Measuring Emotional Processing EPS

Scale validity & reliability

Click here to link to:

  What does the EPS measure?
  Background & development of scale
  How to be included in emotional processing research
  References
  Back to main page

Initial psychometric evaluation

A total of 460 participants (175 males, 284 females, 1 missing) were recruited from a variety of settings to respond to the 45-item version of the questionnaire. Participants included colorectal cancer patients (n =124) and non-patient older adult controls (n = 73), individuals referred to a clinical psychologist or a counsellor for a range of mental health problems (n = 147), university undergraduate students (n = 100), individuals with a diagnosis of chronic back pain (n = 11) and individuals with a diagnosis of ankylosing spondylitis (n = 5).  Participants ranged in age from 17 to 89 years (mean = 47, S.D. = 21). The 460 participants rated themselves on each of the 45 items using the 10-point response scale.  In addition, a number of participants completed several established scales assessing constructs theoretically related to emotional processing. 

Five items were removed prior to the principal components factor analysis:

conceptually weak

and/or low item-total correlations

and/or failed to show any between group differences on independent samples t-tests

Exploratory principal components factor analysis was therefore undertaken on 40 items.

Eight factors emerged (see table 1).  The cumulative variance accounted for by the 8 factor solution was 58.1%

Table 1: Factor structure of the EPS

Concurrent validity

The three factors most related to other theoretically related constructs were egodystonic, dissociation and suppression.  As predicted, several of the subscales of the EPS correlated most highly with the Toronto Alexithymia Scale-20. The suppression subscale of the EPS correlated relatively highly with the Courtauld Emotional Control Scale (r = .53). 

Test Retest

Test-retest reliability was assessed over a 4-6 week interval and was based on a sample of 17 undergraduate students, 3 male (mean age 25.3; SD = 3.2).  The obtained coefficient was .79 (p = <.01).

Dissemination

The Emotional Processing Scale has already been presented at two international emotion conferences:

as a presentation at The XII Conference of the International Society for Research on Emotion, Cuenca, Spain, July 20-24, 2002

in the form of a poster presentation at The Third International Conference on the (Non) Expression of Emotions in Health and Disease, Tilberg, The Netherlands, October 19-21, 2003.

A paper describing the development and initial psychometric evaluation of the Emotional Processing Scale has been published (Baker R, Thomas S, Thomas PW & Owens M (2007) Development of an emotional processing scale, Journal of Psychosomatic Research, 62,167-178

Here is a link to where you can buy the full article

Dorset RDSU
Cornelia House  Poole Hospital NHS Trust  Longfleet Road  Poole  Dorset  BH15 2JB
Email dawn.stevens@poole.nhs.uk