Alissa Sherry

Conducting and interpreting canonical correlation analysis in personality research: a user-friendly primer.

The purpose of this article is to reduce potential statistical barriers and open doors to canonical correlation analysis (CCA) for applied behavioral scientists and personality researchers. CCA was selected for discussion, as it represents the highest level of the general linear model (GLM) and can be rather easily conceptualized as a method closely linked with the more widely understood Pearson r correlation coefficient. An understanding of CCA can lead to a more global appreciation of other univariate and multivariate methods in the GLM. We attempt to demonstrate CCA with basic language, using technical terminology only when necessary for understanding and use of the method. We present an entire example of a CCA analysis using SPSS (Version 11.0) with personality data.

Person factors associated with suicidal behavior among African American women and men.

This study compared person risk factors among the following groups of low-income, African American adults in an urban, public hospital: (a) suicide attempters and nonattempters, (b) male and female attempters, and (c) all 4 groups (50 female attempters, 50 female nonattempters, 50 male attempters, and 50 male nonattempters). Participants completed psychological distress, aggression, substance use, cognitive processes, religiosity/spirituality, and ethnic identity measures. Compared with nonattempters, attempters reported more psychological distress, aggression, substance use, and maladaptive coping strategies; less religiosity/spirituality; and lower levels of ethnic identity. Male attempters endorsed more substance use than female attempters. No person risk factors differentiated among the 4 groups. Assessment of person risk factors and implementation of commensurate culturally competent interventions are recommended.

Discriminant Analysis in Counseling Psychology Research

Social science research often entails the comparison of two or more groups. For example, researchers may explore differences between sexes, races, ethnicities, sexual identities, treatment conditions, diagnostic categories, or various other grouping variables. While there are a number of ways to approach these comparisons, applied social science researchers often become comfortable with only one or two and may lack the time to learn or investigate others that may be more appropriate for their research questions. This article presents discriminant analysis as one of these alternative approaches. More specifically, it describes descriptive discriminant analysis in greater detail, including an example using SPSS (Version 11.0) output.

Internalized homophobia and adult attachment: Implications for clinical practice.

Two hundred eighty-six men and women who identified as lesbian, gay, or bisexual (LGB) were surveyed using measures of attachment, internalized homophobia, shame, and guilt. Results indicated that secure attachment was negatively correlated with internalized homophobia, shame, and guilt. Fearful and preoccupied attachments were positively correlated with these constructs, accounting for 36.9% of the variance in scores. Results are discussed in terms of the therapeutic relationship, with the therapist serving as a secure base for LGB clients working through issues of internalized homophobia and social isolation. (PsycINFO Database Record (c) 2010 APA, all rights reserved).

Gay, Lesbian, and Bisexual Training Competencies in American Psychological Association Accredited Graduate Programs.

The current study examines the extent to which American Psychological Association accredited clinical and counseling doctoral programs incorporate training on issues relevant to gay, lesbian, and bisexual (GLB) clients. One hundred four training directors completed surveys. Results reveal the 1st snapshot of how doctoral programs are addressing GLB issues in existing curricula. For example, programs are incorporating GLB issues in multicultural classes and practicum. However, counseling programs incorporated GLB issues more than clinical programs in several areas, such as requiring multicultural courses (where GLB issues are potentially addressed) and mentoring students in GLB research. Differences in levels of inclusion between clinical and counseling programs could be attributable to the programs’ differential emphases on subspecialization.

A Validation of the Factor Structure of OQ-45 Scores Using Factor Mixture Modeling

This study investigated the Outcome Questionnaire’s (OQ-45) factor structure and demonstrated the use of factor mixture modeling (FMM) for the purpose of score validation. OQ-45 scores did not fit the one-class, one- and three-factor models. Use of FMM to identify a two-class model is detailed. Implications for OQ-45 users are provided.

Evaluating the Appropriateness of College-Age Norms for Use With Adolescents on the NEO Personality Inventory-Revised

The NEO Personality Inventory—Revised (NEO PI-R) measures normal personality characteristics and has demonstrated appropriate score reliability and validity. It is normed for two groups of individuals, college-age individuals 17 to 20 years old and adults 21 and older. Often, personality instruments normed on older individuals have been used with adolescent populations. To examine the appropriateness of this decision, the current study explored the differences between an adolescent sample (n = 79) and a college-age sample (n = 80) on the 30 facets and the five domains of the NEO PI-R. Group differences on the facet and domain scales were analyzed using descriptive discriminant analysis. Results indicated that the adolescent and college groups differed on each of the five domains. As expected, the groups also scored differently using the aggregated domain-level variables as the outcome measures. Suggestions for future research include the development of normative data for the adolescent population.

Psychometric Properties of a Translated Korean Adult Attachment Measure

In this study, psychometric properties of a Korean-adapted version of the Experiences in Close Relationships—Revised scale were investigated using confirmatory factor analysis (CFA) in conjunction with item response theory. In CFA, neither the supposed two-factor model of Anxiety and Avoidance nor two single-factor models for each factor showed a reasonable fit. In addition, the results from the CFA suggested that the negative-worded and positive-worded items form distinct factors. In the item response theory analysis, the item response matrices of the Korean version did not meet unidimensionality and local independence assumptions. The hierarchical nature of the multidimensionality in the data and residual association among items were dealt with by using a bifactor model. Subsequent exploratory bifactor analysis and full-information item bifactor analysis of graded response data provided support for the bifactor structure of the data and other unique features of the Korean version. The differences between the Experiences in Close Relationships—Revised and the Korean version and their probable causes and implications are discussed.

An Attachment Theory Approach to the Short-Term Treatment of a Woman With Borderline Personality Disorder and Comorbid Diagnoses

The current case study highlights the treatment of “Thelma,” a 48-year-old woman of African American and Hispanic descent who was diagnosed with borderline personality disorder (BPD), major depressive disorder, post-traumatic stress disorder, and bulimia. An attachment theory approach was used to elicit core structure change in Thelmas BPD symptoms after traditional modern cognitive approaches proved ineffective. It was hypothesized that treating her Axis II disorder would actually serve to reduce her Axis I symptoms. The attachment approach entailed a shift toward second-order change processes, a shift away from problem-solving approaches, a focus on the quality and intensity of the relationship, and a more relaxed approach to boundary setting in the context of treatment. The approach did appear to reduce many Axis I symptoms including self-mutilation, suicidal ideation, and binging and purging behaviors and appeared to improve her interpersonal functioning.

The Cross-Cultural Validity of the MMPI-2-RF Higher-Order Scales in a Sample of North Korean Female Refugees

We investigated the cross-cultural factorial validity of the three Higher-Order (H-O) scales in the Minnesota Multiphasic Personality Inventory–2 Restructured Form (MMPI-2-RF) among a sample of North Korean female refugees (N = 2,732). Given the importance of the H-O scales in the overall structure of the MMPI-2-RF scales and in interpretation, we were interested in exploring their cross-cultural validity. We conducted an exploratory factor analysis (EFA) on the nine Restructured Clinical (RC) scale raw scores and fitted and compared one- to three-factor models. The three-factor model, akin to the model in Tellegen and Ben-Porath, demonstrated the best fit to the data. Furthermore, the pattern matrices of loadings across the current sample and the U.S. samples were comparable despite some differences, such as the RC2 scale’s salient, negative loading on a factor analogous to the Behavioral/Externalizing Dysfunction scale. We also investigated the unique psychological characteristics of the refugees, possibly resulting from the arduous, perilous journeys out of North Korea taken by this group of female refugees and discussed the results of EFA in light of those singular psychological traits and experiences. Overall, the three H-O scales of the Korean MMPI-2-RF evidenced reasonable cross-cultural factorial validity among the sample of North Korean female refugees.