Task: to evaluate the research design, methods of data collection and analysis, and any other ethical or philosophical issues that arise in the specified research paper.
This assignment will focus on the working paper entitled Child Poverty Dynamics in Seven Nations (Bradbury et al, 2000). It will identify and analyse the research design, methodology, data collection and analysis contained within the paper. There will also be an assessment of philosophical and ethical issues as well as a comparison with other documents of a similar nature. The paper is a comparative study examining how children move in and out of poverty.
The authors of this paper are concerned with child poverty and how children in different countries move in and out of poverty. They maintain that state welfare provision operates more effectively to reduce child poverty when it has prior research knowledge of what causes children to move in and out of poverty The research uses standard relative poverty definitions and examines the mobility rate of the poorest fifth of children from, Britain, the USA, Germany, Ireland, Spain, Hungary and Russia. It is usual at the design stage of a project to decide what approach one is going to take to the research. This is because different epistemological and philosophical assumptions are an inherent part of any approach to research ( Bryman, 2004). At a general level the study is an inter-societal comparison of the dynamics of poverty. Inter-societal comparisons look at the similarities and differences which countries display. Numbers of such studies have been undertaken, a well known recent one of these is Esping-Anderson’s (1990) research into the welfare regimes of different countries. Comparative studies have a long history in sociology, e.g. Weber’s work on religion (1930, 1965) and Durkheim’s work on suicide (1952). Comparative studies are a big part of research into poverty, whether this is the traditional notion of poverty as a lack of disposable income, or whether it is the more contemporary concept of social exclusion (Berghman, 1995). Bradbury et al (2000) acknowledge that they have taken a rather broad brush approach in their study and that there are some problems with this, nevertheless they maintain that charting the flow in and out of poverty cross nationally is useful for policy making that aims to reduce the number of children who are poor. The researchers believe that studying poverty from the vantage point of children needs little or no justification because children represent a country’s future. It might be argued however, that is a rather Eurocentric view and may not necessarily be held in all the countries in the study. The researchers further defend their choice of method in the following way:
If you need assistance with writing your assignment, our professional assignment writing service is here to help!Find out more
Comparisons across countries provide a reference point for assessing the results for any single country, for example whether a particular statistic is large or small. Cross-national analysis also raises provocative questions about whether differences in outcomes are due to, say, differences in policy regimes or differences in population characteristics. Of course the usefulness of crossnational analysis relies on having good data, and making data comparable may require compromises in the depth of analysis which would not be required in a single country study. (The trade-off depends on the number of countries considered.) Our paper illustrates the various strengths – and weaknesses – of taking a cross-national perspective (Bradbury et al, 2000, p.7-8).
The authors thus express a generally held view that comparative research does have integral strengths and weaknesses, the following analysis hopes to demonstrate whether or not the strengths of this research project outweigh any weaknesses it may have..
Methods of Data Collection
The primary data source for this study is panel data drawn from the seven nations involved. This data is largely household survey data and is therefore the income levels are those given by heads of households. The researchers maintain that there study is complementary to one undertaken by Duncan et al (1993). The studies differ in that Duncan et al’s work concentrates on the family as the unit of analysis whereas Bradbury et al (2000) concentrate on the child. The units of analysis are children under the age of 18. The data sets refer to the early 1990s with the most recent year being 1996. The researchers want to use the data to compare poverty dynamics between one year and the next, so they looked at data from two years across the seven nations. The data sets differ, with four countries, Germany, Hungary, Britain and the US providing data spanning five years and data from Germany and the US spanning ten years. This means that some countries receive a deeper analysis than others because they collect more data. The income figures provided by households are used as the income of the child with relevant adjustments for household needs. Table 1 below gives a summary description of each of the data sets used.
Bradbury et al, 2000 pps 10-11
This table gives an overview of the data that the researchers used in their comparative study. The researchers chose the following features from which to compare the data for different countries:
…the type of longitudinal survey, the period to which incomes refer, the definition(s) of income available, and two statistics summarising sample size (Bradbury et al, 2000, p.11).
The chief indicator of whether a child is living in poverty and how a child moves in and out of poverty are the income measures that are available. The researchers point out however that there are differences between countries on how this is assessed i.e. whether income recorded is before or after any deductions for tax etc. They acknowledge that such differences have clear implications for differences in poverty dynamics but they do not elucidate what this is. This lessens any faith that one might have in their methods of data collection because there is no explanation of how this affects poverty dynamics. Income is recorded as net except in the case of Ireland and the US. The researchers say that net income for Britain cannot be recorded in all cases and this causes a reduction in sample size, with possible resulting implications of difference (?).
Only two countries, Spain and Russia, provide evidence of household expenditure in addition to household income and arguably this is a weakness in the data set as levels of expenditure may differ widely from country to country and is a greater indication of the distinction between absolute and relative poverty (Giddens, 2001). Again this might evidence greater discrepancies across nations if more of such data were available and this will raise questions as to the reliability of the findings of the study.The study only uses disposable income as a measurement of poverty, In a sense this is a step back in terms of theoretical development as Berghman (1995) has said the focus has shifted from simply financial poverty to whether a person can fully participate in the society to which they belong. This is not referred to in Bradshaw et al’s study and as such might be said to evidence a weakness in their choice of conceptual indicators. While the measure did work for the variables that the researchers were interested in it is nevertheless a narrow way of measuring child poverty. This is especially the case when one considers the researchers’ acknowledgement that they have no single comparable method of measuring income across the seven nations and this raises questions about the internal validity of their methods.
Data Analysis Techniques
The study uses quantitative data and the study is quite large using household panels from seven nations. The researchers argue that this not only provides them with a cross national comparison of how and why children enter and leave poverty but also allows for any serious discrepancies between nations to be identified. The number of households selected for the analysis is between 1 and 2 thousand per country (see table one above). Statistical comparisons are made between child poverty rates, their relative income levels and income inequality. The statistics are similar to those found in an earlier chapter but no details of this are given. Arguably, one wonders why they mention any similarities here as they then go on to say that:
…they are not fully comparable because there are differences in the definition of the income measure, the year referred to ,the sample, and in most cases even the survey (this is true in Britain, Ireland ,Spain, and the US) (Bradshaw et al, 2000, p.13).
The researchers say that they use the median income of children to measure material well being but because their data set and methodology are rather convoluted they have to explain what this is. Their usage of a median is questionable as they later say (p.15 ibid) that: median income levels provide no guide to how incomes vary among children. Their use of arithmetical averages is also questionable as this can obscure the existence of very high and very low incomes, a fact which they also acknowledge.Their methods include the use of a statistical technique known as the Gini coefficient The technique is named after the person who developed it, an Italian statistician called Corrado Gini. The use of the Gini coefficient does tend to show whether income inequality is increasing or decreasing and so it is often used in comparisons between countries. However, its capacity to measure inequality is also determined by how disposable income is dealt with and this information is not available. It is arguable therefore whether the use of the Gini coefficient gives an accurate representation of the data.
The use of the Gini coefficient tends to suggest that income inequality for children is substantially different across the seven nations. Western Europe has lower inequality levels than do Russia and America, and there may be further discrepancies here because the Russian data also provides details of household expenditure and this is missing from the American data. Bradshaw et al’s (2000) findings show that with the Gini coefficient income inequality in Germany stands at 0.3 while in America and Russia it is 0.4 a rise of 10 percentage points which the authors say is larger than the overall income inequality of Britain and the United States throughout the whole of the 1980s. However, further questions are raised as to the reliability of the data from the US because the US Census Bureau shows that the calculation of the index of the US was changed in 1992 this led to an upward shift of 0.02 in the coefficient making comparisons after that period misleading.
Bradshaw et al (2000) claim that their major finding is a:
… significant (but not total) uniformity in patterns of income mobility and poverty dynamics across the seven countries. The key exception is Russia, where the economic transition has led to a much higher degree of mobility (Bradshaw et al, 2000, p.6).
Having said this, the authors later go on to demonstrate that rather than there being significant uniformity, there are considerable differences between the seven countries. Their findings indicate that the US has the highest rate of income inequality which affects the income mobility but as has already been indicated American figures after 1992 may be misleading. Another factor that makes their finding problematic is that the data for all countries is only available for two years, for some five and for two countries it is available over ten years, as the author’s admit longer time spans significantly affect the dynamics of poverty. Income mobility is greater in those countries where the most data is available it is therefore arguably the case the only significant comparison of income mobility is that between America and Germany. The ways in which the data is analysed, is, like the rest of this paper, rather complex. It is not always easy to decide when they are talking about method and when they switch to data analysis techniques as the information tends to be rather mixed together. The complexity of this paper makes it difficult to assess what the author’s may have missed in their analysis, while the authors claim a number of findings from the research there is also evidence to suggest that they could have found the contrary.
Analysing this research paper has been a problematic task. The authors continually refer back to other papers/chapters in the collection of which this paper is a part but because they do not give sufficient information on what they are referring to it might be argued that it is pointless referring back to work that is not available for comparison. The structure of this paper is confusing it does not give clear indications as to when it goes from research design, to data collection, to analysis. Discussion of the issues is ongoing and at times repetitive. There are significant weaknesses in this paper, there is little discussion of ethics as they use secondary data analysis but by their own admission there are significant differences and discrepancies in the data and this leads to the view that the both the reliability and validity of the methodology and findings are questionable. On the positive side the researchers do recognise that there are weaknesses in the data set and that this may affect the findings. There are points where the researchers (arguably at least) make claims for their data that are difficult to substantiate e.g. that there are significant similarities between patterns of income mobility across the seven countries. This claim does not really hold up in view of their further analysis of the data which lists the differences between countries. In addition to this much of the data relating to the US could, according to the US Census Bureau, be misleading. Finally the constant referring back to other papers without elucidating the issues does not help the researchers to make their case and their use of income as a major conceptual indicator does not really tell the reader how poverty affects the lives and well being of the children who are meant to be the focus of the research.
Berghman, J. 1995 “Social Exclusion in Europe: Policy, context and analytical framework” in
Room, R. ed Beyond the Threshold: The Measurement and Analysis of Social Exclusion University of Bristol, Policy Press
Bradbury, Bruce, Stephen P. Jenkins and John Micklewright (2000), ‘Child Poverty Dynamics in Seven Nations’. Innocenti Working Paper, No. 78 Florence: UNICEF Innocenti Research Centre.
Bryman, A 2nd ed. 2004 Social Research Methods Oxford, Oxford University Press
Duncan, G.J., Gustafsson, B., Hauser, R., Schmauss, G., Messinger, H., Muffels,
R., Nolan, B., and Ray, J.-C. (1993). ‘Poverty Dynamics in Eight
Countries’. Journal of Population Economics, 6: 295–34.
Durkheim, E. 1952 Suicide London, Routledge
Esping-Anderson 1990 The Three Worlds of Welfare Capitalism Cambridge, Polity
Giddens, A 4th ed 2001 Sociology Cambridge, Polity
Weber, M. 1930 The Protestant Ethic and the Spirit of Capitalism London, George Allen and Unwin
Weber, M., 1965 The Sociology of Religion London, Methuen
http://en.wikipedia.org/wiki/Gini_coefficient accessed 28/4/06
http://hampshire.edu/~apmNS/design/RESOURCES accessed 28/4/06
 This is the terminology used by the researchers see page 10 of the study
 This working paper apparently forms part of a collection of such papers on child poverty. The researchers continually refer back to issues mentioned in other papers without elucidating what these issues are.
 Again the paper refers back to issues covered in other papers without indicating what these issues are.
 For example, the UK figures in BJM Chapter 3 (the UK is defined as Britain and Northern
Ireland) are based on the Family Expenditure Survey while the results for Britain in this paper are
based on the British Household Panel Survey.
 By ‘median income for children’ we mean the median of the distribution of children, ranked by
the value of equivalised income of their household (p, 15 ibid).
 This is a measure of inequality of distribution which is often used in the measurement of income levels. It is a measure between 0 and 1 where 0 corresponds to complete equality and 1 to complete inequality
 http://en.wikipedia.org/wiki/Gini_coefficient accessed 28/4/06
Cite This Work
To export a reference to this article please select a referencing stye below:
Related ServicesView all
DMCA / Removal Request
If you are the original writer of this assignment and no longer wish to have your work published on UKEssays.com then please: