next review Flashcards
how do you calculate the mean of a frequency distribution
midpoint of each class x frequency = x sum of all the x = y
x/y = mean
- verify this
The median is also the
50th percentile or Q2
when calculating descriptive statistics for grouped data
treat data values as if they occur at the midpoint of a class (ie in frequency distribution)
the value of the sd can never be
negative
The variance can never be
negative
If all the observed value sin a sample are close to the sample mean the sd will be
small (close to zero)
If the observed values vary widely aroudn the sample mean, then SD will be
large
If all of the vaLues in the sample are identical the sample sd will be
zero
The sum of the deviations from the mean is
zero
Can the variance of a sample be negative
no
Correlation coefficient ranges between
-1 and +1
Range is a measure of
variability
IQR is a measure of
Variability
Variance is a measure of
variability
SD is a measure of
variability
Skewness is a measure of
distribution of shape
z-scores is a measure of
distribution of shape
Covariance is a measure of
linear association b/w 2 variables
correlation coefficient is a measure of
association b/w 2 variables
Covariance is a measure of how two variables change together, but its magnitude is unbounded, so it is difficult to interpret. By dividing covariance by the product of the two standard deviations, one can calculate the normalized version of the statistic. This is the
correlation coefficient.
If SXY is positive
there is a positive linear association b/w x and y
if sxy is negative
there is a negative linear association between x and y
what calculation would you use to dispersed distribution if you have the mean and the standard deviation
coefficient of variation , the higher the number, the more dispersed
When computing the geometric mean, how is it different than computing the mean where the values are concerned
you multiply your values together for Geo where mean you add them
Prob the that you get A and b? independent
Probability of A union B which is P(AnB) = P(A) x p(B)
Prob of at least one
Prob of A or B which is P(AuB) = P(A) + P(B) for independent and P(A)+P(B) - P(AnB) for dependent
do not get either A or B
P(B) + (PAnB)
No A but yes to B
P(Ac) + P(AnB)
