Social data science postgraduate personal statement example

Personal statement example

My sociology degree taught me to ask why people behave as they do; it was the quantitative strand that showed me how hard those questions are to answer well. In my second year I took an optional module on survey analysis, largely because a friend said the seminars were small and the tutor patient. I stayed because I liked the discipline of it: the slow work of understanding how a variable was coded, who had been asked, and who had quietly dropped out of the sample. My final-year dissertation used secondary survey data to look at whether young people in outer suburbs of my city reported greater difficulty reaching further education than those nearer the centre. I found a modest association, and then spent several weeks working out how much of it might be explained by household income rather than distance. The honest answer was: a great deal, and my data could not fully separate the two. Writing that clearly, rather than overstating the finding, was the most useful thing I learned.

Since graduating I have worked as a customer resolution officer for a housing association, handling complaints about repairs, noise and rent accounts. Much of my day involves the case management system, and I have become the person colleagues ask when they want to know how many cases of a particular type we closed last quarter. Building those summaries taught me things no module could: that free-text fields are where the real information hides, that categories are chosen for administrative convenience and then treated as though they describe reality, and that two teams can count the same thing differently for years without noticing. When our team reviewed repeat contacts, I suggested we look at cases by property rather than by tenant, which changed the picture noticeably. It was a small contribution, but it convinced me that I want to work with social data properly rather than incidentally.

I have been preparing for further study in my own time. I completed an introductory R course and have been reworking my dissertation analysis in R rather than SPSS, partly to make it reproducible and partly because I wanted to understand what my earlier software had been doing for me. More recently I have started Python, working through exercises on text data using publicly available local council meeting minutes. My attempts at topic modelling are crude, but the exercise raised questions I would like to study formally: how much interpretive weight a researcher can place on clusters produced by an algorithm, and how such methods sit alongside the close reading I was trained in. Reading Cathy O'Neil's Weapons of Math Destruction and, more recently, D'Ignazio and Klein's Data Feminism has sharpened my sense that decisions about measurement are also decisions about whose experience counts.

Outside work I share the care of my grandmother with my mother and aunt, which mostly means shopping, appointments and two evenings a week, and I help at a community advice drop-in where people bring benefit letters and housing forms. Both have made me steadier and better at explaining things without condescension. They have also given me a practical scepticism about administrative data: I have sat with people whose circumstances did not fit any of the boxes on the form in front of them.

I am looking for a course that combines social theory with computational methods, including causal inference, working with digital trace and administrative data, and the ethics of research using personal information. Afterwards I would like to work in research for a local authority, a housing body or a policy charity, producing analysis that is careful about its own limits.