Magíster en Bioestadística postgraduate personal statement example

Personal statement example

Every weekday at the dairy cooperative where I work, around three hundred milk samples arrive from farms in plastic racks, each tube carrying a barcode that links it to a herd and a collection date. My job is to check that the barcodes, the tanker logs and the laboratory results agree before anything is reported back to farmers. In my first month I noticed that a handful of somatic cell counts each week were implausibly low for herds that were usually high. Tracing them back, I found that one tanker route occasionally scanned racks in reverse order, so results were attached to the wrong farm. Nothing about this required advanced mathematics, but it showed me how much a statistical conclusion depends on the unglamorous structure of how data are collected. I would like to study biostatistics at postgraduate level because I want the methods to match that care.

My degree was in biology, with a minor in statistics that I chose after enjoying a second-year course on experimental design more than I expected. I liked that a well-designed experiment could answer a question that a larger but careless one could not. Later I took courses in linear models and categorical data analysis, and taught myself enough R to move beyond the menu-driven software we used in practicals.

For my final-year project I worked with an anonymised dataset provided through my department, recording the time taken to dispense prescriptions in a hospital pharmacy over several months. A simple comparison of mean times across weekdays suggested Mondays were slowest, but observations were clustered within staff members and shifts, and some pharmacists worked mostly on busy days. Fitting a linear mixed model with random effects for staff member changed the picture considerably: much of the apparent day effect reflected who was on duty. I also had to decide how to handle heavily right-skewed times, and compared a log transformation with a gamma model before choosing the former for interpretability. My supervisor pushed me to explain each decision in plain language for the pharmacy manager, which I found as demanding as the modelling itself.

Outside work, I have looked for ways to keep learning. I have been working through Gelman and Hill's Data Analysis Using Regression and Multilevel/Hierarchical Models, partly to understand the mixed models from my project more deeply, and I have started reading about survival analysis, since time-to-event questions seem central to clinical research and my dispensing-time data were, in a sense, a crude version of one. I would value formal training in this area, along with longitudinal methods and the principles behind clinical trial design, where I currently know only the outline.

On Saturday mornings I help at the results desk of my local parkrun, matching finish tokens to barcodes when the scanners fail. It is a cheerful, small-scale version of my weekday job, and it has made me patient with messy records and with the people who generate them. I also play five-a-side football weekly, which has little to do with statistics but keeps me in contact with friends from outside work.

My strengths are a solid grounding in biology, which helps me understand what a variable actually measures, practical experience of data quality, and a habit of explaining analyses clearly to non-specialists. I am aware that my mathematical preparation is less extensive than that of a mathematics graduate, and over the past year I have revised linear algebra and probability theory through online lecture notes and problem sets to prepare. I hope postgraduate study will give me the rigour to work on health data where errors matter, and to contribute to research teams as someone who understands both the numbers and where they came from.