3 days to go on Belfast

Dear Upper Sixth,

I know many of you are counting down to Tuesday, so this post is another quick reminder of what to do this weekend, and the key meanings and formats for hypotheses for the stats tests we have been working through this week.

Believe me when I say the blog will become super interesting and more relevant to geography as a whole after the Tuesday exam, I have lots of ideas to share for the tectonics module and the final one on climate, but until Tuesday it's Belfast all the way!

So, starting with arguably the hardest thing to maintain interest in, but possibly the easiest way to get good marks, your stats tests. You need to be able to work, use and interpret four key tests, and the basics such as IQR, mean, median and mode.

1) Spearmans Rank - used to test the relationship between two variables and gives three possibly results.
A) a positive correlation - as one goes up, the other goes up (Close to +1)
B) a negative correlation - as one goes up the other goes down (Close to -1)
C) no correlation (close to 0)

Key checks
The answer should not be outside the range 1 to -1
The most common error is the last stage of the equation which is subtracting the result of the division from 1
You need to comment on the relationships strength (closeness to 1 or -1) and its significance, which will be given to you in the table along with the formula

Hypothesis - There will be a significant positive correlation between the number of X religion and the income deprivation of areas in Belfast

2) Standard Deviation - tests the spread of results from a set of data around the mean
A) a high number in comparison to your range of data indicates there is no clustering of results around the mean, and the data is widely dispersed
B) A low number in comparison to your data range indicates most numbers are close to the mean, or are clustered

Key Checks
Make sure you remember to square root the final answer

Hypothesis
Most people in Ward X will fall within the age range of 18 to 25

3) Chi Squared
Tests to see if the observed data from counts in an area is different to that expected. For instance, the number of people employed in different sectors of industry in four different wards. If the result is positive, it means there is a statistically significant difference between observed and expected, you then need to refer back to the raw observed data to analyse the table and find the reason

Key Checks
Your grand total should be the sum of all your rows or your columns which should be the same number, don't add the two together
Go to 2 decimal places, rounding up from the 3rd for all your calculations
When calculating the degrees of freedom in order to read off significance, be aware it is calculated by
(rows -1) x (columns -1)

Hypothesis
There will be no difference between observed and expected frequencies of users of community centres in the four regions assessed

4) Mann Whitney U
This test is normally done for frequencies or values across two data samples, often surveys or results for two different years. It shows whether or not there is a statistically significant difference, you have to work out if it has increased or decreased from one to the other

Key Checks
You add up your MWU values for each row and take the lower one as your answer
This is the only test for which your value has to be lower than the significance level
You can have different amounts of data in each data set
Significance is read off on a table to give you an accurate number for N in both rows

Hypothesis
Due to regeneration work in Belfast, there will be a significant decrease in hate crimes in regions that have been targeted by the Neighbourhood Renewal Strategy

Hopefully this helps a little,

Keep reading the AIB,  come up with your suggestions for the city, be aware of Liverpool and the Docklands for comparative case studies, remember your geography words - Geographerise each sentence and be synoptic!

One final weekend to work then it is all over and we will be back to Tectonics!