Page 198 - PhD GT

As the first part of the Impaired Control Scale was thus thought to measure a behaviour distinct from but correlated with dependence as measured by the LDQ, and the independence of these variables was demonstrated in Principal Components Analysis, reported below in Section 9.3, the first part alone was used in the analysis of predictors of change in dependence, described later in this chapter.
9.2.3 The Coping Behaviours Inventory in the present study
The Coping Behaviours Inventory (discussed in Chapter 6) was thought to be useful in elucidating the nature of change in dependence. It was hypothesised that there would be a difference in the use of coping strategies by high and low dependence individuals. It was proposed that examination of coping strategies as potential correlates of change in dependence, in addition to the other measures, might throw light on the nature of such change.
The null hypothesis, that people with high dependence will use the same coping strategies as people with low dependence, was examined using a scale which contained cognitive and behavioural coping items. The scale, the Coping Behaviours Inventory was originally designed for problem drinkers and was adapted to heroin users as part of the present study. One of the reasons this scale was thought to be suitable for use in the present study was that, in Principal Components Analysis of the original, four factors were found, two cognitive and two behavioural (Litman et al. 1983). Additionally, Litman’s scale was validated in a UK population of help-seekers for problems of alcohol misuse and dependence and was therefore considered relevant to the target population in the current study.
Principal components analysis of responses to the original Coping Behaviours Inventory in the Sample 6 alcohol group (n = 78) of the present main study yielded 10 factors with an eigenvalue over 1 of which the first factor had an eigenvalue of 11.96 and accounted for 33.2% of the variance. Thirty of the 36 items had loadings over .35 on this factor; items 2,3,8,10,12 and 27 (see Appendix 9) had loadings less than .35 on the first factor, but loadings greater than .35 on one of the next three factors.
Factor analysis of the adapted scale produced 11 factors with an eigenvalue greater than1 which accounted for 67% of the variance; the first factor accounted for 26.4% of the variance and 32 of the 38 items had factor loadings over .35 on this first factor. It was not the case that separate factors were found for cognitive and behavioural coping strategies, nor were factors similar to those found by Litman and her colleagues (1983) identified in the sample of heroin users who participated in the validation study of the adapted Coping Behaviours Inventory.
193