Archive for October, 2018

Ohio teacher discipline statistics

October 4, 2018

An article appeared in the Dayton Daily News (DDN) about
teacher misconduct which included this figure:


The information in the figure came from this Ohio Department of
Education report. (The report shows a similar figure on pages 13.)

A quick examination of the chart reveals that the numbers do not
appear to add up. I.e. the total number of cases does not equal
the number of cases resulting in discipline or no discipline.
At first glance this appears to be a significant issue. I made
my own version of the plot, adding a sum that represents the
“Discipline” + “No Discipline” numbers.


Reading through the article, one finds that the number of cases
resulting in a decision are not necessarily the same cases
that were opened that year because sometimes cases take a while
to resolve.

However, the most important fact to note from the OH DOE report
is that 77% of these cases come from people applying for a
teaching license, not people who are actually teachers since
the department investigates people applying for teaching credentials
as well as current teachers referred to for misconduct. It’s also
important to note that the number of applicants increased and if
you normalize the number of cases opened by the number of applications
you get a very different picture.


which looks like the number of cases being investigated is driven
more by the increasing number of applicants, not some broader
problem with teachers.

Ohio Abortion Trends

October 3, 2018

The Dayton Daily News (DDN) published an article discussing the trend of abortions in Ohio. The linked article does not have the same headlines nor include the table from the print version, so I took a picture of it and put it here. Crucially the subheading says:

Numbers buck 10-year trend showing fewer procedures in the state.


The numbers presented don’t show the full 10 year trend referenced in the article. The numbers in the DDN article actually came from this Ohio Department of Health Report,
which includes more detailed information over a longer period of time. So I plotted back
to 2006. There is clearly a dip and rise again from 2010-2012. The size of this exceeds the 2017 rise substantially, making one wonder what trend the author of the DDN article sees that is being bucked. In addition to totals, the report also breaks out numbers for Ohio residents which I also included.


What was also missing in particular from the DDN article was any discussion of the size of Ohio’s overall poplulation relative to the report abortion numbers. For example, the report increase might actually be a decrease if the change in population size is accounted for. I obtained numbers from US census¬†and computed totals per 100,000 people.


Plotted this way, it appears the Total number increase from 2016 to 2107 is only 0.3%, not the 1% mentioned in the article. Looking at just the Ohio resident numbers, the result is even less exciting, showing almost no change.