<%BANNER%>

The Predictive Validity of Self-Reported Emotional Intelligence in Children

Permanent Link: http://ufdc.ufl.edu/UFE0044628/00001

Material Information

Title: The Predictive Validity of Self-Reported Emotional Intelligence in Children
Physical Description: 1 online resource (95 p.)
Language: english
Creator: Ditterline, Jeffrey W
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: academic -- achievement -- adolescents -- behavior -- children -- emotional -- intelligence
Special Education, School Psychology and Early Childhood Studies -- Dissertations, Academic -- UF
Genre: School Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Emotional intelligence (EI) recently garnered interest following claims that the construct may play more of a role than IQ in predicting academic and occupational success. Several concepts of EI emerged, including the ability model of Mayer and Salovey (1997), which is the focus of most research, the emotional-social intelligence model of Bar-On (2006), and the trait emotional self-efficacy model of Pérez, Petrides, and Furnham (2005). Studies, largely conducted with adults, indicate that EI correlates with important external criteria. However, little data from children complement the adult literature. Moreover, debate exists over incremental predictive validity due to correlations between EI, general intelligence, and personality. To address these issues, a series of multiple regression analyses were conducted to achieve the primary aim of evaluating the incremental predictive validity of the total score from the Bar-On Emotional Quotient Inventory: Youth Version over intelligence as measured by the Woodcock-Johnson III Tests of Cognitive Abilities and five scale scores from the Five Factor PersonalityInventory-Children in the prediction of academic achievement in 50 children and behavior in 38 children (aged 10-14 years in grades 5 through 8). Also, a t test and moderated regression analyses were used to explore gender differences in self-reported EI. Results showed a low yet significant relationship between the adaptability scale of the self-report measure of EI and intelligence as well as moderate significant relationships between overall self-reported EI and the personality dimensions of agreeableness, conscientiousness, and emotional regulation. Overall self-reported EI also correlated moderately and negatively with externalizing, internalizing, and overall behavioral problems yet these associations were no longer significant after accounting for the variance explained by general intelligence and personality. Further, after accounting for general intelligence and personality, the inclusion of self-reported EI led to no significant additional variance in the prediction of other academic or behavioral criteria. Finally, data showed that gender significantly moderated the relationships between overall self-reported EI and adaptive skills, behavioral problems, and math skills. Thus, the present study provided evidence that the relationships between self-reported EI and external criteria differ according to gender yet the construct shows limited incremental predictive validity in educational contexts.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jeffrey W Ditterline.
Thesis: Thesis (Ph.D.)--University of Florida, 2012.
Local: Adviser: Kranzler, John H.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044628:00001

Permanent Link: http://ufdc.ufl.edu/UFE0044628/00001

Material Information

Title: The Predictive Validity of Self-Reported Emotional Intelligence in Children
Physical Description: 1 online resource (95 p.)
Language: english
Creator: Ditterline, Jeffrey W
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: academic -- achievement -- adolescents -- behavior -- children -- emotional -- intelligence
Special Education, School Psychology and Early Childhood Studies -- Dissertations, Academic -- UF
Genre: School Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Emotional intelligence (EI) recently garnered interest following claims that the construct may play more of a role than IQ in predicting academic and occupational success. Several concepts of EI emerged, including the ability model of Mayer and Salovey (1997), which is the focus of most research, the emotional-social intelligence model of Bar-On (2006), and the trait emotional self-efficacy model of Pérez, Petrides, and Furnham (2005). Studies, largely conducted with adults, indicate that EI correlates with important external criteria. However, little data from children complement the adult literature. Moreover, debate exists over incremental predictive validity due to correlations between EI, general intelligence, and personality. To address these issues, a series of multiple regression analyses were conducted to achieve the primary aim of evaluating the incremental predictive validity of the total score from the Bar-On Emotional Quotient Inventory: Youth Version over intelligence as measured by the Woodcock-Johnson III Tests of Cognitive Abilities and five scale scores from the Five Factor PersonalityInventory-Children in the prediction of academic achievement in 50 children and behavior in 38 children (aged 10-14 years in grades 5 through 8). Also, a t test and moderated regression analyses were used to explore gender differences in self-reported EI. Results showed a low yet significant relationship between the adaptability scale of the self-report measure of EI and intelligence as well as moderate significant relationships between overall self-reported EI and the personality dimensions of agreeableness, conscientiousness, and emotional regulation. Overall self-reported EI also correlated moderately and negatively with externalizing, internalizing, and overall behavioral problems yet these associations were no longer significant after accounting for the variance explained by general intelligence and personality. Further, after accounting for general intelligence and personality, the inclusion of self-reported EI led to no significant additional variance in the prediction of other academic or behavioral criteria. Finally, data showed that gender significantly moderated the relationships between overall self-reported EI and adaptive skills, behavioral problems, and math skills. Thus, the present study provided evidence that the relationships between self-reported EI and external criteria differ according to gender yet the construct shows limited incremental predictive validity in educational contexts.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jeffrey W Ditterline.
Thesis: Thesis (Ph.D.)--University of Florida, 2012.
Local: Adviser: Kranzler, John H.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044628:00001


This item has the following downloads:


Full Text

PAGE 1

1 THE PREDICTIVE VALIDITY OF SELF REPORTED EMOTIONAL INTELLIGENCE IN CHILDREN By JEFFREY WILLIAM DITTERLINE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREM ENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Jeffrey William Ditterline

PAGE 3

3 ACKNOWLEDGMENTS This dissertation would not have been possible without the encouragement and support of my adviso r and committee members, colleagues, family members, and enriched the project. Recently, James Algina, Diana Joyce, and Nancy Waldron patiently served on my committee yet prev iously they shared their school psychology and statistical expertise with me in classes. I would like to express my appreciation to them and all of my teachers especially Tom Oakland, who primarily helped develop my skills as a presenter, researcher, and w riter. I also want to thank my Dad and Mom for always believing that I would finish the project and for their invaluable and never ending support. Thank you to my brothers and friends for always caring about my progress and never letting their concerns ge t in the way of having a good time. Finally, I want to express my deepest gratitude to my Grandma Jean, my wife Lawrie, and the rest of my family. Feeling your love and strength bolstered my persistence and carried me through.

PAGE 4

4 TABLE OF CONT ENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 3 LIST OF TABLES ................................ ................................ ................................ ............ 6 ABSTRACT ................................ ................................ ................................ ..................... 8 CHAPTER 1 REVIEW OF THE LITERATURE ................................ ................................ ............ 10 Ability Model of Emotional Intelligence ................................ ................................ .... 11 Measurement ................................ ................................ ................................ ... 12 Critical I ssues ................................ ................................ ................................ ... 12 Mixed Model of Emotional Intelligence ................................ ................................ .... 15 Measurement ................................ ................................ ................................ ... 16 Critical I ssues ................................ ................................ ................................ ... 17 Academic Criteria and Self Reported Emotional Intelligence ........................... 20 Comparison of Ability and Mixed Models ................................ ................................ 21 Summary ................................ ................................ ................................ ................ 23 Purposes of the Current Study ................................ ................................ ................ 24 Main Objective 1: Understanding the Predictive Validity of Self Reported Emotional Intelligence among Children ................................ ......................... 25 Main Objective 2: Characterizing Gender Bas ed Differences in Emotional Intelligence ................................ ................................ ................................ .... 26 Summary of the Objectives and Purposes ................................ ....................... 26 2 METHODS ................................ ................................ ................................ .............. 28 Sampling Procedures ................................ ................................ .............................. 28 Participants ................................ ................................ ................................ ............. 28 Instruments ................................ ................................ ................................ ............. 29 Procedure ................................ ................................ ................................ ............... 45 Data Analysis ................................ ................................ ................................ .......... 47 3 RESULTS ................................ ................................ ................................ ............... 50 Descriptive Statistics ................................ ................................ ............................... 50 Criterion Variables ................................ ................................ ............................ 50 Predictor Variables ................................ ................................ ........................... 51 t Test ................................ ................................ ................................ ......... 53 Correlational Analyses ................................ ................................ ............................ 53 Multiple Regression Analyses ................................ ................................ ................. 55 Academic Achievement ................................ ................................ .................... 55 Behavior ................................ ................................ ................................ ........... 57

PAGE 5

5 Moderated Regression Analyses ................................ ................................ ............ 59 4 DISCUSSION ................................ ................................ ................................ ......... 74 Research Questions 1 and 2: Relationships between Self Reported EI and Important Criteria ................................ ................................ ................................ 74 Relationships between Self Reported EI and General Intelligence .................. 74 Relationships between Self Reported EI and Personality ................................ 75 Behavioral Criteria ................................ ................................ ............................ 77 Academic Criteria ................................ ................................ ............................. 79 Research Questions 3 and 4: Understanding Gender Based Differences in an Emotion Re lated Construct in Youth ................................ ................................ ... 81 Gender Differences in Self Reported EI ................................ ........................... 81 Gender Differences in the Relationships between Self Repor ted EI, Academics, and Behavior ................................ ................................ .............. 82 Implications for Future Research ................................ ................................ ............ 84 Limitations ................................ ................................ ................................ ............... 85 LIST OF REFERENCES ................................ ................................ ............................... 88 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 95

PAGE 6

6 LIST OF TABLES Table page 2 1 Composites, Content, and Primary Scales of the Parent Rating Scales of the Behavior Assessment System for Children Second Edition ............................... 49 2 2 Administration and readability of stand ardized measures used in the current study ................................ ................................ ................................ ................... 49 3 1 Descriptive statistics for criterion variables ................................ ......................... 61 3 2 Descriptive statistics for predictor variables ................................ ........................ 61 3 3 Mean Emotional Quotient Inventory: Youth Version performance by gender ..... 61 3 4 Comparison of me ans based on gender ................................ ............................. 62 3 5 Pearson product moment correlations between self reported EI, intelligence as assessed by the BIA scale of the WJIII COG, and KTEA II BF performance ................................ ................................ ................................ ....... 62 3 6 Pearson product moment correlations between self reported EI and behavior .. 62 3 7 Pearson product moment correlations between self repo rted EI and personality ................................ ................................ ................................ .......... 63 3 8 Sequential regression analysis predicting KTEA II BF math achievement ( N = 50) ................................ ................................ ................................ ...................... 63 3 9 Sequentia l regression analysis predicting KTEA II BF reading achievement ( N = 50) ................................ ................................ ................................ .............. 64 3 10 Sequential regression analysis predicting KTEA II BF writing achievement ( N = 50) ................................ ................................ ................................ ................... 65 3 11 Sequential regression analysis predicting KTEA II overall academic achievement ( N = 50) ................................ ................................ ......................... 66 3 12 Sequential regression analysis predicting suspensi ons ( N = 50) ........................ 67 3 13 Sequential regression analysis predicting externalizing problems ( N = 38) ........ 68 3 14 Sequential regression an alysis predicting internalizing problems ( N = 38) ......... 69 3 15 Sequential regression analysis predicting behavioral symptoms ( N = 38) .......... 70 3 16 Sequential regression analysis predicting adaptive skills ( N = 38) ..................... 71

PAGE 7

7 LIST OF FIGURES Figure page 1 1 Operational model. ................................ ................................ ............................. 27 3 1 Interaction of gender on behavioral symptoms and total score for self reported EI. ................................ ................................ ................................ ......... 71 3 2 Interaction of gender on adaptive skills and total score for self reported EI. ...... 72 3 3 Interaction of gender on math achievement and total score for self reported EI. ................................ ................................ ................................ ....................... 73

PAGE 8

8 Abstract of Dissertation Prese nted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE PREDICTIVE VALIDITY OF SELF REPORTED EMOTIONAL INTELLIGENCE IN CHILDREN By Jeffrey William D itterline December 2012 Chair: John Kranzler Major: School Psychology Emotional intelligence (EI) recently garnered interest following claims that the construct may play more of a role than IQ in predicting academic and occupational success. Several con cepts of EI emerged, including the ability model of Mayer and Salovey (1997), which is the focus of most research, the emotional social intelligence model of Bar On (2006), and the trait emotional self efficacy model of Prez, Petrides, and Furnham (2005). Studies, largely conducted with adults, indicate that EI correlate s with important external criteria. However, little data from children complement the adult literature. Moreover, debate exists over incremental predictive validity due to correlations betw een EI, general intelligence, and personality. To address these issues, a series of multiple regression analyses were conducted to achieve the primary aim of evaluating the incremental predictive validity of t he total score from the Bar On Emotional Quotie nt Inventory: Youth Version over intelligence as measured by the Woodcock Johnson III Tests of Cognitive Abilities and five scale scores from the Five Factor Personality Inventory Children in the prediction of academic achievement in 50 children and behavi or in 38 children (aged 10 14 years in grades 5 through 8) Also, a t

PAGE 9

9 test and moderated regression analyses were used to exp lore gender differences in self reported EI. Results showed a low yet significant relationship between the adaptability scale of th e self report measure of EI and intelligence as well as moderate significant relationships between overall self reported EI and the personality dimensions of agreeableness, conscientiousness, and emotional regulation. Overall self reported EI also correlat ed moderately and negatively with externalizing, internalizing, and overall behavioral problems yet these associations were no longer significant after accounting for the variance explained by general intelligence and personality. Further after accounting for general intelligence and personality, the inclusion of self reported EI led to no significant additional variance in the prediction of other academic or behavioral criteria. Finally d ata showed that gender significantly moderated the relationship s be tween overall self reported EI and adaptive skills, behavioral problems, and math skills. Thus the present study provided evidence that the relationships between self reported EI and external criteria differ according to gender yet the construct shows lim ited incremental predictive validity in educational contexts

PAGE 10

10 CHAPTER 1 REVIEW OF THE LITERA TURE Currently, no consensus definition for emotional intelligence (EI) exists. Generally, emotion (Mayer & Salovey, 1997). Specifically, individual differences exist in emotional understanding as well as intrapersonal and interpersonal emotional perception, access to, appraisal, differentiation, feeling, and generation of emotion, and the manag ement and use of emotion to adapt to environmental demands and promote intellectual growth (American Psychological Association, 2007; Bar On, 2006; Goleman, 1995; Mayer & Salovey, 1997). Concepts of EI gained exposure following the publication of a text in which the author, despite little empirical evidence, claimed that EI played a larger role than IQ in predicting the academic success and healthy psycho social development of children as well as the occupational performance of adults (Goleman, 1995). This and similar generalizations led to a high level of popular and scientific interest in EI. Yet, these claims remain largely unsubstantiated. variance in academic achievement (Neisse r et al., 1996), personality variables explain approximately 10% (Furnham & Chamorro Premuzic, 2003) and approximately 5% can be attributed to error variance. This leaves approximately 60 % of the variance unexplained. The recent emergence of concepts of E I formed enticing explanations for the remaining variance. Thus, debate began over whether EI adds significant and unique variance beyond measures of general intelligence and personality in the prediction of educationally important criteria. Before introd ucing the present

PAGE 11

11 investigation, t he following sections review Mayer and Salovey (1997) ability model of EI the emotional social intelligence model of Bar On (2006), and the trait emotional self efficacy model of Prez, Petrides, and Furnham (2005). Abi lity Model of Emotional Intelligence Salovey and Mayer (1989 1990) first proposed a scientific theory of EI. Emotional intelligence, a subtype of social intelligence, was considered to involve individual differences in the ability to perceive, discriminate and regulate inter and intrapersonal emotional information in order to guide behavior and thought (Salovey & Mayer, 1989 1990). This initial conceptualization of EI was divided into three branches: appraisal and expression, regulation, and utilization o f emotion. A later revision placed EI within a hierarchical model in an attempt to associate it with intellectual ability while differentiating it from personality traits (Mayer & Salovey, 1997) The revised conceptualization of EI was divided into four br anches. Emotional intelligence included the ability to: (1) accurately appraise, express, and perceive emotion; (2) access and generate feeling in order to facilitate thought; (3) understand emotional information; and (4) regulate emotion in order to impro intellectual development (Mayer & Salovey, 1997). The first branch includes the inter and intrapersonal identification of emotions and emotional content in behavior, feeling, language, physical appearance and state, sound, and thoug ht. Also, it includes the expression of emotions and needs related to feelings as well as appraisal of the accuracy or honesty of expressed feelings. The second branch includes the availability and generation of emotions and feelings in order to facilitate intellectual processes such as attention, judgment, and memory. Further, it includes changes in dispositional moods that affect whether one considers multiple perspectives or employs different forms of

PAGE 12

12 reasoning in order to approach problems. The third br anch includes knowledge of the words used to label emotions as well as an understanding of the relationships between words, actual emotions, and events. Moreover, it includes analysis of complex feelings and recognition of sequences of emotions. The fourth branch includes awareness of emotional reactions as well as reflective disengagement from feelings in appropriate situations. Last, it includes inter and intrapersonal monitoring of the clarity, influence, and typicality of emotions as well as the enhanc ement or moderation of emotions without neglecting to consider the information that they convey. Measurement Measurement within the ability model is based on the assessment of e motion related cognitive abilities and knowledge through maximum performance (i.e., IQ like) tests. The Mayer Salovey Caruso Emotional Intelligence Test Version 2.0 (MSCEIT; Mayer, Salovey, & Caruso, 2002 ) operationalizes EI based on the four branches described in their model knowledge, perception, regulation, and use of emotion i n a performance based test of problem solving and task execution. A youth version of the MSCEIT currently is under development. The MSCEIT requires that test takers define complex emotional terms; generate and reason with emotions; select the best emotiona l decision making strategies for certain scenarios; solve problems that concern emotions; and use emotions by, for example, judging emotions in designs and faces. Critical issues The subjectivity of emotional experience may preclude measurement within a m aximum performance IQ undermines the development of items or tasks that can be scored according to truly

PAGE 13

13 (Petrides, Furnham, & Mavroveli, 2007, p. 154). Performance based measures that are scored based on consensual norms have problems including restriction of range and skew; those that are scored based on expert judgment have problems, as well, because the n ature of the requisite expertise is unclear (MacCann, Matthews, Zeidner, & Roberts, 2003). Also, there is limited support for the factor structure of performance based measures. The first empirical examination of a performance based measure of ability EI revealed not the four branch model suggested by Mayer and Salovey, but a general EI factor that was divided into three subscales of managing, perceiving, and understanding emotions (Mayer, Caruso, & Salovey, 1999). Other data have not been supportive of th e theoretical factor structure of the MSCEIT ( Ciarrochi, Chan, & Caputi, 2000; Farrelly & Austin, 2007; Gignac, 2005; R ossen, Kranzler, & Algina, 2008 ). This becomes problematic when, for example, significant correlations are found between one or another o f the branch scores and outcome criteria. The use of a total score is more appropriate due to the lack of support for the proposed structure of the measure and because a general factor is more consistent than secondary factors. Ability based measures have a degree of predictive validity. In this review, guidelines for the interpretation of correlations are based on the work of Cohen (1988), who observed that, although the interpretation of correlations depends on their context and purposes, generally low co rrelations are from 0.1 to 0.3, moderate correlations are from 0.3 to 0.5, and high correlations are from 0.5 to 1. Among college students, data showed a positive zero order correlation between MSCEIT total scores and course grades; however, th e correlation became nonsignificant when verbal

PAGE 14

14 Scholastic Aptitude Test ( SAT ) scores were statistically controlled (Barchard, 2003; Brackett & Mayer, 2003). While o ther data showed no correlation between MSCEIT scores and grade point av ittle, 2003), f indings of weak correlations in samples of college students may result from the restricted range of their IQ scores (Rivers, Brackett, Salovey, & Mayer, 2007). Among high school students, data showed that scores from the Spanish version of t he MSCEIT, which was administered at the beginning of the academic year, predicted final grades after general intelligence and personality were statistically controlled (Gil Olarte Mrquez, Palomera Martn, & Brackett, 2006). Among students between the age s of 10 and 18, data showed that scores from the youth version of the MSCEIT, which currently is under development, correlated moderately and significantly with an objective version of reading achievement; this represented the only study thus far that dire ctly measured academic achievement and EI among children and adolescents (Peters, Kranzler, & Rossen, 2009). O ther studies indirectly measured academic achievement using self reported SAT scores, GPA, or broad indicators of achievement such as final grades (Barchard, 2003; Brackett & Mayer, 2003; Gil Olarte Mrquez, Palomera Martn, & Brackett, 2006; To test the relationship between EI and academic achievement, studies should recruit from charter, magnet, private, public and speci al schools to obtain representative samples of students with a wide range of general intelligence scores. Then, associations between academic achievement and EI should be considered after accounting for general intelligence and personality. An intelligence test is expected to correlate with crystallized, fluid, and general abilities (Mayer et al., 1999) However, data indicate that ability based measures of

PAGE 15

15 emotional intelligence do not demonstrate consistent relationships with IQ tests. Among 39 adolescent s in the 11 th and 12 th grades at a residential school for gifted students, results showed a 0.29 correlation between EI, as measured by an unpublished ability based measure, and IQ, as measured by the Test of Cognitive Skills/Second Edition (Woitaszewski & Aalsma, 2004). Among adults, Ciarrochi et al. (2000) found a nonsignificant correlation between EI, as operationalized by an ability based measure, ability based EI corr elated significantly with personality dimensions of empathy and extraversion. Further, EI as assessed by the MSCEIT is associated with crystallized rather than fluid ability (Farrelly & Austin, 2007; MacCann, Roberts, Matthews, & Zeidner, 2004; Zeidner, Sh ani Zinovich, Matthews, & Roberts, 2005). This association may be an artifact of the verbal ability that is required to complete most MSCEIT subscales (Wilhelm, 2005). Nonetheless findings of inconsistent, low, and negative correlations with crystallized, fluid, and general abilities conflict with the speculation that ability based tests measure a specific intelligence. Mayer, Salovey, and Caruso (2000) distinguished between their own ability model of EI and models such as the Bar On model. Mixed Model of Emotional Intelligence Bar On (2006) postulated that because intra and interpersonal competencies comprise EI, a more accurate descriptor is emotional social intelligence. Emotional social intelligence (ESI) includes emotional and social compet encies, facilitators, and skills that determine the effectiveness with which we cope with daily demands, express and understand ourselves, and relate to as well as understand others (Bar On, 2006). The emotional and social competencies, facilitators, and s kills include five components.

PAGE 16

16 and feelings; (2) relate with others and understand their feelings; (3) control and man age emotions; (4) adapt, manage change, and solve int ra and interpersonal problems; and (5) generate positive affect and self motivate. The following paragraph describes each component in more detail. The first component includes intrapersonal acceptance, perception, and understanding of oneself as well as expression, freedom from emotional dependency on others, and the pursuit of personal interpersonal social awareness and understanding of includes cooperation with others, the establishment of mutually satisfying and pro social interpersonal relationships, and identification with a social group. The third component includes the control, management, and regulatio n of emotions, impulses, and stress. The fourth component includes external locus of thought, management of change, and in novel situations, and the resolution of intra and interpersonal problems. The fifth component includes feelings of contentment with life, oneself, and others as well as a self motivated positive outlook. Measurement Measurement within the mixed model is based on self perceptions of typical performance in relation to abilities, competencies, dispositions, personality traits, and skills. The Bar On Emotional Quotient Inventory (Bar On, 1997) and the Bar On Emotional Quotient Inventory: Youth Version (EQ i: YV; Bar On & Parker, 2000) operationalize emotional social intelligence based on the five components described in

PAGE 17

17 Bar adaptability, emotional and stress management, general mood, intrapersonal emotional expression and understanding, and the maintenance of interpersonal relationships in a self report questionnaire regarding emotional understanding and potential for success. The EQ i: YV requires that test takers respond to 60 short sentences by choosing the degree to which each item is true of them on a four point Likert type scale. Critical issues Emotional social intelligence includes broad social skills such as social responsibility, dispositional moods such as optimism, and emotion related abilities such as emotional self awareness as well as non ability personality traits such as assertiveness (Bar On, 2006). Some components are less akin to abilities and more like traits that describe behavioral preferences. Therefore, the descriptor emotional social intelligence seems questionable. Tra its beyond intelligence can predict success, yet most researchers object to classifying these characteristics as components of intelligence (Prez, Petrides, & Furnham, 2005). The inclusion of emotional and non emotional abilities and traits under the labe l emotional social intelligence creates Self report measures of EI have a degree of predictive validity, although the techniques assess dispositional constructs that are corre lated with personality and independent of intelligence (MacCann et al., 2003). Among university students, no significant correlations were found between full scale, performance, or verbal IQ and self reported EI (Saklofske, Austin, & Minski, 2003). Other d ata showed no significant correlations between cognitive ability and self reported EI (Newsome, Day, & Catano,

PAGE 18

18 2000). This supports the claim that emotional telligence as assessed by self reports is a multidimensional collection of at least some non co gnitive abilities. Dimensions of personality likely are the non cognitive abilities that are assessed by self reports of EI. Data showed a multiple correlation of 0.79 between the short version of the Bar On Emotional Quotient Inventory (EQ i: S; Bar On, 2 002) and a 50 item measure of the personality dimensions of the Five Factor Model obtained from the International Personality Item Pool, which indicated considerable redundancy between self reported EI and the five factor model of personality (Grubb & McDa niel, 2007). Among university students, moderate and significant correlations were found between self reported EI and the anxiety, extraversion, independence, and self control dimensions of the Sixteen Personality Factor Questionnaire (16PF; Cattell, Catte ll, & Cattell, 1993; Newsome et al., 2000). Bar On (1997) acknowledged significant correlations between subscales of the adult version of the Emotional Quotient Inventory (EQ i) and the 16PF. To support construct validity, the EQ i manual reported that the Emotional Stability dimension of the 16PF correlated significantly with optimism, self regard, stress tolerance, and total EQ i scores (Bar On, 1997). Oftentimes a high percentage of variance in self report measures of EI is attributed to personality var iables; nonetheless, the degree of association varies and correlations between EI and specific personality dimensions are inconsistent. Among university students, low to moderate correlations were found between self reported EI and all dimensions of the Fi ve Factor Model of personality (Saklofske et al., 2003). In another study of university students, moderate to high correlations were found between a different measure of self reported EI and four personality dimensions of the Five

PAGE 19

19 Factor Model (i.e., agree ableness, conscientiousness, extraversion, and neuroticism; Dawda & Hart, 2000). Despite the overlap between self reported EI and personality, the constructs are not identical, as self reported EI predicts important outcomes beyond the variance accounted for by personality. Among 1,186 business managers, self reported EI correlated substantially with agreeableness, autonomy, emotional stability, and extraversion while still showing significant incremental validity over the Five Factor Model in the predict ion of occupational competency (van der Zee & Wabeke, 2004). Among a sample of adults in the military, self reported EI accounted for incremental variance in job and life satisfaction even after controlling for the personality dimensions of the Five Factor Model (Livingston & Day, 2005). Among college students, self reported EI was associated with emotional distress and psychological well being and predicted health changes over time (Shulman & Hemenover, 2006). These relationships remained significant after statistically controlling for the personality dimensions of the Five Factor Model. Self reported EI also was a significant positive predictor of happiness after controlling for the effects of personality (Furnham & Christoforou, 2007; Furnham & Petrides, 2003). Among adolescents, parent ratings of EI accounted for incremental variance in life satisfaction and powerlessness, and peer ratings of EI accounted for approximately 10% of the variance in job performance (Law, Wong, & Song, 2004). These relationshi ps remained significant even after controlling for the personality dimensions of the Five Factor Model. While data suggest that the constructs measured by parent peer and self reports of EI have incremental validity over

PAGE 20

20 dimensions of personality few studies control for both intelligence and the Big 5 personality traits Academic Criteria a nd Self Reported Emotional Intelligence Studies of self reported EI and academic performance report mixed results. Data show that among university students, cognitiv e ability and the extraversion and self control dimensions of personality correlated significantly with academic achievement whereas self reported EI was not correlated significantly with academic achievement (Newsome et al., 2000). Among 64 college stude nts, a self report measure of EI completed at the beginning of the academic year significantly predicted official cumulative grade point average obtained at the end of the year, with a correlation of 0.32 (Schutte et al., 1998). In comparison, scores on th e same measure of EI showed a significant although low and negative partial correlation with high school rank after controlling for personality dimensions of the Five Factor Model and Verbal SAT scores ( r = 0.16; Brackett & Mayer, 2003). The negative corr elation defies expectations by indicating that students with a higher level of self reported EI achieve a lower high school rank. Other studies of younger populations report similarly mixed results. Among students between the ages of 13 and 21 years from B ahrain, no correlation was found between grade point average and self reported EI (Alumran & Punamki, 2008). In contrast, among students aged 11 and 12 years from England, higher levels of self reported EI were associated with better grade point averages, even after statistically controlling for general ability (Qualter, Whiteley, Hutchinson, & Pope, 2007). Among adolescent students between the ages of 14 and 18 years, overall self reported EI predicted approximately 16% of the variability in grade point a verage (Parker et al.,

PAGE 21

21 2004). However IQ scores, which explain approximately 25% of the variance in academic achievement (Neisser et al., 1996), were not collected for this sample. Also personality variables, which explain approximately 10% of the varianc e in academic achievement (Furnham & Chamorro Premuzic, 2003), were not collected. Presently, the relevance of self reported EI in educational contexts is questionable as no studies have tested the incremental validity of the construct when intelligence a nd personality are accounted for. To confirm the relationship between self reported EI and educationally important criteria, an examination of the incremental validity of a self report measure over measures of IQ and personality is necessary. Comparison of Ability and Mixed Models Low correlations have been found between ability based performance measures and mixed model self report measures of EI ( Brackett & Mayer, 2003; Livingston & Day, 2005 ). A meta analytic review found a correlation of 0.14 between ab ility and mixed me asures, which indicated that the measures evaluated relatively distinct constructs (Van Rooy, Viswesvaran, & Pluta, 2005). Even despite a common underlying theoretical framework, data showed weak correlations between the MSCEIT and the Se lf Report Emotional Intelligence Test (SREIT, based on the Mayer and Salovey model; Schutte et al., 1998), which further indicated a lack of correspondence between performance and self report measures of EI (Goldenberg, Matheson, & Mantler, 2006). In a com parison of the ability and mixed models of EI, measures of the ability model showed a higher correlation with cognitive ability whereas measures of the mixed model showed a higher correlation with personality (Van Rooy et al., 2005). To explain the low co rrelations that are found between ability based performance measures and mixed self report measures, Prez, Petrides, and Furnham (2005) hypothesized that the two

PAGE 22

22 types of measures assess conceptually different phenomena. Ability based performance measure s assess ability EI, which is considered to be embedded within the overall structure of psychometric intelligence. Ability based measures assess the ability to, for example, identify, express, and label emotions. Mixed self report measures assess trait emo tional self efficacy, which is considered to be embedded within the framework of personality. Trait emotional self efficacy is assessed by valid self report questionnaires that measure self perceived abilities, behavioral dispositions, and traits such as a ssertiveness, empathy, and optimism (Prez et al., 2005). Gender. Based on cultural norms and gender role socialization, the display of emotion often is valued in females while emotional control is valued in males (Brody, 2000). Thus, one expects differenc es between females and males on measures of emotional intelligence. Accordingly, females scored significantly higher than males on an ability based test of emotional intelligence ( Brackett & Mayer, 2003; Brackett, Mayer, & Warner, 2004; Brackett, Warner, & Bosco, 2005; Ciarrochi, Chan, & Caputi, 2000). Among university students, a significant negative relationship was found between ability EI, maladjustment, and negative behaviors for males and not females (Brackett, Mayer, & Warner, 2004). Within a mixed model of emotional intelligence, female adolescents reported significantly higher levels of overall emotional intelligence than male adoles cents (Harrod & Scheer, 2005). More specifically, female adolescents and adults reported significantly higher levels of interpersonal emotional intelligence than male adolescents and adults (Alumran & Punamki, 2008; Reiff, Hatzes, Bramel, & Gibbon, 2001). Also, girls were rated by their parents as having significantly higher levels of interpersonal

PAGE 23

23 and total emotional i ntelligence than boys (Santesso, Reker, Schmidt, & Segalowitz, 2006). Further, different relationships exist between emotion related constructs and cognitive ability. Among 873 adults, results of a study using Bar i showed a significant correlation between general mood and fluid intelligence for males; a significant correlation was found between adaptability and fluid intelligence for females (Derksen, Kramer, & Katzko, 2002). Assuming the EI constructs are valid, t he different relationships with fl uid intelligence indicate that self reported emotional intelligence may operate differently in males and females. However, extant empirical data are mixed as other studies reported no gender differences among female and male college students on a self repo rt measure of emotional intelligence (Brackett & Mayer, 2003; Saklofske, Austin, & Minski, 2003). Mixed results indicate that additional research is needed to more fully understand gender based differences in emotional intelligence. Summary The ability mod el, which is assessed by performance based measures, places EI within the structure of human cognitive abilities Carroll (1993) used factor analysis with large data sets to support a three stratum theory of human cognitive abilities. At stratum three, the general level, is g At stratum two, the broad level, are eight factors including crystallized and fluid intelligences as well as auditory perception, cognitive speed, general memory and learning, processing speed, retrieval ability, and visual perception At stratum one, the specific level, are at least 69 narrow abilities (e.g., listening ability, working memory ). Ability based e motional intelligence is hypothesized to be a stratum two ability akin to spatial or verbal intelligence. Mixed models of EI w hich are assessed by self report measures, place EI across the spheres of general abil ity and personality.

PAGE 24

24 intelligence may be a misnomer as emotions interact with competencies, dispositions, skills, and traits. The necessity of fostering stude intuitively and may be based on claims that emotional intelligence affects educational and occupational outcomes. However, claims that EI or similar constructs influence occupational performance may be oversta ted in consideration of a reliance on anecdotal evidence, case studies, expert opinion, and unpublished proprietary surveys (Zeidner, Matthews, & Roberts, 2004). Also, assertions about the relationship between academic success and emotional intelligence may be premature as there is no agreed upon definition of the construct, whether it represents more than personality variables remains unclear, and there is disagreement over whether the construct should be assessed by ability based maximum performance or self report measures (Humphrey, Curran, Morris, Farrell, & Woods, 2007; Matthews, Roberts, & Zeidner, 2004; Mat thews, Zeidner, & Roberts, 2005 ). To address these issues, the present study adopts the theoretical perspective of Prez, Petrides, and Furnham ( 2005) who proposed that ability based EI as assessed by maximum performance measures and emotional intelligence as assessed by self report measures are conceptually distinct. The present study defines self reported emotional intelligence as abilities, beh avioral dispositions, and traits that are assessed by a validated self report questionnaire that measures typical performance. Purposes of the Current Study Debate exists over the extent to which self reported EI adds unique variance beyond measures of gen eral intelligence and personality in the prediction of educationally important criteria (Humphrey, Curran, Morris, Farrell, & Woods, 2007;

PAGE 25

25 Matthews, Roberts, & Zeidner, 2004; Matthews, Zeidner, & Roberts, 2005). To address this issue, the present study eva luated the incremental predictive validity of self reported EI over general intelligence and personality in the prediction of academic achievement in 50 children and parent rated behavior in 38 children (aged 10 14 years in grades 5 through 8) The aim was to determine the extent to which self reported EI added unique variance beyond measures of general intelligence and personality in the prediction of academic achievement, behavior, and other educationally important criteria. To accomplish the purposes of the current study, detailed assessments including objective and subjective information were completed to determine whether self reported EI remained predictive after accounting for general intelligence and personality factors. Tests of the following objec tives and corresponding research questions were conducted. Main Objective 1: Understanding the Predictive Validity o f Self R eported Emotional Intelligence a mong Children The main objective was directed at better understanding the predictive validity of a s elf report measure of EI. Research question 1 addressed the extent to which self reported EI correlated with academic achievement, behavior, general intelligence, and personality characteristics. Research question 2 addressed whether self reported EI incre mentally predicted academic achievement, parent rated behavior, and in school and out of school suspensions for misbehavior beyond the variance predicted by intelligence and personality (see Figure 1 1 ).

PAGE 26

26 Main Objective 2: Characterizing Gender Based Diffe rences i n Emotional Intelligence A secondary objective was directed at better understanding the gender based differences of an emotion related construct in youth. Research question 3 addressed whether females and males performed differently on a self repor t measure of EI. Research question 4 addressed whether between gender differences existed in the relationship of self reported EI and the criterion variables. Summary of the Objectives a nd Purposes Clarification was sought regarding the relationships betw een academic achievement, behavior, general intelligence, personality, and self reported EI. Data were collected from children. Considering that little data from youth exist, a thorough analysis of the construct may help lay the foundation for the study of children and add to the literature evidence for the predictive validity of self reported EI in educational academic development; yet, to improve academic, health, and social outcomes, policies coordinated academic, emotional, and social learning. Results of the current study may provide a comprehensive and research based model of system atic evaluation, which is necessary for the development and sustainability of appropriate education, intervention, and prevention strategies that diminish the risk for negative outcomes.

PAGE 27

27 Figure 1 1. Operational model Predicto r Variables General intelligence Personality Self reported emotional intelligence Criterion Variables Academic achievement Behavioral ratings In school suspensions Out of school suspensions

PAGE 28

28 CHAPTER 2 METHODS Sampling Procedures The sample was drawn from communities and schools in North Central Florida and Central New York A three stage sampling procedure was used (communities, schools within communities, students within schools). Throughout the project, communities, s chools, and students were solicited for participation through advertisements, direct contact with schools, and local postings. The sample included participants from the major ethnic, racial, and socioeconomic categories that represent the diverse populatio n of the State s of Florida and New York Participants Participants were 50 students attending public schools in grades 5 through 8 ( 26 female, 24 male). The a g e of participants ranged from 10 to 14 years ( M = 11.84, SD = 1.1) The majority of participants were White/non Hispanic (50%) The remainder were African American (36%) Latino American (8%) and Asian American (6%). A majority of the sample received fre e or reduced lunches (64%) while full price lunch status was noted for the remaining participants (36%). All participants were treated in accordance Psychological Association, 2002). Exclusionary criteria. Potential participants were drawn from the general education populatio n. Based on a review of records, those who were retained for three or more years were excluded. Also, potential participants with a Full Scale IQ less than 70, a sensorimotor disability (e.g., deafness, blindness), a severe learning disability in reading, or a r eading level below fourth grade were excluded. Reading level was

PAGE 29

29 determined based on grade equivalent scores from the reading subtest of the Kaufman Test of Educational Achievement Second Edition, Brief Form (Kaufman & Kaufman, 2005) I nstruments Demographic questionnaire. A demographic questionnaire was administered to collect information on age, gender, and grade level. Predictor measures. Detailed assessments including objective and subjective information were completed to determine whether self reported emotional intelligence remained predictive of educationally important criteria after accounting for general intelligence and personality factors. Bar On Emotional Quotient Inventory: Youth Version. The Bar On Emotional Quotient Inve ntory: Youth Version (EQ i: YV; Bar On & Parker, 2000) was utilized to measure emotional intelligence for 50 participants The EQ i: YV was designed to assess the adaptability, coping skills, and well being of children and adolescents. Normative data were collected from a sample of 9,172 school aged children and adolescents from several different elementary, middle, and high schools across the United States and Canada. The self report inventory can be administered to individuals from age 7 to 18 years. A to tal score is derived from responses on four scales. The Adaptability Scale, managing change and solving problems. The Interpersonal Scale, which has 12 items, measures appreciate and understand the emotions of others. The Intrapersonal Scale, which has

PAGE 30

30 understand em otions. The Stress Management Scale, which has 12 items, measures events. The EQ i: YV also includes an Inconsistency Index a General Mood Scale, and a Positive Impr ession scale. The Inconsistency Index notes discrepancies in responses to similarly worded items, which may indicate that the respondent answered in a careless way or misunderstood the instructions. The General Mood Scale includes 14 items that are related to overall mood including depression, dysphoria, optimism, and pessimism. The Positive Impression scale, which has 6 items, measures the likelihood that a respondent answered in a way that creates an overly positive self impression. The EQ i: YV, which c an be group or individually administered, takes approximately 20 to 30 minutes to complete and the readability is at a North American fourth grade reading level. Participants respond to statements that best describe the way they act, feel, or think in sit uations. They answer questions about themselves by selecting the most appropriate response from a four point Likert scale. Responses report inventory is administered in paper and pencil format, online using the EQ i: YV Online program, or via the computer using EQ i: YV software. Responses on the paper and pencil format are hand scored by the administrator. Responses on the computer and online formats are computer scored by the test publisher. Reliability. for all domains, with the lower coefficients for the six item Intrapersonal domain ( Ballard, 2003). Three week test retest reliability coefficients vary fro m 0.84 for the Intrapersonal Scale to 0.89 for the total Emotional Intelligence score (Bar On & Parker, 2000). Three

PAGE 31

31 week test retest reliability coefficients varied from 0.77 to 0.89 in an analysis of 60 children (Ballard, 2003). Validity. Bar On and Pa rker (2000) reported that the EQ i: YV has a replicable factor structure and the scales correlate with comparable scales on the adult version of the inventory. Results of research generally support the validity of interpreting the four scales as measures o f their intended construct. Factorial analysis revealed that items produced a four factor solution (Ballard, 2003). Several scales on the EQ i: YV have been found to correlate with academic success (Parker, et al., 2004). N egative correlations exist betwe en the EQ i: YV and measures of depression and psychopathology in children (Ballard, 2003). A correlation of 0.85 was found between the Stress Management domain of the EQ i: YV and anger control problems. Five Factor Personality Inventory Children. The Fiv e Factor Personality Inventory Children (FFPI C; McGhee, Ehrler, & Buckhalt, 2007) was utilized to measure personality for 50 participants The FFPI C was designed to assess the Five Factor Model of personality and is the first such measure to be standardi zed on and appropriate for use with children and adolescents. Normative data were collected from a sample of 1,284 youth from 18 states. The demographic characteristics of the normative sample approximated the total American population based on 2000 and 20 02 census data according to age, ethnicity, exceptionality status, gender, geographic region, and race. The self report inventory can be administered to youth from age 9 to 18 years, 11 months. Five total scores are derived from responses on five scales o f agreeableness, conscientiousness, emotional regulation, extraversion, and openness to experience,

PAGE 32

32 which correspond with the Five Factor Model of personality. Each of the FFPI scales is comprised of 15 items. Each of the 75 items consists of two sentences which describe opposite feelings and are separated by five Likert type response options. The FFPI C, which can be group or individually administered, takes approximately 15 to 40 minutes to complete and the readability is at a third grade reading level. Responses on the paper and pencil self report inventory are hand scored by the administr ator. Then, raw scores can be converted into corresponding percentile ranks, quotients, stanines, T scores, and z scores. Reliability. from 0.74 to 0.86 across the normative sample. T est items are homogeneous and not biased toward different groups (Klingbeil, 200 9 ). Consistency of responding over a two week interval was measured in a sample of 192 youth aged 9 to 18 years. Over a short time period, test retest reliability correlations bet T scores ranged from 0.81 to 0.92 Validity. Support for content validity derives from test construction procedures. An initial item pool of 100 items was based on a literature review of the relevant aspects of each of the Big Five perso nality dimensions. Experts in assessment, child development, education, and personality theory commented on the potential bias, readability, and theoretical utility of each item, which led to the omission of 10 items. Item discrimination studies led to the selection of the 15 most valid items for each scale. Median item discrimination coeffi cients ranged from 0.44 to 0.59

PAGE 33

33 The internal structure was examined through intercorrelations among the scales and conf irmatory factor analysis (CFA), which showed c or relations between the scales. Normative data were analyzed with confirmatory factor analy sis that computed four model fit indices. Results indicate that the structure of the FFPI C and the Five Factor Model are closely aligned. Thus, some evidence suggests that the test produces data that accurately measure the Five Factor Model of personality (Klingbeil, 200 9 ); however, a critical review of the FFPI C indicates that the description of CFA procedures and the presentation of results in the manual are incompl ete. Also, the manual does not mention exploratory factor analyses. Thus, there is limited support for validity and professionals may have difficulty when independently evaluating the adequacy of the factor structure of the scales (Suldo & Stewart, 2008). Convergent validity was assessed through the measurement of the FFPI concurrent relationships with other personality inventories. The NEO Five Factor inventory is a self report inventory developed for use with adults. The FFPI C manual reports concurr ent relationships with the NEO Five Factor inventory in 52 adolescents aged 15 to 18 years. Scales that assess the same dimensions show correlations that range from 0.59 to 0.45. The Junior Eysenck Personality Questionnaire (JEPQ) is a self report inventor y developed for use with children and adolescents; the JEPQ assesses two dimensions of personality that also are assessed by the FFPI C: extraversion and neuroticism. The manual reports concurrent relationships between the FFPI C and the JEPQ in 184 childr en and adolescents aged 9 to 18 years. The extraversion and neuroticism/emotional regulation scales showed correlations of 0.62 and 0.50, respectively.

PAGE 34

34 The concurrent relationship between FFPI C scores and academic achievement was measured. Correlations between conscientiousness scale scores, scores on the Georgia Criterion Referenced Competency Test and grade point averages ranged from 0.40 to 0.55, which indicates there are relationships between FFPI C scores and academic achievement. Concurrent, conv ergent relationships between FFPI C scales and academic achievement, grade point average, and cognitive ability also were measured. The FFPI C manual reports correlations between conscientiousness and indicators of academic achievement including statewide achievement and competency tests as well as overall grade point averages (Suldo & Stewart, 2008). Among a small sample of 25 adolescents in high school, FFPI C scales of agreeableness, conscientiousness, emotional regulation, and openness to experience sho wed correlations with general cognitive, nonverbal, and verbal abilities as measured by the Hammill Multiability Intelligence Test. Thus, the FFPI C manual provides preliminary evidence that supports the validity of interpreting the scale as a measure of i ts intended construct. Woodcock Johnson III Tests of Cognitive Abilities. The Brief Intellectual Ability scale of the Woodcock Johnson III Tests of Cognitive Abilities (WJIII COG; Woodcock, McGrew, & Mather, 2001) was utilized to measure general intelligen ce for 50 participants The WJIII COG was designed to measure general and specific cognitive functions. Normative data were collected from a nationally representative sample of 8,818 individuals (McGrew & Woodcock, 2001). The kindergarten through 12 th grad e sample was comprised of 4,783 children and adolescents. Data came from over 100

PAGE 35

35 geographically diverse U.S. communities. The instrument can be administered to individuals from age 2 to 90 or more years. The Brief Intellectual Ability (BIA) scale score i s derived from performances on three subtests: Concept Formation, Verbal Comprehension, and Visual Matching. The BIA, which is administered individually, takes approximately 15 minutes to complete. The subtests are hand scored by the administrator during t esting to determine basal and ceiling levels. All responses are recorded verbatim. Then, raw scores are converted into age and grade equivalents, percentile ranks, and discrepancy scores through the use of scoring tables or computer scoring software. Reli ability. The mean internal consistency reliability coefficient of the BIA for ages 14 to 18 years is 0.96 (McGrew, Shrank, & Woodcock, 2007). The mean reliability coefficients for individual tests of the BIA for ages 14 to 18 years are 0.95 for Concept For mation, 0.92 for Verbal Comprehension, and 0.86 for Visual Matching. One day test retest reliability of Visual Matching for ages 14 to 17 years is 0.75 (McGrew et al., 2007). For individuals between the ages of 8 and 18 years at first testing, the mean one to two year test retest correlation for Visual Matching is 0.89. For individuals between the ages of 8 and 18 years at first testing, the mean one to two year test retest correlation for Concept Formation is 0.76. Validity. The technical manual reports su fficient evidence to support the validity of interpreting scores from the WJIII COG as measures of their intended constructs (McGrew & Woodcock, 2001). The battery was developed to address the general and specific abilities of the Cattell Horn Carrrol (CHC ) cognitive model the theory of intelligence with the most empirical support, which indicates a general intellectual factor

PAGE 36

36 exists and is comprised of several narrow and specific abilities. The three tests that comprise the BIA were selected because they represent different areas of crystallized and fluid intelligence. Concept Formation primarily represents the broad CHC factor of fluid reasoning and the narrow ability of induction. Verbal Comprehension primarily represents the broad CHC factor of comprehe nsion knowledge and narrow abilities of language development and lexical knowledge. Visual Matching primarily represents the broad CHC factor of processing speed and the narrow ability of perceptual speed. Support for the internal structure of the WJIII CO G was obtained through correlation analysis (McGrew & Woodcock, 2001) For example, tests such as Academic Knowledge, General Information, and Verbal Comprehension, which hypothetically would be highly related according to CHC theory, showed intercorrelati ons of 0.80 or higher. They were less related to tests such as Picture Recognition and Spatial Relations, showing intercorrelations of 0.25 to 0.36. Additional support for the internal structure was obtained through confirmatory factor analyses. In compari son to six alternative models, the CHC model demonstrated the closest fit to the hypothesized structure. Additional support for the validity of the measure derives from correlation studies of the relationships between the WJIII COG and other measures. Conc urrent validity with the Wechsler Intelligence Scales for Children, Third Edition is 0.69. Concurrent validity with the Das Naglieri Cognitive Assessment System is 0.70 (McGrew & Woodcock, 2001) Criterion measures. Objective academic achievement and behav ioral data were included as well as the behavioral ratings of parents to determine whether self reported

PAGE 37

37 emotional intelligence was predictive of important criteria after accounting for general intelligence and personality. Kaufman Test of Educational Achi evement Second Edition, Brief Form. The Kaufman Test of Educational Achievement Second Edition, Brief Form (KTEA II BF; Kaufman & Kaufman, 2005) was utilized to measure academic achievement for 50 students The instrument includes three subtests that asses s the academic domains of mathematics, reading, and written expression for children, adolescents, and adults aged 4 years, 6 months to 90 years. The instrument was standardized on a large representative sample that closely corresponded to the 2001 United S tates Bureau of the Census population survey. The grade norm sample consisted of 1,645 students from kindergarten through 12th grade. The total age sample consisted of 2,495 individuals aged 4 years, 6 months through 90 years. The normative sample matched the U.S. population on education level of the examinee or parent, gender, geographic region, and race/ethnicity. The KTEA II BF yields norm referenced subtest scores for math, reading, and writing as well as a global Brief Achievement Composite (BAC) that is derived from performances on all three subtests (Kaufman & Kaufman, 2005). The math subtest includes 67 application and computation items that assess skills from addition and subtraction to square roots, exponents, and algebra. The use of a paper and p encil is allowed for all items. The reading subtest has two parts: the Recognition part includes 27 letter and word recognition items; the Comprehension part includes 46 passage des a

PAGE 38

38 broad measure of skills in written expression, including 46 items that focus on capitalization, grammar, punctuation, sentence structure, and spelling. The instrument, which is administered individually, takes approximately 15 to 45 minutes to compl ete (Lichtenberger & Smith, 2005). The administrator determines criteria. The KTEA II BF is computer or hand scored. Raw scores are converted to derived scores includi ng standard scores, percentiles, normal curve equivalents, stanines, and age and grade equivalents. Reliability For subtests scored using age norms, average internal reliability values are 0.95, 0.91, and 0.90 for reading, math, and writing, respectively (Kaufman & Kaufman, 2005). For subtests scored using grade norms, average reliability values are 0.94, 0.90, and 0.86 for reading, math, and writing, respectively. The split half reliability value for the BAC is 0.96 for both age and grade norms. Test r etest reliability was examined by administering the test to 327 students twice within a 2 to 8 week period (mean interval of 3.7 weeks; Kaufman & Kaufman, 2005). Across age and grade norm samples, the average adjusted test retest reliabilities were 0.94, 0 .93, 0.90, and 0.81 for the BAC, reading, math, and writing, respectively. Validity To demonstrate the validity of interpreting the KTEA II BF test scores : intercorrelations between composites and subtests were calculated to show the relationships betwe en the academic domains; factor analyses were conducted to show that the structure of the instrument was empirically grounded; correlations with other instruments were calculated to evaluate construct validity; and special population

PAGE 39

39 studies were conducted to show the efficacy of using the instrument to assess children with learning disabilities, mental retardation, attention deficit hyperactivity disorder, and other qualities such as deafness and giftedness (Lichtenberger & Smith, 2005). Correlations also were calculated between the Brief and Comprehensive Forms of the KTEA II in a sample of 1,318 students (Kaufman & Kaufman, 2005). Correlations were 0.89, 0.86, 0.85, and 0.78 for the BAC, math, reading, and writing, respectively. Correlations between the KTEA II BF and Wide Range Achievement Test Third Edition (WRAT3; Wilkinson, 1993) were examined in a sample of 80 students aged 7 to 19 years (Kaufman & Kaufman, 2005). The correlation between the reading subtests of the two measures was 0.84; the correla tion between the math subtests was 0.75; the correlation between the Writing subtest of the KTEA II BF and the Spelling subtest of the WRAT3 was 0.64. Correlations between the KTEA II BF and the Woodcock Johnson Third Edition tests of Achievement (WJIII T A; Woodcock et al., 2001) were examined in a sample of 25 students aged 7 to 16 years (Kaufman & Kaufman, 2005). The correlation between the BAC of the KTEA II BF and the Academic Skills composite of the WJIII TA was 0.89; the correlation between the Writi ng subtest and Broad Written Language was 0.79; the correlation between the Reading subtest and Broad Reading was 0.78; and the correlation between the Math subtest and Broad Math was 0.74. Correlations between the KTEA II BF and the Kaufman Assessment Ba ttery for Children, Second Edition (KABC II; Kaufman & Kaufman, 2004a), Kaufman Brief Intelligence Test Second Edition (KBIT 2; Kaufman & Kaufman, 2004b), and Wechsler Abbreviated Scale of Intelligence (WASI; The Psychological Corporation, 1999) were

PAGE 40

40 calcu lated and showed that the Math and Reading subtests correlated more strongly with global cognitive ability than did the Writing subtest (Lichtenberger & Smith, 2005). Correlations between the BAC and the KBIT 2 IQ Composite, WASI Full Scale IQ 4, KABC II F luid Crystallized Index, and KABC II Mental Processing Index were 0.78, 0.71, 0.70, and 0.70, respectively (Kaufman & Kaufman, 2005). Parent Rating Scales of the Behavior Assessment System for Children Second Edition. The Parent Rating Scales (PRS) compone nt of the Behavior Assessment System for Children Second Edition (BASC 2; Reynolds & Kamphaus, 2004) was utilized to measure the behavior and emotional characteristics of 38 participants. The BASC 2, which is a revision of the widely accepted Behavior Asse ssment System for Children (BASC; Reynolds & Kamphaus, 1992), was designed to evaluate the behavior of children, adolescents, and young adults aged 2 through 25 years. The instrument is multidimensional as it measures various characteristics of behavior an d personality including adaptive and positive as well as clinical and negative dimensions. The PRS component of the BASC 2 is a psychometrically sound and reliable measure of observable behaviors in community and home settings. The general combined sex nor ms are based on a large national sample that is representative of the general population of U.S. children according the 2001 Current Population Survey in terms of clinical or special education classification, geographic region, parent education, and race/e thnicity. The PRS general norm sample was comprised of 4,800 children and adolescents aged 2 through 18 years. Standardization occurred between August 2002 and May 2004.

PAGE 41

41 The PRS, which is designed to be completed by the parent, guardian, foster parent, or custodial caregiver with the most frequent and recent contact with the participant, has three forms that are targeted at three corresponding age levels: preschool (2 to 5 years), child (6 to 11 years), and adolescent (12 to 21 years; Reynolds & Kamphaus, 2 004); in the present study, only the child and adolescent forms were used as they correspon d to the targeted age range of 10 to 13 years. The forms contain descriptors of behaviors that a caregiver rates on a four point scale of frequency ranging from Neve r to Almost Always The child form includes 160 items and the adolescent form includes 150 items. The PRS, which is written at approximately a 4th grade reading level, is available in English and Spanish and takes 10 to 20 minutes to complete. The instrume nt is individually administered in paper and pencil format and can be computer or hand scored. Raw scores are converted to T scores and percentile ranks. Parent ratings on the PRS produce composite as well as primary and content scale sco res, which are lis ted in Table 2 1 The PRS also includes three validity scales: an F or fake bad index detects a negative response pattern; the Response Pattern Index detects forms that may be invalid because the caregiver disregarded the content of the items; and the Cons istency Index detects internally inconsistent forms on which the caregiver frequently responded differently to similar items. If responses elicited Caution or Extreme Caution on these indices, then another caregiver was asked to complete the BASC 2 PRS or the data were excluded from analyses. Reliability Internal consistency reliabilities of the PRS composites and scales are consistent between females and males and across different age levels, which indicates that the composite and scale scores are in dicators that yield reliable scores for their

PAGE 42

42 behavioral dimensions. For the general norm samples for individuals aged 8 through 18 years, mean composite score reliabilities show coefficient alphas of 0.95, 0.95, 0.94, and 0.91 for the Behavioral Symptoms Index (BSI), Adaptive Skills, Externalizing Problems, and Internalizing Problems, respectively (Reynolds & Kamphaus, 2004). Mean individual scale score reliabilities range from 0.74 for Activities of Daily Living to 0.88 for Social Skills, with most coeffi cient alphas equal to 0.81 or higher. Test rete st reliabilities of the PRS indicate to change greatly at different times. On the child form of the PRS, a sample of 77 children (median age of 9 years, 0 months) was r ated twice by the same caregiver with an interval of 10 to 69 days (median interval of 46 days) between ratings (Reynolds & Kamphaus, 2004). Test retest reliabilities for the composite scores generally ranged from 0.78 for Internalizing Problems to 0.92 fo r the BSI. Reliabilities for the individual scales ranged from 0.65 for Somatization to 0.87 for Conduct Problems. On the adolescent form, a sample of 88 children (median age of 15 years, 8 months) was rated twice by the same caregiver with an interval of 23 to 62 days (median interval of 41 days) between ratings. Test retest reliabilities for the composite scores ranged from 0.83 for the BSI to 0.90 for Internalizing Problems. Reliabilities for the individual scales ranged from 0.75 for Adaptability, Hyper activity, and Withdrawal to 0.88 for Conduct Problems. Evaluations of interrater reliability also were conducted. On the child form of the PRS, a sample of 43 children (median age of 9 years, 2 months) was rated by two different caregivers (Reynolds & Kamp haus, 2004). Interrater reliabilities for the composite ranged from 0.68 for the BSI to 0.77 for Adaptive Skills. Reliabilities for the

PAGE 43

43 individual scales ranged from 0.53 for both Aggression and Somatization to 0.80 for Anxiety. On the adolescent form, a s ample of 51 adolescents (median age of 15 years, 2 months) was rated by two different caregivers. Interrater reliabilities for the composite scores ranged from 0.65 for Internalizing Problems to 0.86 for Adaptive Skills. Reliabilities for the individual sc ales ranged from 0.55 for Somatization to 0.84 for Activities of Daily Living. Generally, the correlations indicated that different caregivers are not expected to provide identical ratings as occasions and settings for observations of children vary and chi ldren may behave differently in the presence of different caregivers. Validity Evidence for the validity of interpreting scores from the PRS as measures of their intended constructs is based on score profiles of clinical groups as well as the content of t he scale, which relates to standard diagnostic systems (e.g., the Diagnostic and Statistical Manual of Mental Disorders ), expert opinions, and the perceptions of parents (Reynolds & Kamphaus, 2004). Further evidence for the validity of interpreting scores from the scale as measures of their intended construct is based on empirical support from scale intercorrelations, factor analyses for the grouping of scales into composites, and patterns of correlations with other instruments. As expected, positive correl ations existed within adaptive and clinical scales whereas negative correlations generally existed between adaptive and clinical scales (Reynolds & Kamphaus, 2004). Confirmatory factor and principal axis analyses provided support for the factor structure o f the PRS. All scales of the Externalizing Problems composite had high loadings; the scales most affecting the Internalizing Problems composite were Anxiety, Atypicality, and Depression; and the scales most

PAGE 44

44 affecting the Adaptive Skills composite were Acti vities of Daily Living, Functional Communication, Leadership, and Social Skills. The PRS and the Achenbach System of Empirically Based Assessment Child Behavior Checklist for ages 6 through 18 years (ASEBA CBC; Achenbach & Rescorla, 2001) were completed fo r 65 children aged 6 through 11 years and 67 adolescents aged 12 through 18 years (Reynolds & Kamphaus, 2004). Correlations between s imilarly named composites ranged from 0.67 for Internalizing Problems to 0.84 for the BSI. Correlations between s imilarly n amed scales ranged from 0.34 for Anxiety to 0.77 for Aggression. The PRS and Conners' Parent Rating Scale Revised (Conners, 1997) were completed for 60 children aged 6 through 11 years and 55 adolescents aged 12 through 18 years. Correlations between s imil arly named composites ranged from 0.65 for the BSI on the adolescent form to 0.79 for the BSI on the child form. Correlations between s imilarly named scales ranged from 0.35 for Anxiety to 0.84 for Aggression. The correlations with other instruments provid es support for the validity of interpreting scales of the PRS as measures of their intended constructs For the standardized measures used in the current study, Table 2 2 presents their readability, length of administration, and whether they can be group or individually administered. In the current investigation all measures were individually administered. Records review. Documents and records were reviewed to gather additional information. The analysis of relevant documents and records helped to dete rmine whether self reported emotional intelligence predicted unique variance in educationally important criteria after accounting for general intelligence and personality. Information was collected regarding nd in school and

PAGE 45

45 out of school suspensions for misbehavior. The information on suspensions was collected for the preceding year; for example, for a student in grade 8, the number of Data also w er e collected on free and reduced lunch status as an index of socioeconomic status. Procedure Schools and students in North Central Florida and Central New York that agreed to participate were identified. A purposeful sample of potential participants was dr awn from students who met eligibility criteria (e.g., age and absence of exclusionary criteria). Caregivers and students were contacted for consent. They were informed that their participation was voluntary and they could withdraw at any time. Caregivers and students were briefed about the purposes of the study as well as potential benefits and risks. They were encouraged to ask questions about potential benefits or risks, the study, and their participation. Assent and consent forms were read to caregivers and students. Students were asked for assent; caregivers were asked to sign an informed consent form approved by the University of Florida. Caregivers and students were told that the results of the study may be published. However, all identifying informat ion would be removed. To ensure confidentiality, students were assigned individual identification numbers to be used for the entirety of their participation. Following the receipt of assent and consent, data regarding age, gender, and grade level were coll ected through a demographic questionnaire. Next, a detailed document and records review was conducted. Data regarding parti race and grade retentions were recorded. Data on free and reduced lunch status were recorded as an index of s ocioeconomic status. Information regarding in school and out of school suspensions also was collected from school databases. The BASC 2 PRS

PAGE 46

46 was directly administered or mailed to caregivers along with the following directions: need information from you about their behavior as you see it. Please read the instructions on this form, and respond to all items, even if some are difficult to answer or do not seem to apply. If yo There was a 76% return rate for the BASC 2 PRS forms; thus, all analyses utilizing the BASC 2 PRS were conducted with 38 participants. Then, testing was gro up and individually administered over several testing periods. First, the KTEA II BF was individually administered. Reading level was determined II BF. Participants with a reading level below fourth grade were excluded. Next, t he self report measures of emotional intelligence and personality were group administered whenever possible. The n, the objective measure of intelligence was individually administered. Six test administ ration orders were used to account for order effects. The forms were: ( a) personality, self reported emotional intellige nce, general intelligence ; ( b ) general intelligence, personality, self report ed emotional intelligence; ( c ) self reported emotional inte lligence, general intelligence, personality ; ( d) personality, general intelligence, self reported emotional intelligence; ( e) self reported emotional intelligence, personality, general intelligence; and ( f) general intelligence, self reported emotional int elligence, personality The author conducted all evaluations. Data collection for each participant occurred over a time period as long as 2 weeks; although, collection occurred for several participants concurrently.

PAGE 47

47 Data Analysis Descriptive statistics we re calculated for all demographic, criterion and predictor variables. Analyses with the EQ i: YV, FFPI C, KTEA II BF, and WJIII COG BIA included all 50 participants. Analyses with the BASC 2 PRS included the 38 participants whose caregivers returned the b ehavioral rating forms. T he reliability of the data was determined through the examination of studentized distributions of variance. Pearson product moment correlation coefficients were calculated to examine relationships between all variables and scor es o n the EQ i: YV. To examine group differences based on gender, t tests were conducted with with gender as the independent variable and total score from the EQ i: YV as the dependent variable Next a series of multiple regression analyses were conducted to examine the incremental validity of the total score from the EQ i : YV over the Brief Intellectual Ability score from the WJIII COG and personality scores from the FFPI C in predicting the criterion variables (academic achievement as measured by the KTEA II BF and behavior as measured by the BASC 2 PRS and suspensions). The BIA of the WJIII COG was entered in the first block, five scale scores of the FFPI C were entered in the second block, and the total score from the EQ i: YV was entered in the third block to determine whether the addition of self reported emotional int elligence after general intelligence and personality added significantly to the prediction of the criterion variables. Because multiple measures were used in the regression analyses, increments in R 2 were expected to be small. However, small increments may indicate contribution to the prediction of the criterion (Hunsley & Meyer, 2003). A series of moderated regression analyses also were conducted to examine whether gender moderated the relationship between self reported EI as measured by the total score of

PAGE 48

48 the EQ i VY and the criterion variables of achievement as measured by math, reading, writing, and total achievement scores on the KTEA II BF and behavior as measured by the adaptive skills, behavioral symptoms, externalizing problems, and internalizing pr oblems composite scores of the BASC 2 PRS. Analyses were conducted at the p < 0.05 level of significance.

PAGE 49

49 Table 2 1. Composites, Content, and Primary Scales of the Paren t Rating Scales of the Behavior A ssessment System for Children Second Edition Primary Adaptive /Clinical Scales Content Scales Composites Adaptability Anger Control Adaptive Skills Activities of Daily Living Bullying Externalizing Problems Functional Communication Developmental Social Di sorders Behavioral Symptoms Index Leadership E m otional Self Control Internalizing Problems Social Skills Executive Functioning Aggression Negative Emotionality Anxiety Resiliency Attention Problems Atypicality Conduct Problems Depression Hyperactivity Somatization Withdrawal Table 2 2. Administration and readability of standardized measures used in the current study Measure Administration Length of Administration Readability EQ i: YV a Group or individually administered 20 to 30 minutes Fourth grade FFPI C b Group or indiv idually administered 15 to 40 minutes Third grade WJIII COG BIA c Individually administered 15 minutes KTEA II BF d Individually administered 15 to 45 minutes BASC 2 PRS e Individually administered 10 to 20 minutes Fourth grade Note: a EQ i: YV = Bar On Emotional Quotient Inventory: Youth Version; b FFPI C = Five Factor Personality Inventory Children; c WJIII COG BIA = Woodcock Johnson III Tests of Cognitive Abilities Brief Intellectual Ability; d KTEA II BF = Kaufman Test of Educational Achievement Second E dition Brief Form; e Behavior Assessment System for Children Second Edition Parent Rating Scales

PAGE 50

50 CHAPTER 3 RESULTS Results for this study are presented in five sections. The first section presents the descriptive statistics for the criterion and predicto r variables. The second section presents results of the t tests based on gender The third section presents results of correlation analyses that investigated the relationships between self reported EI and the criterion and predictor variables. The fourth s ection presents results of a series of sequential multiple regression analyses and the fifth section presents results of a series of moderated regression analyses. Descriptive Statistics Criterion Variables Table 3 1 displays descriptive s tatistics for cr iterion variables. The mean scores and standard deviations of the 50 participants on the KTEA II BF (Math M = 99.5 SD = 14.9 ; Reading M = 99.9 SD = 11.3 ; Writing M = 99. 1 SD = 12.9 ; Total Achievement M = 99.3 SD = 12.1 ) were comparable to those from th e standardization sample ( M = 100, SD = 15). Reliability analyses allowed for the study of the properties of the KTEA II BF the items that composed the scales and the scales that composed the total achievement composite score Internal consistency was ca alpha, which calculates a reliability c oefficient that ranges between 1 and 1 and is based on average inter item correlations. Coefficients equal to or greater than 0.70 generally indicate acceptable scale reliability (Kline, 199 9) Coefficients for the Math, Reading, Writing, and Total Achievement scales ranged from 0.71 to 0.94, which indicated that the scales had acceptable internal consistency. With the exception of Math ( = 0.94), coefficients for the scales generally fell below the average internal

PAGE 51

51 reliability values reported from the standardization sample (Kaufman & Kaufman, 2005). This likely is due to range restriction (e.g., standard deviations of 11 to 12 points fo r the current sample versus 15 points for the standardization sample). Sample data for suspensions indicate that among the sample of 50 students, t he average participant had no suspensions while others were suspended between 1 and 6 times during the preced ing year. T he mean scores and standard deviations from 38 participants on the BASC 2 PRS (Adaptive Skills M = 47.2 SD = 9.9; Externalizing M = 51.2, SD = 11.8; Internalizing M = 52.3 SD = 14.9; Behavioral Symptoms M = 53.2 SD = 13.6) were comparable to those from the standardization sample ( M = 50, SD = 10). Caregivers returned behavioral ratings forms for 76% of the sample; thus, all analyses with BASC 2 PRS included data from 38 participants. Reliability analyses allowed for the study of the properties of the BASC 2 PRS and the scales that composed composites area scores Coefficients for the Adaptive Skills, Externalizing, Internalizing, and Behavioral Symptoms indices ranged from 0.87 to 0.89, which indicated that the indices had high internal consist ency. The coefficients for the composite scores fell slightly below the mean reliabilities reported from the standardization sample (Reynolds & Kamphaus, 2004). Predi c tor Va riables Table 3 2 displays descriptive statistics for predictor variables. While t he mean intellectual ability score for the 50 participants on the WJ III COG (BIA M = 103.5 SD = 12.4 ) was slightly greater than the mean of t he standardization sample the standard deviations and mean verbal ability score ( Verbal Ability M = 98.2 SD = 11.8 ) were slightly less than those of the standardization sample ( M = 100, SD = 15).

PAGE 52

52 The mean T scores and standard deviations for the 50 participants on the FFPI C were comparable to those of the standardization sample ( M = 50, SD = 10). Scores were grea test for Agreeableness ( M = 53.1 SD = 9. 8 ), followed by Openness to Experience ( M = 50.8 SD = 11 ), Conscientiousness ( M = 50.8 SD = 8. 3 ), Emotional Regulation ( M = 49.4 SD = 11.1 ), and Extraversion ( M = 46 SD = 9.2 ). Reliability analyses allowed for t he study of the properties of the FFPI C and the items that composed the five dimensions Coefficients for the dimensions ranged from 0.58 to 0.82 and generally were lower than those found across the normative sample, which ranged from 0.74 to 0.86 ( McGhee Ehrler, & Buckhalt, 2007 ). While the mean scores and standard deviations for the 50 participants on the EQ i: YV (General Mood M = 102.2 SD = 12 Adaptability M = 100.7 SD = 17 Interpersonal M = 99.1 SD = 14.3 Total Score M = 97.7 SD = 16.3 and In trapersonal M = 96.7 SD = 15.4 ) generally were comparable to those from the standardization sample ( M = 100, SD = 15), the mean score on the Stress Management scal e ( M = 94.6 SD = 18.6 ) was 0.3 standard deviations less than the mean score from the standa rdization sample. Reliability analyses allowed for the study of the properties of the EQ i: YV and the scales that composed the Total Score. The coefficient for the Total Score was 0.74, which indicated an acceptable level of internal consistency. Table 3 3 displays mean EQ i YV performance and standard deviations by gender. The mean scores of females ( N = 26) were greater on the Intrapersonal, Interpersonal, and General Mood scales. The mean scores of males ( N = 24) were greater on the Adaptability, Stres s Management, and Total Score scales.

PAGE 53

53 t Test Table 3 4 displays results of t tests that were conducted to examine group differences in performance on the EQ i YV. No statistically significant d ifferences were found in the mean performances of f emal es and males on the five scales and the total score of the EQ i YV Correlational Analyses G uidelines for the interpretation of correlations are ba sed on the work of Cohen (1988) who observed that low correlations generally are from 0.1 to 0. 3, moderate correlations are from 0.3 to 0.5, and high correlations are from 0.5 to 1. Table 3 5 displays correlations between self reported EI, intelligence as measured by the BIA, and math, reading, writing, and overall achievement as measured by the KTEA II BF ( N = 50) A low yet significant correlation was found between adaptability and intelligence ( r = 0.31, p < 0.05) ; however, again the accuracy of the results is low as the 95% confidence interval (CI) for the correlation ranges from 0.03 to 0.54. No other statistically significant relationships were found. Table 3 6 displays correlations between self reported EI and behavior as measured by the adaptive skills, externalizing problems, internalizing problems, and behavioral symptoms indices of the BASC 2 PRS ( N = 38) Externalizing problems correlated significantly and negatively with stress management ( r = 0.50 p < 0.0 1 95% CI of r = 0.71 to 0.21 ), general mood ( r = 0.57 p < 0.0 1 95% CI of r = 0.75 to 0.31 ), and the total score for self reported EI ( r = 0.34 p < 0.05 95% CI of r = 0.6 to 0.02 ). Internalizing problems correlated significantly and negatively with all scales and the total score for self reported EI (intrapersonal, r = 0. 38 p < 0.05 95% CI of r = 0.62 to 0.07 ; interpersonal, r = 0.36 p < 0.05 95% CI of r = 0.61 to 0.05 ; stress management, r =

PAGE 54

54 0.40 p < 0.05 95% CI of r = 0.64 to 0.09 ; adaptability, r = 0.5 1, p < 0.0 1 95% CI of r = 0.71 to 0.23 ; general mood, r = 0.45 p < 0.0 1 95% CI of r = 0.67 to 0.15 ; and total score, r = 0.54 p < 0.0 1 95% CI of r = 0.73 to 0.27 ). Behavioral symptoms correlated significantly and negatively with stress management ( r = 0. 49 p < 0.01 95% CI of r = 0.70 to 0.20 ), adaptability ( r = 0.5 1, p < 0.01 95% C I of r = 0.71 to 0.23 ), and the total score for self reported EI ( r = 0.55 p < 0.01 95% CI of r = 0.74 to 0.28 ). Table 3 7 displays correlations between self reported EI and personality as measured by the FFPI C ( N = 50) Agreeableness correlated s ignificantly with the interpersonal ( r = 0.36, p < 0.05 95% CI of r = 0.09 to 0.58 ) and stress management scales ( r = 0.58, p < 0.01 95% CI of r = 0.36 to 0.74 ) as well as the total score for self reported EI ( r = 0.35, p < 0.05 95% CI of r = 0.08 to 0. 57 ). Extraversion correlated significantly with the intrapersonal scale ( r = 0.30, p < 0.05 95% CI of r = 0.02 to 0.53 ). Openness to experience correlated significantly with general mood ( r = 0.29, p < 0.05 95% CI of r = 0.01 to 0.53 ). Conscientiousness correlated significantly with the intrapersonal( r = 0.37, p < 0.01 95% CI of r = 0.10 to 0.59 ), stress management ( r = 0.38, p < 0.01 95% CI of r = 0.11 to 0.60 ), adaptability ( r = 0.54, p < 0.01 95% CI of r = 0.31 to 0.71 ), and general mood scales ( r = 0.47, p < 0.01 95% CI of r = 0.22 to 0.66 ) as well as with the total score ( r = 0.52, p < 0.01 95% CI of r = 0.28 to 0.70 ). Emotional regulation correlated significantly with all scales and the total score for self reported EI (intrapersonal, r = 0.35, p < 0.05 95% CI of r = 0.08 to 0.57 ; interpersonal, r = 0.39, p < 0.01 95% CI of r = 0.13 to 0.60 ; stress management, r = 0.53, p < 0.01 0.30 to 0.70 ; adaptability, r = 0.54, p < 0.01 95% CI of r = 0.31 to 0.71 ; general mood, r = 0.59, p <

PAGE 55

55 0.01 95% CI of r = 0.37 to 0.75 ; total score, r = 0.63, p < 0.01 95% CI of r = 0.43 to 0.77 ). Multiple Regression Analyses A series of sequential multiple regression analyses were used to examine the incremental validity of self reported EI over intelligence and pe rsonality in the prediction of academic achievement and behavior. Intelligence as measured by the BIA was entered in the first block. Personality as measured by the FFPI C was entered in the second block. Overall self reported EI as measured by the EQ i YV total score was entered in the third block. All regressions took the same form. Academic Achievement Table 3 8 displays results of the sequential regression analysis for math achievement as measured by the KTEA II BF ( N = 50) Intelligence was a significa nt predictor of KTEA II BF math performance ( Adjusted R 2 = 0.45 p < 0.01). Personality dimensions were added in the second block; the dimensions predicted an additional 8 % of the variance in KTEA II BF math performance and this increase did not reach stati stical significance ( p > 0.05). When added in the third block, self reported EI also did not significantly predict additional variance in KTEA II BF math achievement ( R 2 = 0.03 p > 0.05). The full model accounted for 49 % of the explained variance in KTEA II BF math achievement. Table 3 9 displays results of the sequential regression analysis for reading achievement as measured by the KTEA II BF ( N = 50) Intelligence was a significant predictor of KTEA II reading p erformance ( Adjusted R 2 = 0.37 p < 0.01) Personality dimensions were added in the second block; the dimensions predicted an additional 8 % of the variance in KTEA II reading performance and this increase did not reach

PAGE 56

56 statistical significance ( p > 0.05). When added in the third block, self repor ted EI also did not significantly predict additional variance in KTEA II reading achievement ( R 2 = 0.00 p > 0.05). The full model accounted for 37 % of the explained variance in KTEA II BF reading achievement. Table 3 10 displays results of the sequential regression analysis for writing achievement as measured by the KTEA II BF ( N = 50) Intellige nce was a significant predictor of writing achievement ( Adjusted R 2 = 0.17 p < 0.01). Personality dimensions were added in the second block; the dimensions predi cted an additional 7 percent of the variance in KTEA II BF writing performance and this increase did not reach statistical significance ( p > 0.05). When added in the third block, self reported EI also did not significantly predict additional variance in KT EA II BF writing achievement ( R 2 = 0.01 p > 0.05). The full model accounted for 14 % of the explained variance in KTEA II BF writing achievement. Table 3 11 displays results of the sequential regression analysis for academic achievement as measured by the brief achievement composite of the KTEA II BF ( N = 50) Intelligence was a significant predictor of overall achievement on the KTEA II BF ( Adjusted R 2 = 0.50 p < 0.01). Personality dimensions were added in the second block; the dimensions predicted an ad ditional 8 % of the variance in overall achievement and this increase did not reach statistical significance ( p > 0.05). When added in the third block, self reported EI also did not significantly predict additional variance in overall achievement ( R 2 = 0.02 p > 0.05). The full model accounted for 54 % of the explained variance in overall achievement.

PAGE 57

57 Behavior Table 3 12 displays results of the sequential regression analysis for behavior as measured by number of suspensions ( N = 50) Intelligence w as a significant predictor of suspensions ( Adjusted R 2 = 0.07 p < 0.05 ). Personality dimensions were added in the second block T he dimensions predicted an additional 2 % of the variance in suspensions T his increase did not reach statistical significance ( p > 0.05) and intelligence was no longer a significant predictor of suspe n sions When added in the third block, self reported EI also did not significantly predict additional variance ( R 2 = 0.02 p > 0.05). In the full regression equation intelligence was no longer a significant predictor of suspensions Table 3 13 displays results of the sequential regression analysis for behavior as measured by the Externalizing Problems Composite o f the BASC 2 PRS ( N = 38) Intelligence was not a significant predictor of externalizing problems ( Adjusted R 2 = 0.03 p > 0.05 ). Personality dimensions were added in the second block; while the dimensions predicted an additional 14 % of the variance in ext ernalizing problems this increase did not reach statistical significance ( p > 0.05). When added in the third block, self reported EI also did not significantly predict additional variance ( R 2 = 0.02 p > 0.05). The full model accounted for only 4 % of the explained variance in externalizing behavior Table 3 14 displays results of the sequential regression analysis for behavior as measured by the Internalizing Problems Composite of the BASC 2 PRS ( N = 38) Intelligence was not a significant predictor of internalizing problems ( Adjusted R 2 = 0.05 p > 0.05 ). When p ersonality dimensions were added in the second block they predicted an additional 49 % of the variance in internalizing problems and this increase was

PAGE 58

58 statistical ly significant ( p < 0.01 ). Emotional regulation emerged as the only significant predictor and the beta was negative, which suggests that as emotional regulation increases, internalizing problems decrease. When added in the thi rd block, self reported EI did not significantly predict additional variance ( R 2 = 0.01 p > 0.05). In the full regression equation emotional regulation was no longer a significant predictor of internalizing problems The full model accounted for 48 % of t he explained variance in internalizing problems Table 3 15 displays results of the sequential regression analysis for behavior as measured by the Behavioral Symptoms Index of the BASC 2 PRS ( N = 38) Intelligence was a significant predictor of behavioral symptoms ( Adjusted R 2 = 0.12 p < 0.05 ). Personality dimensions were added in the second block They predicted an additional 40 % of the variance in behavioral symptoms and this increase was statistically significa n t ( p < 0.01 ). When added in the third bloc k, self reported EI did not significantly predict additional variance ( R 2 = 0.03 p > 0.05). In the full regression equation intelligence was no longer a significant predictor of behavioral symptoms while extraversion and conscientiousness emerged as sign ificant predictors in the model Beta s were negative, which suggests that as conscientiousness and extraversion increase, behavioral symptoms decrease. The full model accounted for 47 % of the explained variance in behavioral symptoms Table 3 16 displays r esults of the sequential regression analysis for behavior as measured by the Adaptive Skills Composite of the BASC 2 PRS ( N = 38) Intelligence was a significant predictor of adaptive skills ( Adjusted R 2 = 0.18, p < 0.01). Personality dimensions were added in the second block; while the dimensions predicted an

PAGE 59

59 additional 11 % of the variance in adaptive skills, this increase did not reach statistical significance ( p > 0.05). When added in the third block, self reported EI also did not significantly predict a dditional variance ( R 2 = 0.02 p > 0.05). In the full regression equation intelligence was no longer a significant predictor of adaptive skills. The full model accounted for 17 % of the explained variance in adaptive skills. Moderated Regression Analyses A series of moderate d regression analyses were used to examine whether gender moderated the relationship between self reported EI as measured by the total score from the EQ i YV and the criterion variables of achievement as measured by math, reading, writing, and total achiev ement scores on the KTEA II BF ( N = 50) and behavior as measured by the adaptive skills, behavioral symptoms, externalizing problems, and internalizing problem s composite scores of the BASC 2 PRS ( N = 38) Because the sample size was too small to conduct m ore detailed analyses, control variables including intelligence and personality dimensions were not included in this model Gender did not moderate the relationship between self reported EI and total achievement ( R 2 [ 1, 45 ] = 0. 0 5, p = 0.14) reading ( R 2 [ 1, 45 ] = 0. 06 p = 0.08 ), writing ( R 2 [ 1, 45 ] = 0.00 p = 0.71 ), externalizing problems ( R 2 [ 1, 33 ] = 0.08 p = 0.08 ) and internalizing problems ( R 2 [ 1, 33 ] = 0.01, p = 0.58) Gender significantly moderated the relationship between self reporte d EI and parent rated behavioral symptoms among the 38 participants whose caregivers returned the BASC 2 PRS ( R 2 = 0.19, F [ 1, 33 ] = 11.45, p < 0.01). The interaction is depicted in Figure 3 1. An increase in overall EI in males was associated with a

PAGE 60

60 d ecrease in behavioral symptoms while an increase in EI in females was not associated with a substantial change in parent rated behavior. G ender also significantly moderated the relationship between self reported EI and parent rated adaptive skills such t hat the relationship between the variables was stronger for males compared to females ( N = 38; R 2 = 0.13, F [ 1, 33 ] = 5.3, p < 0.05). The interaction is shown in Figure 3 2. An increase in overall EI in males was associated with an increase in adaptive skills while an increase in EI in females was associated with a slight decrease in parent rate d a daptive skills Additionally, g ender significantly moderated the relationship between self reported EI and math achievement as measured by the KTEA II BF ( N = 50; R 2 = 0. 12 F [ 1, 45 ] = 6.09 p < 0.0 5 ). The interaction is depicted in Figure 3 3. F emale students with lower overall self reported EI performed better than males with similar levels of EI on the individually administered standardized math achievement test while the math achievement performance of male students with higher overall EI was one standard deviation greater than the math achievement performance of females with a similar level of overall self reported EI.

PAGE 61

61 Table 3 1. Descriptive statistics for criterion variables Variable N M SD Range KTEA II BF Total Achievement 50 99.3 12.1 83 133 0.72 KTEA II BF Reading 50 99.9 11.4 79 134 0.79 KTEA II BF Math 50 99.5 14.9 67 129 0.94 KTEA II BF Writing 50 99.1 12.9 76 134 0.71 BASC 2 PRS Behavioral Symptoms 38 53.2 13.6 34 100 0.88 BA SC 2 PRS Externalizing 38 51.2 11.8 37 84 0.89 BASC 2 PRS Internalizing 38 52.3 14.9 32 98 0.87 BASC 2 PRS Adaptive Skills 38 47.2 10.0 25 64 0.89 Suspensions 50 0.6 1.3 0 6 Table 3 2. Descriptive statistics for predictor variables Variable N M SD Range WJ III COG BIA 50 103.5 12.3 77 133 WJ III COG Verbal Ability 50 98.2 11.8 81 131 FFPI C Agreeableness 50 53.1 9.8 31 73 0.67 FFPI C Conscientiousness 50 50.8 8.3 31 70 0.70 FFPI C Emotional Regulation 50 49.4 11.1 21 72 0.82 FFPI C Ext raversion 50 46.0 9.2 26 73 0.58 FFPI C Openness to Experience 50 50.8 11.0 32 76 0.66 EQ i: YV Intrapersonal 50 96.7 15.4 67 130 EQ i: YV Interpersonal 50 99.1 14.3 65 120 EQ i: YV Stress Management 50 94.6 18.6 65 128 EQ i: YV Adaptability 50 100.7 17.0 65 130 EQ i: YV General Mood 50 102.2 12.0 65 121 EQ i: YV Total Score 50 97.7 16.3 65 130 0.74 Table 3 3. Mean Emotional Quotient Inventory: Youth Version performance by gender Scale Female ( N = 26) M SD Male ( N = 24) M SD Intraper sonal 97.9 13.8 95.4 17.1 Interpersonal 101.7 14.4 96.3 13.9 Stress Management 90.4 20.1 99.1 16.1 Adaptability 99.6 18.4 101.8 15.7 General Mood 103.5 13.2 100.8 10.5 Total Score 96.8 18.0 98.8 14.5

PAGE 62

62 Table 3 4. Comparison of means based on gen der Scale Degrees of Freedom T Significance of t Intrapersonal 44.14 0.55 0.59 Interpersonal 47.89 1.37 0.18 Stress Management 47.10 1.69 0.10 Adaptability 47.69 0.47 0.64 General Mood 47.00 0.82 0.42 Total Score 47.23 0.42 0.68 Table 3 5 Pea rson product moment correlations between self reported EI, intelligence as assessed by the BIA scale of the WJIII COG, and KTEA II BF performance Emotional Intelligence Math ( N = 50) Reading ( N = 50) Writing ( N = 50) Total Achievement ( N = 50) Intelligenc e ( N = 50) Intrapersonal 0.08 0.01 0.08 0.01 0.06 Interpersonal 0.03 0.22 0.04 0.10 0.13 Stress Management 0.00 0.17 0.01 0.04 0.15 Adaptability 0.15 0.19 0.15 0.19 0.31* General Mood 0.06 0.01 0.03 0.04 0.11 Total Score 0.04 0.18 0.09 0.11 0.25 Note: p < 0.05, ** p < 0.01 Table 3 6 Pearson product moment correlations between self reported EI and behavior Emotional Intelligence Adaptive Skills ( N = 38) Externalizing Problems ( N = 38) Internalizing Problems ( N = 38) Behavi oral Symptoms ( N = 38) Intrapersonal 0.12 0.06 0.38* 0.31 Interpersonal 0.16 0.05 0.36* 0.30 Stress Management 0.25 0.50** 0.40* 0.49** Adaptability 0.22 0.20 0.51** 0.51** General Mood 0.07 0.57** 0.45** 0.16 Total Score 0.29 0.34* 0.54** 0.55** Note: p < 0.05, ** p < 0.01

PAGE 63

63 Table 3 7 Pearson product moment correlations between self reported EI and personality Emotional Intelligence Agreeable ness ( N = 50) Extraversion ( N = 50) Openness to Experience ( N = 50) Consci en tiousness ( N = 50) Emotional Regulation ( N = 50) Intrapersonal 0.09 0.30* 0.08 0.37** 0.35* Interpersonal 0.36* 0.17 0.19 0.20 0.39** Stress Management 0.58** 0.09 0.19 0.38** 0.53** Adaptability 0.15 0.16 0.21 0.54** 0.54** General Mood 0.19 0.2 1 0.29* 0.47** 0.59** Total Score 0.35* 0.17 0.22 0.52** 0.63** Note: p < 0.05, ** p < 0.01 Table 3 8 Sequential regression analysis predicting KTEA II BF math achievement ( N = 50) Predictor Variables B SE R 2 Adjusted R 2 Block R 2 Block 1 Intell igence 0.46 0.45 0.46** Intelligence 0.82 0.13 0.68** Block 2 Personality 0.54 0.47 0.08 Intelligence 0.79 0.14 0.66** Agreeableness 0.20 0.19 0.13 Extraversion 0.37 0.23 0.23 Openness to Experience 0.17 0.1 6 0.12 Conscientiousness 0.19 0.27 0.11 Emotional Regulation 0.12 0.22 0.09 Block 3 Self Reported EI 0.57 0.49 0.03 Intelligence 0.81 0.14 0.67** Agreeableness 0.28 0.19 0.18 Extraversion 0.40 0.22 0.25 Openness to Experience 0.16 0.16 0.12 Conscientiousness 0.11 0.26 0.06 Emotional Regulation 0.02 0.23 0.01 Self Reported EI 0.21 0.13 0.23 Note: SE = Standard Error B, p < 0.05, ** p < 0.01

PAGE 64

64 Table 3 9 Sequent ial regression analysis predicting KTEA II BF reading achievement ( N = 50) Predictor Variables B SE R 2 Adjusted R 2 R 2 Block 1 Intelligence 0.38 0.37 0.38** Intelligence 0.57 0.11 0.62** Block 2 Personality 0.46 0.39 0.08 Intelligence 0.54 0.12 0.58** Agreeableness 0.28 0.16 0.24 Extraversion 0.28 0.19 0.2 3 Openness to Experience 0.13 0.13 0.12 Conscientiousness 0.10 0.22 0.07 Emotional Regulation 0.10 0.18 0.10 Block 3 Self Reported EI 0.46 0.37 0.00 Intelligence 0.54 0.12 0.58** Agreeableness 0.28 0.16 0.24 Extraversion 0.29 0.19 0.23 Openness to Experience 0.13 0.13 0.12 Conscientiousness 0.09 0.22 0.07 Emotional Regulation 0.09 0.20 0.09 Self Reported EI 0.02 0.11 0.02 Note: SE = Standard Error B, p < 0.05, ** p < 0.01

PAGE 65

65 Table 3 10 Sequential regression analysis predicting KTEA II BF writing achievement ( N = 50) Predictor Variables B SE R 2 Adjusted R 2 R 2 Block 1 Intelligence 0.19 0.17 0.19** Intelligence 0.45 0.14 0.43** Block 2 Personality 0.26 0.15 0.07 Intelligence 0.39 0.16 0.37* Agreeableness 0.02 0.21 0.01 Extraversion 0.05 0.25 0.03 Openness to Experience 0.33 0.18 0.28 Conscientiousness 0.01 0.29 0.01 Emot ional Regulation 0.01 0.24 0.01 Block 3 Self Reported EI 0.26 0.14 0.01 Intelligence 0.39 0.16 0.38* Agreeableness 0.01 0.22 0.01 Extraversion 0.04 0.25 0.03 Openness to Experience 0.33 0.18 0.28 Conscient iousness 0.02 0.30 0.01 Emotional Regulation 0.06 0.26 0.05 Self Reported EI 0.08 0.14 0.09 Note: SE = Standard Error B, p < 0.05, ** p < 0.01

PAGE 66

66 Table 3 11 Sequential regression analysis predicting KTEA II overall academic achiev ement ( N = 50) Predictor Variables B SE R 2 Adjusted R 2 Block R 2 Block 1 Intelligence 0.51 0.50 0.51** Intelligence 0.70 0.01 0.72** Block 2 Personality 0.59 0.54 0.08 Intelligence 0.65 0.11 0.67** Agreeableness 0.16 0.15 0.13 Extraversion 0.22 0.17 0.1 7 Openness to Experience 0.23 0.12 0.21 Conscientiousness 0.09 0.20 0.06 Emotional Regulation 0.08 0.17 0.07 Block 3 Self Reported EI 0.61 0.54 0.02 Intelligence 0.66 0.11 0.67** Agreeableness 0.21 0.15 0.17 Extraversion 0.24 0.17 0.19 Openness to Experience 0.23 0.12 0.21 Conscientiousness 0.04 0.20 0.03 Emotional Regulation 0.00 0.18 0.00 Self Reported EI 0.13 0.10 0.18 Note: SE = Standard Error B, p < 0.05, ** p < 0.01

PAGE 67

67 Table 3 12 Sequential regression analysis predicting suspensions ( N = 50) Predictor Variables B SE R 2 Adjusted R 2 Block R 2 Block 1 Intelligence 0.09 0.07 0.09* Intelligence 0.03 0.01 0.30* Block 2 Personality 0.11 0.02 0.02 Intelligence 0.03 0.02 0.26 Agreeableness 0.01 0.02 0.11 Extraversion 0.00 0.03 0.03 Openness to Experience 0.00 0.02 0.07 Conscientiousness 0.03 0.03 0.16 Emotional Regulation 0. 02 0.03 0.14 Block 3 Self Reported EI 0.13 0.02 0.02 Intelligence 0.37 0.02 0.27 Agreeableness 0.01 0.02 0.07 Extraversion 0.15 0.03 0.04 Openness to Experience 0.01 0.02 0.07 Conscientiousness 0.03 0.03 0.20 Emotional Regulation 0.01 0.03 0.06 Self Reported EI 0.01 0.02 0.18 Note: SE = Standard Error B, p < 0.05, ** p < 0.01

PAGE 68

68 Table 3 1 3 Sequential regression analysis predicting externalizing problems ( N = 38) Predictor Variables B SE R 2 Adjusted R 2 R 2 Block 1 Intelligence 0.06 0.03 0.06 Intelligence 0.21 0.14 0.24 Block 2 Personality 0.20 0.05 0.14 Intelligence 0.09 0.16 0.11 Agreeableness 0.35 0.23 0.31 Extraversion 0.33 0.27 0 .29 Openness to Experience 0.04 0.21 0.03 Conscientiousness 0.37 0.38 0.23 Emotional Regulation 0.10 0.28 0.10 Block 3 Self Reported EI 0.22 0.04 0.02 Intelligence 0.09 0.16 0.10 Agreeableness 0.32 0. 23 0.28 Extraversion 0.33 0.27 0.28 Openness to Experience 0.02 0.21 0.02 Conscientiousness 0.32 0.38 0.20 Emotional Regulation 0.19 0.30 0.19 Self Reported EI 0.13 0.16 0.19 Note: SE = Standard Err or B, p < 0.05, ** p < 0.01

PAGE 69

69 Table 3 14 Sequential regression analysis predicting internalizing problems ( N = 38) Predictor Variables B SE R 2 Adjusted R 2 Block R 2 Block 1 Intelligence 0.07 0.05 0.07 Intelligence 0.30 0.18 0.27 Block 2 Personality 0.56 0.48 0.49** Intelligence 0.03 0.15 0.03 Agreeableness 0.08 0.21 0.06 Extraversion 0.30 0.25 0.20 Openness to Experience 0.09 0.19 0.07 Conscientiousness 0.42 0.35 0.20 E motional Regulation 0.59 0.26 0.46* Block 3 Self Reported EI 0.58 0.48 0.01 Intelligence 0.02 0.15 0.02 Agreeableness 0.12 0.24 0.08 Extraversion 0.30 0.25 0.20 Openness to Experience 0.08 0.19 0.06 Conscientiousness 0.36 0.36 0.18 Emotional Regulation 0.49 0.28 0.39 Self Reported EI 0.14 0.15 0.16 Note : SE = Standard Error B, p < 0.05, ** p < 0.01

PAGE 70

70 Table 3 15 Sequential regression analysis predicting behavioral symp toms ( N = 38) Predictor Variables B SE R 2 Adjusted R 2 Block R 2 Block 1 Intelligence 0.14 0.12 0.14* Intelligence 0.38 0.16 0.38* Block 2 Personality 0.55 0.46 0.40** Intelligence 0.12 0.14 0.12 Agreeableness 0.31 0.20 0.23 Extraversion 0.52 0.2 3 0.39* Openness to Experience 0.04 0.18 0.03 Conscientiousness 0.84 0.33 0.45* Emotional Regulation 0.03 0.24 0.02 Block 3 Self Reported EI 0.57 0.47 0.03 Intelligence 0.11 0.14 0.11 Agreeableness 0.26 0.20 0.20 Extraversion 0.52 0.23 0.39* Openness to Experience 0.02 0.18 0.02 Conscientiousness 0.78 0.33 0.42* Emotional Regulation 0.09 0.25 0.08 Self Reported EI 0.88 0.13 0.22 Note: SE = St andard Error B, p < 0.05, ** p < 0.01

PAGE 71

71 Table 3 16 Sequential regression analysis predicting adaptive skills ( N = 38) Predictor Variables B SE R 2 Adjusted R 2 R 2 Block 1 Intelligence 0.21 0.18 0.21* Intelligence 0.33 0.11 0.45* Block 2 Personality 0.31 0.18 0.11 Intelligence 0.26 0.13 0.36* Agreeableness 0.28 0.18 0.29 Extraversion 0.25 0.21 0.26 Openness to Experience 0.16 0.16 0.17 Conscientiousness 0.18 0.29 0.13 Emotiona l Regulation 0.19 0.22 0.23 Block 3 Self Reported EI 0.33 0.17 0.02 Intelligence 0.26 0.13 0.35 Agreeableness 0.25 0.18 0.26 Extraversion 0.25 0.21 0.26 Openness to Experience 0.15 0.16 0.16 Conscientiousn ess 0.14 0.30 0.10 Emotional Regulation 0.26 0.23 0.31 Self Reported EI 0.10 0.12 0.17 Note: SE = Standard Error B, p < 0.05, ** p < 0.01 Figure 3 1. Interaction of gender on behavioral symptoms an d total score for self reported EI ( N = 38)

PAGE 72

72 Figure 3 2. Interaction of gender on adaptive skills and total score for self reported EI ( N = 38)

PAGE 73

73 Figure 3 3 Interaction of gender on math achieve ment and total score for self reported EI ( N = 50)

PAGE 74

74 CHAPTER 4 DISCUSSION The first objective of the present study was to better understand the relationships between general intelligence, personality, academic achievement, parent rated behavior, and self re ported emotional intelligence in students. A second and more specific objective was determine the extent to which self reported E I adds unique variance beyond measures of general intelligence and personality in the prediction of educationally important cri teria including academic achievement and behavior. A third and final objective was to develop a better understanding of gender based differences in an emotion related construct in youth Research Questions 1 and 2: Relationships between Self Reported EI an d Important Criteria Research question 1 addressed the extent to which self reported EI as measured by the Bar On Emotional Quotient Inventory: Youth Version (EQ i: YV; Bar On & Parker, 2000) correlated with academic achievement, behavior, general intelli gence, and personality characteristics. Research question 2 addressed whether self reported EI incrementally predicted academic achievement, parent rated behavior, and suspensions for misbehavior beyond the variance predicted by intelligence and personalit y. Relationships between Self Reported EI and General Intelligence Correlational analyses included the five scale area scores as well as the total score of the EQ i: YV and general intelligence as measured by the Brief Intellectual Ability scale of the Woo dcock Johnson III Tests of Cognitive Abilities (WJIII COG; Woodcock, McGrew, & Mather, 2001) Data from 50 participants showed that there was a low yet significant relationship of 0.31 between the adaptability scale of the EQ i: YV and general intelligence No other significant correlations were found between general

PAGE 75

75 intelligence and the remaining four scales or the total score of the EQ i: YV. While t he findings of the present study realistic in managing change and solving problems are positively related to their intelligence the results also corroborate the results of previous studies that generally did not find significant correlations between general intelligence and self reported EI (e.g. Newsome, Day, & Catano, 2000; Saklofske, Austin, & Minski, 2003). The low correlations between intelligence and self reported EI support the claim that emotional as assessed by self reports is a multidimensional collection of at least some non cognitive abilities. Relationships between S elf Reported EI and Personality R esults of previous studies with university students found that correlations between self reported EI and personality dimension s ranged from low to high and several associatio ns reached statistical significance (Dawda & Hart, 2000; Newsome et al., 2000 ; Saklofske et al., 2003 ). Results of the curren t study with students in the 5 th through 8 th grades were similar to the results of previous studies, which indicate that dime nsions of personality likely are the non cognitive abilities that are assessed by self report measures of EI. T he current study included c orrelational analyses for the five scale area scores as well as the total score of the EQ i: YV and personality as measure d t he Fi ve Factor Personality Inventory for Children (FFPI C; McGhee, Ehrler, & Buckhalt, 2007 ). Data from 50 participants showed s ignificant correlations between several scales of the self report measure of EI and various personality dimensions which prov ided evidence of the mixed model of EI that is based on self perceptions of typical performance in relation to a combination of abilities, competencies, dispositions, personality traits, and

PAGE 76

76 skills. Significant moderate positive correlations were found bet ween the interpersonal scale of the EQ i: YV and the personality dimensions of agreeableness ( r = 0.36 ) and emotional regulation ( r = 0.39) ; also, the intrapersonal scale correlate d significantly with extraversion ( r = 0.30), conscientiousness ( r = 0.37), and emotional regulation ( r = 0.35 ); the stress management scale correlated significantly with agreeableness ( r = 0.58), conscientiousness ( r = 0.38), and emotional regulation ( r = 0.53 ); significant correlations were found between the adaptability scale a nd conscientiousness ( r = 0.54) and emotional regulation ( r = 0.54) ; and general mood correlated significantly with openness ( r = 0.29), conscientiousness ( r = 0.47), and emotional regulation ( r = 0.59). Overall self reported EI correlated moderately and significantly with the personality dimensions of agreeableness ( r = 0.35), conscientiousness ( r = 0.52), and emotional regulation ( r = 0.63 ), which indicated that s elf reported EI relates most closely to personality characteristics including altruism, comp etence, compliance, and self discipline as well as self consciousness regarding anxiety depression, and impulsiveness. The two major concepts of EI include the ability model of Mayer and Salovey (1997), which is the focus of most empirical researc h, and the mixed or emotional social intelligence model of Bar On (2006) Measures of the ability model, which are based on IQ like tests of maximum performance, typically show higher correlations with cognitive ability (Van Rooy et al. 2005). Results of the current study provide support for previous findings indicating that s elf report measures of the emotional social intelligence model typically show higher correlations with personality factors ( Dawda & Hart, 2000; Newsome et al., 2000; Saklofske et al. 2003 )

PAGE 77

77 The implications and sources of the high degree of associations between self reported EI and personality are debatable. High associations may occur because the concepts are in fact similar; or, they may occur because of item overlap in the questio ns used to evaluate EI and personality. A moderate c orrelation between for example, the adaptability scale of the EQ i : YV and a personality dimension such as conscientiousness supports the construct validity of the measures ability to adapt or m anage change is related to her or his ability to be conscientious and effective Y et the high degree of association s between aspects of self reported EI and personality leads one to question the extent to which the emotional construct adds unique variance beyond general intelligence and personality in the prediction of important criteria. Behavioral Criteria The results of the current study indicated that self reported EI is correlated with behavior. Correlational analyses included the five scale area sco res as well as the total score of the EQ i: YV and behavior as assessed by the Parent Rating Scales (PRS) of the Behavior Assessment System for Children Second Edition (BASC 2; Reynolds & Kamphaus, 2004). There was a 76% return rate for the BASC 2 PRS; thu s, all analyses utilizing parent rated behavior were conducted with data from 38 participants. Moderate negative and significant correlations were found between overall self reported EI and externalizing ( r = 0.34), internalizing ( r = 0.54), and overall behavioral problems ( r = 0.55). While zero order correlations indicated that higher levels of self reported EI were associated with fewer behavioral problems, multiple regression analyses indicated that these associations no longer remained significant af ter accounting for variance explained by general intelligence and personality.

PAGE 78

78 T he EQ i: YV includes five scale area scores yet only the total score was used for the series of sequential multiple regression analyses that examined the incremental validit y of self reported EI over intelligence and personality dimensions in the prediction of behavior as measured by school suspensions ( N = 50) and parent rated behavior s ( N = 38) Significant relationships were found between intelligence, which was entered in the first block, and adaptive skills, overall behavioral problems, and suspensions. Squared correlation coefficients were low and ranged from 0.09 for suspensions to 0.21 for adaptive skills. No significant relationships were found between intelligence an d externalizing or internalizing behaviors. The inclusion of personality dimensions in the second block of the analyses produced no significant additional variance in the prediction of adaptive skills, externalizing behaviors, or suspensions. The amount of additional variance ranged from a low of 2% for suspensions to 14% for externalizing behaviors. The inclusion of personality dimensions in the second block added significant variance in the prediction of internalizing behaviors and overall behavioral sym ptoms. Squared correlation coefficients were 0.40 for overall behavior and 0.49 for internalizing problems, which indicated that personality dimensions accounted for a significant amount of additional variance beyond general intelligence in the prediction of behavior. In the model accounting for intelligence and personality, extraversion and conscientiousness emerged as significant predictors of overall behavioral problems while intelligence no longer was correlated significantly with the outcome variable a nd the personality dimension of emotional regulation emerged as a significant predictor of internalizing problems. Betas were negative for both cases, which indicated that

PAGE 79

79 students with increased conscientiousness, emotional regulation, and extraversion di splayed fewer internalizing and overall problem behaviors. While scales on the FFPI C are known to correlate with aspects of social emotional functioning (McGhee et al., 2007), results of the current study showed that after accounting for the variance attr ibuted to general intelligence, personality factors emerged as significant predictors of behavioral criteria. In contrast to the results of previous studies showing that after controlling for personality factors, self reported EI accounted for incrementa l variance in the prediction of various behavioral characteristics ( Furnham & Christoforou, 2007; Furnham & Petrides, 2003; Law, et al., 2004; Livingston & Day, 2005; Shulman & Hemenover, 2006; van der Zee & Wabeke, 2004), results of the current study show ed that after accounting for general intelligence and personality, the inclusion of self reported EI led to no significant additional variance in the prediction of parent rated behavior and the amount of additional variance ranged from 1% for internalizing behaviors to 3% for overall behavioral problems. Thus, data do not consistently suggest that the constructs measured by self reports of EI have incremental validity over dimensions of personality in the prediction of behavioral outcomes. Academic Criteria In contrast to the findings of previous research (e.g., Parker et al., 2004; Qualter et al., 2007; Schutte et al., 1998), t he results of the present study provide limited support for the notion that self reported EI is correlated with academic achievement Correlational analyses included the five scale area scores as well as the total score of the EQ i: YV and academic achievement as assessed by three objective measures

PAGE 80

80 D ata showed that among 50 students in the 5th through the 8th grades, self reported EI did not correlate significantly with math, reading, and writing performance or total achievement on an individually administere d standardized achievement test In comparison Peters et al. (2009) found significant moderate correlations ( r = 0.53) betwee n the total score from the youth version of the MSCEIT, an ability based performance measure of EI, and reading achievement as measured by the Stanford Achievement Test Series, Tenth Edition ( SAT 10 ), which is a group administered high stakes statewide exa m Thus, while in the present study the relationships between EI and academic skills as assessed by an individually administered achievement test were limited, other results indicate that emotion related constructs and achievement are more related in high stakes statewide testing situations in which students must manage their emotions and stress. While the EQ i: YV includes five scale area scores, only the total score was used in the present study for the series of sequential multiple regression analyses th at examined the incremental validity of self reported EI over intelligence and personality dimensions in the prediction of academic achievement. Significant relationships were found among 50 participants between intelligence, which was entered in the first block of the analyses, and all areas of academic achievement including math, reading, and writing performance on the individually administered achievement test. Squared correlation coefficients ranged from lows of 0.17 for writing performance on the KTEA II BF to moderate relationships of 0.51 for total achievement on the KTEA II BF. The inclusion of personality dimensions in the second block of the analyses produced no significant additional variance in the prediction of academic criteria. The

PAGE 81

81 amount of additional variance ranged from 7 to 8% in the prediction of math, reading, writing, and overall performance on the KTEA II BF. In the present study, t he inclusion of self reported EI in the third block produced no significant additional variance in the p rediction of academic criteria and the amount of additional variance ranged from undetectable amount s for KTE A II BF reading performance to 3 % in the prediction of KTEA II BF math performance. Results of the present study with students in primary grades ar e similar to the results of Newsome et al. (2000) with students in college, which indicated that self reported EI adds no significant additional variance beyond intelligence and personality in the prediction of academic achievement criteria. Research Que stions 3 and 4: Understanding Gender Based Differences in an Emotion Related Construct in Youth Research question 3 addressed whether females and males performed differently on a self report measure of EI. Research question 4 addressed whether between gen der differences existed in the relationship of self reported EI and the criterion variables. Gender Differences in Self R eported EI Based on cultural norms and gender role socialization, the display of emotion often is valued in females while emotional con trol is valued in males (Brody, 2000). Thus, one expects differences between females and males on mea sures of emotional intelligence. D ata from the current study with students in grades 5 through 8 showed no statistically significant differences in the mea n performances of females and males on the five scale areas or overall self reported EI. While other studies reported no significant gender differences among female and male college students on a self report measure of

PAGE 82

82 emotional intelligence (Brackett & Ma yer, 2003; Saklofske, Austin, & Minski, 2003), extant empirical data are mixed and the results of the current study showed that males scored 2/3 of a standard deviation higher than females on the stress management scale which aligns with the cultural norm described above by indicating that males perceive themselves to be better able to control and manage their emotions and to respond calmly to stressful events. Data from the current study also showed that for students in grades 5 through 8, females scored 1/3 of a standard deviation higher than males on the interpersonal scale. Thus, data from children provide support for previous findings which showed that female s reported significantly higher levels of interpersonal EI than males (Alumran & Punamki, 2008 ; Reiff, Hatzes, Bramel, & Gibbon, 2001). This relations hip also has emerged in parent ratings which indicate that as compared to boys, girls have significantly higher levels of interpersonal EI (Santesso, Reker, Schmidt, & Segalowitz, 2006). While result s are mixed, data indic ate that as compared to males, females perceives themselves as having a greater level of interpersonal social awareness and thus a better Gender Differences in the Relationships between Self Report ed EI, A cademics, and B ehavior A final objective of the current study was directed at developing a better understanding of an emotional construct in youth. Gender did not moderate the relationship between self reported EI and total achievement reading, wr iting, externalizing problems and internalizing problems Other d ata indicated that the relationship between self reported EI and external criteria differed according to gender.

PAGE 83

83 Gender significantly moderated the relationships between overall self reporte d EI and parent rated adaptive skills ( N = 38) and behavioral symptoms ( N = 38) as well as math skills as measured by an individually administered achievement test ( N = 50) Compared to males with lower overall self reported EI females with similar EI dis played higher levels of parent rated adaptive skills y et as self reported EI increased, the adaptive skills of males improved while the skills of females remained fairly consistent. As self reported EI increased, females also remained fairly consistent in their overall parent rated behavioral symptoms. Females and males with lower self reported EI had similar levels of parent rated problem behaviors yet as self reported EI increased, the behavioral symptoms of males decreased while as noted above the behavi ors of females remained consistent. Gender also significantly moderated the relationship between overall self reported EI and math achievement such that for males and females with lower levels of EI, females displayed greater math achievement. Yet for male s and females with higher levels of self reported EI, males displayed math achievement that was one standar d deviation above that of females The data suggest that for male students greater overall self reported EI generally is associated with more positi ve outcomes for example, higher adaptive skills and math achievement and lower behavioral problems. For female students, greater overall self reported EI generally is associated with no change in behavioral outcomes and a lower level of math achievement. Results of the current study are similar to results of a previous study that found a significant negative relationship between EI and misbehavior for males yet not for females (Brackett, Mayer, & Warner, 2004). While one other study showed that gender

PAGE 84

84 based differences exist in the relationships between emotion related constructs and cognitive ability (Derksen, Kramer, & Katzko, 2002), the present study is among th e first to show that an emotion related construct as assessed by a self report measure ope rate s differently in males and females. Implications for Future Research Experimental and l ongitudinal studies with large sample sizes are needed to clarify issues of causality effect, and prediction. Examinations may focus on the stability of EI over th e course of the lifespan and its susceptibility to change following the implementation of interventions. Future studies also should include the results of group administered statewide tests as previous research (e.g., Peters et al., 2009) indicated that em otion related constructs and academic achievement may be more related in high stakes testing situations in which students must manage their emotions and stress. Future research also is needed to further clarify the relationships between gender, importan t criteria, and EI as assessed by self reports and performance measures to determine whether gender based differences can be replicated for ability based and mixed model concepts of EI. Research with larger sample sizes would allow for a determination of w hether main effects for gender remain significant after accounting for IQ and personality. Ideally, future research can address why self perceptions of greater emotional intelligence are associated with more positive outcomes for males yet self perceptions of greater emotional intelligence in females do not seem to be associated with significantly more positive outcomes. Considering the debate and difficulty associated with identifying the appropriate theoretical construct of EI, future studies are needed t o synthesize the ability based and mixed models. A correlation of 0.42 has been found between the overall scores from a

PAGE 85

85 performance measure of the ability based model and a self report measure of the mixed model (Peters et al., 2009). While the correlation can be interpreted to indicate that the tests measure distinct constructs, the moderate and significant rela tionship also indicated that the tests and constructs share d a great deal of variance. Drawing out aspects of the measures that lead to shared vari ance and incorporating together those aspects that are different may allow for a more complete assessment of EI through an integrated measure that includes performance based activities and self report questions. Limitations Several limitations may affect the current research Generalizations are limited to students in Florida and New York State who are between the ages of 10 and 14 years. S tudents often could not be removed from classrooms for research purposes in lieu of particip ation in educational activ ities and so the sample largely is comprised of students who either were members of after school programs or those who could stay after school to participate in the research. Thus, the findings may be limited by sample selection bias as participation was b ased on self selection and almost all students were recruited from after school programs. If the performance of the selected sample did not generally estimate the performance of a random sample from the population, then sample selection bias likely limited The sample also was limited as families, members of school districts, and students may have been reluctant to participate in research regarding a sensitive issue such as emotions due to fears that the resu lts may end up reflect ing negatively upon them. There was a low participation rate of approximately 5% (despite the distribution of over 900 letters of recruitment ) and so the study was limited by a small sample size.

PAGE 86

86 While analyses indicated that the samp le of 50 students was normally distributed and reliable, the number of participants likely was not enough to be generalizable and predictive. A nalyses with three explanatory variables were run for samples of 38 students for behavior as assessed by the Par e nt Rating Scales of the BASC 2 and for 50 students for academic achievement as assessed by the KTEA II BF. A sample size of 53 would achieve the target of 0.95 probability with 10% accuracy for estimating the correlation between criterion variables and EI when other predictor variables were controlled. With smaller sample sizes and when changes in additional variance are expected to be small (e.g., 5% or less), there is limited power to detect changes in predictive power at step 3 of the multiple regressio n analyses and, with 10% accuracy, the direction of relationships can even be misrepresented. An additional limitation of the current study was the use of self report measures of personality and EI. While the attention, effort, and truthfulness of all p articipants were monitored closely, some students may have manipulated their answers. Measures of personality and measures of mixed models of EI often require individuals to complete self descriptive statements or select adjectives or sentences to describe themselves. As such, the measures are prone to the random and systematic errors of self reporting (Schwarz, 1999). Self reflection (Riccio & Rodriguez, 2007) and the falsification of responses and self report bi ases based on social desirability or other causes are impossible to rule out. Thus, differences between individuals may represent valid differences in EI and personality, or differences may result from the manner in which the students respond ed S elf repor t measures were used because a better modality for collecting data was unclear. For example, the

PAGE 87

87 reports of others are limited as parents, peers, and teachers are prone to halo effects and can have inaccurate informatio Also, youth versions of performance measures of the ability based concept of EI are still under development. The present study provided evidence indicating that relationships between a mixed model concept of EI and academic and behavioral criteria differ accordi ng to gender. The present study also provided evidence indicating that self reported EI does not add unique variance beyond general intelligence and personality in the prediction of academic and behavioral criteria. Thus, a mixed model concept of EI likely has little validity within educational contexts yet considering the limitations noted above, additional research is needed with a larger sample size to provide more conclusive evidence.

PAGE 88

88 LIST OF REFERENCES Achenbach, T. M., & Rescorla, L. A. (2001). Manua l for the ASEBA school age forms and profiles Burlington, VT: University of Vermont, Research Center for Children, Youth, and Families. Alumran, J. I. A., & Punamki, R. (2008). Relationship between gender, age, academic achievement, emotional intellig ence, and coping styles in Bahraini adolescents. Individual Differences Research (6) 2, 104 119. American Psychological Association. (2002). Ethical principles of psychologists and code of conduct. American Psychologist, 37 (12), 1060 1073. American Psyc hological Association. (2007). American Psychological Association dictionary of psychology. Washington, DC: Author. Ballard, J. C. (2003). [Review of the Bar On Emotional Quotient Inventory: Youth Version.] In B. S. Plake, J. C. Impara, & R. A. Spies ( Eds.), The fifteenth mental measurements yearbook (pp. 91 93). Lincoln, NE: Buros Institute of Mental Measurements. Barchard, K. A. (2003). Does emotional intelligence assist in the prediction of academic success? Educational and Psychological Measureme nt, 63 (5), 840 858. Bar On, R. (1997). Bar On Emotional Quotient Inventory: Technical manual. Toronto, ON: MHS. Bar On, R. (2002). Bar On Emotional Quotient Short Form (EQ i: Short): Technical manual. Toronto, ON: MHS. Bar On, R. (2006). The Bar On model of emotional social intelligence (ESI). Psicothema, 18 (supl.), 13 25. Bar On, R., & Parker, J. D. A. (2000). Emotional Quotient Inventory: Youth Version (EQ i: YV) : Technical manual. Toronto, ON: MHS. Brackett, M. A., & Mayer, J. D. (2003). Co nvergent, discriminant, and incremental validity of competing measures of emotional intelligence. Personality and Social Psychology Bulletin, 29 (9), 1147 1158. Brackett, M. A., Mayer, J. D., & Warner, R. M. (2004). Emotional intelligence and its relatio n to everyday behaviour. Personality and Individual Differences, 36 (6) 1387 1402. Brackett, M., Warner, R. M., Bosco, J. S. (2005). Emotional intelligence and relationship quality among couples. Personal Relationships, 12 (2), 197 212.

PAGE 89

89 Brody, L. R. (20 00). The socialization of gender differences in emotional expresssion: Display rules, infant temperament, and differentiation. In A. H. Fischer (Ed.), Gender and emotion: Social psychological perspectives (pp. 24 47). New York: Cambridge University Press. Carmeli, A., & Josman, Z. E. (2006). The relationship among emotional intelligence, task performance, and organizational citizenship behaviors. Human Performance, 19 (4), 403 419. Carroll, J. B. (1993). Human cognitive abilities: A survey of factor anal ytic studies. New York: Cambridge University Press. Cattell, R. B., Cattell, A. K., & Cattell, H. E. (1993). Sixteen Personality Factor Questionnaire (5 th ed.). Champaign, IL: Institute for personality and Ability Testing. Ciarrochi, J. V., Chan, A. Y C., & Caputi, P. (2000). A critical evaluation of the emotional intelligence construct. Personality and Individual Differences, 28, 539 561. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2 nd ed.). Hillsdale, NJ: Erlbaum. Conners, C. K. (1997). Revised North Towanda, NY: Multi Health Systems. Dawda, D., & Hart, S. D. (2000). Assessing emotional intelligence: Reliability and validity of the Bar On Emotional Quotient Inventory (EQ I) in university students. Personality and Individual Differences, 28 (4) 797 812. Derksen, J., Kramer, I., & Katzko, M. (2002). Does a self report measure for emotional intelligence assess something different than general intelligence? Personality and Individual Differ ences, 32 (1) 37 48. Farrelly, D., & Austin, E. J. (2007). Ability EI as an intelligence? Associations of the MSCEIT with performance on emotion processing and social tasks and with cognitive ability. Cognition and Emotion, 21 (5), 1043 1063. Florida Dep artment of Education. (2009). Statewide comparison of math and reading scores. Retrieved from http://fcat.fldoe.org/mediapacket/2011/default.asp Furnham, A., & Chamorro Premuzic, T. (2003). Personality predicts academic performance: Evidence from two l ongitudinal university samples. Journal of Research in Personality, 37 (4), 319 338.

PAGE 90

90 Furnham, A., & Christoforou, I. (2007). Personality traits, emotional intelligence, and multiple happiness. North American Journal of Psychology, 9 (3), 439 462. Furnha m, A., & Petrides, K. V. (2003). Trait emotional intelligence and happiness. Social Behavior and Personality, 31 (8), 815 824. Gignac, G. (2005). Evaluating the MSCEIT V2.0 via CFA: Comment on Mayer et al. 2003). Emotion, 5 (2), 233 235. Gil Olarte Mr quez, P., Palomera Martn, R., & Brackett, M. A. (2006). Relating emotional intelligence to social competence and academic achievement in high school students. Psicothema, 18 (supl.), 118 123. Goldenberg, I., Matheson, K., & Mantler, J. (2006). The assess ment of emotional intelligence: A comparison of performance based and self report methodologies. Journal of Personality Assessment, 86 (1), 33 45. Goleman, D. P. (1995). Emotional intelligence: Why it can matter more than IQ for character, health and li felong achievement New York: Bantam Books. Grubb, W. L., & McDaniel, M. A. (2007). The fakability of Bar Inventory Short form: Catch me if you can. Human Performance, 20 (1), 43 59. Harrod, N. R., & Scheer, S. D. (2005). An expl oration of adolescent emotional intelligence in relation to demographic characteristics. Adolescence, 40 (159), 503 512. Humphrey, N., Curran, A., Morris, E., Farrell, P., & Woods, K. (2007). Emotional intelligence and education: A critical review. Educ ational Psychology, 27 (2), 235 254. Hunsley, J., & Meyer, G. J. (2003). The incremental validity of psychological testing and assessment: Conceptual, methodological, and statistical issues. Psychological Assessment, 15 (4), 446 455. Kaufman, A.S., & Kauf man, N. L. (2004a). Kaufman Assessment Battery for Children Second Edition manual Circle Pines, MN: American Guidance Service. Kaufman, A.S., & Kaufman, N. L. (2004b). Kaufman Brief Intelligence Test Second Edition manual Circle Pines, MN: American Guidance Service. Kaufman, A.S., & Kaufman, N. L. (2005). Kaufman Test of Educational Achievement Second Edition Brief Form manual Circle Pines, MN: American Guidance Service. Kline, P. (1999). The handbook of psychological testing (2 nd ed.). Lincoln: Routledge.

PAGE 91

9 1 Klingbeil, D. A. (2009 ). A review of the Five Factor Personality Inventory Children. Assessment for Effective Intervention, 35 ( 1 ), 6 1 6 4. Law, K. S., Wong, C., & Song, L. J. (2004). The construct and criterion validity of emotional intelli gence and its potential utility for management studies. Journal of Applied Psychology, 89 (3), 483 496. Lichtenberger, E. O., & Smith, D. R. (2005). Essentials of WIAT II and KTEA II a ssessment Hoboken, NJ: John Wiley & Sons. Livingston, H. A., & Day, A. L. (2005). Comparing the construct and criterion related validity of ability based and mixed model measures of emotional intelligence. Educational and Psychological Measurement, 65 (5) 757 779. MacCann, C., Matthews, G, Zeidner, M., & Roberts, R. D. ( 2003). Psychological assessment of emotional intelligence: A review of self report and performance based testing. International Journal of Organizational Analysis, 11 (3), 247 274. MacCann, C., Roberts, R. D., Matthews, G., & Zeidner, M. (2004). Consensus scoring a nd empirical option weighting of performance based emotional intelligence (EI) tests. Personality and Individual Differences, 36 (3), 645 662. Matthews, G., Roberts, R. D., & Zeidner, M. (2004). Seven myths about emotional intelligence. Psychological Inquiry, 15 (3), 179 196. Matthews, G., Zeidner, M., & Roberts, R. D. (2005). Emotional intelligence: An elusive ability? In O. Wilhelm & R. W. Engle (Eds.), Handbook of understanding and measuring intelligence (pp. 79 99). Thousand Oaks, CA : Sage Publications. Mayer, J. D., Caruso, D. R., & Salovey, P. (1999). Emotional intelligence meets traditional standards for an intelligence. Intelligence, 27 (4) 267 298. Mayer, J. D., & Salovey, P. (1997). What is emotional intelligence? In P. Salo vey & D. J. Sluyter (Eds.), Emotional development and emotional intelligence: Educational implications (pp. 3 31). New York: Basic Books. Mayer, J. D., Salovey, P., & Caruso, D. (2000). Models of emotional intelligence. In R. J. Sternberg (Ed.), Handboo k of intelligence (pp. 396 420). Cambridge, UK: Cambridge University Press. Mayer, J. D., Salovey, P., & Caruso, D. R. (2002 ). Mayer Salovey Caruso Emotional Intelligence Test (MSCEIT) item booklet, Version 2.0 Toronto, ON: MHS Publishers.

PAGE 92

92 McGhee, R.M ., Ehrler, D.J., & Buckhalt, J. (2007). Five Factor Personality Inventory Children. Austin, TX: Pro Ed. McGrew, K. S., & Woodcock, R. W. (2001). Woodcock Johnson III technical manual. Itasca, IL: Riverside. McGrew, K. S., Shrank, F. A., Woodcock, R. W. (2007). Woodcock Johnson III normative update technical manual Rolling Meadow, IL: Riverside. Neisser, U., Boodoo, G., Bouchard, T., Boykin, A. W., Brody, N., Ceci, S. J., Halpern, D. F., Loehlin, J. C., Perloff, R., Sternberg, R. J., & Urbina, S. (1996). Intelligence: Knowns and unknowns. American Psychologist, 51 (2) 77 101. New York State Education Department. (2010). English/language arts and math listings. Retrieved from http://www.p12.nysed.gov/irs/pressRelease/20100728/home.html Newsome, S., Day, A. L., & Catano, V. M. (2000). Assessing the predictive validity of emotional intelligence. Personality and Individual Differences, 29 (6) 1005 1016. inte lligence: Self report versus ability based measures. Personality and Individual Differences, 35 (8), 1893 1902. Parker, J. D. A., Creque, R. E., Barnhart, D. L., Harris, J. I., Majeski, S. A., Wood, L. M., Bond, B. J., & Hogan, M. J. (2004). Academic achi evement in high school: Does emotional intelligence matter? Personality and Individual Differences, 37 (7), 1321 1330. Prez, J. C., Petrides, K. V., & Furnham, A. (2005). Measuring trait emotional intelligence. In R. Schulze and R. D. Roberts (Eds.), Emo tional intelligence: An international handbook (pp. 181 201). Gttingen, Lower Saxony, Germany: Hogrefe & Huber. Peters, C., Kranzler, J. H., & Rossen, E. (2009). Validity of the Mayer Salovey Caruso Emotional Intelligence Test: Youth Version Research Ed ition. Canadian Journal of School Psychology, 24 (1), 76 81. Petrides, K. V., Furnham, A., & Mavroveli, S. (2007). Trait emotional intelligence: Moving forward in the field of EI. In G. Matthews, M. Zeidner, & R. D. Roberts (Eds.), The Science of Emotion al Intelligence (pp. 151 166). Oxford: University Press. The Psychological Corporation. (1999). Wechsler Abbreviated Scale of Intelligence manual San Antonio, TX: Author.

PAGE 93

93 Qualter, P., Whiteley, H. E., Hutchinson, J. M., & Pope, D. J. (2007). Supporti ng the development of emotional intelligence competencies to ease the transition from primary to high school. Educational Psychology in Practice, 23 (1), 79 95. Reiff, H. B., Hatzes, N. M., Bramel, M. H., & Gibbon, T. (2001). The relation of LD and gende r with emotional intelligence in college students. Journal of Learning Disabilities, 34 (1), 66 78. Reynolds, C. R., & Kamphaus, R. W. (1992). Behavior Assessment System for Children Circle Pines, MN: American Guidance Service. Reynolds, C. R., & Kamph aus, R. W. (2004). Behavior Assessment System for Children Second Edition manual Circle Pines, MN: American Guidance Service. Riccio, C. A., & Rodriguez, O. L. (2007). Integration of psychological assessment approaches in school psychology. Psycholog y in the Schools, 44 (3), 243 255. Rivers, S. E., Brackett, M. A., Salovey, P., & Mayer, J. D. (2007). Measuring emotional intelligence as a set of mental abilities. In G. Matthews, M. Zeidner, & R. D. Roberts (Eds.), The science of emotional intelligence : Knowns and unknowns. (pp. 230 257). New York: Oxford University Press. Rossen, E., Kranzler, J., & Algina, J. (2008). Confirmatory factor analysis of the Mayer Salovey Caruso Emotional Intelligence Test V 2.0 (MSCEIT). Personality and Individual Differ ences, 44 (5), 1258 1269. Saklofske, D. H., Austin, E. J., & Minski, P. S. (2003). Factor structure and validity of a trait emotional intelligence measure. Personality and Individual Differences, 34 (4) 707 721. Salovey, P., & Mayer, J. D. (1989 1990). E motional intelligence. Imagination, Cognition, and Personality, 9 (3), 185 211. Santesso, D. L., Reker, D. L., Schmidt, L. A., & Segalowitz, S. J. (2006). Frontal electroencephalogram activation asymmetry, emotional intelligence, and externalizing behav iors in 10 year old children. Child Psychiatry & Human Development, 36 (3), 311 328. Schutte, N. S., Malouff, J. M., Hall, L. E., Haggerty, D. J., Cooper, J. T., Golden, C. J., & Dornheim, L. (1998). Development and validation of a measure of emotional in telligence. Personality and Individual Differences, 25 (2), 167 177. Schwarz, N. (1999). Self reports: How the questions shape the answers. American Psychologist, 54 (2), 93 105.

PAGE 94

94 Shulman, T. E., & Hemenover, S. H. (2006). Is dispositional emotional inte lligence synonymous with personality? Self and Identity, 5 (2), 147 171. Suldo, S. M., & Stewart, T. N. (2008). Test reviews: McGhee, R.M., Ehrler, D.J., & Buckhalt, J. (2007). Five Factor Personality Inventory Children Austin, TX: Pro Ed. Journal of P sychoeducational Assessment, 26 (2), 202 209. van der Zee, K., & Wabeke, R. (2004). Is trait emotional intelligence simply or more than just a trait? European Journal of Personality, 18 (4), 243 263. Van Rooy, D. L., Viswesvaran, C., & Pluta, P. (2005). An evaluation of construct validity: What is this thing called emotional intelligence? Human Performance, 18 (4), 445 462. Wilhelm, O. (2005). Measures of emotional intelligence: Practice and standards. In R. Schulze & R. D. Roberts (Eds.), The internat ional handbook of emotional intelligence (pp. 131 154). Gttingen, Germany: Hogrefe. Wilkinson, G. S. (1993). Wide Range Achievement Test Third Edition administration manual Wilmington, DE: Wide Range. Woitaszewski, S. A., & Aalsma, M. C. (2004). The contribution of emotional intelligence to the social and academic success of gifted adolescents as measured by the Multifactor Emotional Intelligence Scale Adolescent Version. Roeper Review, 27 (1), 25 30. Woodcock, R. W., McGrew, K. S., & Mather, N. (2 001). Woodcock Johnson III Itasca, IL: Riverside Publishing. Zeidner, M., Matthews, G., & Roberts, R. D. (2004). Emotional intelligence in the workplace: A critical review. Applied Psychology: An International Review, 53 (3), 371 399. Zeidner, M., Sha ni Zinovich, I., Matthews, G., & Roberts, R. D. (2005). Assessing emotional intelligence in gifted and non gifted high school students: Outcomes depend on the measures. Intelligence, 33 (4), 369 391.

PAGE 95

95 BIOGRAPHICAL SKETCH Jeffrey William Ditterlin e was born in 1979 and grew up in Syracuse, New York. Henninger High School in 1997. Jeff earned his Bachelor of Arts degree in p sychology from Colgate University in 2001 and his Ma ster of Education degree in school p sychology from the University of Florida in 2007. Jeff completed his pre doctoral internship in th e Syracuse City School District, became a New York Stat e Certified School Psychologist, and accepted a position as a ful l time school psychologist in the Rush Henrietta Central School District in 201 1. He graduated with his Doctor of Philosophy degree in school p sychology from the University of Florida in 2012.