Roastatistics

I love maths and numbers nearly as much as I love gravy.

The first question I am going to tackle is – “Is it worth paying more for a roast dinner?”.

I listed the 30 roasts reviewed so far in two columns on Excel, one with the score, the other with the price. I then produced a scatter-diagram:

UntitledIt seems straight-forward that there is a positive correlation as you can see by the best-fit line.

But I didn’t want to take that for granted so I ran a correlation test using the Data Analysis package in Excel, and it gave me a score of 0.55.

Now there are more precise statistical methods to determine the significance of a correlation score, however for the sake of my audience (and also partly my brain as it is 17 years since I did A-level Maths) I will just use the general guidelines that anything over 0.5 is a strong correlation.

As it is a positive number, this is a strong correlation and proves that for our sample, you do get what you pay for.

If you want a better roast dinner, pay more money.

I hope you enjoyed this – if you have any suggestions of analysis that you would like to see, please do message me.

 

Like
Like Love Haha Wow Sad Angry

One thought on “Roastatistics

  • July 6, 2015 at 4:03 pm
    Permalink

    0.55 Is actually a really bad correlation! Try something closer to 0.8

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *