3_2_2_standard_deviation_StatCrunch.mov Video Lecture | Statcrunch Tutorials for Beginners - Data & Analytics

this state again represents the
responses to that Sullivan statistics
survey let's go ahead and pick a
variable where we can compute some of
these measures of central tendency and
dispersion like here's one television
this is a number of hours of television
watched in the past week for folks in
the student and in the survey okay so if
I wanted to get some measures of central
tendency and dispersion from this I
would go to stat summary stats columns I
then tell statcrunch that I want the
television data the other thing you'll
notice is because I'm asking for summary
stats it only gives me the quantitative
variables to choose from I could group
by gender if you wanted to see if there
is maybe a difference between say males
and females go undo that click Next and
here are the different measures of
central tendency and dispersion we've
talked about the mean the variance the
standard deviation we don't know what
standard error is yet min and Max
probably a sense of what those are but
we haven't focused on them at all and
then down here unadjusted variance and
unadjusted standard deviation
those are your population parameters
that's the population variance the
population standard deviation we do need
to have we do need to distinguish
between population and sample standard
deviation because the formulas are
different right population standard
deviation divided by and the size of the
population the sample standard deviation
divided by n minus 1 this is sample data
so we'll stick with just those and click
calculate and all the results pop out by
gender so the females are watching about
8.4 hours on average as far as the mean
goes the median has them at 7 hours the
males are watching about ten point nine
hours by the mean
ten hours by the median and by the way
if you look at the mean and the median
for the male's they're pretty close for
the females the mean is a little bit
bigger which might be suggested that
that data is skewed to the to the right
and then there's your variance and
standard deviation and in fact if we
wanted to look at the graphs just for
fun to see if that's the case histogram
television and we need to group by
gender so that we can compare the two I
mean I'm gonna put in all the labels or
the class limits put them right on top
of each other both are pretty skewed to
the right now this one okay both they're
pretty skewed to the right