Averages and measures of spread Flashcards
median equation
(n+1) /2
normal distribution
mean and median approximately equal
positive skew
mean greater than median (peaks to the left)
negative skew
mean less than median (peaks to the right)
what to use for ordinal categorical or continuous variables
mean or median
what to use for categorical nominal variables
no sense to use median or means
what’s a factor
when there’s 2 category variables
how to know what analysis to use
if both are categorical use a contingency table
if group and continuous compare means/medians
percentile
value at or below which a specified percentage of the scores in the distribution fall
eg 60th %tile = 60% of observations lie at or below and 40% at or above
percentile calculation
value of percentile = (percentile/100) x (n+1)th
25th percentile
lower quartile
50th percentile
median
75th percentile
upper quartile
lowest quintile
lowest 20%
highest quintile
highest 20%
lowest decile
lowest 10%
highest decile
highest 10%
lowest tertile
bottom third
highest tertile
top third
measures of spread
range
interquartile range
standard deviation
variance
interquartile range
diff between upper and lower quartile
variance equation
sd squared
lower fence equation
LQ - (1.5xIQR)
upper fence equation
UQ + (1.5xIQR)
residual
difference between a particular observation and mean
standard deviation in relation to residual
square root of sum of squared residuals or in statistical jargon
sd equation
square root of
sum of (y-mean)2
/
n-1
