What are official statistics?
Official statistics are the data collected by government departments in the course of their work (i.e. routine statistics) or collected specifically for statistical purposes. 'Unofficial' statistics are from independent organisations. Official statistics are often used as a data source for secondary statistical analysis, employing a quantitative approach. They are also used in health needs assessments, health status or profiles and health impact assessments. Statisticians who work in Official Statistics not only compile statistics but they also work at how to measure concepts and conditions. An example is how the ethnicity question in the census was refined between 1991 and 2001. They also look at how to improve the data collection of statistics and provide an information service to assist policy makers in decision-making and running government services efficiently. An interesting titbit is that a study using official statistics on New Zealand house prices showed that what were, in estate agent speak, 'up and coming' areas ten years ago were still 'up and coming' now and that areas which were expensive ten years ago remained desirable areas.
Another way of putting it is that 'official statistics' are officially produced statistical representation of social phenomena and they can provide a good resource for researchers.
Check out the Royal Statistical Society's Official Statistics Section for latest news and forthcoming meetings.
- They are of a large-scale nature, offering a basis for generalizing findings, claiming representativeness and for investigating relationships between social variables that cannot be identified at a local level.
- They are useful for placing smaller-scale (perhaps qualitative) data in a broader context.
- Often published in the form of a statistical series which enables a more analytical perspective.
- It is an unobtrusive measure. An unobtrusive measure is any method of observation that directly removes the observer from the set of interactions or events being studied. It removes researcher bias to an extent and 'reactivity' of the respondents. They don't know that they are participating so their behaviour is not influenced by their knowledge that they are being investigated.
- Ecological fallacy - i.e. the error of assuming that inferences about individuals can be made from findings relating to aggregate data.
- Problem of reliability of reporting, especially within an area like health.
- Missing data! (Quite often, topics are added or dropped in accordance to a political agenda)
- Think CART! There can be problems with Completeness, Accuracy, Relevance or Representiveness and Timeliness.
- Numbers are social facts which means they are also acts of interpretation.
- The researcher has little control over the variables. S/he can transform existing variables into derived variables (e.g. combining questions on disability into a disability index and then compare men and women).
Useful Official Statistics for Public Health
NHS Digital for the Health and Social Care indicators including population health indicators, NHS Outcomes Framework, maternity and CCG outcomes etc. Note hospital episodes are no longer being collected.
Public Health England statistics for everything on food to prison health, school populations to infectious disease surveillance.
PHE data and analysis tools for access to data and analysis tools that public health professionals can use.
Public Health Fingertips for health profiles (sexual health/oral health/mental health/alcohol and drugs), children and young people benchmarking and profiles, public health outcomes framework, general practice profiles, atlas of variation, diabetes, cancer, disease and risk factor prevalence profiles and lots more! It hosts 35 tools sharing one codebase and over 2,000 indicators. Fingertips also contains GP profiles and can take it to local authority areas, wards etc. You can pull up performance on an indicator e.g. cancer and benchmark against CCG value and national. You need to use PHOF for immunisations.
NHS England's Data Catalogue which hosts 1266 datasets including A&E statistics, QOF, cancer, waiting times and NHS Outcomes Framework.
ONS for health expectancies, pregnancy, conceptions, geographical patterns of cancer survival, suicides, child mortality, avoidable mortality, national well-being, personal well-being, sexual identity estimates,poverty, health-related lifestyles etc.
MyNHS for GP Practice metrics.
International Statistics including OECD comparisons, Eurostat and WHO European Health and Hospital Morbidity
Bowling, A. Research Methods in Health. (Open University Press, 2002).
Neuman, W.L. Social Research Methods: Qualitative and Quantitative Approaches. (Allyn & Bacon, 1997).
Check out examples at ONS website.