A hierarchy of distress: Mokken scaling of the GHQ-30

Background Hierarchical cumulative scales are common and informative in psychology. The General Health Questionnaire (GHQ) does not appear to have been subjected to an analysis that examines the hierarchical and cumulative nature of its items. We report an analysis of data from the 30-item GHQ (GHQ-30) as part of the Health and Lifestyle Survey (HALS). Method Data from 6317 participants who completed the GHQ-30 as part of the HALS were analysed using the Mokken Scaling Procedure (MSP), which is a computer program that searches polychotomous data for hierarchical and cumulative scales on the basis of a range of diagnostic criteria. Results A final scale consisting of nine items from the GHQ-30 was obtained that, according to the criteria for a Mokken scale, was a reliable and very strong scale. The least difficult item in the scale is ‘been (un)able to face up to your problems?’ and the most difficult item is ‘felt that life isn't worth living?' Conclusions Items from the GHQ-30 form a short hierarchical and cumulative scale. The majority of these items also appear in the GHQ-12. The nine-item GHQ shows better distribution properties than the GHQ-30 and compares very favourably with the GHQ-12.

[1]  K Sijtsma,et al.  A survey of theory and methods of invariant item ordering. , 1996, The British journal of mathematical and statistical psychology.

[2]  D. Goldberg,et al.  A scaled version of the General Health Questionnaire , 1979, Psychological Medicine.

[3]  R. Watson.  The Mokken scaling procedure (MSP) applied to the measurement of feeding difficulty in elderly people with dementia. , 1996, International journal of nursing studies.

[4]  H. Möller,et al.  Depressive, anxiety, and somatoform disorders in primary care: prevalence and recognition , 2007, Depression and anxiety.

[5]  Charles Lewis,et al.  A Nonparametric Approach to the Analysis of Dichotomous Item Responses , 1982 .

[6]  Norman Schofield,et al.  Data analysis and the social sciences , 1984 .

[7]  I. Deary,et al.  Mokken Scaling of the Epworth Sleepiness Scale items in patients with the sleep apnoea/hypopnoea syndrome , 1998, Journal of sleep research.

[8]  T. Suurmeijer,et al.  A Study of the Unidimensionality and Cumulativeness of the MOS Short-Form General Health Survey , 1994, Psychological reports.

[9]  T. Suurmeijer,et al.  Factors influencing professional home care utilization among the elderly. , 1991, Social science & medicine.

[10]  W. Maier,et al.  Identification of items which predict later development of depression in primary health care , 2009, European Archives of Psychiatry and Clinical Neuroscience.

[11]  I. Deary,et al.  Are personality trait items reliably more or less ‘difficult’? Mokken scaling of the NEO-FFI , 2007 .

[12]  S. Stouffer,et al.  Measurement and Prediction , 1954 .

[13]  S. Kaasa,et al.  Measuring quality of palliative care: Psychometric properties of the FAMCARE Scale , 2003, Quality of Life Research.

[14]  T. Suurmeijer,et al.  Psychometric properties of the RAND-36 among three chronic disease (multiple sclerosis, rheumatic diseases and COPD) in the Netherlands , 2004, Quality of Life Research.

[15]  W. Resing,et al.  Constructing Geometric Analogies for the Longitudinal Testing of Elementary School Children. , 1997 .

[16]  W Wieners,et al.  On measuring quality. , 1988, Computers in healthcare.

[17]  Klaas Sijtsma,et al.  Mokken scale analysis for polychotomous items: theory, a computer program and an empirical application , 1990 .