Lane, A., Meyer,
B., Devonport, T., Davies, K., Thelwell, R., Gill, G., Diehl, C.,
Wilson, M. and
Weston, N. (2009) ‘Validity of the emotional intelligence scale for use
in
sport’, Journal
of Sports Science and Medicine (2009) 8, 289-295
The Abstract says it all really:
Abstract
This study investigated the
factorial validity of the 33-item selfrated Emotional Intelligence
Scale (EIS:
Schutte et al., 1998) for use with athletes. In stage 1, content
validity of
the EIS was assessed by a panel of experts (n = 9). Items were
evaluated in
terms of whether they assessed EI related to oneself and EI focused on
others.
Content validity further examined items interms of awareness,
regulation, and
utilization of emotions. Content validity results indicated items
describe
6-factors: appraisal of own emotions, regulation of own emotions,
utilization
of own emotions, optimism, social skills, and appraisal of others
emotions.
Results highlighted 13-items which make no direct reference to
emotional
experiences, and therefore, it is questionable whether such items
should be
retained. Stage 2 tested two competing models: a single factor model,
which is
the typical way researchers use the EIS and the 5-factor model
(optimism was
discarded as it become a single-item scale fiolliwng stage 1)
identified in
stage 1. Confirmatory factor analysis (CFA) results on EIS data from
1,681
athletes demonstrated unacceptable fit indices for the 33-item single
factor
model and acceptable fit indices for the 6-factor model. Data were
re-analyzed
after removing the 13-items lacking emotional content, and CFA results
indicate
partial support for single factor model, and further support for a
five-factor
model (optimism was discarded as a factor during item removal). Despite
encouraging results for a reduced item version of the EIS, we suggest
further
validation work is needed.
‘ Emotional intelligence can be
defined as “the ability to carry out accurate reasoning about emotions
and the ability
to use emotions and emotional knowledge to enhance thought”’ (289). It
has been
connected with performance and with response to coaching. The EIS scale
is a
better predictor than self-report measures. EIS has 33 items on
perceptions,
reduced by comparative factor analyses ( CFA) to a number of smaller
factors --
1, 4, and in this case 6 ( the latter splits stuff on emotions in
oneself 4
items and emotions in others 2) It has not been tested on athletes
until this
study.
At Stage 1, 9 experts were
invited to
assess ‘content validity’. This led to the 6 factor model. Some items
were
rejected as not apparently about emotions (eg I find it hard to read
non-verbal
messages – seen here as cognitive not emotional. Could be rephrased to
refer to
emotions aroused by not being able to read non-verbals they say [so why
not
rephrase them? Because this is a standard scale and they want
parsimony?].
The full model and the
reduced one
were to be tested though at Stage 2 (and another variant), on 1681
student
athletes. CFA used a definite method (never heard of it) and a
statistical
corrective procedure designed to overcome any unwanted variability in
the
sample – the 6 factor jobbie had higher levels of fit than the 1 factor
version.
[What does ‘fit’ measure
exactly? Fit
with the distribution of scores? Could be fit with the scores of other
studies
– this is a technique with validation studies they say]
An alternative is discussed
–exploratory factor analysis where the data themselves yield factors.
This is
criticised because it can misleadingly conflate items – in this case a
mood
regulation item and an optimism item. [Implies some non-empirical
purpose for
the items -=- to yield theory as we see below]. The criticism arises in
this
case because an extreme optimist could simply not realise the need to
regulate
moods [A very weird discussion – this is a purely speculative
possibility? No
-- extrapolated from some study – but they
admit this is ‘purely speculative’. Great example of tactical argument
defending their content validity stage ultimately].
The general argument is that
any
factors must relate to conceptual elements. This is why they are
testing the
validity of the EIS – so it can be used in future theoretical study,
but only
if it is valid, of course. They conclude they were right to use experts
to bin
some of the more dubious items from the original list.
Full table of items on p.
292 with
measures of fit. NB found reverse -scored items did not fit well,
[despite
usual advice to include them to check for guessing – this team chose to
regard
some odd answers on those reverse-scored items as errors arising from
respondent carelessness] – seems to be common for athletes.
[This whole thing must be
pretty well
impossible to judge. A huge literature is selectively cited. Some items
and
measures are used and others rejected. The statistical techniques and
issue are
just uncontrollable and the argument can only be one from authority,
although
even here the articles cited are so obscure that the status of the
supported
must be unclear and the validity of the test unknown. There seems to be
an
acceptance that published articles must be authoritative – and also a
sneaky
attempt to snow readers under with so many references that they could
not
possible check them all – readerly texts! One journal seems to have
been
extensively cited – so it is the authority? Tactical choices of the
items and
the content validity procedure are also clear – here the team even
resort to
speculative possibilities about respondents and their carelessness, or
the
opinions of experts and their theoretical status].
Sample refs: (35 in all)
Austin,
E.J., Saklofske, D.H., Huang, S.H.S. and McKenney, D. (2004)
Measurement of
trait emotional intelligence: Testing and crossvalidating a modified
version of
Schutte et al.’s (1998) measure. Personality
and Individual Differences 36, 555-562.
Besharat,
M.A. (2007) Psychometric properties of Farsi version of the Emotional
Intelligence
Scale-41 (FEIS-41). Personality
and Individual Differences 43, 991–1000.
Brackett,
M.A. and Mayer, J.D. (2003) Convergent, discriminant, and incremental
validity
of competing measures of emotional intelligence. Personality and Social Psychology Bulletin 29,
1147-1158.
Colvin,
C.R., Block, J. and Funder, D.C. (1995) Overly positive evaluations and
personality: Negative implications for mental health. Journal of Personality and Social Psychology, 68,
1152-1162.
Engelberg,
E. and Sjöberg, L. (2004) Emotional intelligence, affect
intensity, and social
adjustment. Personality and
Individual
Differences 37, 533-542.