Secondary Sources: Official Statistics and Documents

Secondary sources are data that already exist and were originally collected or created for purposes other than the researcher's own study. Using secondary sources allows sociologists to analyse large-scale patterns, historical trends, and phenomena that would be impossible or impractical to study through primary data collection. In this lesson, we examine two key types of secondary source: official statistics and documents.

---

Official Statistics

Official statistics are quantitative data collected and published by government agencies and other official bodies. In the UK, key sources include the Office for National Statistics (ONS), the Home Office, the Department for Education, and NHS Digital.

Key Definition: Official statistics — numerical data collected and published by government departments and agencies, covering topics such as crime, health, education, employment, population, and poverty.

Examples of Official Statistics

Source	Data Provided
Census	Comprehensive data on the UK population, collected every 10 years (age, ethnicity, occupation, housing, religion, etc.)
Crime statistics	Recorded crime figures compiled by the police and published by the Home Office
Crime Survey for England and Wales (CSEW)	Large-scale victim survey measuring experiences of crime, including unreported crime
Educational statistics	Data on school performance, exam results, exclusions, and participation in higher education
Labour Force Survey	Data on employment, unemployment, economic activity, and earnings
Vital statistics	Data on births, deaths, marriages, and divorces
Health statistics	Data on illness, hospital admissions, mental health diagnoses, and life expectancy

Practical Advantages of Official Statistics

Advantage	Explanation
Free and accessible	Most official statistics are published online and freely available to researchers
Large scale	They cover the entire population or very large samples, making them highly representative
Allow comparisons	Data is collected using standardised definitions and methods, allowing comparisons over time and between regions or countries
Already collected	Using existing data saves the researcher the time and cost of collecting their own primary data
Identify trends	Longitudinal data (collected over time) can reveal important social trends — for example, changes in divorce rates, crime rates, or educational achievement gaps
Basis for further research	Official statistics can highlight patterns that require further investigation using primary methods

Practical Disadvantages of Official Statistics

Disadvantage	Explanation
Definitions may not match	Official definitions of concepts (e.g. 'unemployment', 'poverty', 'crime') may not correspond to the sociologist's own operationalisation
Changes in definition	Definitions and measurement methods change over time (e.g. the government has changed the way it measures unemployment over 30 times since the 1980s), making longitudinal comparisons unreliable
Not collected for sociological purposes	Official statistics are designed to serve government needs, not sociological research questions
Incomplete	They may not cover the topics or variables the researcher is interested in
Political manipulation	Governments may change definitions or measurement methods to present more favourable data (e.g. changing how unemployment is counted to reduce headline figures)

Theoretical Perspectives on Official Statistics

The Positivist View

Positivists treat official statistics as objective facts — reliable, valid measures of social phenomena that can be used to identify patterns, correlations, and causal relationships. Durkheim's study of suicide (Le Suicide, 1897) is the classic example: he used official statistics on suicide rates across different countries and social groups to argue that suicide was caused by social factors (such as levels of social integration and moral regulation) rather than individual psychology.

The Interpretivist View

Interpretivists argue that official statistics are social constructions — they do not simply reflect reality but are the product of the decisions, interpretations, and activities of the people who create them. This is particularly evident in crime statistics:

• Police recording practices — the police decide whether to record an incident as a crime. Changes in recording practices (e.g. the introduction of the National Crime Recording Standard in 2002) can cause dramatic apparent changes in crime rates that do not reflect actual changes in criminal behaviour.
• The 'dark figure' of crime — official crime statistics only capture crimes that are reported to the police and then recorded by them. A large proportion of crime goes unreported and unrecorded. The British Crime Survey consistently shows that the actual level of crime is much higher than police recorded figures suggest.
• Labelling — interactionist sociologists such as Cicourel (1968) argue that crime statistics reflect the activities and prejudices of the criminal justice system rather than the actual distribution of criminal behaviour. Cicourel showed that police officers used typifications (stereotypical assumptions about what a 'typical delinquent' looks like) when deciding who to stop, arrest, and charge, meaning that working-class and ethnic minority young people were disproportionately represented in crime statistics.

Key Definition: Dark figure of crime — the difference between the actual amount of crime committed and the amount recorded in official statistics. It includes crimes that are not reported to the police and crimes that are reported but not recorded.

The Marxist View

Marxists argue that official statistics serve an ideological function — they reflect and reinforce the interests of the ruling class. For example: