Part V: creating the EQ-i 2.0 and EQ 360 2.0

UK and Ireland Norms

EQ-i 2.0 UK and Ireland Standardisation

This section describes the psychometric properties of the UK and Ireland norms for the EQ-i 2.0, including standardisation, reliability, and validity. The first section is devoted to the standardisation, reliability, and validity of the EQ-i 2.0 General Population Norm sample, followed by a section describing these same properties for the EQ-i 2.0 Professional Norm sample.

All tables and figures representing detailed depictions of the analyses described in this chapter are available in Appendix C.

EQ-i 2.0 UK and Ireland General Population Norm

Standardisation

NORMATIVE SAMPLE

Normative data for the UK and Ireland General Population Norm sample (N = 1,450) were collected in October and November of 2011.

The demographic composition of the normative sample is shown in Tables C.1–C.4. The General Population Norm sample was collected across five age ranges, evenly proportioned by gender within each age interval. The normative sample is very close to UK and Ireland Census targets in terms of race/ethnicity and geographic region. Education level targets, however, deviated somewhat from Census targets; therefore, the normative samples were statistically weighted in order to create a sample that more closely matched Census targets. The weighted normative sample, and therefore the reference group against which EQ-i 2.0 scores are compared, can be considered representative of the UK and Ireland general populations. In terms of employment status, 85.4% of the sample were employed or self-employed, 2.0% were unemployed, 11.2% were retired, and 1.4% indicated “other”.

NORMING PROCEDURES

The first step in the preparation of the UK and Ireland norms was to determine if any age or gender trends existed in the data. Large differences in scores between men and women, or across various age groups, would suggest a need to create an option for separate gender- or age-based norm groups. Conversely, a lack of such differences may dictate the use of a single norm group with genders and age groups combined.

A series of analyses of covariance (ANCOVA; for the Total EI score) and multivariate analyses of covariance (MANCOVA; for the composites and subscales) were used to examine the relationships between gender and age with EQ-i 2.0 scores. Education level and race/ethnicity were used as covariates in order to control for the effects of these demographic variables. In an attempt to control for Type I errors that might occur with multiple analyses, a more conservative criterion of p < .01 was used for all F-tests.

The Wilks’ lambda statistic generated from these analyses ranges from 0.00 to 1.00 and conveys the proportion of variance that is not explained by the effect (gender, age, or the interaction between gender and age) in the multivariate analyses. These values were all close to 1.00, suggesting that only a small amount of variance could be explained by the effects of these variables. However, F-tests revealed significant effects of gender, age, and the interaction of gender and age (see Table C.5). Given these results, the univariate effects are described in detail below.

Gender Effects. Results of the gender analyses showed that males and females did not differ significantly on the Total EI score, indicating that overall emotional intelligence as measured by the EQ-i 2.0 is the same for males and females. However, small to medium effects were seen on a number of scales (see Table C.6 for effect sizes and Table C.7 for descriptive statistics and significance test results), with a pattern quite similar to what was observed with the North American normative sample (see Standardization, Reliability, and Validity for more information on North American results). The largest gender difference seen in the UK and Ireland sample was on Empathy, with women scoring higher than men with a moderate effect size (d = -0.67). Smaller differences were found with women scoring higher than men on the Interpersonal Composite (d = -0.49), Emotional Self-Awareness (d = -0.45), Emotional Expression (d = -0.35), and Interpersonal Relationships (d = -0.30). Men scored higher than women with small effect sizes on Problem Solving (d = 0.28), Stress Tolerance (d = 0.21), and Independence (d = 0.25).

Age Effects. Significant but small effects were found across age groups for most scales. See Table C.6 for effect sizes and Table C.8 for descriptive statistics and significance test results. Though patterns across the age groups were not consistent across scales, scores were generally lowest for the 18–29 year-old age group and highest for the 60+ age group, showing a clear pattern of increases in EI with age. Differences between adjacent age groups were typically only a few standard score points in magnitude, but differences between the 50–59 year-old group and the 60+ group were larger for some scales. On some scales (Social Responsibility, Problem Solving, and Reality Testing), differences of up to 9 standard score points were observed between the youngest and oldest groups. Emotional Expression and Impulse Control were the only subscales that failed to show at least a small effect size.

Gender x Age Interaction. For most scales, interaction effects were not observed between age and gender. Only two scales reached significance, and met the minimum partial η2 criterion for a small effect size. For the Stress Management composite, males scored higher than females in the 18–29 year-old group, but for the 50–59 year-old group it was the females who scored higher. For Flexibility, a larger gender difference was seen in the 50–59 year-old group (with females scoring 6 standard score points higher) compared to the other age groups. Overall, however, age effects were largely consistent across males and females, and any gender effects were consistent across age groups.

Norm Groups and Norm Construction. Overall, similar to what was found in the North American Norm sample, the age and gender analyses revealed significant, but relatively small effects. Therefore, specific Age and Gender Professional Norms, as well as an Overall Professional Norm (i.e., collapsed across ages and genders), were both developed. Results revealed that skewness and kurtosis values were close to 0 (skewness values ranged from -0.77 to 0.00; kurtosis values ranged from -0.38 to 0.53), and an examination of the scale histograms did not reveal any significant departures from a bell-shaped (Guassian) curve. A histogram for the EQ-i 2.0 UK and Ireland Total EI score is provided in Figure C.1. This figure displays the shape of the distribution for the Total EI score. The shapes of the distributions for the other scales are comparable to this one. Since the data closely match a bell-shaped curve, artificial transformation of scores was deemed unnecessary. Actual construction of the norms was conducted in the same manner as the North American Norms, including the use of statistical smoothing (see Standardization, Reliability, and Validity for more information on the construction of the North American General Population Norm).

Comparison of UK and Ireland General Population Norms to North American General Population Norms. The UK and Ireland sample was compared against the North American normative sample by computing standard scores for the EQ-i 2.0 scales with the North American norms, and comparing these scores against a mean of 100. Overall, no difference between the UK and Ireland sample and the North American norms was observed on the Total EI score. Although for a number of scales the effects did reach statistical significance because of the large sample size being used, all differences were within 3 standard score points, and none reached a meaningful effect size (all Cohen’s d values were below 0.20). Results are presented in Table C.9.

Internal Consistency

Internal consistency, a measure of reliability, conveys the degree to which a set of items are associated with one another. High levels of internal consistency suggest 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 for the UK and Ireland normative sample are presented in Table C.10. Although there is no universal criterion for a “good” alpha level, informal cutoffs for evaluating alpha are typically .90 is “excellent”, .80 is “good”, .70 is “acceptable”, and lower than .70 are “questionable”. Most of the values found in Table C.10 demonstrate excellent or good reliability, and these values are particularly favourable 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 were .75 or higher for all subscales. Similar patterns were seen across the age and gender normative groups, including a Total EI alpha value of .96 or higher for each normative group. The high level of internal consistency found in the Total EI score supports the idea that the EQ-i 2.0 items are measuring a single cohesive construct, namely emotional intelligence. The same can be said of the individual components of emotional intelligence that make up the EQ-i 2.0 (i.e., the composites and subscales).

Factor Validity

EXPLORATORY FACTOR ANALYSIS

Exploratory factor analysis (EFA) was used to determine whether the subscales established with the North American EQ-i 2.0 normative data empirically emerge from the UK and Ireland normative dataset. Five EFAs were conducted, analysing the items within each composite scale separately. In each EFA, a three-factor solution was forced to examine whether the items corresponding to each subscale within the composite also loaded together in the UK and Ireland normative data. As with the North American 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 had loadings within .10 of each other on these factors.

For the Self-Perception Composite EFA, items for the Self-Regard, Emotional Self-Awareness, and Self-Actualisation subscales loaded together as expected by the established factor structure. All items loaded significantly onto their respective factors, with no cross-loadings. For the Self-Expression Composite EFA, all items loaded significantly onto their respective factors for the Independence, Emotional Expression, and Assertiveness subscales. Two items also cross-loaded with other factors: one Emotional Expression item cross-loaded onto Independence, and one Independence item cross-loaded onto Assertiveness. For the Interpersonal Composite EFA, items for the Empathy, Interpersonal Relationships, and Social Responsibility subscales all loaded significantly onto their respective factors with no cross-loadings. For the Decision Making Composite EFA, although most of the Problem Solving, Impulse Control, and Reality Testing items loaded significantly onto their respective factors with no cross-loadings, there were a few exceptions. One Reality Testing item cross-loaded with Problem Solving, and factor loadings for two Impulse Control items fell slightly below the cutoff of ± .30 (i.e.,.29 and .28). For the Stress Management Composite EFA, other than a single Optimism item that cross-loaded with Flexibility, all other Optimism, Flexibility, and Stress Tolerance items loaded significantly onto their respective factors with no cross-loadings.

To summarise, the EFAs generated solutions that strongly correspond to the established EQ-i 2.0 factor structure, with the items for each subscale empirically grouping together onto the expected factors. Of the 126 items entered in the EFAs, only four items showed cross-loadings, and only two items fell just below significance criteria for the factor loadings.

CORRELATIONS AMONG EQ-i 2.0 COMPOSITE AND SCALES AND SUBSCALES

Correlations among the EQ-i 2.0 composite scales and subscales were examined to determine the degree of cohesiveness among them. It is expected that these correlations will generally be high, given that they are all measuring the same underlying construct of emotional intelligence, but they should not be so high as to indicate redundancy between the scales. Correlations observed in the UK and Ireland normative sample are presented in Tables C.11 (composite scales) and C.12 (subscales). These results are similar to what is seen with the North American normative sample.

The composite scale correlations ranged from r = .40 (Interpersonal/Decision Making) to r = .75 (Self-Perception/Stress Management). For the subscales, the correlations ranged from r = .19 (Reality Testing/Impulse Control) to r = .65 (Self-Regard/Self-Actualisation). 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 for the multidimensional nature of the assessment.

EQ-i 2.0 UK & Ireland Professional Norm

Standardisation

NORMATIVE SAMPLE

Collection of data for the UK and Ireland EQ-i 2.0 Professional Norm sample took place between October 2011 and March 2012. The UK and Ireland Professional Norm sample includes 1,150 individuals (N = 1,035 [90%] from the UK, and N = 115 [10%] from Ireland). The sample includes an equal number of men and women, with a good spread across the age groups (see Table C.13 for the age and gender distribution of the sample). The individuals in this sample were all employed or self-employed, and worked in a variety of professional occupations (see Table C.14 for a breakdown of employment areas).

NORMING PROCEDURES

The first step in preparation of the norms was to determine if any age or gender trends existed in the data. Large differences in scores between men and women, or across various age groups, would suggest a need to create and option for separate gender or age-based norm groups. Conversely, a lack of such differences may dictate the use of a single norm group with genders and age groups combined.

A series of analyses of covariance (ANCOVA; for Total EI) and multivariate analyses of covariance (MANCOVA; for the composites and subscales) was used to examine the relationships between gender and age with EQ-i 2.0 scores. In an attempt to control for Type I errors that might occur with multiple analyses, a more conservative criterion of p < .01 was used for all F-tests.

The Wilks’ lambda statistic generated from these analyses ranges from 0.00 to 1.00 and conveys the proportion of variance that is not explained by the effect (gender, age, or the interaction between gender and age) in the multivariate analyses. These values were all close to 1.00, suggesting that only a small amount of variance could be explained by the effects of these variables. However, F-tests revealed significant effects of gender and age for both the composite scale and subscale analyses (see Table C.15). Given these results, the univariate effects are described in detail below.

Gender Effects. Results of the gender analyses showed that males and females did not differ significantly on the Total EI score, indicating that overall emotional intelligence as measured by the EQ-i 2.0 is the same for males and females; however, small to medium effects were found for some composite scales and subscales (see Table C.16 for effect sizes and Table C.17 for descriptive statistics and significance test results). The largest difference was on Empathy, with women scoring higher than men with a moderate effect size (d = -0.54). Smaller differences were found with women scoring higher than men on the Interpersonal Composite (d = -0.40) as well as the Emotional Self-Awareness (d = -0.36), Emotional Expression (d = -0.35), and Interpersonal Relationships (d = -0.23) subscales. Small effects were found with men scoring higher than women on the Self-Regard (d = 0.25), Independence (d = 0.23), Problem Solving (d = 0.28), and Stress Tolerance (d = 0.28) subscales.

Age Effects. Results of the age analyses revealed small effects of age on EQ-i 2.0 scores for most scales (see Table C.16 for effect sizes and Table C.18 for descriptive statistics and significance test results). Although score patterns across age groups did differ slightly between scales, the typical pattern was an increase in scores with age. For the majority of scales, the lowest scores were observed in the 18–29 year-old group, and the highest scores were seen in the 50+ group.

Gender x Age Interactions. There were no significant interactions between age and gender, and partial η2 values all fell below .01 (see Table C.16), indicating that age effects were consistent within males and females, and any gender effects were consistent across age groups.

Norm Groups and Norm Construction. Overall, similar to what was found in the UK and Ireland General Population Norm sample, the age and gender analyses revealed significant but relatively small effects. Therefore, both specific Age and Gender Professional Norms as well as an Overall Professional Norm (i.e., collapsed across ages and genders) were developed. Also similar to results found with the UK General Population Norm sample, skewness and kurtosis values were not large enough to suggest a normalising transformation was necessary for the EQ-i 2.0 scores in the UK and Ireland Professional Norm sample (skewness values ranged from -1.08 to -.23; kurtosis values ranged from -.29 to 1.59). A histogram for the EQ-i 2.0 Total EI score is provided in Figure C.2. The shape of the scores for the Total EI score is comparable to the shape of the scores for the other scales. Actual construction of the norms was conducted in the same manner as the North American Norms, including the use of smoothing techniques (see Standardization, Reliability, and Validity for more information on the construction of the North American General Population Norm).

Comparison of UK and Ireland Professional Norms to North American Professional Norms. The UK and Ireland Professional Norm sample was compared against the North American Professional Norm sample by computing standard scores for the EQ-i 2.0 scales with the North American Professional norms, and comparing these scores against a mean of 100. The UK and Ireland Professional Norm sample scored lower than the North American Professional Norm sample on all but two scales, and although most scales did reach statistical significance, meaningful effect sizes (i.e., |d| ≥ 0.20) were obtained for five of the scales. All differences were within 5 standard score points. Results are presented in Table C.19.

Internal Consistency

Cronbach’s alpha values for the UK and Ireland EQ-i 2.0 Professional Norm sample are presented in Table C.20. Most values shown demonstrate excellent or good reliability. For the overall sample, the alpha value of the Total EI scale was .97, values for the composite scales ranged from .87 to .92, and values were .77 or higher for all subscales. Similar patterns were seen across the age and gender normative groups, including a Total EI alpha value of .96 or higher for each normative group. The high level of internal consistency found in the Total EI score supports the idea that the EQ-i 2.0 items are measuring a single cohesive construct, namely emotional intelligence.

Comparing the General Population and Professional Norm Groups

Because the individuals in the Professional Norm sample have professional careers and are more educated than those in the General Population Norm sample, it was hypothesised that the professionals should score higher than the general population. In order to test this expectation, the UK and Ireland Professional Norm sample data were scored using the UK and Ireland General Population norms. As expected, the Professional Norm sample scored between 2.2 and 8.7 standard score points higher than the General Population sample across the EQ-i 2.0 scales (mean difference = 5.1 standard score points); results are presented in Table C.21.

Large Scale Analysis of Professional Norms

In order to determine what the distribution of scores from UK and Ireland EQ-i 2.0 customers would look like when scored with the Professional Norms, data from a sample of 615 customers (collected from June 2011 to February 2012) were scored using the UK Professional Norms (see Table C.22). Individuals in this sample came from both the UK (69.3%) and Ireland (30.7%). The sample was 51.1% male, 37.4% female, and 11.5% did not indicate their gender. The customers’ age ranged from 19 to 68 years (M age = 40.9; SD = 8.0 years). Results from this sample revealed that when using the Professional Norms, the average scores were close to 100 (M = 100.6 to 103.3), and the standard deviations were close to 15 (SD = 13.6 to 15.5). Furthermore, skewness and kurtosis values for this sample were small; skewness values ranged from -1.02 to -0.14, and kurtosis values ranged from -0.42 to 2.07 (with only one value above 1.1), indicating that the distribution of most scores approximates a bell-shaped (Gaussian) curve. These values, combined with an examination of the scale histograms, indicate a slight negative skew, with no significant departures from a bell-shaped (Guassian) curve (see Figure C.3 for the Total EI score histogram for this sample of customers).