Week 4

So then lads and lasses you may be wondering what we got up to this week in Geography, well like the previous week we talked about general stuff for a bit and then moved swiftly onto the most exciting topic known to man... statistics. I have to admit, music in Wednesday's lesson (The Bloodhound Gang, Chesney Hawkes, Rick Astley etc.) and cake did make it more bearable. 


Always remember that in statistics n=the number of pieces of data.


Mean, Median, Mode
So basically, there are three really easy ones; the mean, the median and the mode. The mean is all of the pieces of data added together and then divided by the number of pieces of data so say you had 2, 5, 7, 3, 6, 10, 1 the mean would be 34/7=4.86 (to 2 d.p). 
The median is the piece of data which is in the centre, once all of the pieces of data have been put in order so the median would be 1, 2, 3, 5, 6, 7, 10. If there is a lot of data then a quick way of figuring out the centre value is (n+1)/2, so for our example it would be (7+1)/2=4 so the median is the fourth value. Simple right? 
The mode is just the number which appears the most, if they all appear the same number of times then there is no mode. 


The Interquartile Range (IQR)
The interquartile range is relatively straightforward to figure out but people seem to struggle with it. To figure out the range then you need to first figure out the Lower and Upper Quartiles. To figure out the Lower Quartile then you need to do the same for figuring out the value for the median but divide it by 4 instead. So it's (n+1/4), thus for our data the Lower Quartile would be data piece number two because 7+1/4=2... so the Lower Quartile for our set of data would be 2. 
The Upper Quartile is exactly the same apart from we have to multiply by 3, so the equation is 3(n+1/4). Thus the Upper Quartile for our data would be data piece 6 (3(n+1)/4=6) which is 7. 


After this figuring out the range is simple, just subtract 2 from 7 which gives you 5. (It's just a coincidence that is equals the same as the median, it won't always. 


Standard Deviation 
As hard I tried I couldn't figure out how to write this up on my laptop, if you want more help feel free to ask but this website explains it brilliantly. 
http://www.gcseguide.co.uk/standard_deviation.htm


Spearman's Coefficient Rank of Correlation 
Can't do this on a laptop, I know I know I suck. Here's a link instead 
http://www.wikihow.com/Calculate-Spearman's-Rank-Correlation-Coefficient
If they don't give you significance figures then the closer to 1 or -1 your answer then the more correlated. For example a result of 0.96 would show strong positive correlation, whereas a result of -0.96 would show strong negative correlation. The closer to 0 the less correlation. Generally anything below 0.4/above -0.4 has little correlation. 


Chi Squared
Everyone's favourite test, the one which tells you so little and takes so long. Link time, this time find the section entitled "Getting Results"
http://www.geography-site.co.uk/pages/skills/fieldwork/stats/chi.html#n6
Make sure you know this one, don't rely on the internet if you're stuck find out how you do it! 


Mann Whitney U 
Unfortunately I can't find a link for this one and I can't type it up. Sorry guys. 


Yeah so that's what week 4 comprised of...