11
tested for inter-coder reliability and determined that coders agreed on approximately 94% of
coding decisions.
To create our measure the school discipline court climate, we considered the direction of
all previous discipline-related court decisions in a public schools’ jurisdiction. That is, a given
state’s school discipline court climate is captured by examining decisions handed down in all
courts at the state and federal level that hold sway over The court climate measure is constructed
by 1) treating all relevant court decisions as having equal weight, 2) assigning the court decisions
values of +1, 0, or -1 depending on whether the ruling was, respectively, pro-school, neutral, or
pro-student, and then 3) calculating a value for the average prior direction of decisions found in
the cases for a given location or time period.
For scholars interested in using the technique in other research, alternative approaches to
measuring court climates are also possible, depending on one’s interests. For example, rather
than treating all cases equally, researchers could weight their court climate measures to account
for how much influence a given case outcome had on other legal decisions. Shepardizing a case
on the Lexis-Nexis database allows one to assess how many times a given case has been cited in
subsequent decisions, thus quantifying its influence on later cases. Supreme Court decisions
would thus be likely to “count more” toward the court climate than state high court decisions
because they set precedent for every state rather than only one. Another alternate measure might
consider cases in nearby jurisdictions that may be reported in local newspapers and thus possibly
influence the behavior of institutional actors. Court climates could also be measured using an
instrumental variable which controls for variation in details of the cases and is thus a court
climate relatively uninfluenced by the specifics of each case.
Convention