JOURNAL OF RESEARCH IN SCIENCE TEACHING VOL. 44, NO. 2, PP. 327–348 (2007) Students’ Mental Models of the Environment Daniel P. Shepardson, 1,2 Bryan Wee, 1 Michelle Priddy, 1 Jon Harbor 2 1 Department of Curriculum and Instruction, Purdue University, 100 North University St. West Lafayette, Indiana 47907-2098 2 Department of Earth and Atmospheric Sciences, Purdue University, 100 North University St. West Lafayette, Indiana Received 27 June 2005; Accepted 21 June 2006 Abstract: What are students’ mental models of the environment? In what ways, if any, do students’ mental models vary by grade level or community setting? These two questions guided the research reported in this article. The Environments Task was administered to students from 25 different teacher-classrooms. The student responses were first inductively analyzed in order to identify students’ mental models of the environment. The second phase of analysis involved the statistical testing of the identified mental models. From this analysis four mental models emerged: Model 1, the environment as a place where animals/plants live—a natural place; Model 2, the environment as a place that supports life; Model 3, the environment as a place impacted or modified by human activity; and Model 4, the environment as a place where animals, plants, and humans live. The dominant mental model was Mental Model 1. Yet, a greater frequency of urban students than suburban and rural students held Mental Model 3. The implications to environmental science education are explored. ß 2007 Wiley Periodicals, Inc. J Res Sci Teach 44: 327–348, 2007 Historically American’s viewed the environment as a commodity, an inexhaustible resource for their use and profit; few individuals were concerned about the environment (Nash, 1990): Thoreau, in the 1850s, who gave significance to nature; Muir’s efforts in the early 1900s led to a national sentiment for preserving wilderness lands; and Leopold in the mid 1900s pioneered the land ethic, an understanding of the interrelationships of organisms and the environment (Nash, 1990). The opinions, findings, and conclusions or recommendations expressed in this paper are those of the authors and do not necessarily reflect the views of the NSF. Contract grant sponsor: National Science Foundation (NSF); Contract grant number: 9819439-ESI. Correspondence to: D. P. Shepardson; E-mail: [email protected]DOI 10.1002/tea.20161 Published online 9 January 2007 in Wiley InterScience (www.interscience.wiley.com). ß 2007 Wiley Periodicals, Inc.
22
Embed
Students’ Mental Models of the Environmentkcarmel/CC_BehavChange_Course...mental models are always under construction and based on new knowledge, ideas, conceptions, and experiences;
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
JOURNAL OF RESEARCH IN SCIENCE TEACHING VOL. 44, NO. 2, PP. 327–348 (2007)
Students’ Mental Models of the Environment
Daniel P. Shepardson,1,2 Bryan Wee,1 Michelle Priddy,1 Jon Harbor2
1Department of Curriculum and Instruction, Purdue University,
100 North University St. West Lafayette, Indiana 47907-2098
2Department of Earth and Atmospheric Sciences, Purdue University,
100 North University St. West Lafayette, Indiana
Received 27 June 2005; Accepted 21 June 2006
Abstract: What are students’ mental models of the environment? In what ways, if any, do students’
mental models vary by grade level or community setting? These two questions guided the research reported
in this article. The Environments Task was administered to students from 25 different teacher-classrooms.
The student responses were first inductively analyzed in order to identify students’ mental models of the
environment. The second phase of analysis involved the statistical testing of the identified mental models.
From this analysis four mental models emerged: Model 1, the environment as a place where animals/plants
live—a natural place; Model 2, the environment as a place that supports life; Model 3, the environment as a
place impacted or modified by human activity; and Model 4, the environment as a place where animals,
plants, and humans live. The dominant mental model was Mental Model 1. Yet, a greater frequency of urban
students than suburban and rural students held Mental Model 3. The implications to environmental science
education are explored. � 2007 Wiley Periodicals, Inc. J Res Sci Teach 44: 327–348, 2007
Historically American’s viewed the environment as a commodity, an inexhaustible resource
for their use and profit; few individuals were concerned about the environment (Nash, 1990):
� Thoreau, in the 1850s, who gave significance to nature;
� Muir’s efforts in the early 1900s led to a national sentiment for preserving wilderness
lands; and
� Leopold in the mid 1900s pioneered the land ethic, an understanding of the
interrelationships of organisms and the environment (Nash, 1990).
The opinions, findings, and conclusions or recommendations expressed in this paper are those of the authors and do
not necessarily reflect the views of the NSF.
Contract grant sponsor: National Science Foundation (NSF); Contract grant number: 9819439-ESI.
The Environments Task is a two-part task. First students are asked to draw a picture of the
environment and to explain their drawing in their own words. Second, students are shown a series
of seven photographs (Figure 2) and asked: (a) to indicate if they think each photograph depicts the
environment and (b) to justify their responses in writing. The seven photographs represent
STUDENTS’ MENTAL MODELS 333
Journal of Research in Science Teaching. DOI 10.1002/tea
different natural and human-managed environments: desert, urban homes and sky scrapers, bears
in a stream, woodland stream, aerial view of a cornfield, industrial plant, and a deciduous forest.
The students’ drawings are conceptual visualizations or representations of their conceptions
of the environment. Drawings as representations are an active, deliberate meaning-making
process and, like words, are embodied with meaning (Alerby, 2000; Kress et al., 2001). The
students’ drawings, then, are representations of their mental models (Glynn & Duit, 1995) and
‘‘reveal qualities of understandings that are hidden from other procedures’’ (White & Gunstone,
1992, p. 99). Furthermore, it allows students who have difficulty expressing their ideas verbally or
in writing an opportunity to reveal their ideas (Rennie & Jarvis, 1995). The written portion allows
students to explain the drawings in their own words, and clarifies their conceptions of the
Figure 2. Photographs from the environments task.
334 SHEPARDSON ET AL.
Journal of Research in Science Teaching. DOI 10.1002/tea
environment for the authors. These written responses also allow the authors to validate meanings
constructed from students’ drawings.
A prototype of the Environments Task was administered to a sample of 20 seventh, 23 eighth,
and 25 ninth grade students as a field test. These students were primarily Caucasians from a rural
school in an agricultural community. As part of the field test students were interviewed about the
task and their responses. Based on this field test the task was modified slightly in its wording and
several photographs were deleted.
Data Collection
The Environments Task was administered to the students during the month of September.
Students completed the task during their regularly scheduled science time (elementary) or science
class. Each teacher was familiar with the task and its administration and had previously completed
the task themselves. The teachers administered the task to students prior to providing any formal
classroom instruction about the environment. As noted above, it is unknown what formal or
informal educational experiences these students had prior to completing the Environments Task.
Data Analysis
Data analysis involved two phases. The first phase involved a content analysis of students’
responses resulting in the identification of students’ mental models, and this process was inductive
in nature. The second phase of analysis involved the statistical testing of the identified mental
models within and across tasks and classrooms. The Chi-square test was used to statistically
determine the independence and goodness-of-fit of the mental models. These two phases of
analysis are described in detail below.
Content Analysis. The open-endedness of the Environments Task required an inductive
approach as students described through words and drawings what was meaningful and salient to
them. In order to identify students’ mental models the data were content analyzed using methods
of inductive analysis; that is, instead of searching for pre-determined patterns, themes were
allowed to emerge from the data as the authors interpreted students’ conceptions of the
environment (Patton, 2002). The following process details the analytical procedure described by
Rubin and Rubin (1995). We first coded the draw and explain portion of the task and then coded the
photograph portion of the task. From the first reading of the students’ responses core concepts
(codes) were identified. These initial codes were revised after a second reading. Descriptive
themes were constructed using the core concepts that emerged from students’ responses at
different grade levels, from different geographical locations, and community settings within and
across the two components of the task. The codes were placed into categories and the categories
were grouped into typologies that reflected the students’ mental models. The data were analyzed
for confirming and discrepant situations.
The two parts of the Environments Task ensured credibility (Erlandson, Harris, Skipper, &
Allen, 1993) in the data collection and analysis process. This also provided synchronic reliability
(Kirk & Miller, 1986) in that the different parts of the task allowed us to interpret the consistency in
students’ responses. Each part of the task is equivalent in its content focus, yet different in the
manner by which it elicits students’ conceptions of the environment. This difference allows
students’ concepts to be checked against each other, providing a degree of triangulation. Students’
conceptions were also triangulated across different grade levels, geographic regions, and
STUDENTS’ MENTAL MODELS 335
Journal of Research in Science Teaching. DOI 10.1002/tea
community settings. To ensure consistency in coding, an inter-rater reliability coefficient was
calculated by comparing the authors’ coding of 10 randomly selected Environments Tasks. The
inter-rater reliability coefficients for the draw and explain and photograph components of the task
were 0.80 and 0.89, respectively. Coding was monitored throughout to ensure consistency and
reliability.
Statistical Analysis. We employed the Chi-square test to determine statistical significance of
the frequency of the identified mental models across task and classrooms. All Chi-square tests
were run using Microsoft Excel. We first compared the frequency of the mental models across both
tasks to determine if the tasks (i.e., draw and explain, photographs) were similar in eliciting
students’ mental models using a 2� 4 matrix (task�mental model). Next we compared the
frequency of the mental models for each task across classrooms (teachers) to determine the
stability of the mental models using a 25� 4 matrix (classroom�mental model) for each task.
These initial Chi-square tests allowed us to determine the validity of the identified mental models
based on statistical significance; that is, if the mental models reflected the data, then there should
be no significant difference in the frequencies of the mental models across task or classrooms.
Although there should be no significant difference in the overall frequencies of the mental
models, it is possible that the frequencies of the mental models may differ by grade level or
community setting. Thus, to answer our second research question we conducted a series of
Chi-square tests partitioning the data by grade level and community setting. To compare the
frequency of the mental models across grade level (i.e., upper elementary, middle school, and high
school) we constructed a 3� 2 matrix (grade level� task) for each mental model. To compare
the frequency of the mental models across community setting (i.e., rural, suburban, urban) a 3�2 matrix (community setting� task) was constructed for each mental model. In order to determine
if a significant difference existed between the mental models we conducted a Chi-square test
comparing the frequencies of each mental model. This would statistically determine the goodness-
of-fit of the data and if students held a prevalent mental model of the environment.
Results
We first present the results of our inductive analysis, identifying the 12 emergent categories
that reflected the students’ mental models. Next we report the results of our statistical analysis of
the distribution of students’ mental models across different data partitions: grade level and
community setting.
Students’ Mental Models of the Environment
From the inductive analysis we identified 32 codes or concepts that we grouped into
12 different categories of student responses on the draw and explain and photograph portions of
the Environments Task (Table 1). Next we grouped the 12 categories into 4 typologies that
reflected these students’ mental models of the environment:
� Model 1: a place where animals/plants live—a natural place
� Model 2: a place that supports life (animal, plant, and human)
� Model 3: a place impacted or modified by human activity or intervention
� Model 4: a place where animals, plants, and humans live
336 SHEPARDSON ET AL.
Journal of Research in Science Teaching. DOI 10.1002/tea
Description of Mental Model 1. Mental Model 1, a place where animals and plants live—a
natural place reflected student conceptions that the environment consists of living and nonliving
components (not including humans), or was a place where energy is transformed and matter
cycled. This mental model consisted of the following five categories: (1) a place where plants and
animals live; (2) the environment as a natural place; (3) the environment as consisting of living and
nonliving components; (4) the cycling of matter; and (5) energy transfer. In all cases, the
environments drawn were natural places which excluded humans and built or managed environ-
ments (Figure 3). The category, a place where plants and animals live, reflected 35% of the student
responses (Table 1) and is illustrated in Figure 3. The student has drawn a picture of trees, birds, an
owl, and a butterfly, a natural place where plants and animals are found or live. There are no
symbols or signs of humans or human built objects or artifacts included in the drawing. The student
writes, ‘‘my drawing is in an environment because it [h]as trees, animals.’’ The next dominant
category within this typology, the environment as a natural place, reflected 11% of the students’
responses (Table 1). This category is slightly different in that the students’ written responses
explicitly describe the environment as a ‘‘natural place’’ or that it is ‘‘nature.’’ The students’
drawings consisted of plants, trees, and animals, but not humans or human built objects or artifacts.
The living and nonliving category reflected 6% of the students responses and is similar to the
previous two categories (Table 1). The difference here is that students explicitly wrote about or
labeled parts of their drawing as consisting of living or nonliving factors. For example, students
would draw and identify soil and air as nonliving components and trees and animals as living
components of the environment. Again, humans were not drawn or mentioned in the students’
responses. The remaining categories (cycling of matter, energy transfer) reflected 3% of the
student responses. In these cases, student drawings would show the sun and describe the sun as
providing energy for plants (photosynthesis) and that the energy was transferred from plants to
animals and animals to animals. Again, humans were not part of these processes, nor were the
processes explicitly linked to the survival of plant and animals; they were simply processes of
nature.
Table 1
Relationship between mental models and categories
Mental Model Categories Percentage of Responses
Model 1A place where animals/plants Place where animals/plants live 35
live, a natural place Natural place, nature 11Living and nonliving 6Cycling of matter 2Energy transfer 1
Model 2Supports life Supports animal life 12
Supports human and animal/plant life 2Supports human life 2Supports animal/plant life 2Supports human and animal life 2
Model 3A place impacted or modified by
humansPlace were only people live (Built
environment)14
Polluted environment 1Model 4
A place where animals, plants,and humans live
Place where animals, plants, andpeople live
10
STUDENTS’ MENTAL MODELS 337
Journal of Research in Science Teaching. DOI 10.1002/tea
Description of Mental Model 2. Mental Model 2, a place that supports life is comprised of
five categories, all indicating that living organisms need the environment to survive or live, that the
environment provides the resources necessary for life (Table 1). These five categories vary only by
the degree that student responses explicitly reference animals, plants, and humans. This mental
model of the environment as supporting life is illustrated in Figure 4. The student has drawn and
labeled the environment consisting of a ‘‘tree,’’ ‘‘duck,’’ ‘‘clouds,’’ ‘‘sun,’’ ‘‘pond’’ (water), and
‘‘grass/soil.’’ The student writes, ‘‘It provides oxygen, water, and sunlight. It is an environment
that has everything you need to live on’’. The student conceptualizes the environment as providing
the resources that animals, plants, and people need to live; the environment supports life. In this
mental model, the environment is viewed as a natural place that includes abiotic and biotic factors,
but it emphasizes the environment as a natural resource for supporting life—plants, animals, and
humans. The cycling of matter or energy transfer is not made explicit.
Description of Mental Model 3. Mental Model 3, a place impacted or modified by human
activity or intervention is illustrated in Figure 5. Here the student has conceptualized the
environment as a natural place impacted (in many cases polluted) by humans. For example, the
student has drawn and labeled ‘‘happy family,’’ ‘‘healthy bird,’’ ‘‘beautiful pond,’’ ‘‘pollution,’’
‘‘sewage plant,’’ and ‘‘tire factory.’’ The student writes, ‘‘My drawing is an environment because
this used to be a peaceful little pond which has turned into a polluted [sic] area.’’ This category of
polluted environment was reflected in 1% of the student responses (Table 1). The other category
within this mental model conceptualizes the environment as a place where only people live. This
category reflected 14% of the student responses (Table 1). Student drawings represented the built
environment: houses, buildings, and various commercial and retail structures.
Description of Mental Model 4. Mental Model 4, a place where animals, plants, and humans
live (Figure 6) places humans in the environment living in harmony with other organisms
Figure 3. Example student response: Model 1, a place where animals/plants live.
338 SHEPARDSON ET AL.
Journal of Research in Science Teaching. DOI 10.1002/tea
Figure 4. Example student response: Model 2, supports animal life.
Figure 5. Example student response: Model 3, a place impacted or modified by humans.
STUDENTS’ MENTAL MODELS 339
Journal of Research in Science Teaching. DOI 10.1002/tea
(e.g., plants and animals). This Mental Model was based on a single category and accounted for
10% of the student responses (Table 1). As illustrated in Figure 6, this student, for example, has
drawn and labeled an environment consisting of a ‘‘human,’’ ‘‘bird,’’ ‘‘fish,’’ ‘‘flower,’’ and writes
‘‘An environment is a place where living things live. . .’’ This mental model differs from the other
mental models in the following ways:
� Mental Model 1 does not include humans, separating humans from the environment
whereas Mental Model 4 conceptualizes humans as part of the environment, living with
plants and animals.
� Mental Model 2 emphasizes the environment as providing resources to support life;
Mental Model 4 does not conceptualize the environment as providing resources, but
simply as a place to live.
� Mental Model 3 views humans impacting (polluting) or modifying (building) the
environment; Mental Model 4 conceptualizes the environment as a natural place
providing organisms (including humans) with a place to live.
Descriptive Results. The descriptive statistics for the four mental models by task are shown
in Table 2. The dominant mental model for these students was Mental Model 1: the environment as
a place where animals and plants live. This mental model separates humans from the environment.
In other words, environments are natural places and humans do not live in, rely on, or impact the
environment. The next most prevalent mental model was Mental Model 2: the environment as a
place that supports life, including humans. This mental model views the environment as a resource
that living organisms need to survive. It views the environment as natural capital available to living
organisms and humans, and does not view the use of resources as impacting the environment but
rather as necessary for animal and human survival. This is followed by Mental Model 3, a place
Figure 6. Example student response: Model 4, a place where animals and humans live.
340 SHEPARDSON ET AL.
Journal of Research in Science Teaching. DOI 10.1002/tea
impacted or modified by humans and where the environment is a place that humans pollute and/or
modify to live. It views human activity as impacting the environment in some way. Mental Model
4, a place where animals and humans live, is similar to Mental Model 1, but includes humans as
part of the environment. It does not view the environment as providing resources for supporting
life or that animals or humans have an impact on the environment, but simply a place where
animals and humans co-exist. It suggests that humans live in harmony as part of the environment,
and it was the least common mental model held by students.
Statistical Results. Each task elicited students’ mental models in similar frequencies;
there was no significant difference among the frequencies of the students’ mental models by task,
X2 (3 df)¼ 3.17, p¼ 0.366. The comparison of the frequency of the students’ mental models by
teacher-classroom also showed no significant difference: task 1,X2 (72 df)¼ 2.02, p¼ 0.999; task
2, X2 (72 df)¼ 0.08, p¼ 0.999. This further supports the plausibility of the existence of the
students’ mental models. There was, however, a statistically significant difference among
the frequencies of the mental models: X2 (3 df)¼ 6.42, p¼ 0.093; thus Mental Model 1 was more
prevalent than the other mental models.
Grade Level and Community Differences
The percentage of each mental model by grade level and community setting are shown in
Figures 7 and 8. The dominant mental model across all situations is Mental Model 1, the
Table 2
Mental model frequencies by task
Mental Model Draw and Explain Photographs
Model 1. A place where animals/plants live 650 (55%) 548 (62%)Model 2. A place that supports life 237 (20%) 155 (18%)Model 3. A place impacted or modified by humans 180 (15%) 83 (9%)Model 4. A place where animals, plants, and humans live 115 (10%) 91 (10%)
0
10
20
30
40
50
60
70
Per
cen
tag
e
Model 4Model 3Model 2Model 1
Mental Model
High Sch
Mid Sch
Up El
Figure 7. Mental Models by grade level.
STUDENTS’ MENTAL MODELS 341
Journal of Research in Science Teaching. DOI 10.1002/tea
environment as a place where animals and plants live. Mental Model 1 also tends to be more
prevalent in students from upper elementary grades in suburban settings. On the other hand,
Mental Model 3, the environment as a place impacted or modified by humans, is more dominant in
urban students’ conceptions relative to suburban and rural students (Table 3). A greater frequency
of students in urban schools held Mental Model 3. In other words, students in suburban and rural
schools were less likely to view the environment from the perspective of human impact or
intervention.
Discussion
It is important to stress that the four mental models are descriptions of students’ conceptions
of the environment as a whole and not of individual students. It is possible that an individual
student, under a different context, might convey a different mental model. The four mental models
are an attempt to characterize the different conceptualizations students hold about the
environment and to summarize these in such a way as to inform practice and to further our
understanding about how students make meaning of the natural world. Furthermore, the mental
models are not intended to represent a conceptual hierarchy. They are meant to distinguish the
varied ways in which students make sense of the environment. Therefore, no particular mental
model is conceptually more complex or ‘‘better’’ than the other.
These students’ mental models about the environment incorporate similar conceptions found
in other studies (e.g., Loughland et al., 2002; Strommen, 1995), that the environment provides a
0
10
20
30
40
50
60
70
Per
cen
tag
e
Model 4Model 3Model 2Model 1
Mental Model
Urban
Suburban
Rural
Figure 8. Mental Models by community setting.
Table 3
Chi-square and probability values for each mental model by grade and community