Voice
in the Classroom: How an Open Classroom Climate Fosters Political
Engagement Among Adolescents
David
E. Campbell
Published
online: 17 April 2008 Ó Springer Science+Business Media, LLC 2008
Classroom
Climate
Obviously,
the key independent variable is the openness of the classroom
climate. CIVED includes a series of items, which suggest that
adolescents are frequently exposed to varying political opinions from
their classmates, and that many teachers do indeed cultivate a
climate of respect for political differences.3 Using principal
components factor analysis on the individual- level data within
CIVED, these items have been combined into an index which measures
the degree to which differing political views are discussed within
the classroom.4 all six items load cleanly onto a single dimension.
However,
previous studies share a common weakness in determining whether
political discussion in the classroom is having the presumed effect.
Past research has relied on respondents to indicate the degree to
which social and political issues are discussed in their classes,
leaving the strong possibility that it is being politically engaged
that leads adolescents to perceive a greater degree of political
discussion in their schools, rather than the other way around. The
analysis below works around this problem by not relying solely on an
individual’s self-report regarding the level of openness within the
classroom. Instead, the openness of the classroom climate is
estimated by averaging the perceptions of multiple respondents in the
same classroom. The use of the classroom mean to measure exposure to
political discussion in the classroom also does not account for the
fact that within a given classroom, students will differ in their
assessment of the degree to which there is discussion of contested
issues. It is likely that adolescents who perceive that there is more
discussion are also going to be more politically engaged. Therefore,
both the classroom average and each individual’s own assessment of
the classroom climate are included in the analysis.
The
complication with including both measures of classroom climate in a
single equation is that they are highly correlated with each other.
To separate individuals’ own perception from the aggregate
(classroom-level) values, therefore, the two measures have been
‘‘purged’’ of any correlation. For the individual
respondent’s assessment, this has been done by regressing the
individual’s own classroom environment score on the class mean, and
saving the residuals (Individual Perception of Classroom Climate).
Since the residuals reflect the degree to which an individual’s own
score deviates from the aggregate value, the two are by definition
uncorrelated. Likewise, the classroom mean was regressed on the
individual’s perception of classroom climate, and the resulting
residuals were entered into the model as Classroom Climate. Note that
this procedure produces a different classroom climate score for each
respondent, which means that it must be treated as an
individual-level variable. However, when the ‘‘unpurged’’
measure of classroom climate is used instead, and entered into the
model as a classroom-level variable (i.e. each respondent within a
given classroom receives the same score), the results are
substantively identical (available upon request).
Admittedly,
these corrections are not a cure-all for potential endogeneity, as
students within a single classroom may collectively be both highly
engaged in politics and more likely to perceive political discussion.
Accordingly, the models control for other characteristics of the
individual, classroom, school, and district which are likely to
foster greater political engagement.
0 comments:
Posting Komentar