John C. Duffy

A Comparison of Computer Interviewing Techniques and Traditional Methods in the Collection of Self-Report Alcohol Consumption Data in a Field Survey

Summary A sample survey to obtain self-report alcohol consumption data from adult males in Edinburgh was conducted. Two matched half samples were assigned to a computer interview or a direct interview at random. The aims of the study were to assess the feasibility of using computer interviewing techniques in a field survey of the general population and also to ascertain whether the computer interview method would show a significant increase in the reported consumption values when compared with direct interview. The mean total consumption reported in the direct interviews was consistent with findings from other recent surveys of this demographic subgroup (Knight & Wilson, 1980; Kendell et al., 1982). The mean total consumption reported in the computer interviews was some 30% higher, a difference which remained highly significant even after controlling for other variables known to affect consumption. The percentage increase for wine and spirits at 50% was greater than that for beer (25%). These results suggest that the bias of underreporting known to exist in survey measurement of alcohol consumption may be reduced by replacing the interviewer by a computer. Further studies are required to substantiate this preliminary finding, but the potential of the methodology has been demonstrated.

Scotland's liquor licensing changes: An assessment

The impact of the changes in Scotlands liquor licensing laws was assessed by comparing trends in alcohol related problems in Scotland with those in England and Wales. The study showed no appreciable effect on the level of alcohol related morbidity and mortality, though some improvements were noted in relation to the rates of convictions for drunkenness. The changes introduced since 1976 appeared to be popular and there was widespread perception that public drunkenness had become less common-place.

Breast cancer and breast self-examination: What do Scottish women know?

To aid the design of a health education campaign centred around a breast screening programme, a survey of knowledge about breast cancer and breast self-examination was carried out in 810 women selected randomly from the electoral registers in Edinburgh and Aberdeen. On the whole, the level of knowledge was poor, but was strongly related to age, social class, previous experience of breast problems and preventive health behaviour. Very few women knew of signs to look for other than a lump or of any treatments other than mastectomy or radiotherapy. Although 57% had tried or were practising self-examination, only a small minority knew of the correct technique and when to do it. There may be several reasons for our findings, including previous lack of appropriate information from professional sources, avoidance due to fear and anxiety about breast cancer or lack of motivation because this disease is not seen as an important single threat. These results will be compared with a later survey as part of the evaluation of our health education campaign.

Changes in Licensing Law in England and Wales and Indicators of Alcohol-Related Problems

Trends in alcohol-related problems were examined in the light of the 1988 amendments to the Licensing Act in England & Wales. Data concerning accidents and absenteeism in the workplace, road traffic accidents, drunken driving and criminal offences were collected and compared with control data obtained from Scotland. The data were analysed by fitting either a logistic or loglinear model as appropriate using the GLIM statistical software package.After accounting for economic factors, significant changes in the levels of certain alcohol problem indicators were observed. In several instances these changes related to an increased risk in Scotland and were clearly not a result of the liberalisation of licensing hours in England & Wales. Reports of non-sexual crimes of violence and slight accidents in the workplace have increased in England and Wales concurrently with the law change but the causal relationship, if any, remains a matter of speculation.

Fallacy of the Distribution of Alcohol Consumption

It is pointed out that it is fallacious to expect alcohol consumption to follow any particular statistical distribution. Emphasis on the distribution of alcohol consumption in research is therefore misguided.