-measures of central tendency: give single number to tell what is typical of a distr?n, only tells part of the story Mean, median, and mode -variability measures: describe how much variation, diversity, or heterogeneity there is in a distribution Range Inter-quartile range Variance Standard deviation -dispersion: the extent to which the data deviate from the mean -variability: quantitative measure of the degree to which the scores in a distribution are spread out or clustered together -range: distance b/w max & min scores used with I/R variables clunky statistic( only based on 2 scores, lowest or highest may be extreme, yields no info about variation of scores -percentile: point below which a specific percentage of cases fall First Quartile (Q1): score that separates the lower 25% of the distribution from the rest Second Quartile (Q2): score that has exactly 2 quarters, or 50%, of the distribution below it Third Quartile (Q3): score that separates the lower 75% of the distribution from the rest -Inter-quartile range (IQR): width of middle 50% of distr?n (Q3-Q1) value of Q1 = N (.25) = score of # case (ex: 30th case) value of Q3 = N (.75) = score of # case (ex: 10th case) IQR = Q3 - Q1 = 30th case ? 10th case = 3 ? 0 = 3 used with I/R variables not influenced by extremes all benefits of the median( well for skewed ignores a lot in distr?n( only based on 2 scores, no info about variation of the other scores -boxplot: graphic representation of the range, IQR, and median scores of a distribution box represents middle 50% of data (in other words, IQR is length of box) difference between ends of whiskers is the range useful for comparing 2 or more samples compare the centers (median lines) compare variation (length of the boxes or whiskers) -standard deviation( best measure of variability & most used characterize how far all deviations from mean( how spread out from the mean? score ? mean (Y ? Y-bar) variance: the mean squared deviations, find each deviation score square each sum up all squared deviations (to make them positive) divide by N standard deviation: square root of the variance, puts back into same units as original data indicates how far away from the mean the scores are on avg uses all scores in distr?n (all info) describes avg deviation of scores can be tedious if many scores -MCT alone shows similarities & differences, but may stereotype ex: Changing Incomes 1967-2001 look at changes in MCT & MV over time changes in mean income tells about avg income of all Americans changes in median income tells about income for avg or middle Americans std dev is preferred for I/R variables like income 1967 middle Americans earned $32,000, rose to $42,000 in 2001( marked increase in std of living but also look at percentiles 20th percentile( not much change for lower income group ($20,000 to 25,000) 80th( dramatic increase for higher income group ($45,000 to $70,000)