You are Here: Home

# srandard deviaton and correlation coefficient watch

Announcements
1. hi all!! wats standard deviationa dn correlation coeffienct? how will this help me in my coursework? how do box and whisker plots help me???
2. Correlation Coeffients can help you put number to the relationship between two things. The one that examiners love is the "Product Correlation Coefficient", this is the one where you calculate Sxx, Syy and Sxy.
3. Standard Deviation is the "spread about the mean". It kinda means the average distance from the mean of the values. It is kina difficult to explain.
4. Box Plots are a graphical representation of the highest and lowest values. It also shows the median, upper and lower quartiles.
5. We did pearson's correlation coefficient in psychology to test the strength of the correlation between two variables. The higher the number, the closer the scores are together so the higher the correlation is.

Standard Deviation is the number of points scores are away from the mean. I measures how much scores fluctuate from the mean.
6. Standard deviation is one measure of spread about the mean. As Leekey says, it's a measure of the average difference from the mean. But how do you average these differences, given that some are negative and some are positive, and their sum will be zero? Standard deviation does it by squaring them (always giving non-negative results), then taking the mean of the squares, then taking the square root at the end.

There is another way of doing this which is simpler. This is called the mean absolute deviation. This involves just taking the absolute values of the differences from the mean, and then taking the mean of those values.

Two of the big reasons why standard deviation (SD) is used more than mean absolute deviation (MAD) are:

1) SD is easier for computers to update, because they have to store fewer figures in their memory. The best way to understand this is...check it out . Whether or not this is mentioned is a good indicator of whether an introductory statistics textbook is the real McCoy or whether it's just been hacked out.

2) you can use a 'false zero' with SD, which allows the alternative formula

SD = square root of ([mean of squares] minus [square of mean].

You can think of this as moving the mean to zero, then using the differences from there , i.e. the values themselves, then taking away the square of the mean, etc.

Many undergraduate courses on quantitative methods give social science students (the subset who hate stats and are just doing it because they have to) the impression that SD is some sort of 'best' measure of spread, but this isn't true. As with other statistics (e.g. average), it depends on what you want, and it's worth bearing in mind that ease of calculation is often a factor in choosing SD rather than MAD.

Dave

TSR Support Team

We have a brilliant team of more than 60 Support Team members looking after discussions on The Student Room, helping to make it a fun, safe and useful place to hang out.

This forum is supported by:
Updated: June 5, 2003
Today on TSR

### Christmas GAME

Hurt or Heal? You decide

Poll
Useful resources

## Articles:

Debate and current affairs forum guidelines