The General Health Questionnaire (GHQ) is a subjective measure of psychological wellbeing.
It was developed by Goldberg who described the measure as a psychological well-being measure:
“The GHQ is specifically concerned with the hinterland between psychological sickness and psychological health.”
Goldberg was keen to develop a measure of well-being that did not require a full psychiatric assessment. His premise was that psychological strain is evenly distributed throughout the population in varying degrees of severity ranging from severe to hypothetical normality, and that a questionnaire ought to be able to screen the top end of the distribution.
His ultimate aim was to establish a quantitative estimate of an individual’s degree of psychological strain. This score could then be used in quantitative analysis techniques, for instance, comparing two populations, or the same population on two different occasions. The scores could also be correlated with other clinical and social variables.
One of Goldberg’s main problems in determining illness from a normally distributed variable was in trying to establish where normality ends and clinically significant disturbance begins. His solution was to ask what proportion of the population would be thought to have a clinically significant disturbance if they were interviewed by a clinical psychiatrist.
A comparison from the same sample of the results of a full clinical assessment and the results from the questionnaire acts as a construct validity test. A whole series of construct validity tests (see Goldberg & Williams, 1988) has indicated that the GHQ is extremely sensitive in detecting psychological strain, and established the efficacy of the epidemiological approach towards determining the prevalence of psychological disturbance in a normal population.
Of significance to the debate about what is stress and what is ill-health, the GHQ concerns itself with two major classes of phenomena: an inability to continue to carry out one’s normal “healthy” functions, and the appearance of new phenomenon of a distressing nature such as an inability to sleep, or headaches etc. Such ill-health indicators have been assumed to represent relevant stress-related outcome variables, and hence, the GHQ has been used in a variety of studies to represent the stress response.
The rating scale is a behaviourally anchored scale with four options; Better than usual, Same as usual, Worse than usual, and Much worse than usual. A Likert style scoring procedure can be used on a four point scale with Better than usual representing 1 point and Much worse than usual representing 4, with high scores equating to poor health. Alternatively a “GHQ-scoring” method can be adopted in which the first two anchors, that is, Better than usual and Same as usual, are scored as zero, and Worse than usual and Much worse than usual scored as one.
In this procedure, the first two anchors represent non-presentation of symptoms and so are both scored zero. The next two responses represent a presentation of symptoms and are therefore scored one.
|