Analysing illness absence trends

The data for our school fluctuates a lot so it is hard to see a clear trend but if a curve of best fit was drawn, it could be deduced that there was a rise in absences as the number of people ill increased. What we need is to have the absences rise just before the number of people ill increases. The timing for the flu outbreak was unfortunate because it occurred during the Christmas where there is no school absences data. Due to this, it is difficult to draw any accurate conclusion from the data. We just have to wait for another widespread outbreak during school time.

One Response to Analysing illness absence trends

  1. Dr Rob says:

    I agree that the timing of the peak in the national data was unfortunate from the perspective of our research, but that’s just the way things go. We’ve still got lots of data we can analyse which is exciting.

    In relation to not being able to see a clear trend in your data due to the weekly fluctuations: one thing you might want to try in order to smooth things out is to calculate three week rolling averages. It’s a trick we use in epidemiology quite a lot when we want to smooth out the data. There’s a nice clip about how to do this on you tube here – The clip shows how to do this for sales results, but it’s the same method for our epidemiological data. Let me know if you get stuck and I can try and help out.