Measures of central tendency
- This an average of a set of scores.
- Used for interval or ratio data
- But, it can be affected by extremes/outliers
- This is the 'middle score' of an ordered set.
- Used for ordinal, interval and ratio data.
- Is resistant to extreme values and outliers.
- The most frequent value in a set of scores
- Use any kind of data!
- The difference between the highest and lowest values in a data set.
- The variability of scores
- Calculated by the sums of deviations from the mean(squared) divided by n-1.
1 of 4
STD DEV and SEM
- Standard deviation represents the difference from the mean, how much scores differ from the mean.
- Standard Error of the Mean (SEM) is how well the sample mean estimates to the population mean. Therefore variation in the sample = the 'error' in estimating the population mean.
- The equation for SEM is STDDEv of the sample/square root of N.
- A larger sample will yield a smaller SEM value.
- SEM is used to calculate confidence intervals and error bars in graphs.
2 of 4
The Normal Distribution
- This is a family of distribution, with the area under the curve representing 100% of the sample.
- The bell curve (aka the normal curve) can be seen with continuous variables, with scores closer to the mean appearing more frequent.
- The most extreme values are found at the tails of the curve, the tails approach x but never actually touch.
- The mean, mode and median are all very similar.
3 of 4
- Smaller samples
- Lead to HIGH sampling errors
- Relatively varied sample means
- Larger samples
- Lead to LOW sampling errors
- Have generally stable sampling means
- The sample distribution of the mean is a normal distribution.
- Larger samples mean that there are smaller confidence intervals. This is because there need to be more specific estimates.
- A confidence interval is an interval estimate of the population parameters. 95% of the area under a normal curve is in the interval 1.96 SD away from the mean.
4 of 4