Chapter 17 Descriptive Statistics Part 2

Both variance & std dev will be bigger if the data are spread out

Skewness

Positively skewed

Negatively skewed

tail on the right

e.g. difficult test

mean > median > mode

tail on the left

e.g. easy test

mean < median < mode

Measures of Variability (how spread out the data values are)

Range

Variance

Standard Deviation

H-L

limited - severely affected- extreme numbers

average deviation from mean - squared units

square root of variance

how far - numbers - vary - from mean

Both variance & std dev will be bigger if the data are spread out

Measures of Relative Standing (where a score falls in comparison to other scores in the data)

percentile ranks

z- scores

% of scores - reference group - falls below - a raw score

raw score -> standard dev. units

how many std devs - raw score- from the mean

raw score above mean : + z score

raw score = mean : 0 z score

can combine z-scores and percentile ranks if normally distributed

Examining r/s among variables

Contingency Tables

Regression

calculated down, compare across

calculated across, compare down

to see r/s between variables

Simple

Multiple

predict the values of DV based on 1 IV

predict - values of DV based on 1 / more IV

Y intercept

Regression Co-efficient

where the line crosses Y axis

predicted change in Y per unit change in X

slope