New Approaches to Data on Food Choices
Libby Bishop, the manager of ESDS Qualidata, has explored change in British food choices. In everyday life the change can be seen in the enormous supply of processed food in the shops. Neither do family members necessarily eat dinner together at home anymore. One reason for the change may be a growing arrogation of individual choice through neoliberalism. The title of Bishop's study "Convenience Food and Choice" reflects this.
Libby Bishop wanted to explore the beliefs and practises from which current uses of convenience food may have emerged by analysing two historical qualitative data sets. One data set was Mildred Blaxter's Mothers and Daughters (Blaxter 2004), which consisted of grandmother interviews and was collected in the 1970s. The interviews covered themes like relationships between generations and health of the interviewees and their families. The aim of the original study was to analyse inter-generational transmission of deprivation.
The other data set analysed was a large-scale interview study The Edwardians, collected by Paul Thompson and Trevor Lummis in the 1970s (Thompson 2005). The original research studied social change. "I analysed what the interviewees had said about foods eaten, home processing of food, and whether meals were eaten together. The original research themes were not connected to food but nevertheless the subject came up in the interviews."
According to Bishop, the possibility to find a new point of view and new research questions is one of the most characteristic features of secondary analysis.
Archiving sparked the interest
Libby Bishop moved from the US to the UK in 2003 and started working as a manager at ESDS Qualidata. The move sparked her research interest in food choices. "At first I felt that it was difficult to find anything else except large amounts of processed food from the grocery aisles. Another inspiration was my exposure to Blaxter's data as I helped to prepare Mothers and Daughters for deposit. I noticed that tinned food was frequently mentioned in the interviews. Since Thompson's and Lummis' data also contained descriptions of foods eaten and social practices at meals, it seemed natural to choose it as the second data set for analysis."
After having studied the two data sets more in detail, Libby Bishop generated a list of questions for her research project. The questions included attitudes expressed toward convenience food, was the use of convenience food always condemned and if not, under what conditions its use was accepted, and what reasons were given for its use. The Edwardians data contained hundreds of interviews, allowing Bishop to study how answers varied by time period, age of respondent, and social class.
The results show that convenience and other processed food have long been viewed as inferior, and women interviewed in the 1970s felt they needed to give reasons for their use of tinned food. They did not feel they needed to give specific reasons why homemade food was superior, and only mentioned briefly that homemade was always better. This fact was regarded as self-evident both by the interviewees and the interviewers.
Contrary to the 1980s and 1990s, people in the 1970s did not justify the superiority of homemade by saying it is more nutritious or healthy. During the periods that the grandmothers describe, themes of tradition, economy and care were dominant in food discourses. Nowadays convenience food advertising tends to focus on variety, ease and freedom. Neither do present-day convenience food users feel the need to justify their choices.
Reuse methodology another area of interest
In addition to food choices, Libby Bishop was also interested in the reuse methodology of archived data. During her research process, she considered the difference between primary analysis, which is based on self-collected data, and secondary analysis, which utilises archived data. In her earlier studies, Bishop had mainly used self-collected data.
"In my research, I made use of computer software which was designed for qualitative data management. Computer technology was really helpful in analysing a large amount of data, although I still had to familiarise myself with both data sets by reading several interviews from beginning to end. It was only after reading them that I could deduce which interview questions were most likely to bring up food issues."
The most essential difference between primary analysis and secondary analysis lies in the relationship between the researcher and the researched. A researcher who has collected the data him/herself possesses memories, mental images, and sensations that may come back during the analysis, to add to his/her data recordings and notes. A researcher who uses archived data does not have these memories. In this respect, the primary researcher has better resources to contextualise the data. Otherwise, both are in the same situation: the existing data need to be analysed, which usually means transcription.
Researchers who are critical to reusing data often claim that knowing the context is a prerequisite for understanding and analysing the data. Libby Bishop also sees contextual information as important. The archived data she used included plenty of documentation, even including the interview transcripts of the primary researchers.
Libby Bishop contemplated the situation of the grandmother interviews a lot. In those interviews, female academic interviewers had interviewed poor, elderly women. The interviewees had most likely perceived the interviewers as social workers rather than researchers. The potential effect of this misinterpretation on the contents of the interviews should not be neglected.
Own research improves co-operation with researchers
Even though the context of the original project is important, Bishop says that secondary analysis involves the process of recontextualising data. Contextual information is needed across time and across two sets of research questions. In addition to the contextual information of the original project, secondary researchers need to add information on the context at the time of reuse as this is the context from which the new research questions have emerged.
Libby Bishop thinks it is good for data archivists to conduct research involving data reuse because it enables them to more effectively address concerns of prospective data depositors about archiving.
» Bishop, L. (2006) "A Proposal for Archiving Context for Secondary Analysis", Methodological Innovations Online [Online], 1(2).
» Bishop, L. (May 2007) "A Reflexive Account of Reusing Qualitative Data: Beyond Primary/Secondary Dualism", Sociological Research Online [Online], Special Section on Reusing Qualitative Data, 12(3).
» Blaxter, M. (2004) "Mothers and Daughters: Accounts of Health in the Grandmother Generation", 1945-1978 [computer file]. Colchester, Essex: UK Data Archive [distributor], July. SN: 4943.
» Thompson, P. & Lummis, T. (2005) "Family Life and Work Experience Before 1918", 1870-1973 [computer file]. 5th Edition. Colchester, Essex: UK Data Archive [distributor], April. SN: 2000.
Crime and Health Most Popular in Qualidata
Qualidata was founded in 1994 as a unit within the Department of Sociology at the University of Essex. At that time, the unit concentrated on charting the location and condition of qualitative data sets collected for what could be defined as classic studies. Recovery plans were made for located data. Indeed, many important data sets were archived, the earliest dating from the 1950s. For instance, the data sets collected by Peter Townsend, Paul Thompson, Frank Becchofer, and Mildred Blaxter were successfully recovered, and have been in active reuse since then.
In 2001, Qualidata began a new life as a specialist unit housed within the UKDA (The UK Data Archive) at the University of Essex. In the past few years, Qualidata's focus has been on archiving of electronic qualitative datasets and promoting their reuse.
There are altogether nine employees in Qualidata, some of whom work with both quantitative and qualitative data. Senior qualitative data support services officer John Southall says there are already 113 electronic datasets available in the archive's collections. All research themes have attracted interest among re-users, although data sets related to crime and health have been the most popular. Customers vary from researchers to post-graduate and undergraduate students.
In the UK, archived data are used in positioning new studies at the pilot stage, in comparative research, and in methodology teaching.
"Nowadays there is a growing tendency among researchers to conduct studies which aim at comparisons over time and longitudinal dimensions. For example, the data for a new study are often collected within the same geographical area and using a similar target group as in the archived data set in its time," John Southall says.