variation and measure of dispersion

will never be asked to calculate SD!

what are the 2 types of variation?

• continuous variation

• discrete variation

what is continuous variation and how can it be displayed? give an example:

• features showing continuous variation can have any value w/in a range

• most of the time, most individuals are close to the mean and few are found at either extremes - normal distribution

• traits often regulated by multiple genes and influenced by environment

• plotted on histogram overlaid w/ a curve

• e.g. height

what is a normal distribution?

data points are grouped symmetrically around central mean, resembling a bell shaped curve

what is discrete/discontinuous variation and how can it be displayed? give an example:

• features showing discrete variations have distinct categories w/o intermediate values

• traits often regulated by a single gene and not influenced by environment

• plotted on bar charts

• e.g. blood group

what is standard deviation?

the spread of data around the mean

when is a difference significant/not significant?

• SDs don’t overlap = significant

• SDs overlap = not significant

what does a smaller/larger SD mean in terms of variation?

• smaller SD = less variation

• larger SD = more variation

what does a smaller/larger SD mean in terms of reliability?

• smaller = greater reliability (as data has a narrower range and points are more closely grouped around mean)

• larger = lower reliability (as data has a wider range and points are less closely grouped around mean)

evaluate the use of the range as a measure of dispersion:

• strength: easier to calculate (than SD)

• limitation: distorted by extreme values

evaluate the use of SD as a measure of dispersion:

• strength: not distorted by extreme values

• limitation: more difficult to calculate (than range)

how can we evaluate SD data?

• find +ves in data

• find +ves in method/procedure

• find -ves in data

• find -ves in method/procedure