The effects of definitions in social issues changing may affect official statistics reliability and validity over time. Using crime statistics the notion of a 'criminal' not being a static definition is hugely important when analysing crimes statistics over time. In the past there were certain crimes that were seen as 'criminal' and today is not, for example, before 1994 it was legal for a man to rape his wife however, after the law was changed there was an increase in the about of rape cases.
One can argue whether it was do with there being more rapes or the fact that women that were raped by their husband where able to classify it as rape consequently increasing rape statistics. The fact that official definitions can shift over time is a significant problem with time trends in data, which is also shown in statistics about unemployment produced by the Labour Force Market (LFM). The LFM when measuring unemployment still calculates it based on pre-1982 definitions.
Yet, the definition of poverty has been altered over twenty times since 198012, consequently, can one take the statistics allocated by the labour force market at face value. Since the definition has been changed from the one used in the survey, how can this be a representation of unemployment in society when society now uses a different definition from the survey. Once again government manipulation is brought in, Ruth Leivtas (1996) argues that the 'government uses this strategy of measuring unemployment to manipulate the statistics'13.
However, measures of employment should not be under rated because they illustrate perhaps more clearly than any other area of statistics the contradictory trends of the last fifteen years, unlike most research studies which tend to instead offer simply a 'snapshot' of society at a given moment. Also statistics on unemployment should be greatly valued because the volume of alternative information available particularly from the labour force survey is much greater than it was in 1979 and despite the short comings of the data, there is considerable scope for secondary analysis.
The previous paragraphs have discussed many criticism placed on official statistics however, official statistics are often under-rated, under-valued and over-criticised. Many criticisms of official statistics are primarily based on ambiguous social issues such as crime and unemployment. Therefore, it would be impossible for official statistics to illustrate clear unflawed statistics, when the information collected is unclear. Yet, in certain less ambiguous social factors, 'British data deprived from birth and death registrations, for instance is probably among the highest- quality data currently available'14 Bulmer (1984).
Thus, realist would argue that official statistics do have their uses, depending on the types of data that are utilized. Although there are many problems with official statistics certain data cannot simply be blamed or criticised and therefore dismissed, for example the extent of income misdistribution in the United Kingdom is so striking that it can not be account for as methological artefact (the gap between rich and poor increasing). Therefore highlighting how official statistics can still reveal some crucial patterns and relationships.
In conclusion, official statistics are dominated and produced by the government and in social ambiguous issues, the official statistics may be affected by political considerations such as when they are used to assist the image of the government of the day. Although they are many flaws in official statistics, this is the same with most research studies; therefore, there is no logical reason for why official statistics should lead to their outright rejection. They produced tremendous about of information for example the census and reveal crucial patterns and relationships.
Realist simply state that official statistics are helpful for some aspects of society and institutionalist argue that they simply show the process of data, either way this shows that official statistics are useful, therefore should not be over criticised as in some social issues official statistics are the only data available.
Allan, G. and Skinner, C. (1993) Handbook for Research Students in the Social Sciences Bristol: British library Coleman. C and Moynihan. J, 1996 Understanding crime data: haunted by the dark figure Buckingham: Open University Press, Kershaw. C, 2000 The 2000 British crime survey: England and Wales London: Home Office, Research, Development and Statistical Directorate Levitas, R. and Guy, W. (eds) 1996 interpretating Official Statistcs. London: Routledge May, T. (1997) Social Reserch issues, methods and process: second edition Philadelphia: British library Povey. D, 1999 Recorded crime statistics: England and Wales, Prime London: Home Office, Research, Development and Statistics Directorate