EQ-i 2.0 Higher Education Norms Supplement
The release of the Higher Education Norms provides users with the ability to score their clients against data collected from a population enrolled in post-secondary education. This chapter is designed to provide normative and psychometric information particular to the Higher Education population in North America. The EQ-i® 2.0 assessment remains unchanged, but an additional normative sample is now available (the original North American normative samples are described in detail in Standardization, Reliability, and Validity).
This chapter describes the development of the EQ-i 2.0 Higher Education (EQ-i 2.0:HEd) normative sample. For information on the EQ-i 2.0, including administration, interpretation, and development of the North American Norms, please refer to Parts I–V of the EQ-i 2.0 User’s Handbook. |
This supplement begins with a description of the relationship between emotional intelligence and higher education. The features and benefits of each of the EQ-i 2.0:HEd report types (i.e., Student Summary Report, Student Comprehensive Report, Counselor’s Report) are also described. The remainder of the supplement is devoted to the development, standardization, reliability, and validity of the EQ-i 2.0 Higher Education Norms. The data were collected from 1,800 college and university undergraduate students enrolled in a variety of 3- or 4-year programs within the United States and Canada; data were evenly proportioned by gender and year of study (i.e., Year 1 to Year 4).
Overall norms (collapsed across gender and year of study) are provided; however, gender-specific norms are also available as a scoring option because several small differences in EQ-i 2.0 scores were found between male and female students. Women scored higher than men on several subscales (i.e., Empathy, Social Responsibility, Self-Actualization, Emotional Self-Awareness, Emotional Expression, Interpersonal Relationships, and Happiness), while men scored higher than women on the Stress Tolerance subscale. In contrast, EQ-i 2.0 scores did not change as year of study increased, therefore separate year of study norms are not provided.
Strong evidence of reliability and validity for the use of the EQ-i 2.0 in higher education samples was found. EQ-i 2.0 scores were found to be highly reliable in the Higher Education normative sample (with both strong internal consistency and test-retest reliability), and the factor structure that was developed with the North American General Population normative data was replicated with the Higher Education sample. As expected, a strong link was observed between emotional intelligence and academic achievement; students with low grade point averages (GPA) had lower EQ-i 2.0 scores than did students with moderate or high GPAs. Finally, results indicated that the EQ-i 2.0: HEd norms can be used across different programs of study (i.e., negligible differences were found between different programs of study including Arts, Science, Business, and Other), and with students of different race/ethnicities (i.e., scores were approximately the same for students of different racial/ethnic backgrounds).
Academic success is not just determined by cognitive intelligence skills, but is also a function of factors that are emotional and social in nature (Parker, Summerfeldt, Hogan, & Majeski, 2004). Many potentially stressful lifestyle changes take place during the first years of post-secondary education, such as learning to live independently, developing new friendships, and keeping up with the demands of a new academic environment. Students are also laden with increased financial stress, in addition to balancing part-time work and studying, as well as competing for fewer jobs (with higher education requirements than ever before). In fact, almost a third of first generation students (i.e., students whose parents did not attend college) report having major concerns about their ability to continue financing their college education (Your First College Year, 2012). Successful adaptation to these lifestyle changes is directly related to emotional intelligence (EI).
In the largest, most representative student survey that has been conducted by the Higher Education Research Institute (HERI), over 38,000 first-year students were surveyed from more than 144 post-secondary institutions (Your First College Year, 2005). This survey indicated that a large portion of first-year students felt overwhelmed (39%) or lonely (49%), or were worried about meeting new people (42%). They indicated concerns about having to break away from their families in order to succeed. Students described themselves as going through serious changes in self-concept during the first year of college or university. It is clear that the majority of these issues are not about academic skills or cognitive intelligence, but pertain directly to emotional and social skills that are connected to EI.
EQ-i 2.0:HEd results provide a framework for understanding one’s current EI skills in order to foster academic success. Areas of strength and areas for improvement are pinpointed. Finally, results can link directly to strategies and curriculum to improve facets of EI (e.g., through self-improvement techniques and training), and thus improve the likelihood of academic success.
Similar to the EQ-i 2.0, the EQ-i 2.0:HEd was designed based on feedback from customers who use EI in higher education settings. Some of the new report features were designed to help better address everyday challenges, such as gaining buy-in for EI, customizing reports for the institution (i.e., brand and campus resources), and saving time in the administration and debriefing process.
EQ-i 2.0: HEd Report Types
The EQ-i 2.0:HEd contains three reports that are all included in the cost of a single assessment: a Student Summary Report, a Student Comprehensive Report, and a Counselor’s Report. Students meeting certain criteria (Alert status, described below) require follow-up by the institution, and it is strongly recommended that these students are given the information and support provided in the Student Comprehensive Report.
The Student Summary Report
The Student Summary Report is a 7-page report that describes the entire EQ-i 2.0 model and provides an interpretive summary of the student’s results on 6 of the subscales (their top three and bottom three scores). Results for the remaining subscales are shown with specific development strategies for all 15 subscales to ensure the students have plenty of resources to improve all of their EI skills, should they desire to work on areas outside of their top and bottom scores. Actual scores are not included in the Student Summary Report; only interpretive graphs are presented.
The Student Summary Report can be automatically generated and emailed to the student upon item completion. This report is best incorporated into a feedback session with the student and an appropriately qualified counselor; this session could include sharing details from the Student Comprehensive Report.
The Student Comprehensive Report
The Student Comprehensive Report is a 13-page report that contains results, interpretive text, and development strategies for all 15 subscales. One report option is to include scores that provide specific, numeric feedback. Due to the depth of information in this report and the relative maturity required to effectively plan and carry out the developmental strategies, this report should only be given to the student under the direct supervision of an appropriately qualified counselor.
The Counselor's Report
The Counselor’s Report is a 10-page report that contains all the information required to debrief the student on his/her results. The validity indicators are the same as those in the Workplace and Leadership Reports. Please refer to Step 1: Assess the Validity of the Results in Part IV for a step-by-step guide to assessing the validity of results.
A unique feature of the Counselor’s Report is the addition of “Alert status”.
You will see the following statement at the top of page 2 in the Counselor’s Report if a student has an Alert status: |
In the EQ-i 2.0 Portal, there is an Alert column when managing and generating reports. Alerted profiles indicate a need for further investigation on the part of the administrator. Alerting is a preliminary check of students’ results designed to help prioritize feedback.
For administrators who are familiar with the Higher Education Report for the Bar-On EQ-i, the Alert status is similar to “Flagged Students”; however, the criteria for receiving an alert is slightly different than a flag in the original report. Although Bar-On’s original Self-Contentment subscales are still being used, a more holistic approach now includes identifying students who are struggling in any of the areas measured by the EQ-i 2.0. |
An Alert status is based on a set of 5 primary subscales: Happiness, Optimism, Self-Regard, and Self-Actualization, which according to Bar-On (1997) all relate to self-contentment, as well as a fifth subscale, Interpersonal Relationships.
The first four of these subscales are important to consider because evidence shows they are indicative of physical and psychological health (Bar-On, 1997; Krivoy, Weyl Ben-Arush, & Bar-On, 2000) and academic functioning (Parker, Creque, Harris, Majeski, Wood, & Hogan, 2004; Parker, Summerfeldt, Hogan, & Majeski, 2004). Interpersonal Relationships is also included among the primary scales because there is evidence that strong relationships and social support act as buffers against negative affect associated with chronically stressful conditions, and also bolster physical health and survival (Taylor, 2011). College years can certainly be considered chronically stressful as most students deal with the difficulties of forming new relationships, living apart from loved ones, and adjusting to new study habits. Parker, Duffy, Wood, Bond and Hogan (2005) also found a link between interpersonal skills and academic success; successful students use their interpersonal skills to form new relationships and maintain older relationships as they adjust to college life.
Keefer, Parker, and Wood (2012) provide evidence supporting the use of an Alert status in higher education. Their research suggests that targeted intervention efforts will offer the greatest impact when focusing on students with low EI scores. Individuals with low EI are more likely to be unsuccessful in completing their post-secondary education when compared to those with average scores. Yet, individuals with high EI do not have an increased likelihood of completing their post-secondary education when compared to those with average scores. Therefore, directing efforts toward raising students’ low EI scores into average ranges will provide the highest return in post-secondary retention rates.
In addition to the primary subcales, an Alert status takes into consideration lower functioning across a student’s entire EQ-i 2.0 profile. This holistic approach ensures that administrators are made aware of students who are struggling in other areas of EI, which could lead to possible disengagement in college.
An Alert status identifies students who are low in specific EI skills; as a result, they may be less likely to do well in school, or worse, they may not graduate. It is strongly recommended to meet with any student with an Alert status and use the debriefing guide contained in the Counselor’s Report to investigate any areas of concern.
A student is marked as “Alert” if at least one of the following conditions is met:
- At least one of the primary subscales (Happiness, Optimism, Self-Regard, Self-Actualization, Interpersonal Relationships) has a score below 70.
- At least three of the primary subscales have scores between 70 and 84.
- One or two of the primary subscales have scores between 70 and 84, and Total EI is less than 85.
- At least one of the other subscales has a score below 70, and Total EI is less than 85.
This section describes the psychometric properties of the EQ-i 2.0 Higher Education Norms, including standardization, reliability, and validity information.
All tables and figures representing detailed depictions of the analyses described in this chapter are available in Appendix G. |
NORMATIVE SAMPLE
Normative data for the EQ-i 2.0 Higher Education Norm sample (N = 1,800) were collected from March, 2010 to May, 2010. All participants in the normative sample were 18–24 years of age (M = 20.3 years, SD = 1.5 years), and were full-time students enrolled in post-secondary education. The normative data were collected evenly across year of study (i.e., Year 1 to Year 4) and evenly proportioned by gender (see Table G.1). The sample breakdown by program of study is as follows (see Table G.2): Arts (35.6%), Sciences (35.6%), Business/Management (20.0%), and Other (including Fine Arts; 8.9%). The majority of students (97.2%) were enrolled in a 3–4 year college/university program, with 2.8% enrolled in a 2–3 year college/university program (see Table G.2).
Data were gathered from all 50 U.S. states and the District of Columbia, as well as from 9 Canadian provinces (see Table G.3 for regional distributions). The race/ethnicity distribution of the normative sample is very similar to the U.S. and Canadian Census targets (i.e., Statistics Canada, 2006; U.S. Bureau of the Census, 2010; see Table G.4).
NORMING PROCEDURES
The first step in the preparation of the Higher Education Norms was to determine if demographic trends existed in the data, particularly with regard to gender and year of study. Age analyses were not conducted, as the sample was confined to a limited age range (i.e., 18–24 year-old college/university undergraduate students).
Large differences in scores between men and women, or across years of study, would suggest a need to create an option for separate gender- or year-based norm groups. Conversely, a lack of such differences may dictate the use of a single norm group with genders and year of study groups combined.
A series of analyses of variance (ANOVA) was used to examine the relationship between EQ-i 2.0 scores and both gender and year of study. To better control for Type I errors that might occur with multiple analyses, a more conservative criterion of p < .01 was used for all F-tests to test for statistical significance.
Gender Effects. Results of the gender analyses showed that men and women in the Higher Education normative sample did not differ significantly on the Total EI score, indicating that overall emotional intelligence (as measured by the EQ-i 2.0) is about the same for men and women. However, significant differences were seen on a number of scales and subscales, most of these having small or small-medium effect sizes (see Table G.5 for effect sizes and Table G.6 for descriptive statistics and significance test results).
The largest gender differences seen in the Higher Education normative sample were on the Empathy subscale (d = -0.36) and the Interpersonal composite scale (d = -0.34), with women scoring higher than men. Smaller differences were found with women scoring higher than men on Social Responsibility (d = -0.28), Self-Actualization (d = -0.25), Emotional Self-Awareness (d = -0.24), Emotional Expression (d = -0.24), Interpersonal Relationships (d = -0.22), and Happiness (d = -0.22). Men scored higher than women with a small effect size on Stress Tolerance (d = 0.33). These gender differences found in the Higher Education sample are comparable to the results found in the EQ-i 2.0 General Population normative sample (see Standardization, Reliability, and Validity). However, it is important to note that these effects were small and represent only a few absolute standard score points.
Year of Study Effects. Only the Decision Making composite scale attained a notable effect size (η2 ≥ .01). See Table G.5 for effect sizes and Table G.7 for descriptive statistics and significance test results.
Gender × Year of Study Interaction. None of the scales showed a significant interaction effect between gender and year of study, or reached the minimum partial η2 criterion for a small effect size. Overall, year of study effects were largely consistent within men and women, and gender effects were largely consistent across years of study.
Norm Groups and Norm Construction. The gender analyses revealed a number of significant effects that were small in size. Therefore, specific gender norms as well as overall norms (i.e., collapsed across genders) were both developed. The lack of meaningful differences across years of study indicated that separate norms were not required on this basis.
Results revealed that skewness and kurtosis values were not large enough to suggest that a normalizing transformation was necessary (skewness values across scales ranged from -0.81 to 0.07; kurtosis values ranged from -0.31 to 0.43). An examination of the scale histograms did not reveal any significant departures from a bell-shaped (Gaussian) curve (Figure G.1 shows a histogram for the EQ-i 2.0 Higher Education Total EI score). Actual construction of the norms was conducted in the same manner as the North American General Population Norms (see Standardization, Reliability, and Validity for more information on the construction of these norms).
Comparison of Higher Education Norms to North American General Population Norms. The Higher Education normative sample was compared to the North American General Population normative sample by computing standard scores for the EQ-i 2.0 scales with the North American General Population norms, and comparing these scores against a mean of 100. Mean differences ranged from 8.3 standard score points below (Independence) to 2.5 standard score points above (Self-Actualization) the North American General Population mean of 100. Due to the large sample size, significant differences were observed between both samples on most scales, however only the following seven scales reached notable effect sizes, all indicating lower scores for the Higher Education normative sample: the Independence (d = -0.54), Problem Solving (d = -0.42), Flexibility (d = -0.34), and Impulse Control (d = -0.20) subscales, and the Self-Expression (d = -0.28), Decision Making (d = -0.28), and Stress Management (d = -0.20) composite scales. Results are presented in Table G.8. These results indicate the usefulness of separate Higher Education Norms, which provide a more relevant basis of comparison for this population.
Reliability
INTERNAL CONSISTENCY
Internal consistency, a measure of reliability, conveys the degree to which a set of items are associated with one another. A high level of internal consistency suggests that the set of items are measuring a single, cohesive construct. Internal consistency is typically measured using Cronbach’s alpha (Cronbach, 1951). Cronbach’s alpha ranges from 0.0 to 1.0 and is a function of both the interrelatedness of the items in a test or scale and the length of the test (John & Benet-Martinez, 2000). Higher values reflect higher internal consistency.
Cronbach’s alpha values for the EQ-i 2.0 scales in the Higher Education normative sample are presented in Table G.9. Although there is no universal criterion for a good alpha level, informal cut-offs for evaluating alpha are typically as follows: .90 is “excellent,” .80 is “good,” .70 is “acceptable,” and lower than .70 is “questionable.” The values shown in Table G.9 demonstrate favorable internal consistency, especially given the small number of items included in most subscales. For the overall sample, the alpha value of the Total EI scale was .97, values for the composite scales ranged from .86 to .92, and values for the subscales ranged from .73 to .91. Similar patterns were seen across gender groups. The high level of internal consistency found in the Total EI score supports the idea that the EQ-i 2.0 items are measuring an overarching cohesive construct, namely emotional intelligence. Additional items allocated to specific composite scales and subscales also show high reliability, supporting the individual components of emotional intelligence that make up the EQ-i 2.0.
TEST-RETEST RELIABILITY AND STABILITY
The test-retest reliability of an assessment refers to the stability or consistency of scores over time. This type of reliability is typically calculated by examining the correlation between individuals’ scores on the same assessment at two different times. This time interval must not be too long (Anastasi, 1982) to ensure that certain factors (e.g., developmental changes) do not overly obscure the assessment of the instrument’s reliability, and must not be too short as to be contaminated by memory effects (Downie & Heath, 1970).
When test-retest reliability is assessed at the group level, high correlations indicate that the rank-order of individuals’ assessment scores have remained consistent over time. However, differences in mean scores may confound the interpretation of the results. For example, if each individual’s score increases or decreases in a dramatic, but uniform, manner over time, the test-retest correlation will be high, even though the scores have changed. Test-retest stability analyses can be used to determine whether the actual scores remain stable over time.
EQ-i 2.0 test-retest data was available for 191 Higher Education students who were assessed approximately two to three weeks apart (mean interval = 16.8 days, SD = 1.0 day). The sample was 51.3% female, with a mean age of 20.6 years (SD = 1.3 years). With respect to race/ethnicity, the sample was 73.8% White, 13.6% Hispanic, 8.4% Black, and 4.2% Other.
Test-retest correlations (see Table G.10) were very high for the Total EI (r = .90) and composite scale scores, ranging from r = .77 (Decision Making) to r = .89 (Stress Management). Test-retest correlations across subscales were also high, ranging from r = .69 (Impulse Control) to r = .87 (Self-Regard).
Results demonstrating the stability of the EQ-i 2.0 scores are presented in Table G.11. The results suggest scores remained highly stable over time. Positive mean differences between Time 1 and Time 2 scores indicate an increase in scores over time; negative mean differences indicate that scores decreased over time. For the majority of the scales and subscales, over 90% of the individuals’ scores did not change by more than one normative standard deviation (i.e., 15 standard score points) between administrations. Finally, the 95% confidence intervals surrounding the mean difference were consistently small, and for all scales this interval encapsulates zero, indicating no significant difference in the means across the two administrations. These results provide support for the temporal stability of the EQ-i 2.0 in the Higher Education population.
Validity
EXPLORATORY FACTOR ANALYSIS
Exploratory factor analysis (EFA) was used to determine whether the subscales established with the North American EQ-i 2.0 General Population normative data empirically emerge from the Higher Education normative data. Five EFAs were conducted; the items within each composite scale were analyzed separately. In each EFA, a three-factor solution was forced to examine whether the items corresponding to each subscale within the composite loaded together in the Higher Education normative data. As with the North American General Population normative data, principal axis factoring extraction was used, with direct oblimin (i.e., oblique) rotation, as the factors within each composite are expected to correlate with each other. Reverse scoring was applied to relevant items prior to the analysis. Factor loadings were considered significant if they reached at least ± .30, and an item was defined as cross-loading if it was significant on more than one factor and the loadings were within .10 of each other.
For the Self-Perception Composite EFA, all items for the Self-Regard, Self-Actualization, and Emotional Self-Awareness subscales loaded together as expected by the established factor structure (i.e., items loaded significantly onto their respective factors, with no cross-loadings).
For the Self-Expression Composite EFA, all items for the Emotional Expression, Assertiveness, and Independence subscales loaded significantly onto their respective factors, with the exception of one Independence item that cross-loaded with Assertiveness.
For the Interpersonal Composite EFA, items for the Interpersonal Relationships, Empathy, and Social Responsibility subscales loaded onto their respective factors, with the exception of one Interpersonal Relationships item that loaded onto Empathy.
For the Decision Making Composite EFA, Problem Solving, Reality Testing, and Impulse Control items loaded onto their respective factors with no cross-loadings, with the exception of two Impulse Control items that had factor loadings of .29 and .26, which were very close to the cut-off.
For the Stress Management Composite EFA, all Flexibility, Stress Tolerance, and Optimism items loaded onto their respective factors with no cross-loadings.
To summarize, the EFAs generated solutions that strongly correspond to the established EQ-i 2.0 factor structure.
CORRELATIONS AMONG EQ-i 2.0 COMPOSITE SCALES AND SUBSCALES
Correlations among the EQ-i 2.0 composite scales and subscales were examined, and it was expected that these correlations would generally be high, given that they are all measuring the same underlying construct of emotional intelligence. However, they should not be so high as to indicate redundancy between the scales. Correlations observed in the Higher Education normative sample are presented in Tables G.12 (composite scales) and G.13 (subscales). These results are similar to what was found with the North American General Population normative sample.
All but one composite scale correlation reached a large effect size, ranging from r = .42 (Interpersonal/Decision Making) to r = .72 (Interpersonal/Self-Perception), with an average correlation of r = .62. Subscale correlations were also of the expected magnitude. As highlighted in Table G.13, most subscale correlations within a composite reached at least a medium effect size and a third of these correlations reached a large effect size, ranging from r = .23 (Reality Testing/Impulse Control) to r = .63 (Self-Regard/Self-Actualization). These results support the notion that a single, underlying dimension is being represented in the EQ-i 2.0, yet the values are not overly high and there is enough variation in the correlations to provide clear evidence of the multidimensional nature of the assessment, and support the existence of composite scales and subscales.
EMOTIONAL INTELLIGENCE AND ACADEMIC ACHIEVEMENT
A series of analyses of variance (ANOVA) was used to examine the relationship between grade point average (GPA) and EQ-i 2.0 scores. The normative sample was divided into groups based on self-reported GPA scores. The High GPA group (N = 706) comprised students with GPAs greater than or equal to 3.5 (i.e., an average grade of A). The Middle GPA group (N = 969) comprised students with GPAs between 2.5 and 3.4 (i.e., an average grade of B). The Low GPA group (N = 96) comprised students with GPAs equal to or lower than 2.4 (i.e., an average grade of C or lower). Twenty-nine students indicated that they did not know their GPA, and were therefore excluded from this analysis.
To better control for Type I errors that might occur with multiple analyses, a more conservative criterion of p < .01 was used for all F-tests. Results showed a significant effect for the Total EI score, as well as for all composite scales and most subscales. See Table G.14 for descriptive statistics, significance test results, and overall effect sizes (partial η2).
To more closely examine the relationship between EQ-i 2.0 scores and GPA, effect sizes were computed for each pairwise comparison between GPA groups (Cohen’s d; see Table G.15). The typical pattern indicated that students with higher GPAs tended to have higher EQ-i 2.0 scores than did students with lower GPAs. Small to large effects were observed between the lowest GPA group and the higher GPA groups on all scales, with all effects showing lower EQ-i 2.0 scores for the Low GPA group. The strongest effects were seen for Self-Actualization (d = 1.05 Low vs. High GPA group, and d = 0.72 Low vs. Middle GPA group). Medium-sized effects were observed on a number of scales when comparing the Low GPA group against the other two GPA groups, including the Total EI score (d = 0.60 Low vs. High GPA and d = 0.54 Low vs. Middle GPA), the Self-Perception (d = 0.73 Low vs. High GPA and d = 0.60 Low vs. Middle GPA), Decision Making (d = 0.50 Low vs. High GPA and d = 0.41 Low vs. Middle GPA), and Stress Management (d = 0.44 Low vs. High GPA and d = 0.53 Low vs. Middle GPA) composites, and the Social Responsibility (d = 0.56 Low vs. High GPA and d = 0.39 Low vs. Middle GPA), Optimism (d = 0.53 Low vs. High GPA and d = 0.55 Low vs. Middle GPA), and Happiness (d = 0.55 Low vs. High GPA and d = 0.52 Low vs. Middle GPA) subscales. Small effects were observed on all other scales. Little difference was observed between the High and Middle GPA groups; only Self-Actualization (d = 0.27 Middle vs. High GPA) and Flexibility (d = -0.23 Middle vs. High GPA) showed meaningful effects.
These results demonstrate a strong link between emotional intelligence and academic achievement, across all facets of social and emotional functioning as measured by the EQ-i 2.0. These results clearly show that those with low GPAs display lower levels of EI skills. In fact, the Low GPA group scored below 100 on all scales (ranging from 88.4 to 97.1), whereas the Middle and High GPA groups scored above 100 on most scales (the lowest score obtained by the High GPA group was 98.2 on Flexibility).
PROGRAM OF STUDY EFFECTS
A series of analyses of variance (ANOVA) was used to examine the relationship between program of study and EQ-i 2.0 scores. The four program of study groups compared in these analyses were Arts (N = 640), Sciences (N = 640), Business (N = 360), and Other (including Fine Arts; N = 160). To better control for Type I errors that might occur with multiple analyses, a more conservative criterion of p < .01 was used for all F-tests.
Most scales did not show any significant effects, demonstrating negligible differences between students of different programs. The only scale showing a significant effect with at least a small effect size was Self-Regard. See Table G.16 for descriptive statistics, significance test results, and overall effect sizes (partial η2). To more closely examine the relationship between EQ-i 2.0 scores and program of study for Self-Regard, effect sizes were computed for each pairwise comparison between groups (Cohen’s d). Results indicated that Business students, as well as those in the Other/Fine Arts group, scored slightly higher than Arts students on Self-Regard (d = 0.27 and 0.20, respectively).
Other than these small effects on Self-Regard, there did not appear to be a relationship between program of study and EI skills. These results indicate that the EQ-i 2.0 can be used effectively with students enrolled in different areas of study.
COMPARISONS AMONG RACIAL/ETHNIC GROUPS
The examination of potential racial or ethnic differences or biases is of critical importance in the development of an assessment. It is vital to ensure that assessment scores do not show large differences among racial/ethnic groups that are not expected since that may indicate test bias. A series of analyses of variance (ANOVA) was used to examine the relationship between race/ethnicity and EQ-i 2.0 scores. To better control for Type I errors that might occur with multiple analyses, a more conservative criterion of p < .01 was used for all F-tests.
Results demonstrated that for most scales, the effect of race/ethnicity on EQ-i 2.0 scores was not statistically significant. The only scales showing significant effects with at least a small effect size were Independence and Emotional Self-Awareness. See Table G.17 for descriptive statistics, significance test results, and overall effect sizes (partial η2).
To more closely examine the relationship between EQ-i 2.0 scores and race/ethnicity for these two subscales, effect sizes were computed for each pairwise comparison between groups (Cohen’s d). No meaningful differences were observed between the White and Hispanic groups, but there were differences between the Black group and the other two groups (with the Black group scoring higher). Small-medium effect sizes were observed between the Black group and the White group on both Independence (d = -0.43) and Emotional Self-Awareness (d = -0.32). Small effects were also seen between the Black group and the Hispanic group on both Independence (d = -0.28) and Emotional Self-Awareness (d = -0.21).
These results demonstrate that the EQ-i 2.0 does not show strong differences among racial/ethnic groups in the Higher Education normative sample, and there was no evidence of test bias toward minority groups.