5 Potter

NATIONAL FORUM OF APPLIED EDUCATIONAL RESEARCH JOURNAL VOLUME 21, NUMBER 1, 2007--2008

THE IMPACT OF AN EMOTIONAL INTELLIGENCE INTERVENTION PROGRAM ON FRESHMEN UNIVERSITY STUDENTS George R. Potter Texas A&M University--Kingsville ABSTRACT The study examined the impact of a campus wide Emotional Intelligence intervention program on freshmen students at a South Texas institution of higher education. Results of the study indicated significant improvement in the selfreported emotional intelligence skills of the intervention group (Total EI skill scores). Pre and post-test results for the comparison group were not significant. When controlling for pre-test differences (ANCOVA) the intervention group demonstrated significantly higher scores on the Commitment Ethic scale than the comparison group. These findings suggested an overall positive effect of the EI intervention program on freshmen skill levels.

s

Introduction

ince the early 1900s, intellectual intelligence and standardized scholastic aptitude tests were the foundation for evaluating student academic achievement as well as determining entrance into college. Assessment tools screen applicants to postgraduate programs. A plethora of standardized assessments have been used to make predictions about student success, retention, and success in employment following graduation. There exists a growing body of evidence suggesting that these tests are poor predictors of future behavior and have often been misused (Epstein 1998; Sternberg 1996).

66

67

NATIONAL FORUM OF APPLIED EDUCATIONAL RESEARCH JOURNAL

Many talented and successful students, leaders, and individuals have not been selected because of reliance on these types of tests (Kincheloe, Steinburg, & Gresson, 1996). Measurement of student achievement has been an increasing trend in education today. Texas, like many other states, has mandated levels of proficiency for promotion into the next grade. The problem associated with this trend has been that most, if not all of these measurements, are cognitive factual tests that are singular in dimension (Epstein, 1998). Tests that are of a singular dimension have assessed only one aspect of cognition or recollection as compared to a test that is multi-dimensional and measures several or many aspects of learning or traits. Many studies have indicated that students of low socio-economic, ethnic, and culturally diverse backgrounds have done poorly on these single dimension test types like the SAT (The College Board, “Dimensions of the Minority High Achievement Problem”, 2000). Traditional teaching methods in higher education have centered on the distribution of knowledge held solely by the professor or instructor and imparting this knowledge to students by means of lecturing. Students have had the total responsibility for absorption of the information. Many talented and gifted students, oftentimes first generation college students and students from culturally, ethnically, or economically diverse backgrounds, have failed or struggled with this type of instruction. The loss of potential talent and future leadership is a resource seldom accounted for, and may be lost forever. Lectureonly instruction has been directed toward auditory learning style and the cognitive domain (Francis, K., McDaniel, M., & Doyle, R., 1987). . This type of instruction may have augmented students fluent in the same language as the presenter. Lecture only instruction has not met the needs of students with a different type of learning style such as the students with a different first language. Additionally, lecture style instruction has favored students with college-educated parents and

George R. Potter 68

similar socio-economic, cultural, and ethical backgrounds to the instructor. Sternberg (1996) has suggested that different cultures have a different perception of intelligence. Background of the Problem Promise for advancement in the understanding of factors in student success has been balanced by a lack of consensus on what defines emotional intelligence (Epstein, 1998). Additionally, few universal intervention programs have existed to aid in student success. More research was needed that explore the effects of emotional intelligence on future academic success. Goleman’s work served as a model that evaluated the application of emotional intelligence to success in the business world. Institutions serving a high percentage of minority and lowincome based students have especially needed intervention programs. Additional assets, such as intervention programs, are needed at regional or community colleges. Low income and minority students at the secondary education level, if retained, fail twice as often as middle-class white students. This pattern of failure has continued into the freshman college year if intervention was not provided (Marshall, 2003). Purpose of the Study The study was designed to examine the impact of an emotional intelligence intervention program on freshmen students enrolled in higher education.

69

NATIONAL FORUM OF APPLIED EDUCATIONAL RESEARCH JOURNAL

Research Methodology and Design The study was a mixed method of qualitative and quantitative measures. “Mixed methods in evaluation have [sic] been extensively explored and have been widely accepted as a useful and valid means of gathering diverse information” (Brace-Govan, 2002, p. 50). A qualitative approach was used to discover and identify patterns or themes emerging from data collected from open-ended questionnaire collected during interviews. Quantitative research was used to determine the relationships between pre-test/post-test scores and intervention/control groups. The study consisted of a self-administered pre-test and a post-test assessment with the intervention program given to all subjects participating in the intervention program. A control group received pre-test and post-test without the intervention program, but instead received only an introduction about emotional intelligence. The study was descriptive and inferential and applied a quasiexperimental, nonequivalent control group as a quantitative design. A deductive logical approach was used and the theoretical concept of Emotional Intelligence was narrowed to three hypotheses that were tested using data collection (Trochim, 2002). Quantitative information and data were collected by means of an emotional intelligence assessment and records from a secure database from the computer services of Texas A&M University-Kingsville. The purpose of the study was to evaluate statistically the impact of this intervention program. Personal interviews were used to construct alternative realities that might emerge. The interviews allowed students to discuss the strength and weaknesses of the EI intervention program. The blending of these two research methods, quantitative and qualitative, provided numeric and human data to the study.

George R. Potter 70

Total EI scores utilizing ten subtests were statistically analyses to determine the impact of the intervention program. The within analysis was used to determine differences if any between the intervention group pre-test and post-test , non-intervention pre-test and post-test, and between analyses for differences in the intervention and non-intervention groups. Population and Sample The population for the study consisted of college students enrolled in the Javelina Emotional Intelligence intervention program at an institution of higher education in South Texas. The population consisted of 310 students enrolled in the Spring semester. The student sample was divided into two groups of intervention (n = 179) and nonintervention (n = 131). The total EI sub-test and subsequent statistical computations were based on this original group. Next, because of missing or incomplete data the demographic information was reduced to a number of 133 for the intervention group (n = 133) and nonintervention of 102 (n = 102). Finally, because of missing data, instructors’ non-participation and selection for ANCOVA the sample was further reduced to 72 for intervention group (n = 72) and 52 for the non-intervention group (n = 52). The intervention group sample consisted of 133 participants, 50 were males (n = 50), and 83 females (n = 83). Approximately 70 percent of the participants were Hispanics (69.5%), Whites represented 22 percent (22.1%) and the remaining ethnics composition were 7 percent Blacks (6.9%) and 2 percent others (1.5%). The non-intervention group sample consisted of 102 participants, 69 were males (n = 69), and 33 females (n = 33). Approximately 65 percent of the participants were Hispanics (64.7%), Whites represented 17 percent (16.7%) and the remaining ethnics composition were 16 percent Blacks (15.7%) and 3 percent others (2.9%).

71

NATIONAL FORUM OF APPLIED EDUCATIONAL RESEARCH JOURNAL

For the qualitative portion of the study, interviews were carried out to answer the question: What value does the Javelina Emotional Intelligence intervention program as an academic success model provide as perceived by students? During an interview, five students were asked to respond to questions (Appendix B). All students interviewed had participated in the Javelina Emotional Intelligence program. Students were selected from the total pool of students in the educational orientation class and were selected from three different professors. The interview was held on two separate days within a one week period. The beginning of each interview was initiated by a brief period of time to allow for comfort and rapport to develop. The session lasted for a period of 40 minutes to one hour. Data Collection and Recording The pre-test began with permission being obtained from all participating students and parental permission being obtained for students under the age of eighteen. University personnel for the Javelina Emotional Intelligence intervention program collected the study data. Students were given an explanation and directions for assessment prior to administration and distribution of the assessment. Assessments were given to both control and intervention groups and collected by the individual instructors. Assessment was blinded and scored by university personnel and graduate students trained and supervised under the EI Program. The Emotional Intelligence Intervention Program was given to the intervention groups and a placebo intervention program to the control group. A post-test was given during the last month of the semester and conformed to the procedures of the pre-test. Once scored, the subtest results were entered into a database without inclusion of name to protect anonymity.

George R. Potter 72

Demographic data (gender, age, and ethnicity) were obtained from the computer center of the university. Statistical data reported only summative information, and personal information was not reported. Respondents were asked to respond to the 5 open-ended questions. Each interview session began with a statement to the respondents about the motives and the purposes of the interview, specifically that the interviews were part of a study on the EI intervention program. Additionally, the respondents were told the expected length of the session, the number of sessions, and who had final control over the content of the interview. Observations as well as responses to questions and probes were recorded as field data. Participants The Participants consisted of freshmen and other students enrolled in orientation classes (e.g. EDED-1301) with instructors who choose to participate. The design of the study was to sample approximately 400 students in the intervention arm and 50 to 60 students in the control arm. The Non-Equivalent Group Design (NEGD) was selected. Consideration was given to the statistical implication in the use of NEGD design instead of a randomized experimental design (Trochim, 2002). Because the NEGD was selected as the appropriate design, the use of Analysis of Covariance (ANCOVA) and other appropriate statistical operations were used to compensate for a non-random selection of subjects. The study design assumed that the control group represented an approximate match to the intervention group, yet differences were controlled statistically. The study assumed as acceptable practice the invoking of statistical significance in testing when samples from an approximate representative sample are used (Thompson, 1994).

73

NATIONAL FORUM OF APPLIED EDUCATIONAL RESEARCH JOURNAL

Purposive sampling (Erlandson, Harris, Skipper, & Allen, 1993) was used to select subjects because the major goal of the qualitative portion of the study was discovery of patterns, themes, and problems and not generalization of the findings. The subjects consisted of five students chosen from the total student population. Subjects who agreed to participate were selected from EDED-1301 classes that participated in the Emotional Intelligence intervention program. Quality over quantity is generally an accepted practice in naturalistic inquiry with emphasis on richness and depth of information (Erlandson, et al., 1993).

Analysis of Data A within-group analysis was conducted to analyze the total EI scores of the intervention group, comparing pre-test to post-test EI skill scores. The total score is the cumulative scores of the ten subtests (assertion, comfort, empathy, decision making, leadership, drive strength, time management, commitment ethics, self-esteem, and stress management). A general linear model repeated-measures ANOVA indicated a significant time difference, Wilks’s Λ = .938, F (1, 71) = 4.70, p = .034, multivariate η2 = .062. The students who received the intervention showed significant improvement in their scores between the first and second administration of the EI skill scores. The results are shown in Table 1. Based on the significant improvement of the intervention group as demonstrated by total EI scores and statistical implications indicated by ANOVA, the first null hypothesis was rejected.

George R. Potter 74

Table 1 Within-Group Analysis of Variance for Total Emotional Intelligence Measures and Test (Pre/Post), Intervention Group Effect Time3

Wilks' Lambda

Value

F

df

df2

p

η2

.938

4.698

1

71

.034

.062

A within-group analysis was conducted to analyze the total EI scores of the non-intervention group pre-test compared to post-test scores. The total score is the cumulative scores of the ten subtests (assertion, comfort, empathy, decision making, leadership, drive strength, time management, commitment ethics, self-esteem, and time management). A general linear model repeated-measures ANOVA indicated no significant difference, Wilks’s Λ = .929, F (1, 51) = 3.875, p =.054, multivariate η2 = .071. The results are shown in Table 2. The non-intervention group indicated no significant change in EI scores between the first and second administration of the instrument. Based on no significant improvement the non-intervention group as demonstrated by total EI scores as analyzed by ANOVA, the second null hypothesis failed to be rejected.

75

NATIONAL FORUM OF APPLIED EDUCATIONAL RESEARCH JOURNAL

Table 2 Between-Group Analysis of Variance for Total Emotional Intelligence Measures and Test (Pre/Post), Non-Intervention Group Effect Tim3

Wilks' Lambda

Value

F

df

df2

p

.929

3.875

1

51

.054

η2 .071

To answer the research question of whether or not there was a significant difference between students score in pre-test, scores between the intervention and control group, the General Linear Model (GLM) procedure analysis of covariance (ANCOVA) was conducted. The factor was group (intervention and non-intervention) and the dependent variable was the post-score on the EI scale. The covariate was the pre-test score for the EI scale. The Exploring and Developing -Emotional Intelligence Skills (EDEIS) scales included, assertion, comfort, empathy, decision making, leadership, drive strength, time management, commitment ethics, self-esteem, and time management. Additionally, the total EI score was analyzed. The ANCOVA indicated a statistically significant difference in scores on the commitment ethic scale, F (1,121) = 5.62, p = .019, partial η2 = .044, while the differences in the other scale items were not significant. The results are shown in Table 3. The results of the ANCOVA demonstrated significant difference in the adjusted means between the post-test scores of the intervention and non-intervention group for the commitment ethics scale, therefore Null Hypothesis 3 was rejected for that scale only.

Table 3

George R. Potter 76

Analysis of Covariance Results Source

F

Assertion Comfort Empathy Decision Leadership Drive

0.10 0.40 1.55 0.03 0.44 0.31

Time Commit SelfStress * p < .05

0.05 5.62 0.02 0.04

f

d

df2

p

η2

1 1 1 1 1 1

120 120 120 119 121 121

0.752 0.526 0.215 00.856 0.510 00.577

.001 .003 .013 .000 .004 .003

1 1 1 1

121 121 121 121

0.823 00.019* 0.890 0.851

.000 .044 .000 .000

Summary and Conclusions In summary, data from the emotional intelligence intervention program were analyzed using General Linear Model (GLM) procedure repeated-measures, Analysis of Variance and Analysis of Covariance. The following is a summary of the findings: 1. The students who received the intervention showed significant improvement in their scores between the first and second administration of the EI skill scores. In general, the results suggested a change in measurable attitude, perception, and behavior. Since this change is not observed or measured in the non-intervention group, other factors like maturation, exposure, or learning that occurred as a result of the instrument itself is not the probable cause. The improvement of the

77

NATIONAL FORUM OF APPLIED EDUCATIONAL RESEARCH JOURNAL

intervention group is suggested to be related to the EI intervention program. 2. The non-intervention group demonstrated no significant change in EI scores between the first and second administration of the instrument. The results of the nonintervention group suggested a lack of change in attitude, perception, and behavior because of limited exposure to meaningful concepts and activities that are provided in the EI intervention program. 3. The results of the ANCOVA demonstrated significant

difference in the adjusted means between the post-test scores of the intervention and non-intervention group for the commitment ethic scale. The concept of commitment ethic is founded in ideas of succeeding in difficult situations, persistence, and a focus on success. An example of this is question 6 of the assessment survey that states: “When I decide to do something, I carry through and do it”. Although not central to the EI intervention program commitment ethic is a value that is implied and the concept of following through and not quitting is presented in the EI intervention program. The differences between the intervention and non-intervention groups is important in the fact that with varied presenters of the EI intervention program this concept still permeated and was significant across the entire group. 4. Qualitative results supported quantitative finding that time management had impact and was reported as a central theme in all interviews. All participants interviewed stated that they were positively influenced by the EI intervention program. Other areas identified as meaningful were stress management, assertiveness, and goal setting. The importance of the qualitative results is that a year after the EI intervention program students still demonstrated changes in their behavior

George R. Potter 78

with regard to time management, could remember cognitive changes, and decisions they made based on information learned from the EI intervention program. 5. Areas central to the emotional intelligence intervention program had demonstrated short term impact as demonstrated by quantitative data and long term impact as demonstrated by qualitative data. Emotional intelligence as an intervention program appeared to have short term and long term impact on student behavior. The major focus of the Javelina Emotional Intelligence intervention program was a seven step program that focused on assertion, drive strength, time management, and commitment ethics. The within analysis conducted with the intervention group demonstrated significant changes in the time between pre-test and post-test for the total scale scores. The same within group analysis did not demonstrate significant changes between the time of the pre-test and post-test for the non-intervention group. The logical assumption is that the EI intervention program helped participants make positive changes in attitudes, conception and behaviors that are related to student success. Skills like study habits that were derived from the presentations on time management and other topics. Other skills like the importance of assertion not aggression in aiding in class success seemed to have influenced participants. When the two groups were compared by between group analysis (ANCOVA) significance was identified in the area of commitment ethic. Commitment ethic is an area that is indirectly presented by the EI presenters and the concept of persistence, not giving up, and achieving success is interwoven throughout the concept of Emotional Intelligence.

79

NATIONAL FORUM OF APPLIED EDUCATIONAL RESEARCH JOURNAL

With the exception of assertion, the areas of significance were implicated as areas of change as measured from pre-test to post-test on the Emotional Skills Assessment Process assessment. Both the intervention group and the non-intervention group demonstrated significance. A higher level of difference demonstrated by the intervention group may be a reflection of the fact that longer more intense instruction and intervention was given to the intervention group. Qualitative data demonstrated changes in assertion, time management from direct themes and drive strength was assumed to be implied based on interviews. The theme of time management was expressed by all five respondents. Student 1 discussed changes in her studies directly related to her experiences as part of the emotional intelligence program. Student 2 expressed “time management helped me a lot. Lisa P. described that the one activity that stood out was time management and Student 4 stated that “time management really helped”, especially the activity tracking time expenditure for one week. Additionally, 4 out of the 5 students expressed help from the theme of assertion.

George R. Potter 80

REFERENCES Brace-Gavan, J. (2002). Qualitative Data for a Quantitative Audience: An Evaluation Framework of Analysis. Qualitative Journal, Vol.2 No.1, 47-55. College Board, (2000). Advancing minority high achievement: national trends and promising programs and practices. Retrieved on 7-3-04 from http://www.collegeboard.com /about/association/academic/taskforce/MinorityHighAch.pdf Erlandson, D., Harris, E., Skipper, B. and Allen, S. (1993). Doing naturalistic inquiry: A guide to methods. CA: Sage Publications, Inc. Epstein, S. (1998) Constructive thinking: The key to emotional intelligence. West Port, CT: Greenwood Publishing Group, Inc. Francis, K., McDaniel, M., and Doyle, R. (1987). Training in role communication skills: Effect on interpersonal and academic skills of high risk freshmen. Journal of College Student Personnel. March Vol 28(2) 151-156. Kincheloe, J.L., Steinburg, S., Gresson, A.D., (1996). Measured lies, New York: St. Martin’s Press. Marshall, R. L. (2003). The pivotal year how freshmen can become sophomores. Lanham, Maryland: Scarecrow Press. Sternberg, R., (1996). Successful intelligence, NY: Simon & Schuster. Trochim, W. (2002). Research methods knowledge base. Retrieved on 10-12-2003 from http://www.socialresearchmethods.net/kb/index.htm