Ecological Risk
Ecological Risk
Ecological Risk
5, 1995
Relatively little attention has been paid to the role of human perception and judgment in ecological
risk management. This paper attempts to characterize perceived ecological risk, using the psychometric paradigm developed in the domain of human health risk perception. The research began by
eliciting a set of scale characteristics and risk items (e.g., technologies, actions, events, beliefs)
from focus group participants. Participants in the main study were 68 university students who
completed a survey instrument that elicited ratings for each of 65 items on 30 characteristic scales
and one scale regarding general risk to natural environments. The results are presented in terms
of mean responses over individuals for each scale and item combination. Factor analyses show
that five factors characterize the judgment data. These have been termed impact on species, human
benefits, impact on humans, avoidability, and knowledge of impacts. The factor results correspond
with initial expectations and provide a plausible characterization of judgments regarding ecological
risk. Some comparisons of mean responses for selected individual items are also presented.
1. INTRODUCTION
575
0272-4332/95/1ooo6575$07.50/1 0 I995 Society for Risk Analysis
576
2. METHOD
2.1. Participants
577
2.2.1. Items
2.2.2. Scales
The final list included 65 items that were perceived
as posing some level of ecological risk. The set of items
derived from the focus groups was extremely broad, and
included a vast range of human endeavors and natural
phenomena. Four general groupings in these items could
be identified, although some overlap among classifications is unavoidable. One grouping contained natural
disasters (i.e., earthquakes, volcanos, drought, floods,
and meteors colliding with Earth); a second involved
technologies and their applications (e.g., fertilizers, pesticides, mass production farming practices, oil transportation, incineration). A third grouping included human
practices that were seen as potentially having some negative environmental impact (e.g., poaching, disposal of
different kinds of waste products, driving automobiles,
cigarette smoking, beef production, tourism and travel,
scuba diving). The fourth grouping included human beliefs and political/social systems (e.g., capitalism, consumer-oriented society, disconnection of modem life
from natural environments, human dominion over naTwo researchers took notes in each group. Each person then coded
their notes as to potential items and scales. A graduate student compiled the content of each set of notes into a comprehensive list of
items and scales using a simple union procedure. The researchers
then reviewed the lists to eliminate redundancies. This process determined the lists of potential scales and items, which were edited
slightly to insure clarity and make the judgment task feasible for
individuals to complete at one administration.
578
Table 1. Description of Scales and Response Categories in Order Presented in the Survey
Scale end points
Description
of scale
Low (1)
High (7)
Certainty
Please rate how certain it is that the event has an impact on natural environments
Not at all
Very
Adaptability
Please rate how well natural environments maintain their health and productivity
in response to the current level of each event
Cannot at all
can hlly
Avoidability
Please rate how avoidable the event is, in terms of how easy or difficult it would
be to avoid the Occurrence of the event
Not at all
Completely
Relevance to life
Please rate how relevant the event is to your life, in terms of its impact on natural
environments
No
Direct
Controllability
Please rate how controllable is each event, in terms of peoples ability to control
its impact on natural environments
Not at all
Very
Duration of impacts Please rate the duration of the impacts that each event has on natural environments
Short-term
Long-term
Societal benefits
Please rate how much you think the event may benefit the functioning of your
society
No
Great
Personal benefits
Please rate how much you think that you personally can or do benefit from the
event
No
Great
Scope of impacts
Please rate the scope of the impacts of the event, in terms of the size of the area
affected
Small
Widespread
Number of people
Please rate how many people are, or could be, affected by the impact the event
may have on natural environments
Very few
A great number
Species loss
Please rate the impacts of each event in terms of any potential for loss of animal
or plant species
No species
Many species
Destructiveness
Please rate how destructive the event is or can be, in terms of its impacts on
affected natural environments
No adverse
impacts
Complete
destruction
Emotionality
Please rate how much negative emotion (ie., anger, fear, disgust) you feel when
you think about the event and its impacts on natural environments
No
High
Equitableness of
outcomes
Please rate the equity of each event in terms of whether those who receive the
benefits are the same people who incur the costs
Inequitable
Equitable
Ethicality of event
Please rate how ethical you perceive each event to be, in terms of its impact on
natural environments
Very unethical
Completely ethical
Immediacy of
effects
Please rate the immediacy of each event, in terms of how soon its effects on
natural environments may be experienced
Immediately
Infringement on
rights
Please rate to what extent the event infringes on the rights of nonhuman species
Does not
Greatly
are specific to ecological risk (e.g., ability of natural environments to adapt, species loss). In addition, two other
types of scales were derived. First, in keeping with recent research,(14J5)
some focus group members suggested
that the benefits associated with an item may influence
the perceived risk associated with that item. Thus, three
scales were included (i.e., benefits to society, benefits to
persons, and overall goodness). Second, ethical dimensions were mentioned in each of the focus groups. In
response, several scales were developed that addressed
these considerations (i.e., ethicality, infringement on the
rights of nonhuman species, extent of suffering to humans and nonhuman species). Finally, the general risk
posed by each item to the health and productivity of
579
Description
of scale
Revetsibility of
impacts
Please rate the extent to which the impacts on natural environments associated
with the event are reversible (i.e., the ability of natural environments to return to
pre-event conditions)
[rreversible
Reversible
Human suffering
Please rate how much human suffering could result ftom the event as a result of
its impact on natural environments
Please rate how much suffering by animals or plants could occur as a result of
the event
No
Great
No
Great
Understandability
Please rate how easy or difficult it is to understand the impacts each event has
on natural environments
Simple to
Hard to
Predictability
Please rate how well impacts on natural environments associated with the event
can be predicted
Not at all
Very
Recognition of
impacts
Recently
Observability of
impacts
Please rate how observable are the impacts on natural environments associated
with the event
Not at all
Very
Media attention
Please rate how much attention the media has given to the event, in terms of its
impact on natural environments
No
A great deal
Regulatability
of risk
Please rate the extent to which the event can be regulated by governments
Cannot be
Can be completely
Availability of
alternatives
Please rate the extent to which there are reasonable alternatives to the event, or
to the practices that lead to the event
Not available
Are available
Goodness
Human health
risk
Please rate whether you think,in general, the event is good or bad
Please rate the extent to which the event and its impact on natural environments
pose a risk to human health
Very bad
Very good
No risk
A great risk
General acceptability of
event
Please rate the acceptability of each event, in terms of its general impact on
human life and natural environments
Not at all
Completely
General riskiness
Please rate how risky in general you think each event is in terms of its impacts
on the health and productivity of natural environments
Poses no risk
AnimaVplant
suffering
one characteristic of the items. Participants rated the entire set of 65 items on one characteristic scale before
going on the next scale. A final section of the questionnaire collected demographic information (i.e., age, sex,
major area of study) and attitudinal information.
3. RESULTS
3.1. Mean Ratings
An initial step in analyzing the data was to recode
the 1-7 scale into a scale with endpoints of -3 and 3,
and a midpoint of 0. This recording procedure was done
to highlight the relationship between responses and the
scale midpoint. A second step was to create a data matrix
of mean responses over all individuals, for each com-
580
Characteristic
Mean
Certainty
Animaliplant suffering
Infnngement on rights
Duration of impacts
General acceptability of risk
Number of people
Availability of alternatives
Destructiveness
Predictability
Scope of impacts
Species loss
Human health risk
Relevance to life
Controllability
Observability of impacts
Regulatability of risk
Emotionality
Human suffering
Recognition of impacts
Avoidability
Media attention
Reversibility of impacts
Immediacy of effects
Adaptability
Equitableness of outcomes
Ethicality of event
General acceptability of event
Societal benefits
Understandability
Personal benefits
Goodness
1.54
1.31
1.31
1.25
1.13
1.02
1.oo
.90
.8 I
.73
.73
.68
.68
dents were differentiating risks to nature from risks to humans. Another notable difference can be observed by
comparing the social benefit and personal benefit scales,
where on average respondents indicated they benefitted less
personally from the group of items (M = - 1.29) than did
society as a whole ( M = - 1.OO, t = 6.04, p < .01).
In terms of the variability of responses across items,
the highest standard deviations were found for the two
benefit scales (societal = 1.31, personal = 1.32), followed closely by the regulatability scale. The emotionality scale also had a high standard deviation (1.23), with
some items eliciting very negative emotional responses
and other items being perceived as emotionally benign.
The emotionality scale also had the widest range of mean
responses with a low score of -2.43 (outdoor recreation)
as compared to a high score of 2.93 (nuclear war).
Table 111 presents the 65 items ordered in terms of
their mean rating of overall risk to natural environments.
These means were also very diverse, ranging from a low
.64
.64
.64
.60
.59
.39
.27
.I8
-.21
-.71
- .72
-.76
- .83
- .92
- 1.oo
-1.14
-1.29
-1.33
Standard
deviation
.83
.97
I .04
.99
1
.oo
1.01
1.13
.99
.58
1.19
1.05
I .09
.89
I .07
.85
I .30
1.23
1.05
.69
1.19
1.12
.83
.69
.89
.61
I .05
1.10
1.31
.73
1.32
1.05
581
Table 111. Continued
Mean
Nuclear war
Loss of animal species
Depletion of ozone layer
2.69
2.53
2.51
2.5 1
2.5 1
2.43
Loss of wetlands
Air pollution
Disposal of untreated sewage in oceans
2.42
2.26
2.25
2.22
2.1 1
2.06
Acid rain
Conventional warfare
Production and disposal of toxic chemicals
1.99
1.99
1.99
1.77
1.68
1.63
Population growth
Lack of regard for nonhuman rights
Nuclear power plants
1.61
1.60
1.57
1.50
1.44
1.43
Driftnet fishing
Energy production from nonrenewable resources
Drought
1.39
1.38
1.33
Driving automobiles
Earthquakes
Urbanization (continued growth of large cities)
1.28
1.26
1.15
1.15
1.11
1.07
1.07
1.28
1.28
high associations between some scales (e.g., social benefit and personal benefit, r = .96) and no association
between others (e.g., social benefit and availability of
alternatives, r = .03). More than half of the characteristics had correlations of .80 or higher with general riskiness, and five scales (i.e., certainty of impacts,
destructiveness, emotionality, goodness, and acceptability) had correlations of .90 or higher.
The bottom row of Table IV shows that several
scales had relatively low correlations with general risk
Mean
Soil erosion
Floods
Large scale/multinational business
1.06
1.04
1.04
I .03
.99
.99
.92
.90
.89
Dams on rivers
Hunting of animals
Volcanos
36
.85
.82
.82
.76
75
Air conditioning
Mining
Capitalism
.75
.74
,151
Fertilizers
Urban water usage
lmgated agriculture
.53
SO
.47
.32
- .53
-.56
- .72
-36
- 1.42
Scuba diving
Outdoor recreation (e.g., skiing, hiking, climbing)
-1.78
-1.85
16.-
69'-
E8'
56'
16-
IZ'
5P'Z0'-
LL'-
IZ-
8L.88'
L8'
08'-
5L'
L5'
16
8L'
LL'-
9P'
81'-
8P'
E8'-
W'
18'-
6P'
P5'W'
8L'
9s
99-
II'
EL'
ZS'L9'
E8'
P8'-
EL'56'
18'96'
9E'-
LO'-
0Z'08'-
P6'99'
81'
55'-
EO
EL'
65'6L'
85.
EL'PS'
EL'88'-
OZ'
L5'98'
8P'
15'-
L8'
If5Z'
89
LP'ZS'
L5'-
91'SO'
P5'-
0P'-
09'1E'E5'-
tZ'0Z'ZE'
EE'
9P'-
90'
91'
-
W'9L'
89-
LO'10'
ZL'
-
E5'
LE'
6P'
LL'
ZZ'-
EP'
El'
E6'
08'
68'-
0L'-
~~
W'-
99'
OL'
L8'
56'
ZP'ZL'
95'
8P'-
pp'
L9'
PL'
P8'-
EE'-
t8'
00'
E8'-
58'-
EZ-
ZP'60'-
08'-
8t'
65'8L'
LE'-
LL'
L9'-
ZL'
Z9PL'-
6P'
PS
Z9'-
18'
21'Z5'
85.-
p9'
E9'
E8'
18'80
5P'-
IS'-
LZ
6L'
Po'
51'
2 6-
88'
Z Z88'
E8'
ZL'
65'P8'
PZ
8P'- 19.LS'
Z5'-
EL'E5'
19'-
89'
W '-
18'
8P'
5E'
PZ'
PZ'
6L'
PS'
PL'-
IL'
28'51'
I['LZ'S9'-
0Z'-
L9'
00'
LE'
Z5'
8P'
LE'
61'-
IL'
09'59-
LL'LL'
Z1'-
SI'
SP'
0P'-
58'
P8'
0p'S9'
82'96
-
&I.-
6E'
82'
ZP'
EO
62
95'-
08'
PO'
PZ.
6P'-
W'LE'58'-
9L'28'
LL'
8E'
91'
LP'
9s'
If'9E'61'
8L'
P8'W.-
61'
I I'
LS'
81'
90'
69'
10
51'
LL'PL'
8P'-
I Z'
PI'-
98'
6L'LP'
9L'E9'L8.
25'-
LZ'
PE'
LI'Lt'
SZ'
08'
69
88'-
01'-
PE'LE'
01'
95'PZ'9E'
L2
5E'I0
19-
EZ'
09'
ZS'
06'
80-
L1'LZ'
EO
26L8'E5'-
LE.
Po'
P5'
PL'9P'-
90
PZ
60-
05'09'95'-
L9'
Z8'
LV
LV
LL'
LP'L5'55'-
59'
08'
9Z'Po'11'-
69'
ZI'
Zl'80'-
EC
p9'-
9L'
ZO'
81'
ZO'
5Z'
50'0Z'LP'
OL'
19'ES'22-
8E'
89'
If'
9L'16'
L9'E6'
E9'
60' -
91'-
LO-
8P'-
5s'-
18'SP' - Z 9 - 19'
6 9 - LZ'IV- 6L'
65'
89
65'-
88'
PL'
PL.
06.
58'
El
ZP'
ZE'
ZO'
PS'
LL'
89'
55'
If'
LL'
IZ'0Z'PS'-
LO
90'
8 9 - Z9'
28'19'-
E5'LP'-
62
EL'
P8'85'-
SP'
81'
P8'
08'
05'- 19'EP'
65'
SZ-
OL'
IV- El'-
LL06
85'
19'
PL'-
ss-
P8'8L'
8L'EL'68'
E8'
E6'9L'-
5P'-
IL'-
ZL'
8L'89'-
IZ'
ZL'
59
6L'-
19'
8L'85'-
6E'
IP'
9E'0Z'99'-
8P'
EL'
08'
LP'EZ'19-
8L'EL'E9
55'
65'
88'-
IL'
85'
59'
ZL'85'SL'
85'W-
65'
-
LL'
-
59'69'
09'68'
EP'
59-
99'
Z9'89
Lt'
LZ-
6L'
69'
28'LE'
E6
-
69'
09'
-
EZ'62-
OZ
12-
ssautqsu pauat)
583
Characteristic
Species loss
Infringement on rights
AniiaVplant suffering
Destructiveness
Adaptability
Reversibility of impacts
Duration of impacts
Emotionality
Ethicality of event
Certainty
.I1
.I3
-.I2
.I 1
.63
-.62
.54
Societal benefits
Personal benefits
Goodness
-.46
Equitableness of outcomes
General acceptability of event -.58
Human suffering
Number of people
Relevance to life
Scope of impacts
Human health risk
.45
-.I2
- .43
.59
-.41
.58
.49
.88
.88
.I6
.I3
.65
-.60
.59
-.59
.91
.80
.I4
.65
.56
Controllability
Avoidability
Availability of alternatives
Regulatability of risk
Observability of impacts
Pre&ctability
Recognition of impacts
Understandability
Immediacy of effects
Media attention
a
SO
.92
.89
.88
.86
.48
-.45
-.41
.45
.48
.41
.I4
.I3
.69
- .69
-.59
.56
Loadings with absolute values below .40 are omitted from the table.
Names for the factors are discussed in the text.
health risk perception studies to identify such dimensionsJ2)Thus, we conducted a factor analysis of the correlation matrix in Table IV.
Table V presents the summary of a principle components factor analysis with varimax rotation performed
on the interrelation among the mean responses for the
30 risk characteristics. Five orthogonal factors with eigenvalues greater than 1.0 emerged from the analysis.
In the unstated results the first factor accounted for
slightly under 56% of the variance in the data; factor 2
accounted for 18%, followed by 9%, 5%, and 3% for
factors 3, 4, and 5, respectively. Even though the last
three factors accounted for substantially smaller amounts
of variance than the first two, they were retained because
of their conceptual relevance and to maintain clarity and
comprehensiveness in the factor structure.
584
Factor I
Impact of species
Cigarettes
Scuba diving
Fireplaces
Earthquakes
Television
Recreation
Floods
Drought
Burning of waste
Travel
Climate change
Deforestation
Poaching
Nuclear war
Disregard rights
Belief in dominion
Loss of habitat
Animal loss
Plant loss
Wetland loss
Factor 2
Human benefits
-3.43
- 1.78
- 1.77
I .74
-1.67
- I .57
-1.29
- .95
p.87
- .86
-
34
1.oo
1.03
1.26
1.30
1.56
1.60
2.01
2.03
2.06
Cigarettes
CFC emissions
Conventional war
Ozone depletion
Aerosol cans
Nuclear war
Acid rain
Earthquakes
Floods
Poaching
Technology
Dams
Mass farming
Imgation
Urbanization
Urban water use
Automobiles
Travel
Housing
Recreation
-2.38
-1.60
-1.52
- 1.43
- 1.37
- 1.32
-1.13
-1.10
- 1.08
- 1.03
1.06
1.19
1.22
I .29
1.44
1.78
1.83
2.13
2.29
2.55
Collect souvenirs
Scuba diving
Golf courses
Poaching
Hunting
Fireplaces
Volcanos
Driftnet fishing
Dams
Meteors
Waste production
Monetary values
Urbanization
Technology
Population growth
Climate change
CFC emissions
Air pollution
Ozone depletion
Automobiles
~~~~
Factor 4
Avoidability
Factor 3
Impact on humans
Factor 5
Knowledge of impacts
-2.45
-2.29
-2.28
-1.97
-1.95
-1.66
-1.36
- 1.04
-1.03
- .89
Meteors
Volcanos
Earthquakes
Floods
Drought
Economic growth
Climate change
Population
Capitalism
Technology
-3.45
-2.92
-2.58
-2.57
-2.44
- .99
-.93
- .84
- .78
- .64
1.02
1.14
1.14
1.32
1.36
1.37
1.47
1.55
1.73
I .75
Hunting
Air conditioning
Poaching
Deforestation
Driftnet fishing
Untreated sewage
Clearcutting
Golf courses
Aerosol cans
Cigarettes
.85
36
.87
.92
I.05
I .09
I .09
1.26
1.52
1.55
Television
Biotechnology
Meteors
Scuba diving
Air conditioning
Fireplaces
Big business
Climate change
Monetary values
Disconnection
Hunting
Loss of habitat
Floods
Air pollution
Housing
Drought
Automobiles
Deforestation
Clearcutting
Earthquakes
- 1.84
- 1.74
-1.72
- 1.50
- 1.47
- 1.34
- 1.23
- 1.20
-1.20
-1.18
1.08
1.19
1.39
I .40
1.49
1.51
1.53
1.78
1.90
I .90
" The table entries are factor scores calculated using regression procedures, as described in the text.
585
2.50
1.50
Factor 2 score
Human benefits
-1.50
-2.50
-3.50
-3.50
*
-2.50
-1.50
-0.50
0.50
150
2.50
3.50
586
Factor 1 sco?
Impact on speaes
3.50
2.50
150
050
Factor 3 SCOW
Impact on
humans
-0.50
-1.50
etdNmm
-2 50
-3 50
-3.50
-2.50
-1 50
-0.50
--
0.50
1.50
2.50
3.50
587
588
REFERENCES
1. P. Slovic, Perception of Risk, Science 236, 28&285 (1987).
2. P. Slovic, Perception of Risk: Reflections on the Psychometric
Paradigm, in S. Krimsky and D. Golding (eds.), Social Theories
of Risk (New York, Praeger, 1992), pp. 117-152.
3. National Research Council (NRC), Improving Risk Communication (National Academy Press, Washington, D.C., 1989).