Date of Award

5-2023

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Mathematical Sciences

First Advisor

Thomas Marcinkowski

Second Advisor

Joo Young Park

Third Advisor

Ryan T. White

Fourth Advisor

Thomas H. Harrell

Abstract

Due to the small number of research studies that have explored emotional intelligence (EI) in African-American (AA) samples (i.e., twelve studies, with six involving college students), the first purpose of this study was to describe conceptual and empirical dimensions of EI within a sample of AA college students (Research Question 1). To date, only one EI measure has been developed for use with AA samples, the African American Emotional Intelligence Survey or AAEIS (Funderburk, 2007), so the second purpose was to further explore the validity and reliability of the AAEIS (Research Question 2). The third purpose was to explore the relationship of selected demographic, experiential, and familiar factors to scores on the AAEIS and on a second measure of EI, the TEIQue Short Form or SF (Petrides, 2009; Research Question 3). This study’s target population included AA college students who resided in one Central Florida county in 2020, or AA college students whose characteristics resemble those of students in these counties. The accessible population included AA undergraduate students enrolled in any of the branch campuses of the state college in that county in 2019-20. This state college had a total of full-time 14,597 students, of whom 1,605 (11%) were self-identified as AA (males = 534, females = 1,065). A stratified random sample was invited to participate in this study, but was adjusted to allow for an even number of AA male students (50%: 268) and AA female students (25%: 266). A total of 69 usable responses were received and included (n = 69; male = 13, female = 56). Permission was obtained to use the AAEIS, TEIQue-SF, and the Schedule of Racist Events or SRE (Klonoff & Landrine, 1999). The researcher also developed items based on prior research to collect data on selected demographic (age, gender), college experience (number of terms completed, student engagement in campus activities), and familial factors (mother’s and father’s highest level of education, family SES). The Participant Consent Form, along with these items and scales, were developed in Qualtrics for online data collection. Following IRB approval, this sample of AA male and female student received an e-mail message, which introduced the researcher and this study, presented an invitation to participate, and presented the link to this Consent Form and online survey. For Research Question 1, results of an Exploratory Factor Analysis (EFA) of Items 1-15 in the AAEIS indicated that (Factor 1) Self-Control and Conflict Avoidance (4 items), and (Factor 2) Conflict Engagement (5 items) were prominent EI characteristics of this sample, explaining 29% of the variance. Further, results indicated that (Factor 3) Willingness to Understand Others (2 items), and (Factor 4) Willingness to be Responsive to Others (3 items) were influential EI characteristics, explaining an additional 21% of the total variance. It was noteworthy that all of these items reflects dimensions of EI that pertained to self-in-relationship. For exploratory purposes, an additional EFA was conducted, including those 15 AAEIS items and the 30 TEIQue-SF items. The results of that EFA were noticeably different. Factors 1 and 2 explained 20% of the variance in this data set, and included only TEIQue-SF items, nearly all of which focused on healthy dimensions of EI which pertained to one’s self (Well-Being, Personal Motivation, Emotionality, and Self-Control). Further, only Factor 4 included more than one Conflict Avoidance and/or Conflict Engagement item (i.e., 3 and 1 of those 5 items, respectively), although this factor explained only 7.7% of the variance. These EFA results indicated that healthy dimensions of EI which pertained to one’s self were more prominent in this sample than were dimensions of EI which pertained to self-in-relationship, notably those that involved adversity and conflict (i.e., as in the AAEIS). From the analyses for Research Question 2, the results of the Cronbach’s alpha (n=69) was .585, which is lower than minimum thresholds for measures such as this (e.g., Nunally, 1978). As an indicator of concurrent validity, the correlation between AAEIS total scores and TEIQue-SF total scores was .608. Although the size of this correlation is lower than anticipated, few sources indicate whether this value is above or below an acceptable threshold. Nonetheless, this correlation appears to reflect differences in the design of the AAEIS and TEIQue-SF: (a) Funderburk’s reliance on Goleman’s model (3 of 5 components) and Mayer, Salovey and Caruso’s model (3 of 4 components) vs. Petrides’ reliance on the work of he and his colleagues on their own model (e.g., Petrides & Furnham, 2003); and (b) differences apparent in EFA results noted above. For construct validity purposes, convergent, but not divergent, validity was explored. These results indicated that the AAEIS appears to be construct valid as a measure of dimensions of EI which emphasize self-in-relationship, notably those which involve adversity and conflict (Factors 1 and 2) as well as empathy and responsiveness toward others (Factors 3 and 4). However, the evidence from the EFA for AAEIS and TEIQue-SF items indicates that the AAEIS does not appear to be construct valid as a measure of those dimensions of EI which emphasize one’s self, notably well-being and motivation. Post-hoc power analyses for Research Question 3 involving (3a) the AAEIS (n=69) and (3b) the TEIQue-SF (n=42) indicated that each sample was too small relative to the number of independent variables in each (i.e., 7 and 9, respectively). To increase power to a more acceptable level (power = about .75), multiple regression analyses were conducted, and those results were used to reduce the number of independent variables in each regression model. The regression analysis for (3a) indicated that three variables had t values with p values < .20: Father’s Level of Education (p = .05); Age (p = .081), and Number of Terms Completed (p = .154). When only those three variables were included, the regression model was significant (F = 2.874, p = .045), although this model explained only 11.7% of the variance in AAEIS scores, and only Age was statistically significant (t = 2.191, p = .032). The regression analysis for (3b) indicated that only one variable had a t value with a p value < .2: Age. When only Age was included, the regression model was significant (F = 5.437, p = .024), and this single-variable model explained 12.1% of the variance in TEIQue-SF total scores. It was noteworthy that Age was found to be a statistically significant predictor of both AAEIS and TEIQue-SF total scores. When coupled with data on the age range of study participants (ages 18-63), the results appear to support hypotheses regarding the influence of age and associated developmental and experiential factors on EI. A number of recommendations for further research were offered, including for replication, and attention to study limitations and delimitations. Recommendations for research based on findings of this study, included uses of the AAEIS in combination with other measures of EI, and the development of new measures for exploring dimensions of EI that pertain to one’s perceptions of and responses to adversity and conflict. Finally, following from a discussion of implications for practice, recommendations were offered regarding a college’s support for the development of students’ EI, including attention to the college’s climate and culture, and its collaboration with the surrounding community.

Download

Share

COinS