Statistics Project

Authors Avatar

Statistics Project

Plan

Aim

My general aim from this investigation is to see if there is a relationship between the unemployment rate and the turnout of the 2005 general election in a random sample of different constituencies.

Method

I will randomly select 60 (about 10%) pieces of data as a sample by assigning each constituency a number and generating random numbers to choose the constituencies that are included in my sample. To make sure that my sample is as helpful as possible, it should be in the same proportion of winning parties to the actual thing, so the chosen constituencies were filtered so that each category had the right number of each party. I shall then plot the unemployment rate of these chosen constituencies against the turnout for that constituency in the 2005 general election in a graph, and note what I can discover from it and whether there is any correlation. If there is positive correlation I will know that places where more people are unemployed are likely to have a higher turnout for the general election, and if there is a negative correlation that it will seem the opposite. To carry out this investigation further, I plan to make two box and whisker plots to discover whether the constituencies who elect Labour are more likely to be unemployed. I will take a new random sample, taking 20 random constituencies from each of Labour and Conservative using the method described before and plotting each of these onto a separate box and whisker plot. I can then use these two box and whisker plots to compare them and make a judgment as to whether it seems Labour or Conservative constituencies are more likely to have high levels of unemployment.

Join now!

Hypothesis

I believed at first that in my initial investigation, the higher the level of unemployment was then the higher the turnout would be because they would have the incentive to come and vote to make a change in their lives, and wouldn’t be at work anyway so would probably find it easy to find time to vote. However, when I made a small preliminary chart from the first couple of data entries, the correlation seemed to suggest that this hypothesis was wrong and it was actually the opposite. To work with this I have revised my hypothesis ...

This is a preview of the whole essay