Applied Data Science (Online Learning) MSc postgraduate personal statement example

Personal statement example

Most complaints about buses are about lateness, but in my job as a service controller for a regional bus operator, the figure that worries me more is early running. A bus that leaves a stop two minutes ahead of schedule strands anyone who arrived on time. Last year I noticed that one rural route showed early departures clustered at three stops. I exported the ticket machine and GPS logs and plotted departure times against scheduled times by hour. The pattern was not driver behaviour. The timetable assumed school-run congestion that disappears during holidays. I shared the plots with our scheduling team, and the holiday timetable for that route now has adjusted running times. It was a small change, but it showed me that the data our systems already collect can answer practical questions if someone has the time and method to ask them properly. I want to develop that method, and an online MSc would let me do it while continuing in a role where I use it every week.

My statistical foundation comes from a BSc in Biology, which I completed with a 2:1. For my dissertation I counted bumblebee visits to flowering verges across twelve urban sites over a summer, comparing verges mown on the council's standard cycle with those left uncut. Because the data were counts and repeated visits to the same sites were not independent, I fitted generalised linear mixed models in R using a Poisson distribution with site as a random effect. I spent longer than I expected checking for overdispersion and learning why a model that ignores site would have overstated my confidence. The result was modest: uncut verges had more visits, but weather explained much of the variation. Writing that honestly was one of the more satisfying parts of the project.

Since graduating I have tried to fill the gaps in my preparation. I worked through an introductory Python course in the evenings and now use pandas to clean the weekly exports at work, replacing a fragile set of spreadsheet formulas that broke whenever a route number changed format. I have also taught myself enough SQL to query our depot's reporting database rather than waiting for monthly summaries. I am aware that this is self-taught and uneven. I can write a working join, but I have not studied database design, machine learning beyond simple regression, or how to deploy and maintain a model others rely on. Those are the areas I most want structured teaching in.

Outside work I volunteer at my local parkrun, where I help tidy the results after each event: matching finish tokens to barcodes when a runner forgets to scan, or a token is dropped. It is unglamorous, but it has taught me how easily small recording errors propagate, and how much a clear process for correcting them matters. I also keep an allotment, which mostly teaches me patience and that a spreadsheet of sowing dates does not stop slugs.

Studying online suits how I learn. My dissertation involved long stretches of working alone with data, and I am used to organising my own time around shift patterns. My employer has agreed that I can use anonymised operational data for coursework projects where appropriate, which would give me realistic, messy problems to apply new techniques to, from forecasting passenger numbers to understanding where delays begin.

In the longer term I would like to work in transport analytics, where decisions about timetables and routes affect people who depend on public services and rarely see the reasoning behind them. I want to be able to build analyses that are rigorous, explain their limits clearly, and are useful to the colleagues who have to act on them.