| by Kenneth Chase | No comments

StatCrunch: Creating Boxplots with a Group by Column

In this video, you will
learn to create box plots using StatCrunch. The data set I will be using
in this video is called Asking Prices for
4-Bedroom Homes in Bryan-College Station, Texas. This dataset compares
four-bedroom homes listed for sale in the
two adjoining cities– Bryan, Texas, and
College Station, Texas. They consist of a random
sample of 15 homes listed for sale in Bryan,
and 15 homes listed for sale in College Station. The square foot column
contains the square footage of each home, and
the location column lists the city where
the home is located. For this example, I’m interested
in comparing the square footage home of the four-bedroom homes
listed for sale in Bryant to those of College Station. To do that, I will
begin by constructing box blocks grouped by column. Under the Graph menu,
choose Box Plot. Going to select the
Square Footage column, and then in the Group By
box, I’ll choose location. Go ahead and click Compute. The resulting box plots
display the five-number summary for each location. In our box plot, the
outliers are not identified. To identify the outliers, we
need to modify our box plot. To do so, under Options
choose use Edit. This takes us back to
the original window where we built our box plots. In the windows,
StatCrunch allows the user to customize their box plot. Under the other
options, I’m going to check Use Fences
to Identify Outliers. I’ll click Compute. The modified box
plots are now shown, and notice there is an outlier
indicated for Bryan, Texas. If we click and drag to
highlight the outlier, the corresponding row in the
data table is now highlighted. This highlighting can be cleared
at any time using the Clear button in the bottom
left hand corner. StatCrunch also allows
the user to draw the box plots horizontally
instead of vertically. To do so, go back under
Options, choose Edit. Mark the option that says
Draw Boxes Horizontally, click Compute. The resulting box
plots are now shown with a horizontal orientation
with the numerical scale on the x-axis.

Leave a Reply