Quantitative: Official Statistics (Secondary)

Question 1

What is a developed definition of official statistics?

Answer 1

• Secondary data.
• Can be produced by the government or other official bodies.
• Governments collect such data for their own purposes, such as policy making.
• Different types of sources used to create official statistics:
  Registration - births, deaths.
  Official surveys - census or General Household survey.
  Administrative records - hospitals, courts, schools. Include records of illnesses, convictions, truancy etc.
• Other non-state agencies produce ‘non-official’ statistics. For example, churches produce membership and attendance statistics, Shelters produce statistics on homelessness.
• Not collected by sociologists, meaning the purpose of their collection may not match that of the sociologist.

Question 2

P.E.T.
What are practical advantages of official statistics?

Answer 2

• Free source of huge amounts of quantitative data. It would cost millions to conduct such large-scale data.
• Only the state has the power to compel individuals to supply certain data - reduces the problem of non-response. Refusal rate in the last census was 5%.
• Allow us to make comparisons between groups.
• Collected at regular intervals, showing trends and patterns over time. Can be used for ‘before and after’ studies to identify correlations between variables and suggest possible cause and effect relationships.

Question 3

What are practical disadvantages of official statistics?

Answer 3

• Government creates statistics for its own purposes - not for the benefit of sociologists so none may be available on the topic they are interested in.
  For example: France does not collect data on the race, religions or ethnicity of its citizens. Therefore produces no official stats on issues such as religion of people who commit suicide or of prisoners in its jails
• May be mismatches between sets of statistics.
  For example, if we want to compare stats on ill health with stats on unemployment for the people of a particular town, we might find two sets of data that cover slightly different areas and therefore different populations. This makes it impossible to establish the degree of correlation between ill health and unemployment precisely.
• Definitions that the state uses may be different from those that sociologists would use. It may lead to different views of how large the problem is.
• The State might change the definitions it uses over time and different states may define the same term differently. This makes comparison over time or between countries difficult. Some stats are collected infrequently, such as the ten-yearly Census, and therefore don’t always give an up-to-date picture of social trends.

Question 4

What are theoretical issues of official statistics?

Answer 4

• Positivists favour official stats as objective facts about society.
• Interpretivists see them as social constructs, and Marxists and feminists regard them as performing an ideological function.

Question 5

How does positivism and stats relate to official statistics?

Answer 5

• Take it for granted that official stats are reliable, objective social facts.
• Very important resource in the scientific study of society.
• Major source of representative, quantitative data that allows the sociologist to identify and measure behaviour patterns, test hypotheses and develop casual laws to explain the patterns of behaviour that the statistics reveal.
• For example, by using official stats to identify patterns in mental illness, positivists can establish correlations.
• A testable hypothesis can be put forward to explain a possible causal link between the variables.
  For example: If stats show a correlation between being a full-time housewife and a high risk of depression, we might hypothesise that this is due to social isolation.

Question 6

Are official statistics high or low in representativeness?

Answer 6

• Official stats often provide a more representative sample than surveys conducted with the limited budget available to the sociologist.
• Large scale often covering the entire population.
• Stats gathered by compulsory registration and likely to cover virtually all cases and therefore be extremely representative.
• Statistics from official surveys are likely to be less representative due to only being based on a sample of relevant population.
• However are still much bigger than most sociologists can afford to carry out. For example, the Crime Survey for England and Wales.
• Great care is taken with sampling procedures when conducting official surveys = sound basis for making generalisations.

Question 7

Are official statistics high or low in reliability?

Answer 7

• Compiled by trained staff who use standardised categories and collection techniques, following set procedures that can be easily replicated by others.
• Particularly true of official surveys, such as the Census. The survey is carried out using a standardised measuring instrument which is administered in the same way to all respondents. Therefore results are reliable because any other researcher could repeat it and get the same results.
• Registration data - compiling death rates for different social classes, use the occupation recorded on each person’s death certificate to identify class.
• Census coders may make errors or omit information when recording data or forms may be filled in incorrectly = not wholly reliable.

Question 8

How does interpretivism and statistics relate to official statistics?

Answer 8

• Reject the positivist claim that official statistics are real, objective social facts.
• Statistics are merely constructs that represent the labels officials attach to people.
• Should be treated as a topic in themselves and how they are socially constructed.

Question 9

What are soft statistics and what do they do?

Answer 9

• Give a less valid picture of reality.
• Often compiled from administrative records created by state agencies.
• They represent the record of decisions made by those agencies rather than a picture of the world. For example, truancy stats represent the number of pupils that a school has defined as truanting - this isn’t necessarily the same as the number who actually truanted.
• Often neglect an unknown or ‘dark figure’ of unrecorded cases. For example, schools may keep a record of racist incidents, but this doesn’t reflect the number of cases that go unreported and there may be no way to discover the number of cases that do.

Question 10

What are hard statistics and what do they do?

Answer 10

• Provide a more valid picture.
• Include stats on births, deaths, marriages and divorces.
• Little dispute as to how to define the categories used to collect the data.
• They are often created from registration data - for example there is a legal requirement to register births and deaths. Very few go unrecorded.

Question 11

How does marxism and statistics relate to official statistics?

Answer 11

• Reject the positivist claim that official statistics are objective facts.
• However do not see them as merely the outcome of the labels applied by the officials.
• Regard official statistics as serving the interests of capitalism.
• The statistics that the state creates are part of what Althusser terms the ideological state apparatus - a set of institutions that produce ruling-class ideology. As ideology, the function of official statistics is to conceal or distort reality and maintain the capitalist class in power.

Question 12

What are ideological functions for official statistics?

Answer 12

• Politically sensitive data that would reveal the unequal, exploitative nature of capitalism may not be published. For example, since the 1980s, data derived from analysis of Census returns no longer includes class differences in death rates
• Definitions used in creating official stats also conceal the true reality of capitalism.
  For example: the state has frequently changed its definition of unemployment, and this has reduced the numbers officially defined as unemployed. Social class categories used in official stats are based on occupation. This gives the impression of a gradual hierarchy of several classes, rather than a conflict between two opposed classes.
• Conceals the existence of a ruling class whose position is based on ownership of cast wealth, not on occupation.

Question 13

How does feminism and statistics relate to official statistics?

Answer 13

• Criticise official statistics.
• Those like Oakley and Graham reject the use of quantitative survey methods because they regard them as a ‘masculine’ or patriarchal model of research.
• As official stats are often created using these methods, this is also a criticism of the statistics they produce
• Official statistics are created by the state, which feminists regard as maintaining patriarchal oppression, therefore official stats are a form of patriarchal ideology → they conceal or legitimate gender inequality and maintain women’s subordination.
  For example, few statistics collected on women’s unpaid domestic labour, thereby maintaining its invisibility and giving the impression that it is of little importance. Full-time housewives are defined for statistical purposes as ‘economically inactive’.
• Official statistics underestimate women’s economic contribution and they reflect the patriarchal nature of the state.
• However, some stats can show clear evidence of gender inequality, such as those on earnings from paid work.
• Changes in the definitions used in official stats that may reveal women’s position more clearly.
  For example - a family’s class used to be determined by the occupation of the male head of household. This changed in 2001 so that the person who owns or rents the home is now the ‘household reference person’ and it is this person’s occupation which is used to define the family’s class.if the home is jointly owned the person with the highest income is used.
• As men are still more likely to earn more, official stats continue to give distorted pictures of gender and social class. Therefore many feminists argue that official stats should allocate women and men to a social class as individuals, not as households.