates, while Item 24 (“I worry that I won’t measure up to other people”) correlated marginally with hope, moderately with SOC, and poorly with SWB.

A close examination of Item 22 (“I worry that others won’t care about me as much as I care about them”) and Item 24 indicates that Item 24 had higher factor loading (.63 vs. .67) and a higher item-total correlation (.50 vs. .58). However, Item 22 outperformed Item 24 in its relationship with hope (−.37 vs. −.19), SOC (−.43 vs. −.40), and SWB (−.34 vs. .05). In addition, while Items 18, 22, 29, and 32 are categorized in the “preoccupation with relationships” factor that describes anxious attachment in the original ASQ scale, Item 24 is categorized in “need for approval.” Therefore, Item 22 appeared to be a better fit item for the short-form anxious attachment subscale. With regard to Item 29, although it did not correlate highly with the external correlates, it received the highest rating of “2” for content validity, and it had the highest factor loading for this subscale. Based on these indices, we decided to retain Items 18, 22, 29, and 32 for the short-form anxious attachment subscale. The short-form anxious attachment subscale correlated highly with the long-form anxious attachment subscale (r = .88) and the internal consistency reliability of the short-form anxious attachment subscale was computed to be .79 (vs. .84 for the long-form). Both the short- and long-form anxious attachment subscale correlated moderately with hope and SWB and correlated highly with SOC. The results provide strong evidence for the psychometric properties for both the short- and long-form anxious attachment subscale.

Avoidant attachment

Avoidant attachment represents primarily discomfort with closeness and difficulty trusting other people. The four items with the highest factor loadings items (16, 17, 25, and 26) also had the highest item-total correlation coefficients, and the reliability of the subscale would be reduced if any of these items were removed from the long-form avoidant attachment subscale. These four items also correlated moderately with SOC. With regard to external correlates, both Item 5 (“I prefer to keep to myself”) and Item 6 (“To ask for help is to admit that you’re a failure”) ranked higher than Items 17 (“I find it difficult to depend on others”), 25 (“I have mixed feelings about being close to others”), and 26 (“While I want to get close to others, I feel uneasy about it”). Item 5 appears to better represent avoidance behavior than Items 6 and 17, while having acceptable factor loading, moderate item-total correlation, and good contribution to the reliability index. Stanton et al. (2002) suggested when there are conflicts among the criteria, researchers should choose items that preserve the optimal characteristics of the full-length scale. Based on these indices, we decided to retain Items 5, 16, 25, and 26 for the short-form avoidant attachment subscale.

The short-form avoidant attachment subscale correlated highly with the long-form avoidant attachment subscale (r = .86) and the internal consistency reliability of the short-form avoidant attachment subscale was computed to be .78 (vs. .85 for the long-form). Both the short- and long-form avoidant attachment subscale correlated highly with SOC and moderately with hope. The results provide support for the psychometric properties of both the short- and long-form avoidant attachment subscale.

Study 2: Cross-Validation of the Short-Form ASQ

Method

Participants (Sample B)

Data for this sample were collected as part of a larger study focused on investigation of relationships between attachment, hope, and adaptation to treatment for traumatic injuries. One hundred forty participants were recruited from among individuals receiving inpatient care in a large Level 1 Trauma Center. Participants were aged between 18 and 90 years (M = 49.98; SD = 19.53). Most participants were men (74.3%), White (89.3%), and had been admitted for treatment of traumatic injuries caused by motor vehicle accidents (56.5%) or falls (26.5%). Participants received inpatient treatment for between 2 and 28 days (M = 7.5; SD = 4.8).

Measure

Hospital Anxiety and Depression Scale (HADS)

The HADS is a 14-item measure of anxiety and depression experienced during the last week (Zigmond & Snaith, 1983). It is composed of two subscales (anxiety and depression), each containing seven items and scored with a 4-point Likert-type rating scale. Scores for each subscale (anxiety and depression) range from 0 to 21 with scores categorized as follows: normal = 0–7, mild = 8–10, moderate = 11–14, and severe = 15–21. Scores for the entire scale (emotional distress) range from 0 to 42, with higher scores indicating more distress. Results from a systematic review of 71 studies utilizing the HADS indicated that internal consistency reliability coefficients (Cronbach’s alpha) ranged from .63 to .93 for the HADS-anxiety subscale and from .67 to .90 for HADS-depression subscale (Bjelland, Dahl, Haug, & Nettlemann, 2002). The instrument has been used with patients with somatic illnesses and found to be a reliable and valid measure of anxiety and depression among individuals with traumatic injuries, including SCI (Woolrich, Kennedy, & Tasiemski, 2006).

Results

Confirmatory factor analysis (CFA) is frequently used to cross validate the factor structure of a psychological measure (van Prooijen & van der Kloot, 2001). This allows researchers to evaluate the fit between the postulated model and the observed data. Therefore, the three-factor structure of the 12-item ASQ was estimated using a CFA with the traumatic injury data set. As suggested by Weston, Gore, Chan, and Catalano (2008), the goodness of fit of the measurement model was evaluated using the chi-square goodness-of-fit test, χ²/df ratio, the standardized root mean square residual (SRMR), and the comparative fit index (CFI). A nonsignificant chi-square, a relative chi-square ratio (χ²/df) less than 3, SRMR values less than .08, and values greater than .90 indicate an acceptable fit (Hu & Bentler, 1999; Weston et al., 2008). In addition, a root mean square error of approximation (RMSEA) with 90% confidence interval was reported, where a value of less than .05 is considered a close fit and values up to .08 are considered reasonable errors of approximation in the population (Byrne, 2001). The initial CFA model did not adequately fit the data: χ²(51, N = 140) = 92.49, p < .001; χ²/df ratio = 1.814 (<3; acceptable), SRMR = .078 (<.08, acceptable), CFI = .896 (<.90), and RMSEA = .076 (<.08; acceptable). Examination of modification indices suggested that one pair of error term should be correlated: Item e3 (“I feel confident that other people will be there for me when I need them”) with Item e32 (“I often feel left out or alone”). Correlated error terms indicate that knowing the residual of one item helps in knowing the residual associated with another item. Results of the respecified CFA model indicated a relatively good fit between the three-factor measurement model of the short-form ASQ and the data: χ²(50, N = 140) = 81.08, p < .001; χ²/df ratio = 1.636 (<3; acceptable), SRMR = .073 (<.08, acceptable), CFI = .92 (>.90; acceptable), and RMSEA = .67 (<.08; acceptable). All factor loadings were significant at p < .01, ranging from .38 to .87 for secure attachment, between .61 and .81 for anxious attachment, and .41 to .84 for avoidant attachment. The internal consistency reliability coefficients (Cronbach’s alpha) for secure attachment, anxious attachment, and avoidant attachment were computed to be .68, .76, and .69 respectively. The internal consistency reliability coefficients for the construction sample ranged from .74 to .78, while in the cross-validation sample, the range was between .68 and .76. The mean ratings for the secure, anxious, and avoidant attachment subscales were 4.09 (SD = 0.99), 2.92 (SD = 1.22), and 3.36 (SD = 1.04), respectively; these ratings were similar to the average ratings for the three subscales in the construction sample. The short-form ASQ subscales correlated highly with the corresponding long-form ASQ subscales (secure attachment: r = .96, p < .01; avoidant attachment: r = .85, p < .01; and anxious attachment: r = .81, p < .01). In addition, the secure attachment subscale (short- and long-form) was found to negatively associated with emotional distress (short-form: r = −.280, p < .01; long-form r = −.279, p < .01), whereas both avoidant attachment (short-form: r = .332, p < .01; long-form r = .299, p < .01) and anxious attachment (short-form: r = .383, p < .01; long-form r = .405, p < .01) were positively related to emotional distress.

Discussion

Attachment styles shape the development of people’s internal working model of the self and others, which affects their views of people as trustworthy or not trustworthy, their potential to love and be loved, and their affective self-regulation (Wilson et al., 2013). The positive or negative working models derived from secure versus insecure attachment styles have implications for the health and well-being of persons with disabilities. In rehabilitation counseling, Blake and colleagues (Blake et al., 2017; Umucu et al., 2016) demonstrated that adult attachment styles have a direct effect on employment and community participation outcomes of people with disabilities, and attachment styles also have indirect effects on employment through hope (agency thinking and pathways thinking) and self-efficacy. Attachment styles also influence the development of social functioning, social support, coping skills, and SWB (Bucci et al., 2015; Wilson et al., 2013). Research evidence also indicates that attachment has an indirect effect on mental health through SOC (Ying et al., 2007). Thus, adult attachment styles may serve as a protective or vulnerability factor affecting psychosocial and vocational adjustment of people with chronic illness and disability. Given the relatively high rate of insecure attachment styles in the general adult population (approximately 44%), as well as potentially higher rates among persons with disabilities (Howe, 2006; Mikulincer & Orbach, 1995), evaluation of attachment in clinical rehabilitation counseling practice can provide valuable and helpful information, particularly for clients with low levels of social functioning, generalized self-efficacy, hope, social support, coping ability, and SOC.

The two studies described in this article provided preliminary evidence for the validity of the short-form version of the ASQ using sound item selection statistical methodology. The three subscales of secure, anxious, and avoidant attachment from the short-form version of the ASQ demonstrated high correlations with the three subscales from the long-form ASQ. The reliability of the subscales for the short-form version of the ASQ was adequate and similar to the long-form ASQ in both samples. The short-form attachment subscales for Sample A correlated well with the external correlates of hope, SOC, and SWB in the expected directions and correlated well with emotional distress in the expected direction for Sample B. Of the three external correlates, attachment had the highest correlations with SOC, suggesting that attachment plays a significant role in rendering their life challenges comprehensible, meaningful, and manageable to people with disabilities in the current study. It is likely that secure attachment increased hope and SOC leading to higher levels of SWB. Conversely, insecure attachment would lead to lower levels of SWB. People with anxious attachment are particularly vulnerable to emotional distress. Our findings are consistent with expectations regarding the relationships between attachment and hope, SOC, SWB, and depression. Finally, CFA using a cross-validation sample of individuals with traumatic injuries also supported the three-factor measurement structure of the short-form ASQ. The average ratings for the three subscales were similar in the construction and cross-validation samples. However, the internal consistency reliability coefficients in the cross-validation sample for both short- and long-form were lower than the reliability coefficients demonstrated in the construction sample.

Study Limitations

This study involved several limitations that should be addressed. First, the construction sample was relatively small, limiting our ability to confirm the three-factor measurement structure of the short-form ASQ using a split-half cross-validation design. The construction sample was composed entirely of individuals with SCI and the cross-validation sample comprised individuals with traumatic injuries. The sample size of the cross-validation sample was also relatively small (N = 140) for CFA. Although experts have suggested varying guidelines for computing sample sizes for CFA, N ≥ 200 is commonly recommended (N. D. Myers, Ahn, & Jin, 2011). The reliability for both the short- and long-form in the cross-validation sample was lower than the construction sample. It is imperative that future research focus on obtaining a large sample of people with chronic illness and disability, as well as participants from the United States with a range of educational attainment and functioning levels to better analyze the psychometric properties and generalizability of the short-form ASQ. The findings of this study must be considered preliminary only and they should be interpreted with caution.

Implications for Rehabilitation Counseling and Research Practice

Although attachment is an important construct to consider in rehabilitation assessment and planning, the length of the 40-item ASQ limits the likelihood that it will be widely used in the professional practice of rehabilitation counseling. Therefore, to address concerns related to the length of ASQ, we developed a 12-item version of the ASQ to identify items that are most representative of secure attachment, anxious attachment, and avoidant attachment. This study indicates that participants in this study had marginally secure attachment styles and marginally low anxious and avoidant attachment styles in both the short- and long-form ASQ. Although attachment styles are primarily established through the earliest caregiver experiences, research suggests that a sense of others as being trustworthy, loving, and caring can be developed by a strong working alliance with the rehabilitation health professionals (Bordin, 1979; Chan, Shaw, McMahon, Koch, & Strauser, 1997; Wilson et al., 2013). Assessing attachment allows rehabilitation counselors to integrate their knowledge of the client’s attachment style into intervention programs involving the development of the counselor–client working alliance, thus contributing to a stronger bond, more consistent agreement upon goals, and ultimately more positive outcomes. Furthermore, rehabilitation counselors could use ASQ results to inform the ways in which they approach psychological and social interventions surrounding the development of hope, self-efficacy, coping skills, and social functioning. These connections will not only contribute to better rehabilitation outcomes, but may also help to mitigate the negative consequences of an insecure attachment style.

Overall, this study demonstrated that the short-form ASQ subscales are appropriate for use by rehabilitation counselors to identify and assess attachment in people with disabilities. The short-form ASQ can also be used by rehabilitation researchers who need a brief and reliable measure to test the relationships between attachment and other biopsychosocial constructs in rehabilitation and public health research.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The contents of this article were developed with support from the Rehabilitation Research and Training Center on Evidence-Based Practice in Vocational Rehabilitation (RRTC-EBP VR) at the University of Wisconsin–Madison and the University of Wisconsin–Stout and with funding provided by the U.S. Department of Health and Human Services, National Institute on Disability, Independent Living and Rehabilitation Research (Grant H133B100034).

References

Ainsworth

M. D. S.

Blehar

M. C.

Waters

Wall

(1978). Patterns of attachment: A psychological study of the strange situation. Hillsdale, NJ: Lawrence Erlbaum.

Al-yagon

(2012). Subtypes of attachment security in school-age children with learning disabilities. Learning Disability Quarterly, 35, 170–183. doi:10.1037/t17560-000

Antonovsky

(1979). Sense of Coherence Scale. San Francisco, CA: Jossey-Bass.

Antonovsky

(1987). Unraveling the mystery of health: How people manage stress and stay well. San Francisco, CA: Jossey-Bass. doi:10.4135/9781446221129

Avagianou

P. A.

Zafiropoulou

(2008). Parental bonding and depression: Personality as a mediating factor. International Journal of Adolescent Medicine and Health, 20, 261–269. doi:10.1515/IJAMH.2008.20.3.261

Barbaro

Boutwell

B. B.

Barnes

J. C.

Shackelford

T. K.

(2017). Rethinking the transmission gap: What behavioral genetics and evolutionary psychology mean for attachment theory. A comment on verhage et al. (2016). Psychological Bulletin, 143, 107–113. doi:10.1037/bul0000066

Bjelland

Dahl

A. A.

Haug

T. T.

Nettlemann

(2002). The validity of the Hospital Anxiety and Depression Scale: An updated literature review. Journal of Psychosomatic Research, 52, 69–77. doi:10.1016/S0022-3999(01)00296-3

Blake

Brooks

Greenbaum

Chan

(2017). Attachment and employment outcomes for people with spinal cord injury: The intermediary role of hope. Rehabilitation Counseling Bulletin, 60(2), 77–87. doi:10.1177/0034355215621036

Blake

Norton

C. L.

(2014). Examining the relationship between hope and attachment: A meta-analysis. Psychology, 5, 556–565. doi:10.4236/psych.2014.56065

10.

Bordin

E. S.

(1979). The generalizability of the psychoanalytic concept of the working alliance. Psychotherapy: Theory, Research and Practice, 16, 252–260. doi:10.1037/h0085885

11.

Bowlby

(1988). A secure base. New York, NY: Basic Books.

12.

Bradburn

N. M.

(1969). The structure of psychological well-being. Chicago, IL: National Opinion Research Center.

13.

Bucci

Roberts

N. H.

Danquah

A. N.

Berry

(2015). Using attachment theory to inform the design and delivery of mental health services: A systematic review of the literature. Psychology and Psychotherapy: Theory, Research and Practice, 88, 1–20. doi:10.1111/papt.12029

14.

Byrne

(2001). Structural equation modeling with AMOS: Basic concepts, applications and programming. Hillsdale, NJ: Lawrence Erlbaum Associates.

15.

Chan

Shaw

L. R.

McMahon

B. T.

Koch

Strauser

(1997). A model for enhancing consumer-counselor working relationships in rehabilitation: A theoretical and conceptual framework. Rehabilitation Counseling Bulletin, 41, 122–137.

16.

Ciechanowski

P. S.

Walker

E. A.

Katon

W. J.

Russo

J. E.

(2002). Attachment theory: A model for health care utilization and somatization. Psychosomatic Medicine, 64, 660–667. doi:10.1097/01.PSY.0000021948.90613.76

17.

Diener

(1984). Subjective well-being. Psychological Bulletin, 95, 542–575.

18.

Diener

Emmons

R. A.

Larsen

R. J.

Griffin

(1985). The Satisfaction With Life Scale. Journal of Personality Assessment, 49, 71–75.

19.

Drageset

Eide

G. E.

Nygaard

H. A.

Bondevik

Nortvedt

M. W.

Natvig

G. K.

(2009). The impact of social support and sense of coherence on health-related quality of life among nursing home residents: A questionnaire survey in Bergen, Norway. International Journal of Nursing Studies, 46, 65–75. doi:10.1037/t12396-000

20.

Eriksson

Lindström

(2005). Validity of Antonovsky’s Sense of Coherence Scale: A systematic review. Journal of Epidemiology & Community Health, 59, 460–466. doi:10.1136/jech.2003.018085

21.

Feeney

J. A.

Noller

Hanrahan

(1994). Assessing adult attachment. In Sperling

M. B.

Berman

W. H.

(Eds.), Attachment in adults: Clinical and developmental perspectives (pp. 128–152). New York, NY: Guilford Press.

22.

Feldt

Leskinen

Kinnunen

Ruoppila

(2003). The stability of sense of coherence: Comparing two age groups in a 5-year follow-up study. Personality and Individual Differences, 35, 1151–1165. doi:10.1016/S0191-8869(02)00325-2

23.

Feldt

Lintula

Suominen

Koskenvuo

Vahtera

Kivimaki

(2007). Structural validity and temporal stability of the 13-item Sense of Coherence Scale: Prospective evidence from the population-based study. Quality of Life Research, 16, 483–493. doi:10.1007/s11136-006-9130-z

24.

Fibel

Hale

W. D.

(1978). The Generalized Expectancy for Success Scale: A new measure. Journal of Consulting and Clinical Psychology, 46, 924–931. doi:10.1037/0022-006X.46.5.924

25.

Fordyce

(1988). A review of research on the happiness measures: A sixty second index of happiness and mental health. Social Indicators Research, 20, 355–381. doi:10.1007/BF00302333

26.

Fraley

R. C.

Hudson

N. W.

Heffernan

M. E.

Segal

(2015). Are adult attachment styles categorical or dimensional? A taxometric analysis of general and relationship-specific attachment orientations. Journal of Personality and Social Psychology, 109, 354–368. doi:10.1037/pspp0000027

27.

Gana

Garnier

(2001). Latent structure of the sense of coherence scale in a French sample. Personality and Individual Differences, 31, 1079–1090. doi:10.1016/S0191-8869(00)00205-1

28.

Gnilka

P. B.

Ashby

J. S.

Noble

C. M.

(2013). Adaptive and maladaptive perfectionism as mediators of adult attachment styles and depression, hopelessness, and life satisfaction. Journal of Counseling & Development, 91, 78–86. doi:10.1002/j.1556-6676.2013.00074.x

29.

Grotmol

K. S.

Ekeberg

Finset

Gude

Moum

Vaglum

Tyssen

(2010). Parental bonding and self-esteem as predictors of severe depressive symptoms: A 10-year follow-up study of Norwegian physicians. Journal of Nervous and Mental Disease, 198, 22–27. doi:10.1097/NMD.0b013e3181c8189c

30.

Han

(2005). Relationship among perceived parental trauma, parental attachment, and sense of coherence in Southeast Asian American college students. Journal of Family Social Work, 9(2), 25–45. doi:10.1300/J039v09n02_02

31.

Hazan

Shaver

(1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52, 511–524. doi:10.1037/0022-3514.52.3.511

32.

Howe

(2006). Disabled children, parent–child interaction and attachment. Child & Family Social Work, 11, 95–106. doi:10.1111/j.1365-2206.2006.00397.x

33.

L. T.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6, 1–55. doi:10.1080/10705519909540118

34.

Jiang

X. U.

Huebner

E. S.

Hills

K. J.

(2013). Parent attachment and early adolescents’ life satisfaction: The mediating effect of hope. Psychology in the Schools, 50, 340–352. doi:10.1002/pits.21680

35.

Kenny

M. E.

Sirin

S. R.

(2006). Parental attachment, self-worth, and depressive symptoms among emerging adults. Journal of Counseling & Development, 84, 61–71. doi:10.1002/j.1556-6678.2006.tb00380.x

36.

Klepp

O. M.

Mastekaasa

Sørensen

Sandanger

Kleiner

(2007). Structure analysis of Antonovsky’s sense of coherence from an epidemiological mental health survey with a brief nine-item sense of coherence scale. International Journal of Methods in Psychiatric Research, 16, 11–22. doi:10.1002/mpr.197

37.

Kumpfer

(1999). Factors and processes contributing to resilience: The resilience framework. In Glantz

M. D.

Johnson

J. L.

(Eds.), Resilience and development: Positive life adaptations (pp. 179–224). New York, NY: Kluwer Academic/Plenum Press. doi:10.1007/b108350

38.

Lavy

Littman-Ovadia

(2011). All you need is love? Strengths mediate the negative associations between attachment orientations and life satisfaction. Personality and Individual Differences, 50, 1050–1055. doi:10.1016/j.paid.2011.01.023

39.

Lin

C. P.

Chiu

C. Y.

Sung

Lai

Chan

da Silva Cardoso

(2009). Structure of the SOC-13 in alcohol and other drug abuse clients: A confirmatory factor analysis. The Australian Journal of Rehabilitation Counselling, 15, 28–39. doi:10.1375/jrc.15.1.28

40.

Mikulincer

Orbach

(1995). Attachment styles and repressive defensiveness: The accessibility and architecture of affective memories. Journal of Personality and Social Psychology, 68, 917–925. doi:10.1037/0022-3514.68.5.917

41.

Myers

D. G.

(2000). The funds, friends, and faith of happy people. American Psychologist, 55, 56–67. doi:10.1037/0003-066X.55.1.56

42.

Myers

N. D.

Ahn

Jin

(2011). Sample size and power estimates for a confirmatory factor analytic model in exercise and sport: A Monte Carlo approach. Research Quarterly for Exercise and Sport, 82, 412–423. doi:10.1080/02701367.2011.10599773

43.

National SCI Statistical Center. (2016). Facts and figures at a glance. Retrieved from https://www.nscisc.uab.edu/Public/Facts%202016.pdf

44.

Park

Peterson

Seligman

M. E.

(2004). Strengths of character and well-being. Journal of Social & Clinical Psychology, 23, 603–619. doi:10.1521/jscp.23.5.603.50748

45.

Pavot

Diener

Colvin

C. R.

Sandvik

(1991). Further validation of the Satisfaction with Life Scale: Evidence for the cross-method convergence of well-being measures. Journal of Personality Assessment, 57, 149–161. doi:10.1207/s15327752jpa5701_17

46.

Peterson

(2000). The future of optimism. American Psychologist, 55, 44–55. doi:10.1037/0003-066X.55.1.44

47.

Ravitz

Maunder

Hunter

Sthankiya

Lancee

(2010). Adult attachment measures: A 25-year review. Journal of Psychosomatic Research, 69, 419–432. doi:10.1016/j.jpsychores.2009.08.006

48.

Rosenberg

(1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press. doi:10.1515/9781400876136

49.

Scheier

M. F.

Carver

C. S.

(1985). Optimism, coping, and health: Assessment and implications of generalized outcome expectancies. Health Psychology, 4, 219–247. doi:10.1037/0278-6133.4.3.219

50.

Schmidt

Nachtigall

Wuethrich-Martone

Strauss

(2002). Attachment and coping with chronic disease. Journal of Psychosomatic Research, 53, 763–773. doi:10.1016/S0022-3999(02)00335-5

51.

Shorey

H. S.

Snyder

C. R.

Yang

Lewin

M. R.

(2003). The role of hope as a mediator in recollected parenting, adult attachment, and mental health. Journal of Social & Clinical Psychology, 22, 685–715. doi:10.1521/jscp.22.6.685.22938

52.

Sim

T. N.

Yow

A. S.

(2011). God attachment, mother attachment, and father attachment in early and middle adolescence. Journal of Religion & Health, 50, 264–278. doi:10.1007/s10943-010-9342-y

53.

Simmons

B. L.

Gooty

Nelson

D. L.

Little

L. M.

(2009). Secure attachments: Implications for trust, hope, burnout, and performance. Journal of Organizational Behavior, 30, 233–247. doi:10.1002/job.585

54.

Simmons

B. L.

Nelson

D. L.

Quick

J. C.

(2003). Health for the hopeful: A study of attachment behavior in home health care nurses. International Journal of Stress Management, 10, 361–375. doi:10.1037/1072-5245.10.4.361

55.

Snyder

C. R.

(1994). The psychology of hope: You can get there from here. New York, NY: Free Press.

56.

Snyder

C. R.

(2000). Handbook of hope: Theory, measures, and applications. New York, NY: Academic Press.

57.

Snyder

C. R.

Harris

Anderson

J. R.

Holleran

S. A.

Irving

L. M.

Sigmon

S. T.

. . . Harney

(1991). The will and the ways: Development and validation of an individual-differences measure of hope. Journal of Personality and Social Psychology, 60, 570–585. doi:10.1037/0022-3514.60.4.570

58.

Stanton

J. M.

Sinar

E. F.

Balzer

W. K.

Smith

P. C.

(2002). Issues and strategies for reducing the length of self-report scales. Personnel Psychology, 55, 167–194. doi:10.1111/j.1744-6570.2002.tb00108

59.

Taylor

S. E.

Kemeny

M. E.

Reed

G. M.

Bower

J. E.

Gruenewald

T. L.

(2000). Psychological resources, positive illusions, and health. American Psychologist, 55, 99–109. doi:10.1037/0003-066X.55.1.99

60.

Umucu

Lee

J. R.

Chan

Blake

Brooks

Catalano

(2016). Self-efficacy as a mediator for the relationship between secure attachment style and employment status in individuals with spinal cord injuries. Journal of Vocational Rehabilitation, 45, 97–106. doi:10.3233/JVR-160814

61.

van Prooijen

J. W.

van der Kloot

W. A

. (2001). Confirmatory analysis of exploratively obtained factor structures. Educational and Psychological Measurement, 61, 777–792. doi:10.1177/00131640121971518

62.

Wei

Mallinckrodt

Russell

D. W.

Abraham

W. T.

(2004). Maladaptive Perfectionism as a mediator and moderator between adult attachment and psychological distress. Journal of Counseling Psychology, 51, 201–212. doi:10.1037/0022-0167.51.2.201

63.

Weston

Gore

P. A.

Jr. Chan

Catalano

(2008). An introduction to using structural equation models in rehabilitation psychology. Rehabilitation Psychology, 53, 340–356. doi:10.1037/a0013039

64.

Widaman

K. F.

Little

T. D.

Preacher

K. J.

Sawalani

G. M.

(2011). On creating and using short forms of scales in secondary research. In Trzesniewski

K. H.

Donnellan

M. B.

Lucas

R. E.

(Eds.), Secondary data analysis: An introduction for psychologists (pp. 39–61). Washington, DC: American Psychological Association. doi:10.1037/12350-000

65.

Wilson

Catalano

Sung

Phillips

Chou

C. C.

Chan

J. Y. C.

Chan

(2013). Attachment style, social support, and coping as psychosocial correlates of happiness in persons with spinal cord injuries. Rehabilitation Research, Policy, and Education, 27, 186–205. doi:10.1891/2168-6653.27.3.186

66.

Woolrich

R. A.

Kennedy

Tasiemski

(2006). A preliminary psychometric evaluation of the Hospital Anxiety and Depression Scale (HADS) in 963 people living with a spinal cord injury. Psychology, Health & Medicine, 11, 80–90. doi:10.1037/t03589-000

67.

Ying

Y. W.

Lee

P. A.

Tsai

J. L.

(2007). Attachment, sense of coherence, and mental health among Chinese American college students: Variation by migration status. International Journal of Intercultural Relations, 31, 531–544. doi:10.1016/j.ijintrel.2007.01.001

68.

Zigmond

A. S.

Snaith

R. P.

(1983). The Hospital Anxiety and Depression Scale. Acta Psychiatrica Scandinavica, 67, 361–370. doi:10.1037/t03589-000

69.

Zimprich

Allemand

Hornung

(2006). Measurement invariance of the abridged sense of coherence scale in adolescents. European Journal of Psychological Assessment, 22, 280–287. doi:10.1027/1015-5759.22.4.280