Methods
We administered an online survey to a sample of Oregon State University
undergraduate students recruited from NR-related courses during the
2017-2018 academic year. A total of 260 students from a variety of
majors completed the survey; 218 responses were used for analysis based
on their completion of the survey in its entirety (Table
1). Approximately 21% (n=45) were
forestry majors, which
approximates their percentage among natural-resource-related areas of
study nationally (Sharik et.al 2015). The survey consisted of 40
questions, but only a subset of these were used for the present
analysis.
To measure the ethical/metaphysical elements of environmental
worldviews, we used 11 Likert-type items drawn from three established
scales, including the New Environmental Paradigm scale (Dunlap et al.
2000), the Connectedness to Nature scale (Mayer and Frantz 2004), and
the Environmental Identity scale (Olivos and Aragonés 2011). None of the
psychological constructs these scales were designed to measure fully
encompasses the philosophical worldview construct, as we conceptualize
it, and in appropriating items from these scales it was not our intent
to measure the New Ecological paradigm, nature connectedness, or
environmental identity, per se. Rather, we chose select items from these
scales because they were also suitable to measure certain (ethical and
metaphysical) content of the environmental worldview construct; and
because, as tested and widely-used survey items, we were confident that
they were clearly worded and so had minimal likelihood of generating
response error. Items were rated from 1 to 5 (strongly disagree to
strongly agree). To group these items for analysis we used principal
components factor analysis with a varimax rotation, retaining only
variables with loadings of 0.40 or higher and eigenvalues greater than
1.0 (Kaiser 1974). This procedure returned three factors, which we
labeled Moral Inclusion (MI), Bond with Nature (BN), and Human’s Role
(HR) (see Appendix).
Internal reliability, measured using Cronbach’s alpha (Vaske 2008), was
acceptable for all three factors (MI: α = 0.77; BN: α = 0.60; HR: α =
0.56), so we averaged the scores of items loading on each
factor.11One item was removed from the HR score to improve
reliability. This produced three composite scores for each student,
with lower scores signifying more anthropocentric (MI), more dualistic
(BN), and more hierarchical (HR) beliefs. We calculated overall MI, BN,
and HR means for forestry and non-forestry students, and compared them
using independent sample Student’s t-tests (Table 2).
To capture ethical/epistemological aspects of environmental worldviews,
we developed a measure to assess the extent to which students deviate
from the dominant utilitarian mode of moral reasoning. “Moral
reasoning” refers to the reasons people invoke to explain how they
believe they ought to behave. We presented five statements, asking
students to indicate the extent to which they agreed (or not) that each
expresses an appropriate way to approach an environmental decision
(Table 3). Items were inspired by five ethical theories identified in
the environmental ethics literature (see Des Jardins 2001, Nelson and
Vucetich 2012), each representing a different mode of moral reasoning.
According to natural law theory, what is natural is good, and should
therefore be maintained. Rights of nature refers to the idea nature has
certain moral rights, which humans should uphold. Utilitarianism
suggests humans should interact with the environment in ways that
maximize benefits.22Although the item allowed a
non-anthropocentric interpretation, we expect most students
interpreted “benefit” in anthropocentric terms. According to virtue
theory, humans should manifest certain virtues, such as care and
humility, when they interact with the environment. Finally, in divine
command theory, humans should interact with the environment as commanded
by a divine figure. Students rated each statement on a 0-100 sliding
scale, where 0 signifies neutrality and 100 signifies full agreement
(Table 3).
Although we piloted this measure informally among colleagues, we
acknowledge this was a highly exploratory section of the survey, which
is a limitation of the study. Based on suggestive evidence generated
from this measure, reported below, we highlight the development and
validation of a measure of environmental moral reasoning as a direction
that merits attention in future research. For analysis, we used
independent t-tests to compare mean utilitarianism ratings between
forestry and non-forestry students. We also calculated the proportion of
students who rated utilitarianism higher than or equivalent to other
modes of moral reasoning for both groups, and compared these proportions
using chi-square tests (Figure 2).
A final epistemology measure assessed perceptions of non-scientific
(i.e., creative, artistic, philosophical) ways of knowing. We used
Likert-type items developed by Goralnik et al. (2015), to have students
rate a statement about the value of the humanities. A standard
definition of the humanities was included for reference (Stanford
Humanities Center:http://shc.stanford.edu/what-are-the-humanities).
Survey items were scored from 1 to 5 (strongly disagree to strongly
agree). To group the items, we used principal components analysis,
following the specifications noted above. All five items loaded on a
single factor, which we call Attitudes toward Humanities (AH). Internal
reliability was good (α = 0.88), so we averaged the five item scores,
producing one composite measure for each student (Vaske 2008). Higher
scores correspond to more positive attitudes toward the humanities,
suggesting students acknowledge the legitimacy of forms of knowledge
other than the Western scientific approaches that have generally
dominated in NR fields. We averaged AH scores for forestry and
non-forestry students, and compared the two groups using independent
sample Student’s t-tests. Students were also asked whether they want
their academic program to incorporate the humanities (yes/no/unsure). We
used chi-square to compare “yes” versus “no/unsure” responses
between forestry and non-forestry students (Figure 3).