Thought I would slightly not follow the rules by doing the second activity this week (analysing the paper by Kan and Laurie) on my blog. If I am totally honest this was probably the first time I have ever read a quantitative paper properly, as in actually trying to understand the tables of statistics and numbers. Normally I just skip through that bit and go straight to the results and discussion part of the paper. By trying to actually understand the ‘numbers’ bit I now realise two things:
Data Source Kan and Laurie used secondary analysis of an existing large dataset (The UK Household Longitudinal Study). One of the main advantages of using such a service (as well as saving time and resources on collating primary data) is the scale of the sample. As the study covers over 40,000 households the amount of data available to there researchers was vast though the amount of information based on ethnicity is quite smaller (1000). Both the large number of participants and the longitudinal nature of the data validates the research in terms of its scope. However, one of the main disadvantages is the lack of control the researchers have over the survey design. Basically with secondary analysis researchers ‘get what they are given’ and have to deal with it accordingly. For example, a limitation the researchers mention several times is there is no way to determine how the participants determine ‘housework’ which could have had an important effect on the responses. Intersectionality Previous research in this area has primarily focused on differences between gender in domestic division of labour which though useful does not reflect differences in relation to ethnic group and gender. Issues such as inequality within the genders and between the genders based on ethnicity are not reflected but, as this research attempts to show, can indicate wide and unexpected differences. Understanding (or not) of statistical terms In actually trying to understand the methodology used rather than just skipping to the summary of results I realised there are many, many terms that are currently beyond my comprehension - in doing some quick reading on some of the concepts (OLS regression, for example) it really made me appreciate the complexity of quantitive methods. There were however other terms I found much easier to get to grips with (perhaps because the language seems a lot more ‘generic’ and not specific to ‘statistical vocabulary’. One of these was cross sectional analysis. Cross-sectional analysis is often contrasted with longitudinal studies (for example here). Cross-sectional can be seen as analysis which happens ‘in one point in time’ rather than across time. In this instance this means that analysis was taken within the The UK Household Longitudinal Study data and not across the different waves of the study. Though the research uses data from two waves of the study, these are ‘pooled’ together rather than analysis with would look at similarities and differences between the two waves. Findings The researchers make a compelling argument for the importance in their research in extending the range of previous research beyond merely gender to incorporate differences between ethnic groups and gender attitudes, especially in light of other contributing factors such as inequality of opportunity (educational attainment, employment status, etc). By using intersectionality to analyse these different areas the reseachers make an attempt to ‘pull apart’ the data to indicate the relationships between the different elements in creating a more detailed picture of the Household Survey that incorporates more than merely male v female comparisons. One result which was very interesting was based on the assumption that the ‘White. . .’ ethnic groups would have the most equal results which was not the case with the Black Caribbean group identified as having the most equal share in housework. However, I also think that this also reflects one to the weaknesses of this research - there seems to have been some ethnic centred assumptions made regarding the expected results. Furthermore, as other students on the course have highlighted, there seems to be a generic relation made between some ethnic groups and religion which seems unfounded.
0 Comments
Leave a Reply. |
AuthorLisa Peel - MSc student, Librarian and permanently exhausted ArchivesCategories |