Official stats Flashcards
Practical ++++
✓ Availability/access: Already in existence, so are readily available in an accessible form. Some statistics are published monthly, most annually (yearly), & so can be analysed for patterns & trends. Data is already categorised & in immediately usable form so it is not time consuming to collect & is mostly free (from the internet).
✓ Comparisons/trends/patterns: Their quantitative nature makes it easy to make comparisons & identify trends & patterns – e.g. crime rates between classes, genders or ethnic groups. Sociologists can use them for ‘before & after’ studies to try & show cause-&-effect relationships. A sociologist can analyse statistical data to examine the effects in the changes that laws can have on human behaviour – e.g. the amount of alcohol-related violent behaviour before & after the 24-hour drinking law was introduced. Statistical data can allow a sociologist to examine social trends in a society & identify how they have changed over a period of time, e.g. religious attendance or marriage trends.
✓ Useful background data: Often provide a useful starting point, prompt, or background material for a sociologist – e.g. statistics on ethnic differences in crime rates could lead to a more in-depth study.
✓ Useful for governments: Statistical data is very useful for governments as it can help them formulate social policies & intervention strategies. E.g. the analysis of fertility rates has shown a rise in teenage pregnancies in economically deprived areas. This allows the allocation of funds & resources to tackle the problem of teenage pregnancies in such areas.
Practical —-
× Definition & measurement: Definitions of the concepts used, how they have been operationalised & how the data is presented may differ from those of the sociologist. E.g. different definitions of ‘social class’ – occupation or ownership of property? Also, definitions may change over time making comparisons difficult or impossible.
× May be none available on a topic: The government collects statistics for its own purposes, & not for the benefit of sociologists, so there may be none available on the topic we are interested in.
ethics
✓ Raise little/no ethical issues: Limited/no issues with deception, right to withdraw, informed consent & protection from harm for the sociologist as the data already exists and is in the public domain & personal data is anonymous so these ethical concerns will already have been dealt with.
reliability
✓ High in reliability: The same categories & mode of collection are usually used each time statistics are gathered by trained staff, so the research can be replicated by others. As the way in which the statistics have been collected is also usually publicly stated, a sociologist can check whether the same procedures have been followed each time.
× Errors/Inaccuracies: May not be as reliable as positivists claim. Even in the Census, recording errors are made, households missed out & people may complete the form inaccurately.
validity
✓ ‘Hard’ statistics: Some statistics do measure what they claim to measure accurately – e.g. statistics on the number of births, deaths, marriages & divorces generally give a very accurate & therefore valid picture.
× Low in validity: ‘Soft’ statistics give a much less valid picture – e.g. police statistics do not record all crimes. Attempts are made to compensate for this by using self-report or victim surveys & these show that police statistics underestimate the ‘real rate’ of crime (known as the ‘dark figure of crime’) which shows us that we have to be very wary of ‘soft’ statistics.
× Political bias: Marxists claim that official statistics are not politically neutral, but reflect ruling class interests & ideology. Definitions used, areas of social life covered & how statistics are presented are all political decisions. E.g. ‘massaging’ / manipulating crime statistics.
× Statistics don’t tell us why something is the way it is. E.g. we know that Chinese students in the UK are the highest achievers in the education system & that men commit more crime than women but the numbers don’t tell us why this is so. This means that statistics lack depth & a true insight into human behaviour.
examples
• 10-yearly UK Census completed by all households in the UK for all people in that household on one specified night (therefore it does not include the homeless which is an issue). Those that do not complete it face a fine.
Representativeness
✓ Often highly representative: Usually based on a very large sample, so the characteristics of the sample are often the same as the wider group & so generalisations can be made & a hypothesis tested. Even with smaller government surveys, the sample is still much larger than sociologists can normally obtain.
Theory
✓ Favoured by positivists: Assume that statistics measure an objective social facts. Take a scientific approach using standardised (set) methods to get large-scale, reliable, quantitative, representative data that allows for generalisations & cause-&-effect statements. E.g. Durkheim’s study of suicide.
× Rejected by interpretivists: Argue that official statistics are social constructs (created by people), not objective ‘truth’ or fact. E.g. for a crime to appear in official statistics, decisions have to be made by people (victims, police, etc.) that affect whether it will be officially recorded.
