My ultimate task for this summer was to get a descriptive summary of the care work industries in different countries around the world, including the proportion of the care work industry relative to the total industry size and the average personal income in care work industry vs non care work industry within each country. The data source came from the Luxembourg Income Study, which contains harmonized micro datasets for 51 countries spanning five decades. This work will be combined with unpaid caregiving research to observe their correlation and interdependence and ultimately test how unpaid and paid caregiving explain the feminization and devaluation of care work in different countries. My work this consisted of three steps. First, I defined care work for each country and year based on the dataset’s available labels in occupation and industry variables. I defined “broad” to be the definition based on the documentation provided by Budig and Misra’s “How Care Work Employment Shapes Earnings in a Cross-National Perspective.” A more “concise” definition of care work included primarily those involved in direct involvement with children and elders and personal care work. I documented these labels in a spreadsheet. Second, with the use of Stata I printed the frequency and the mean and standard deviation of personal labour income for each country and year. Third, I created time plots of the relative size of care work industry and average income for each country. I also produced similar time series plots for the United States using the Current Population Survey from the Integrated Public Use Microdata.
Through this experience, I learned a great deal about this field of work and I found many of the theories and explanations found in other academic papers very intriguing. I was able to get a taste of what data work would look like and was able to learn a new statistical software language, STATA, and also expand on my experience with R. I learned how to work efficiently and streamline as much as I can, especially with handling with hundreds of datasets and documentations. I learned to work diligently especially in the earlier stages of data labeling and cleaning where it required a lot of attention to detail and hours of repetitive work.
As an Economics student, I was able to experience working with data and see how this work can help lead to the next step that requires many of the regression techniques that I’ve learned in my Econometrics class. Further, it gave me a good understanding of much of the data work that social scientists did in research and has rekindled my interest in pursuing independent research myself sometime during my last two years at Penn.
My ultimate task for this summer was to get a descriptive summary of the care work industries in different countries around the world, including the proportion of the care work industry relative to the total industry size and the average personal income in care work industry vs non care work industry within each country. The data source came from the Luxembourg Income Study, which contains harmonized micro datasets for 51 countries spanning five decades. This work will be combined with unpaid caregiving research to observe their correlation and interdependence and ultimately test how unpaid and paid caregiving explain the feminization and devaluation of care work in different countries. My work this consisted of three steps. First, I defined care work for each country and year based on the dataset’s available labels in occupation and industry variables. I defined “broad” to be the definition based on the documentation provided by Budig and Misra’s “How Care Work Employment Shapes Earnings in a Cross-National Perspective.” A more “concise” definition of care work included primarily those involved in direct involvement with children and elders and personal care work. I documented these labels in a spreadsheet. Second, with the use of Stata I printed the frequency and the mean and standard deviation of personal labour income for each country and year. Third, I created time plots of the relative size of care work industry and average income for each country. I also produced similar time series plots for the United States using the Current Population Survey from the Integrated Public Use Microdata.
Through this experience, I learned a great deal about this field of work and I found many of the theories and explanations found in other academic papers very intriguing. I was able to get a taste of what data work would look like and was able to learn a new statistical software language, STATA, and also expand on my experience with R. I learned how to work efficiently and streamline as much as I can, especially with handling with hundreds of datasets and documentations. I learned to work diligently especially in the earlier stages of data labeling and cleaning where it required a lot of attention to detail and hours of repetitive work.
As an Economics student, I was able to experience working with data and see how this work can help lead to the next step that requires many of the regression techniques that I’ve learned in my Econometrics class. Further, it gave me a good understanding of much of the data work that social scientists did in research and has rekindled my interest in pursuing independent research myself sometime during my last two years at Penn.