I get that standard deviation is a measure of spread, but when is it appropriate to use it? Right now I'm doing a piece of Statistics coursework on the relationship between the wealth of a country and the life expectancy of its inhabitants (my hypothesis = the richer the country, the higher the life expectancy)...
- Can I use standard deviation for this investigation? I've heard you have to use it if you want a decent mark.
- If I don't use standard deviation, can I make up for it using various other higher level techniques?
- What techniques can I use?
- Should I just start a whole new investigation focusing on the Mayfield High School thing?
I'm pretty sure I have to hand this in tomorrow (managed to avoid handing it in on Friday cause I was in an Art exam)! Please someone take pity on me before I go completely round the bend and set my Statistics textbook on fire or something!!
Turn on thread page Beta
WHEN is standard deviation useful/relevant?? watch
- 06-04-2003 16:51
- 06-04-2003 19:53
As you said the std dev is a measure of spread - if a data point is further than 2 std devs from the mean it is considered an outlier, so you could use it to find ouliers - maybe to remove them and see how it affects your results.
- 06-04-2003 19:57
Wahey, I'll do that, thanks v.much gilyan u r fab!
- 15-06-2003 14:45
hi, my advice is stick at it, dont start mayfield its stupidly hard if u want above a b grade, and it takes ages!
- 15-06-2003 14:51
96% of results are contained within 2 standard deviations of the mean. This means that a result on either end of the scale would be in the bottom or top 2% of all possibilities regardless of the range. This is useful in the outlier sense, as mentioned. But it can be used to set cut off points in exams etc. Which is how they do UMS marking in our exams