Orientation Observation In-depth interviews Document analysis and semiology Conversation and discourse analysis Secondary Data Surveys Experiments Ethics Research outcomes



Social Research Glossary

About Researching the Real World



© Lee Harvey 2012–2020

Page updated 29 April, 2020

Citation reference: Harvey, L., 2012–2020, Researching the Real World, available at
All rights belong to author.


A Guide to Methodology

7. Secondary data

7.1 Introduction to secondary analysis
7.2 Extent of re-analysis of secondary data

7.3 Nature of the data

7.4 Data sources

7.5 Examining data sources
7.6 Methodological approaches

7.6.1 Positivism and secondary analysis
7.6.2.Phenomenology and secondary analysis
7.6.3.Critical approach to secondary analysis Critical analyis of published statistics Secondary data an example of a critical approach

7.7 Summary and conclusion

Activity 7.6.3

7.6.3 Critical approach to secondary analysis
Although published statistics are a resource for researchers they need to be approached cautiously. It is necessary to consider carefully how they were collected, why they were collected and exactly what is being measured.

Published statistics, from whatever source, are dependent on the definitions used and are affected by the way in which they are collected. As the Government Statisticians' Collective (1979, p. 149) so aptly pointed out, 'statistics do not, in some mysterious way, emanate directly from the social conditions they appear to describe'. They are not self-evident but rely on the assumptions, conceptions and priorities of the agencies that collect them.

Furthermore, the actual statistics are compiled by people, usually within imperfectly functioning bureaucracies, and the result is not a perfect measure but the best that time, money, and the pressures of the job allow.

Published statistics, then, should be seen not as 'facts' but as evidence for the social researcher. In the end, published statistics are data like any other data and they cannot be detached from the theoretical context within which they were generated. (See theory-laden nature of observation)

It is important to read statistics, and the commentaries on them, as critically as any other forms of evidence.

Activity 7.6.3
Locate the appropriate official statistics to answer the question, why in the first half of the 2010s, do more girls go to unversity than boys? Prepare a brief report that: 1. Fully and clearly documents sources (also indicate sources that proved unfruitful when consulted). 2. Answers the question (making sure that tables and diagrams are specific to the answer). 3. Comments on the results and analysis, suggesting reasons for trends or differences. 4. Relates the discussion to sociological theory. 5. Assesses the adequacy of the statistical sources used.
Time: about six hours

Top Critical analysis of published statistics
Critical sociologists argue that elements of the social world cannot be separated from the broader social structures in which they are located. Published statistics need to be seen in a wider historical and social context. Government statistics come in for particular criticism. Marxists, for example, argue that government statistics are collected to facilitate the day-to-day running of the state. Far from being a neutral observer, the state plays a political and economic role in sustaining and reproducing the status quo and it is argued that the collection and presentation of statistics reflects this (Miles and Irvine, 1979).

Critical sociologists argue that social structures act as oppressive mechanisms by which one group or class holds power over another. Government statistics thus reflect the concerns of the state in legitimating the existing sets of power relations. They do not provide a picture of what really exists, but reflect the priorities of the state. Thus government statistics are subject to political manipulation (see Section 7.5.1 for examples) through selective redefinition of social phenomena in ways that support the prevailing social structure. The state reflects ruling class ideology and interests, and these influence the statistics it produces.

Official statistics are then indicators of the state's role and priorities. It is extremely difficult, if not impossible, to make a really radical criticism of society using available statistical sources, which imprison us in the concepts and concerns that dominate official politics and economic life. (Government Statisticians' Collective, 1979, p. 138).

Feminists argue that published statistics, especially official statistics, are sexist. This sexism does not arise out of any deliberate attempt by government statisticians to oppress women, but because the conceptual schemes used in government statistics embody a particular sexist mode of thinking (Oakley and Oakley, 1979). Women are rendered invisible in many official statistics such as employment and unemployment statistics. This is not because the data cannot be collected but because dominant patriarchal ideology allocates women to specific roles and activities and the official statistics reproduces this ideology.

For example, the New Earnings Survey at one time looked at part-time employment among women, but not men. The Family Intentions Survey was based on a survey of women only, although it refers to intended family size. Marital status is often included in the analysis of statistics for women but not for men (as in employment figures). Housework, an area of work in which women are over represented, is not included in official statistics on work because it is not paid employment. Men, rather than women, are usually defined as 'head of household' and the socio-economic characteristics of many women have tended to be defined on the basis of their husband or father. Women's own occupational identities have effectively been deemed irrelevant where they are not categorised as head of the household. Such statistics reproduce the idea of women as the primary carers and nurturers within families, working only to supplement the man's income.

As noted (in Section, administrative crime statistics appear to under-represent the real level of crime. This distortion is even more acute for women than for men. The picture from crime statistics is that women are far less likely than men to commit violent or aggressive crimes but far more likely to commit crimes like shoplifting or prostitution. This reflects the patterns of socialisation to which the sexes are exposed and the different social roles that they play as adults (Oakley, 1972). However, such statistics also reflect sexist bias at the level of data collection because the processes by which a person becomes a criminal are particularly vulnerable to the influence of sexist attitudes. Much of the sexism in published statistics is a reflection of sexist social reality. This applies particularly to official statistics but also can be found in unofficial statistics and social research generally (Eichler, 1988).

Not all the problems with government statistics, however, can be explained by the general sexism within society. Official statistics impose specific sexist models and attitudes on the data. Thus, Oakley and Oakley (1979) argued that sexism limits the usefulness of official statistics since it is impossible to subtract the influence of sexism from the data available. It is often difficult for the social scientist to establish just what the position of women is from the official statistical data. However, as official statistics are collected to satisfy the needs of government and these needs are intimately bound up with the preservation of a sexist social order, the presence of sexism in official statistics serves to reproduce patriarchal ideology.

Assess the adequacy of official statistics from a anti-racist point of view. For example, how adequate are official definitions of ethnic minority groups? To what extent are official statistics available for ethnic minority groups? Do official statistics take account of cultural differences?
Time: about two hours.

Top Secondary data an example of a critical approach
Tony Kushner's (2004) study concentrated on ethnic minorities in Britain in the 1930s and 1940s and demonstrates the complexity of majority attitudes towards ethnic minorities during the height of European racism. In the period up to, and during, the Second World War, Jews formed the largest ethnic minority group in Britain, followed by several smaller ones including Italians and people from the British Empire, who, at that time, had made little impact. Kushner used material from the Mass Observation Archive at Sussex University because he wanted to understand what 'ordinary' members of the ethnic majority thought about minorities in Britain. The Mass Observation archive is a unique source that provides historical insights.

The Mass Observation movement was a populist process of recording everyday activities. It was established as a distinctive approach by the anthropologists Charles Madge and Tom Harrisson during the 1930s and aimed to provide more than just an opinion poll-style snapshot of public attitudes towards events of the day. There were three different ways of collecting people's views. First, specific projects, such as the study of attitudes in "Worktown" (Bolton) during 1939 that generated a sample of working-class views. Second, a more middle-class group were asked to keep diaries. Third, people were asked to give their views on particular issues of the day, again these tended to be more-middle class respondents. Views elicited included those on race and anti-Semitism both in 1939 and 1943. Some of those who kept diaries did so over many years and provided a broad range of opinions, from the outright anti-Semitic to the desire to always act positively towards ethnic minorities.

Kushner's analysis thus provides a history of everyday attitudes towards ethnic minorities in the 1930s and 1940s. Kushner corrects the often-cited view that ethnic diversity in Britain began with the post-1945 migration. There were black and Asian minorities but they Jewish population totalled about half a million and they were well-established as was the smaller Italian community, which faced internment and was the subject of riots following Mussolini's declaration of war on Britain in June 1940.

Kushner's approach moved away from the race-relations approach that dominated the study of minorities by social scientists in the 1960s1990s. Attitudes towards ethnic minorities were not simplistic but rather complex, nuanced and subtle. For example, Kushner identified several reactions, such as the curiosity towards the rarely encountered black and Asian populations. Kushner focuses on the attitudes of people from Worktown towards so-called 'exotic' performers in Blackpool. There was also hostility, which is evident in the interviews and diaries. This emerges, amongst other things, in comments on the appearance of black people. It is also evident in the hostility towards Jews, which was prevalent across Europe. Indeed, some comments had a fascist flavour; one diarist wrote that he agreed with Hitler wanting to be rid of the Jews although disliking his methods. There were, though, positive views expressed in some diaries, which included shock at events in Nazi-occupied Europe.


Next 7.7 Summary and conclusions