( .
' v

   

 

 

 

 

     

 

   

   

 

 

 

 

 

 

 

 

. . I .

 

W‘Es ‘
1e 0““ ER

’3

This is to certify that the

thesis entitled

An Examination of a Distance Decay Function
Relating Air Quality to
Power Plant Location

presented by

Christopher Paul Brown

has been accepted towards fuifillinent
of the requirements for

Masters—degree in _Geogr_aphy_

 
   

Major professor

Date M

0-7 639 MS U is an Affirmative Action/Equal Opportunity Institution

 

f _\

LEBRARY
Michigan State

University
w J

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

iii“ :—
i

MSU Is An Affirmative Action/Equal Opportunity Institution
cMMM.‘

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

An Examination of a Distance Decay Function
Relatin Air Quality to
Power lant Locatlon

By

Christopher Paul Brown

A Thesis

Submitted to
Michigan State University
in partial fulfillment of

MASTER OF ARTS

Department of Geography
1991

ft'.
1

. 41" 59’

I\
t1 v’

ABSTRACT
An Examination of a Distance Decay Function
Relating Air Quality to
Power Plant Location
by
Christopher Paul Brown

Fossil Fuel Combustion (FFC) plays a pre-eminent role in meeting domestic
US. energy needs; in doing so, it generates a wide range of negative externalities,
with significant effects on the structure of urban environments and residents who
live in urban areas.

This research uses a survey technique to examine the existence of a distance
decay function regarding how urban residents perceive the negative externalities
resulting from a large FFC power plant in the Lansing, Michigan area. The results
indicate that a distance decay function does act on perceptions of these externalities
and that certain socioeconomic variables -i.c., level of education, level of support
for environmental legislation, and level of expressed health concern also
significantly influence the relationship between distance to a FFC facility and

perceived negative externalities.

To my mother and father,
who gave me the talents necessary to
pursue a graduate degree.

ACKNOWLEDGEMENTS

I wish to thank Dr. Joanne Westphal for her continued support, guidance,
and patience in the preparation of this thesis, as well as my graduate career.
Thanks are also due the other members of my research committee, Dr. Bruce
Pigozzi, and Dr. Tom Edens, for their time and effort in this project.

An endeavor such as this would not have been possible without the
financial support of the Geography Department; my sincere gratitude goes to Dr.
Gary Manson and Dr. Judy Olson, for their extraordinary help in this area.

Several other people have provided invaluable assistance in this project.
Ellen White and Charlie Radar went out of their way in assisting me in the trials
of preparing graphic materials. Mike Lipscy provided me with the computer skills
needed to undertake this effort, as well as others, in my graduate career. The
support staff of the Geography Department, particularly, Marilyn Bria and
Carolyn Brookins-Williams, provided the necessary office support that often goes
overlooked.

I would also like to thank Cynthia Wheatley and Eleanor Love of the
Lansing Planning Department, who assisted me in selecting and mapping the
study area. Lastly, I wish to thank. the people living in Lansing who participated in
the survey used in this research, particularly the presidents of the neighborhood
associations involved; without them, this project would never have been

completed.

iv

TABLE OF CONTENTS

List of Tables vii
List of Figures viii
Chapter I. Introduction 1
The Role and General Effects of FFC 2
The Effects of FFC on Human Health 2
The Effects of FFC on the Natural Environment 5
Interactions Between Externalities and 6
Human Populations
Societal Implications of the SEF Conce t and 9
Locally Unwanted Land Uses (LULU’sg
Review of Other Pertinent Literature 10
Summary 22
Chapter H. Problem Statement-Hypotheses 24
Problem Statement 25
Formal Statement of Hypotheses 25
Justification of the Hypotheses 27
Chapter III. Methods 32
Evaluation of Environmental Quality 32
Survey Development 36
Survey Structure 39
Introduction of Survey 39
Survey Explanation 40
Bidding Exercise 40
Socioeconomic Data 42
Environmental Awareness and Support Queries 44
Enumerator Recorded Data 46
Selection of the Research Area 46
Pretest of the Survey Instrument 50
Sampling Strategy and Implementation 51
Data Analysis 54

Chapter IV. Results and Discussion

Results
The Main Research Question
Related Research Questions
Discussion
The Main Research Question
Related Questions

Chapter V. Summary and Conclusions
Sum of the Research
Overall ignificance of the Study
Recommendations for Future Research
Glossary
Appendix

Bibliography

58
58
58
59
65
68
68
69
72
75
78

89

LIST OF TABLES

Table Page
1. Relative importance of sources of domestic air pollution 3
2. Relative significance of major domestic air pollutants 5
3. Predictive equations relating mortality to sulfur oxide 5
' and particulate pollution
4. Size and rationale of NRC’s Emergency Planning Zones 12
5. Six largest lants in the Lansing area reviewed as possible 49

focus Sites or sampling areas

6. Coefficients, t values, R squared values and indications 60
of significance for the regression equations reported in the text

7. T test results comparing mean bids for certain sub-groups 64

vii

LIST OF FIGURES

Figure page

1. Path analysis of the air pollution-mortality model 3

2. General distance decay function 8

3. Spatial externality field concept, and its components 9

4. Concentrated costs and dispersed benefits associated with facilities 10
that generate negative externalities

5. Emergency planning zones 12

6. Distribution of ori 'n of evacuees during the TMI accident, 15
percentage of resi ents evacuating indicated in boxes

7. Destinations of TMI evacuees 16

8. Distance decay functions of facilities generating negative externalities 19

9. Distance decay functions of facilities generating negative externalities 21

10. Research hypothesis 26
11. Null hypothesis I. 27
12. Theory of Reasoned Action 37
13. Map of 5 potential power plants to serve as focus of study area 48
14. Sampling areas relative to power plant location 52
15. Various regression structures explored 57
16. The script used to guide the interviews 78
17. The consent form 80
18. 6 x 8 inch reduction of the original 8 x 10 inch color print of the 81

Eckert Station Plant

19. Reduction of the original map used to aid the respondents in
identifying the plant

20. Subject response form
21. NEPA question form
22. Coverletter

82

83
87
88

Wen:

The present means of meeting domestic US. energy needs generate many
different negative externalities, which are physically distributed over, and therefore
affect urban areas in a variety of ways. Individuals in these areas will perceive
these externalities and may react to them by altering their location in relation to
each other and the externalities in question. Through this process, the very
structure of the urban areas will reflect the effects of these externalities on the
individuals involved.

While it is beyond the scope of this research to directly monitor the
physical effects that various negative externalities have on environmental quality
or the direct effect that they have on the human condition, it is possible to study
the perceptions that people have concerning the value of air quality within the
structural context of an urban area. In fact, mounting population pressures and
competing interests that affect the quality of life of urban residents argue for this
type of research.

Therefore, this thesis examines the effects of fossil fuel combustion (FFC)
facilities on peoples’ perceptions of environmental quality, with a focus on the
spatial dynamics of these effects. The general organization of the thesis is as
follows.

The balance of Chapter One introduces the role that FFC has had in
meeting domestic energy needs and the results of this energy source on the human
and natural environment. Further discussion then moves to the interactions
between these externalities and human populations in a spatial context. With this
societal relevance in place, the literature dealing with past research into these

interactions is reviewed. Chapter Two presents a formal statement of the research

2

problem and hypotheses. Chapter Three provides a description of methods and the
statistical analysis that were part of the research effort. Lastly, the manuscript
closes with a presentation of the results as they relate to the formal statement of
the research problem and a discussion of the associated societal and theoretical

implications.
WM:

FFC plays a pre-eminent role in meeting domestic energy needs.
According to the Energy Information Administration’s Monthly Energy Review
(1990), FFC accounts for approximately 90% of domestic energy use. In fulfilling
this role, FFC generates a myriad of negative externalities with varied, significant
effects on both the human and natural environment.

Specifically, FFC is a major source of air pollution, generating a wide range
of pollutants .ncluding particulates, carbon dioxide, and sulfur oxides. An
examination of pollutants presented by Miller (1980), indicated that ”stationary
fuel combustion (primarily at fossil fuel power plants) emerges as the most
dangerous source" of air pollution, well ahead of both general industry and
transportation (Table 1).

W:

Over the past twenty years, a substantial amount of research has
established cause and effect relationships between air pollution generated by FFC
and subsequent human mortality rates. Lave and Seskin (1970, 1977) were among

the first researchers to examine the negative relationships between various air

3

pollutants and human health (Figure 1). In a 1970 article published in wise,
they conducted an extensive review of the literature and found that air pollution
had statistically significant effects (regression results were reported at the .05 level)
on mortality rates associated with bronchitis, lung cancer, nonrespiratory cancers,
and cardiovascular disease. Lave and Seskin (1970) concluded that ”there is an
important association between air pollution and various mortality and morbidity

rates”.

Table 1. Relative importance of sources of domestic air pollution (CEQ, 1975).

Scum AnnuaLEnissicns Belatimtleamfiflem

% of total Rank % of total Rank
Stationary Fuel Combustion 16.9 2 43.0 1
Industry 15.3 3 25.7 2
Transportation 54.5 1 22.2 3
Agricultural Burning 7.3 4 4.4 4
Solid waste Disposal 4.2 5 3.0 5
Miscellaneous 1 .8 6 1 .7 6
Total 100.0 100.0

Personal habits
Home inning
/ : Mortality rate

 

Air pollution

/

\\

Occupation mix
Wind speed
Precipitation

Figure 1. Path analyzis of the air pollution - mortality
model ( ve and Seskin, 1977).

4

Further research by Lave and Seskin (1977) began to address the issue of
quantitatixe improvements in human health that resulted from lowering ambient
levels of certain air pollutants. They estimated that a 58% decrease in suspended
particulates and an 88% decrease in sulfur oxides would bring about a 7%
decrease of air pollution induced mortality. In addition, their work reviewed the
general relationship between air pollution and human health in light of the nine
criteria that Hill (1965) developed to determine causality between environmental
influences and diseases. This review, when coupled with the association discussed
above, supports an argument for a causal link between air pollution and negative
effects on human health.

The prominence of snspendedpafliculates and sulmrgxides in these causal
relationships is evidenced in research by Lynn (1976), in which the mail: impacts
of certain pollutants on human health are examined. The occurrence of "killer"
smogs in the mid-twentieth centuries in Donora, Pennsylvania (1948) and London,
England (1952) highlighted the severe impact that these pollutants can have on
human health. Further research into these effects revealed that suspended
particulates and sulfur dioxides present the two greatest threats to human health of
any of the airborne pollutants (Table 2); only the combination of these two
pollutants can pose a greater threat.

Research presented by Lynn (1976), was one of the first attempts to
develop a predictive model that examined the relationship between certain
pollutant concentrations and the mortality rates of people in various age groups
(Table 3). This work further established the pre-eminent impact of suspended
particulates and sulfur dioxides on human health.

Table 2. Relative significance of major domestic air pollutants (Lynn, 1976).

Ellll E" El || Bll'E"
106 T/yr. % of total Wma % of total

Sulfur oxides 33.9 12.7 80 34.4
Particulates 26.1 9.8 75 28.2
Carbon Monoxide 148.6 55.8 10,000 1.2
Hydrocarbons 34.9 13.1 160 17.7
Oxides of Nitrogen 22.8 8.6 100 18.5

Totals . 266.3 100.0 100.0

Table 3. Predictive equations relatin mortality to sulfur oxide and particulate
pollution (Hodgson, 1 70, as cited in Lynn, 1976).

Am Erma
All ages R-H deaths - 150.5 + 7.7(COH) + 20.7(802) + 0.7(ternp)
over 65 R-l-l deaths a 101.8 + 5.3(COH) + 18.9(802) + 0.5(terrp)
45 to 65 R-H deaths :- 40.8 + 1.9(COH) + 0.9502 + 0.2(tenp)
under 45 Fl-H deaths - 8.0 + 0.5(COH) + 0.9(802) + 0.03(temp)
3.38351}; .. daily mortaity from respiratory and heart disease
COH - daily mean particulate pollution in coefficient at haze units
802 - daily average 802 concentration in parts per mlllon
term - departure at daily mean tenperature from 65' in degrees F

111W:

It is an established fact that FFC is causing an increase in the concentration
of carbon dioxide (C02) in the atmosphere. In one of the earliest studies into this
phenomenon, researchers found that C02 concentrations have risen approximately
18% from the late 1800’s to the 1970’s as a result of FFC and limestone kilning
(Keeling, et al, 1973). Further research into C02 levels in this century has revealed

an increase in atmospheric C02 concentrations of approximately 3.5% in the

6

period from 1959 to 1971; the results also confirmed the long term trend
mentioned above (Keeling, et al, 1976a and 1976b).

Although the effects of C02 on the natural environment are not as clearly
established as the effects of particulates and sulfur dioxides on human health,
there is nonetheless a large body of literature that suggests that increasing levels of
C02 can have harmful effects on global and regional environments via climate
change. Kellogg and Schware (1981) did an extensive review of a variety of climate
models .based on a wide range of assumptions concerning the complex oceanic and
atmospheric interactions between C02 levels and the global energy balance. The
vast majority of the models indicate that increasing levels of C02 and projected
energy use patterns (generating a doubling of C02 concentration) will result in an
increase of global temperatures ranging from 1.5 to 4.5 degrees Celsius over the
next century. The potential end results of this warming include disruption of food
production due to shifting regional climate zones and pest populations, poleward
spread of tropical disease via increasing vector habitat, and extensive flooding of
coastal areas due to rising sea levels (Kellogg, et al, 1981). While the certainty of
these results varies with the accuracy of the global temperature change predictions,
the impacts that they may have on the human and natural environment argue for

research into the underlying processes.

 

As cited in the literature, FFC plays a pre-eminent role in meeting
domestic energy needs and also generates negative externalities that have
significant effects on both human and natural environments. These externalities

interact differentially across the various populations of existing urban areas;

7

therefore, the actual end results of FF C on society is difficult to measure. One can
measure, however, the perceived effects of FFC on urban populations.

The current issue of where to site certain facilities (i.e., landfills, hazardous
waste facilities, power plants, nuclear facilities, etc.) that generate negative
externalities highlights the importance of spatial and lpcational variables in the
interactions between these facilities and people. The very processes by which
humans interact with their physical environment across space to yield the strpanre
of our urban areas are mirrored by the manner in which negative externalities are
perceived by people. Just as people may wish to locate in proximity to a market or
other amenity to reduce travel time, they may also perceive a facility that
generates negative externalities (e.g., a factory) differently according to the
proximity of this facility. Papageorgiou (1978) sees spatial externalities being
transmitted to the urban form by a process of diffusion as people perceive these
effects and interact with each other over space. He goes so far to suggest that
urban form is the prmpct of ”two interacting surfaces unfolding over the landscape
- a population surface and an externality surface”, with the result being a dynamic
distribution of population across space (Papageorgiou 1978).

The role that space and distance have in relating environmental
interactions with people underlies a considerable portion of the field of geography.
For example, over the years, geographers have demonstrated that a general
distance decay function exists, such that certain types of interactions diminish as
one moves farther away from the source of these interactions (Figure 2). This
distance decay function is explained by the notion of concrete space generating
"friction"; the phenomenon of increasing transport costs of goods and services as
one moves away from urban centers is one example of this concept (Foust and de

Souza, 1978).

Interaction
or
Activity

 

 

Distance
Figure 2. General distance decay function.

The specific issue of how space and distance affect the perception of
facilities that generate negative externalities has been documented in the past. In
research dealing with perceptions of community mental health facilities, Dear, et
a1 (1980) introduced the concept of a "spatial externality field" (SEF), which
describes the explicit manner in which distance to a given facility affects the
perception of the facility. According to this concept, the SEF has three dimensions:
intensity, extent, and rate of distance decay, as indicated in Figure 3. The intensity of
the SEF is a measure of the total impact of the externality; specifically, it can
equate to the total loss in property value due to the activity in question. The extent
of the SEF is the actual spatial dimension of the field, which is in turn determined
by the rate of distance decay. (This is the rate at which the perception of the
negative extemality associated with the facility in question diminishes over
distance). Hence, this concept very closely parallels the traditional view held by

geographers regarding the effect of distance on given interactions.

I?! Ill
9“ ‘u '- t ' t V 'I“

10

OS‘MNCE

f

 

 

 

DIST“! - r
MCAV
(“1'0”

 

 

 

" E2:

Figure 3. Spatial Externality Field, and its components (Dear et a1, 1980).

 

F
W51

 

. Dear and Taylor (1982) note that when externalities exist with
demonstrable consequences, the presence of these externalities (and by extension,
the facility that generates them) is a source for conflict. Davis (1984-1985)
reviewed the literature regarding procedures for the siting of hazardous waste
facilities; he found that policy formulations that required adequate siting decisions
often were seriously impeded by institutional barriers. Davis went so far as to
suggest that political obstacles resulting from public perception of such locally
unwanted land uses (LULU’s) may be the most serious barriers to reaching
adequate siting decisions. Davis (1984-1985) described this situation as one of

"concentrated costs and dispersed benefits” - i.e., the provision of the site in

10

question yields a collective good for a group of people dispersed spatially, while
those living in immediate proximity to the site receive the vast majority of the costs

(Figure 4).

 

Figure 4. Concentrated costs and dispersed benefits associated with facilities that
generate negative externalities (Getz and Walter, 1980).

It is clear that the manner in which people perceive these facilities, via the
externalities that they generate, has tremendous impact on the degree to which
they are accepted by society. Therefore, it is important to establish how people
perceive these facilities, and what implications these perceptions have for siting a

LULU in a given area.

With a basis for the societal relevance of this work in place, a more formal

review of the literature reveals an extensive amount of research dealing with a

11

variety of facilities that generate negative externalities. One of the earlier pieces of
research was a study by Odland and Balzer ( 1978) that examined the localized
processes that determined the spatial distribution of condemned housing. They
derived a predictive model, lagged across space, that predicted the occurrence of
new housing condemnations based on the current spatial distribution of existing
condemnations. Statistical analysis found that within a hexagonal cell of 400
meters, the interaction between existing and new condemnations was maximized.
Accordingly, the Odland and Balzer (1978) study provided empirical evidence of
the spillover effects surrounding the negative externality of being proximal to a
condemned house. In light of the effect of these buildings on the quality of life in
urban areas, this study was also an early attempt to examine the quality of life
dimension of perceptions of negative externalities.

A concept of a "zone of interaction" surrounding negative externalities was
the basis for a set of policy prescriptions developed by a joint task force of the
Nuclear Regulatory Commission and the Environmental Protection Agency
(NRC/EPA, 1978). As detailed in Figure 5 and Table 4, the NRC established two
zones surrounding a nuclear power plant, the plumppxppsurppaflmay (extending
from 0 to 10 miles from the plant) and the ingesjjpnpathway (extending from 10 to
50 miles from the plant). The rational for these zones was the view that effects
from even the most serious accident, a ”class 9 accident” (Le. a core meltdown),
would dramatically decline at a distance of 10 miles from the plant. Past this
distance, the risk to humans was felt to exist only through the ingestion of food and

drink products effected by fallout, not through direct exposure.

12

Table 4. Size and rational of Emergency Planning Zones (USNRC\EPA, 1978).

E El . 2 CT IQ IE EEZB l'
Pathway
Plume Exposure Pathway Whole body (external) approximately 10 miles
Thyroid (inhalation)
Other organs (inhalation)
Ingestion Pathway Thyroid, whole body. bone approximately 50 miles
marrow (ingestion)
an M1
§%PTNL\\
“It.

I

 

 

Figure 5. Emergency Planning Zones (NRC, 1980).

These zones in turn were the basis for further policy prescriptions involving
information dissemination within certain distances of nuclear plants. According to

the NRCS Public Information Planning Objectives (NRC/FEMA, 1980),

13

information about radiation, sheltering, and evacuation must be made available to
"the permanent and transient adult population within about 10 miles of the site".
This recommendation is clearly based on the results of the above mentioned
NRC/ EPA study. These two pieces of NRC policy are evidence that federal policy
formulation bodies acknowledged the existence of a distance decay function that
acts on the SEF’s surrounding large scale nuclear facilities.

The accident that occurred at the Three Mile Island (TMI) nuclear power
plant on March 29, 1978 and the ensuing crisis that followed offered researchers an
excellent opportunity to study another facet of negative externalities surrounding
nuclear plants. The policy that the NRC developed to handle emergencies like this
was based on the following assumptions (Ziegler and Johnson, 1989; Johnson and

Ziegler, 1983).

1) Adequate information would be available upon which public

authorities would base their emergency notifications.

2) This information would result in zones of prescribed activity based on

the two emergency zones described above.

3) People living within these zones would obey the official directives of

evacuating, sheltering, or continuing normal activities.

Actual behavior of people living in the vicinity of TMI ran counter to the
third assumption; namely, many more people evacuated than were instructed to do
so. According to the official notifications, approximately 3500 people should have

evacuated; in actuality, it has been estimated as many as 200,000 people living

14

within 25 miles of the plant evacuated (Johnson and Ziegler, 1986). This instance
of extreme human behavior has been termed the "evacuation shadow”
phenomenon (Ziegler, Brunn, and Johnson, 1981), and it was found to extend 25
miles from the plant (Johnson and Ziegler, 1983). The areal extent of this shadow
and the sheer number of people involved are excellent evidence of the dramatic
spatial dimension that can accompany human perceptions of facilities that
generate negative externalities.

Further work by this team of researchers explored the exact nature of the
spatial dimension of the evacuation shadow, as well as the effect that
socioeconomic factors had on this phenomenon. All of the studies found distance
to the plant to be the most significant variable in determining not only whether
people evacuated, but when they evacuated and how long they stayed away. As
opposed to people living farther away from the plant, people living closer to the

plant:

1) were much more likely to evacuate (Flynn, 1979) (Figure 6);

2) evacuated earlier and stayed away longer (Ziegler, Brunn, and
Johnson, 1981);

3) were more likely to view TMI as a very serious threat during the
emergency period (Flynn, 1979);

4) were more likely to have experienced a significant or severe episode of
stress during the emergency (Flynn, 1979); and

5) were more likely to have experienced disruption of normal household

activity (Flynn, 1979).

15

Clearly, peoples’ intentions and feelings, as well as their actions, reveal a distance

decay function regarding their perceptions of the externalities associated with the
accident at TMI.

 

 

 

Figure 6. Distribution of origin of evacuees during the TMI accident, percentage of
residents evacuating indicated in boxes (Flynn, 1979).
Direction of peoples’ evacuation destination relative to the plant was
another facet of the spatial dynamics of their evacuation behavior. As Figure 7

reveals, there is a strong directional bias that favored sites to the northwest of the

16

plant at TMI (Ziegler, et al, 1981). The authors hypothesized that this was due to a
combination of people desiring to be upwind from the plant and in the mountains,
an area felt to possess a sense of safety amidst the chaos and danger of the area
near the plant. In examining the location of peoples’ residence prim; to the
accident, one should note that there does not seem to be a directional bias present
(i.e. the same percentage of people living downwind versus upwind from the plant
evacuated) (Flynn, 1979). This author suggests that peoples’ perceptions of the
dangers associated with the accident were so severe that diregfipn to the plant was
far surpassed by manic to the plant as the primary motivator to evacuate.
Irrespective of direction to the plant, anyone living proximate to the plant

perceived a risk and acted accordingly.

 

  

W OF TMI EVACUEES

 

f
MUM“

 

 

 

 

 

 

Figure 7. Destinations of TMI evacuees (Ziegler et al, 1981).

Socioeconomic variables associated with the population around the plant

also were examined by the above mentioned researchers, and several variables

17

were found to have a significant effect on evacuation behavior and the perceptions
of the people living in the vicinity of the plant. The following groups of people
were mpsuikely to evacuate (Johnson and Ziegler, 1986):

1) families with a head of household under 35 years of age and having
children;

2) families with a head of household having more than 12 years of
education;

3) families with a head of household having a higher a priori opposition to
nuclear energy; and/ or

4) families with a head of household having a higher apn'on' perception of

the risk of a nuclear accident.

Whereas all of the above follow from logic and therefore are not surprising,
Johnson and Ziegler (1986) did npt find any significant relationship between
married families with a pregnant woman in the household and tendency to
evacuate; one would have expected this relationship to follow from item number
one above.

Related to the above research was another set of studies that examined
peoples’ intentions to evacuate given the mzppthejicfl scenario of a TMI class
accident occurring at the Shoreham nuclear power plant on Long Island, New
York (Johnson and Ziegler, 1983, 1984, 1989; Johnson, 1985). For the most part,
this body of literature supports the findings of the TMI research, with distance and
direction to the facility, place in life cycle, education, and a priori views of nuclear
energy and its associated risks having similar effects on peoples’ intention to

evacuate. Of particular interest, however, is a study by Johnson and Ziegler (1983)

18

that found people living nearer, to a plant are actually morelikely to underseaet
and W in the event of a protectory advisory. This result hints at the
possibility that either there is some type of hypothetical bias to the instrument or
people close to the plant had become acclimated to the effects of the plant. Both
of these possibilities will be addressed later in this manuscript.

Clearly, the literature dealing with peoples’ perceptions of the TMI and
Shoreham nuclear plants reveals a distance decay function that acts upon the
SEFs associated with these plants. This relationship is also influenced by various
socioeconomic factors including peoples’ age, place in lifecycle, education, and a
priori perceptions of an energy source and its attendant risks.

In 1980, as part of a general public opinion poll regarding the environment,
the Council on Environmental Quality (CEQ) queried people regarding their
views about living proximate to certain land use facilities (CEQ, 1980). The results
of this portion of the survey are depicted in Figure 8; this graph relates the
cumulative percentage of those surveyed to the distance at which they would be
willing to accept these facilities. Examination of this diagram indicates several
items of interest regarding the spatial dimension of peoples’ perceptions to these
facilities:

1) peoples’ perceptions of these facilities vary across type, as indicated by
the distinct curves for each of the facilities;

2) acceptance of these facilities increases as the distance to them
increases, as indicated by the positive slopes of the curves; and

3) the rate of change of peoples’ acceptance itself varies across facilities,

as revealed by the different shapes of the curves.

19

 

 

 

 

l l l 1 1 L4 1 l

 

. J _l l L i A
Vauldn't I... I 2 3 4-5 84 III- 15- 10- 30' 40 IO 61- 1” 101+ Don't
matter than M I! 2a 3! 4! C mt
can way an at any
at to ale din-tea
a“ lb

Figure 8. Distance decay function of facilities generating
negative externalities (CEQ, 1980).

20

These results not only support the existence of a distance decay function regarding
peoples’ perceptions of facilities that generate negative externalities, but the
results also provide empirical evidence that this function performs differently when
perceptions of different facilities are involved.

This basic concept was expanded upon by Furuseth (1989), who examined
community acceptance of a hazardous waste facility (HWF) located near a
residential neighborhood of Charlotte, N.C.. Distance to the facility was the most
significant factor in explaining acceptance of it, accounting for 54% of the
variance. Of several socioeconomic variables queried, only level of education was
found to have a significant effect on acceptance of the facility. While not verified
statistically, Furuseth (1989) contended that the tenure of this facility and its
benign appearance were responsible for the high level of community acceptance.
As indicated in Figure 9, these factors may have effected the acceptance curve of
this facility; from the figure, one can see that the curve for the facility lay
somewhere between the acceptance curves of other HWF’s and those of coal fired
power plants and large factories. The author closes with the contention that
optimal siting decisions require policy makers to be aware of how the public

perceives these facilities, and what factors influence these perceptions.

Percent

21

 

 

 

 

 

Figure 9. Distance decay functions of facilities enerating
negative externalities (Furuseth, 1989 .

22

51mm:

Certain large scale energy conversion and industrial facilities generate a
wide range of negative externalities, which in turn have a pronounced effect on the
human and natural environment, and the resultant quality of life for urban
residents. FFC facilities are one such type of facility instrumental in satisfying
domestic energy needs, and they have their own set of externalities and impacts.
Nonetheless, certain parallels can be drawn between FFC facilities and other
facilities that generate negative externalities. The perception of these externalities
by the public can significantly influence the acceptance of these facilities, as well as
the ability to site future facilities.

A review of the literature supports the concept of a distance decay function
acting on the SEF’s of these facilities, such that the degree to which people
perceive the negative externalities declines as the distance to the facilities
increases. This distance decay function is one facet of the interaction of the
population surface and the externality surface that has been hypothesized to
determine the structure of the urban form (Papageorgiou, 1978). Congestion and
both stationary and mobile sources of pollution have been shown to influence the
degree of dispersal and concentration of populations within urban areas
(Papageorgiou, 1978). In addition, the manner in which this interaction occurs over
space may also determine the degree of acceptance of certain facilities (F uruseth,
1989).

The above review of the literature and resulting discussion have attempted
to link the issue of peoples’ perceptions of facilities that generate negative
externalities to both a theoretical relevance in regards to classic distance decay

relationships, and the related societal significance that these perceptions can have

23

on acceptance and siting of these facilities in contemporary society. With this
linkage established, the discussion will now turn to the problem statement and

formal statement of hypotheses.

W:

Although the research cited supports the existence of a distance decay
function regarding peoples’ perceptions of the externalities that certain land uses
generate, the literature is not without exception on the direction of this
relationship. As noted earlier, Johnson and Ziegler (1983) found that people living
in elpse proximity to a nuclear power plant actually would have W to an
evacuation advisory; this situation brings up the possibility that they may have
acclimated to the risks and negative externalities that the plant presented. One
might argue, therefore, that the relationship between distance to a facility and the
perception of the attendant negative externalities may be a positive one.

With the exception of the CEO (1980) research that compared peoples’
perceptions of a variety of types of facilities, little research has been done
regarding how peoples’ perceptions of EB; faeilujes is effected by distance to
these facilities. Several factors at present make this issue a matter of renewed
importance.

First, legislation regarding clean air is pending in several legislative bodies
at the federal, state, and regional level. Second, numerous municipalities are
wrestling with siting issues concerning where to site waste to energy conversion
facilities (solid waste incinerators). Presently, municipalities in and around
Lansing, Michigan are involved with pending legislation and permits concerning a
proposed incinerator, while the corporation that is operating a large, existing
facility in the Detroit, Michigan area is wrestling with the denial of a permit from
the Michigan Air Pollution Control Commission to continue operations.
Therefore, an incentive exists to study the perceptions of individuals that live

adjacent to these types of facilities; as FFC facilities are both directly involved in

24

25

the legislation and have similar effects as waste to energy conversion facilities,

research into the perceptions specific to these facilities is warranted.

Emhlcmfitatcmem:

This research will examine the relationship between distance to a FF C
facility and the perception that people living downwind from the facility have
concerning the negative externalities that the plant is generating. Specifically, what
is the relationship between the distance at which people live from a point source of
air pollution (namely, a FFC power plant) and the perceived evaluation of air
quality at the location of their residence? Two possibilities for this relationship

exist.

1) Is the relationship a positive one, indicating some sortof assimilation
process, whereby the value that peoplehold for air quality increases as
the distance to theplant Lngeases?

2) Is the relationship a negative one, corresponding to a classic distance

decay function, whereby the value that people hold for air quality
declines as the distance to the plant ipqeases?

W:

A tWO-tfiilCd research hypothesis is proposed corresponding to the two

facets of the problem statement mentioned; a null hypothesis is included, which

26

incorporates the possibility that no relationship exists between distance to a FFC
facility and peoples’ perceived evaluation of air quality.

W13 Evaluation of air quality at sites downwind
from a FFC will be negatively related to distance to this facility. Evaluation of air
quality will be at a maximum very close to the facility and will degease as the
distance to the facility inezeases (Figure 10a).

W15: Evaluation of air quality at sites downwind
from a FFC will be pnsitjxely related to distance to this facility. Evaluation of air
quality will be at a minimum very close to the facility and will inerease as the
distance to the facility inepeases (Figure 10b).

W Evaluation of air quality at sites downwind from a
F C will have no relation with distance to this facility. Evaluation of air quality
will be either constant across distance to the facility, or it will be distributed

randomly in relation to distance to the facility (Figures 11a and 11b).

V110 V110

OR

 

 

 

 

Distance Distance

VAQ = -f (Distance) VAQ -.-.- «14‘ (Distance)

Figure 10. Research hypothesis.

27

 

 

 

 

 

VAo VAQ
x x x
x
x w y X x xxx & x
x X X X
X x x xx x X x x Xx
X X x x
a) 9)
Distance Distance

Figure 11. Null hypothesis.
I 'E . [l H l :

Research hypothesis HIA is supported by the literature cited in this

manuscript, including:

1) the work by Odland and Balzer (1978) that examined housing
condemnations;

2) the study by the CEQ (1980) that examined public acceptance of certain
land use facilities;

3) the research by Dear, et a1 (1980; 1982) that examined community
acceptance of mental health care facilities;

4) the work by the team of Johnson and Ziegler (1984; 1986; 1989) that
examined evacuation behavior around nuclear power plants; and

5) the study by Furuseth (1989) that examined community
acceptance of HWF’s.

28

This hypothesis also follows from the generally accepted principle that a
good in scarcity (such as clean air near a FFC facility) will be valued highly at this
location; as the distance to the facility increases, the effect of the facility on air
quality will decrease, relieving the scarcity of clean air, and causing the valuation
of air quality to decline.

Research hypothesis HlB originated from two pieces of literature that
suggested an assimilation effect when individuals are proximal to a facility that
generates negative externalities. As previously mentioned, Johnson and Ziegler
(1983) found that people living proximate to a nuclear facility were more likely to
underreact to an evacuation advisory than those who live at a greater distance
from a plant. While the possibility of some sort of hypothetical bias of the
instrument cannot be discounted, this finding points to the possibility of an
assimilation effect existing for those people living proximal to the plant.

In a similar vein, a study by Vleeming (1985) examined the possibility of
people living near a nuclear plant becoming acclimated to the risks associated with
the plant. During the early 1980’s, protest behavior concerning the Zeeland,
Netherlands plant increased markedly, and Vleeming (1985) hypothesized that the
protesters were actually ”outsiders” or people that lived at a greater distance from
the plant. In addition to this somewhat political premise, the hypothesis concerning
an assimilation effect also had, as an underlying basis, Festinger’s Theory of
Cognitive Dissonance (1957).

According to this theory, individuals strive for a sense of internal
consistency; when an external source of dissonance or inconsistency arises, an

individual will undertake various options to relieve the dissonance. These include:

29

1) physically removing the source of dissonance or relocating hirn/ herself
from the dissonant environment,

2) changing the nature of the source of dissonance to eliminate the
inconsistency, or

3) changing their belief system to eliminate the inconsistency.

In the case of the Zeeland nuclear plant, these three options translated to

the following possible actions:

1) individuals relocating their residence far enough away from the plant to
eliminate it from their practical environment,

2) individuals protesting to shut the plant down, thereby changing its nature
enough to eliminate the dissonance,

3) individuals changing their belief that the plant poses a risk, via some
process of rationalizing or becoming acclimated to the risks, or

4) a combination of the above three actions.

While the results of Vlemming’s (1985) study did not verify the existence of
a ”dissonance” driving an assimilation effect, the underlying premises to this
possibility seem rather logical. Many people who have lived adjacent to a
prominent source of noise (i.e., an airport or rail facility) can attest to the
phenomenon of ”not even hearing the noise after a while”.

These two studies point to the real possibility that people living proximal to
facilities that generate negative externalities may become assimilated to them.
Regarding the specific questions of this study, this translates to the possibility that
people living proximal to FFC facilities may become acclimated to the effects of

30

these plants on air quality. This in turn drives the second research hypothesis, HlB
as follows. People living in very close proximity to a plant will be very acclimated
to the effects and therefore, will not value the absence of these effects (i.e., clean
air) very highly. As the distance to a FFC facility increases, this assimilation will
diminish and the evaluation of air quality will rise accordingly.

In addition to the assimilation concept, another factor may be contributing
to the possibility that perceived evaluation of air quality is positively related to
distance of a FFC facility. It is a commonly known fact that areas either downwind
or highly proximal to facilities generating negative externalities tend to be areas
with lower incomes and\or lower valued housing stock. The existence of a positive
relationship between perceived evaluation of air quality and distance may argue
for the existence of a "bid rent function for pollution" (Pigozzi, 1990).

As distance from a FF C facility increases, incomes increase and people
may be willing and able to spend this income on higher quality housing stock, with
cleaner air. People living farther away from FFC facilities may actually have been
able and willing to "bid" to live at this location. This concept, as suggested by
Pigozzi (1990) could be viewed as the inverse of the established concept of a bid
rent function, as originated by von Thunen and cited by Losch (1954). While not to
be empirigally tested, this concept was used to formulate the second research
hypothesis.

The null hypothesis is proposed strictly as an alternative to the two
research hypotheses to fit the traditional hypothesis testing framework. It is
anticipated that the null hypothesis will not be borne out by the data. The effects
of FFC facilities and the increasing public awareness of the entire issue of negative
externalities of various facilities argue that distance to FFC facilities will have

same effect on peoples’ perceptions of these externalities. Further conjecture is

31

posited as to the direction of this relationship; while the possibility of an
assimilation effect has theoretical merit, it is anticipated that the negative
relationship between distance to FFC facilities and the perception of the
evaluation of air quality will be born out by the data. It is suggested that any
assimilation effect will be overridden by the dynamics underlying the phenomenon
of a distance decay function. With the formal statement of the research problem
and hypotheses in place, discussion now turns to the methods used in the design

and implementation of the research.

El'fE' 10].:

In order to address the research problem and test the above mentioned
hypotheses, a method was needed to assess the value that people have for air
quality. Environmental economists have wrestled with the problem of assigning
economic values to the nonmarket good of environmental quality for years, with
varying degrees of success. A brief review of these methods will be presented after
Randall, et a1 (1974) in order to develop justification for the method that this
author chose for this research. Then, a more formal description of the means that
were used in this research will be undertaken.

Three general classes of methods have been developed to ascribe economic
value to the class of nonmarket goods that comprise environmental quality. These
are direct cost techniques, revealed demand techniques, and contingent valuation
(CV) techniques.

The direct cost techniques simply attempt to aggregate the total costs that a
loss or degradation of environmental quality would entail. Randall, et a1 (1974)
presented a marginal value of damage avoided by abatement (MVDA) curve; this
curve estimated the value that given levels of abatement of water pollution would
yield. Failure to include all relevant costs and difficulty in obtaining certain types
of information may lead to problems with this technique.

Lave and Seskin (1970, 1977) also explored this general technique with
their attempts to find the economic value that abatement-generated improvements
to human health would provide. While their research demonstrated considerable

utility for this method in approximating the optimal level of pollution abatement

32

33

for both stationary and mobile sources, two major shortcomings were evident.
First, this method failed to capture the aesthetic and quality of life aspects of the
damages of air pollution. A second, related problem was the same problem that
Kneese and Bower (1972) encountered - i.e., the inability to include all of the
relevant costs of the abatement in the analysis; this problem resulted in the total
value of the abatement being underestimated. The authors conclude "the relevant
measure is what peOple would be willing to pay (WTP) to reduce mortality and
morbidity" (Lave and Seskin, 1970). This comment suggests the concept of
attempting to internalize these costs in some sort of market mechanism.

The revealed demand technique seeks a suitable proxy for the value of
environmental quality. Much of the work in this area has used property values as a
surrogate for differing levels of air quality in urban areas. One study presented by
Nourse (1967) examined differences in property values and levels of sulfur trioxide
in St. Louis via a regression framework; the results indicated a strong, positive
relationship between the two variables. Another study by Nourse (1967) examined
this question in the same geographic area; while these results were promising, they
were not statistically significant at the .05 level. He closed with a note of caution
regarding using this technique in other areas without modification, due to the
confounding influences of differences in housing stock, income distributions, and
individual preferences of housing characteristics.

Wieand (1973) reviewed this research more critically and found two major
problems with this type of analysis. First, he suggested that the use of total
expenditures on housing as a proxy for the value of housing does not include all
relevant factors. Accordingly, the regression equation is not completely specified.
This fact leads to another problem with the analysis. While the premises -
underlying the technique seem to be valid, Wieand proposes the technique lacks

34

necessary statistical rigor due to the fact that the coefficients are not statistit‘afly
significant at the .05 level.

The third technique used in the past to estimate the economic value of
differing levels of environmental quality is the contingent valuation (CV)
‘ technique. This method is based on the idea mentioned briefly in the close of the
discussion of Lave and Seskin’s work (1970), that of internalizing the costs of
environmental damage in a market-like mechanism.

The CV method is based on the concept of bidding games that query
people about:

1) how much they would be willing to pay (WTP) to receive or retain a
given level of environmental quality, or
2) how much money they would need to receive to accept a given level of

environmental degradation (their willingness to accept or WTA).

The literature proposes that this method yields a hypothetical market that serves
as a viable surrogate to an actual market, but the hypothetical market that results
is not without its possible problems.

Several biases may result from the ‘creation of the hypothetical market that
CV seeks to establish (Schulze, et al, 1981). These include:

1) mategiejjas - the result of the subject giving false bids in order
to further a personal view in a strategic manner;

2) infpmatipnjzias - the result of the information given in the CV

scenario differing significantly from that which would exist in actuality;

35

3) instmmentjias - the result of a bias being introduced via either
the payment vehicle or the starting point of the bidding game;

4) hypnthetieaihias - the result of the CV scenario not being
believable enough to the respondent to elicit a valid bid; and/ or

5) WWW - biases common to this

and all other survey research.

In addition, research by Knetsch and Sinden (1983) challenged the notion of WTA
being approximately equal to WTP and called into question which of these
measures was best to use in this type of research.

Further review of the literature concerning CV addresses the majority of
these concerns. Thayer (1981) examined the issues of information, starting point,
and hypothetical biases via a survey instrument that queried people regarding the
value they held for preserving the aesthetic quality of areas in the Sante Fe
National Forest in New Mexico. His results indicated that ”in cases in which the
commodity is well-defined and the questionnaire requires routine behavior,
starting point bias does not exist" (Thayer, 1981). Via a comparison of the bidding
game results with those of a site substitution exercise, he also was able to discount
the possibility of information and hypothetical biases. Thayer (1981) closes with
the comment that "the survey procedure can provide accurate estimates of the
individual and aggregate welfare losses associated with environmental
degradation".

The issue of strategic bias being introduced into CV scenarios was explored
in a study done by Brookshire, et a1 (1976) that examined peoples’ evaluation of
aesthetic qualities being impacted by the Kaiparowits power plant near Lake
Powell in the southwestern United States. The results of this research indicate

36

strategic bias was not a significant problem, as evidenced by a low percentage of
either zero bids or extremely high bids. In addition, Brookshire, et al’s (1976)
research examined the question of whether WTA and WTP provide equal
measures of the welfare loss associated with environmental degradation; the
results support the theoretical notion that these measures are roughly equivalent.
A study by Coursey, et al, (1984) expanded on this last issue, specifically
exploring which of the two measures, WTP or WTA, was a more accurate

predictor of actual values. Their results indicate that:

1) WTA and WTP do not differ significantly; and
2) hypothetical measures of value obtained by WTP (similar to those

obtained via a CV scenario) may be more accurate predictors of actual

values than WTA.

In light of these findings, this author concluded that the CV technique is
the most accurate means of estimating the value of a nonmarket good like air
quality. Based on this view, this author chose the CV method as the means by
which to gauge the value that people hold for the quality of air in areas downwind
from a FFC facility. With this determination having been reached, discussion now

turns to the development of the survey instrument used in this research.

W:

Given the above, a survey was developed that sought to capture the
maximum willingness to pay (WTP) of a respondent to effect an improvement in

air quality at sites downwind from a large FFC facility in the Lansing, Michigan

37

area. WI‘P was chosen as the measure to use based on the above arguments by
Brookshire, et al (1976) and Coursey, et a1 (1984) concerning the ability of this
measure to most accurately approximate the actual value of this good. In addition,
the survey was developed according to certain principles and guidelines, as
outlined below.

The first principle focus’s on the general question of the validity of surveys.
The Theory of Reasoned Action (Ajzen and F ishbein, 1980) states that a person’s
behavior is determined by the interaction of their a prion“ attitude with other
peoples’ subjective norm; this in turn forms their intention to act in the future, as
depicted in Figure 12. Both this research and a study by Bowman and Fishbein
(1978) examining public reaction to energy proposals, concluded. that peeples’
intention to act is an immediate determinant of their action. While the relationship
does not argue for a perfect ability to predict behavior, it dpes describe a regularity
that can be used in the development of survey instruments that seek to determine

peoples’ future actions.

 

 

 

 

 

 

olltfilldnallnd wmr‘ Inhabit I HI I

ghouldmgt'iow 01 7
or n
orrnlhebehavlorand‘ Wm

 

 

 

 

 

 

 

 

Figure 12. Theory of Reasoned Action (Ajzen and Fishbein, 1980).

38

With the question of general survey validity in place, a survey was
developed according to guidelines initially proposed by Crespi (1971) and used
extensively by Randall, et al (1974). In order to obtain expressions of intent that
are valid predictors of actual actions, the survey instrument must convey certain

information.

1) The situation portrayed in the survey must be highly
institutionalized or routine.

2) The situation portrayed in the survey must be likely.

3) The situation portrayed in the survey must require a low level of
abstraction to visualize (i.e., it must be a concrete situation).

4) The behavior that is queried must be a specific one (i.e. in order to
determine if a person would pay a pollution abatement tax, the
questions in the survey must focus on that pantigiiar tax, not the general

issue of taxation).

The first three of these conditions assures that the situation portrayed is believable
and reduces the likelihood that a hypothetical bias may be introduced; the last
condition assures that the instrument is obtaining data that is valid in the sense of
providing responses about the smeifie action being examined.

The last general concern to be addressed in the design of the survey
instrument is the ”free rider" phenomenon, which arises when an individual
believes that he / she may be able to receive the benefits of a publicly provided
good without paying for it. Individuals who may receive the benefits of public
services without paying their share of the tax burden are one example of this

phenomenon. While previous research into the existence of this phenomenon

39

suggests that perhaps only economists freeride (Marwell and Ames, 1981), this
author made explicit efforts to insure that this potential problem did not surface in
the administration of the survey.

W:

The survey instrument was composed of several parts, the first of these
being a script (Figure 16 in the Appendix) that the interviewer used to guide the
interviews in a consistent manner, thereby avoiding any interviewer or sampling

bias.

Immdmimmfthdurxext

The script began with a very general introduction that:

1) introduced the general purpose of the survey, that of querying peoples’
views of power plants and air pollution;

2) informed them of the hypothetical nature of the survey;

3) disclaimed any involvement of the power company in the research;

4) assured them that their responses would be treated with the strictest
of confidentiality; and

5) advised them of the need to sign a consent form (Figure 17 in the
Appendix), which stated that their participation in the survey was

completely voluntary.

40

Smfimlanaticn:

After dealing with the consent form, the script moved onto the more
detailed explanation of the survey. The wording of this portion of the survey was
written with a concerted effort on behalf of the researcher to provide the subjects
with enough information to participate without biasing their responses.
Specifically, the explanation portion of the survey was designed to address the

following concerns.

1) It identified the plant by mentioning a popular set of nicknames and
presenting the subjects with both an 8 X 10 color print of the plant and a
map of the Lansing area that indicated the location of the plant (Figures
18 and 19 in the Appendix).

2) It presented them with an unbiased list of the advantages and
disadvantages of the plant.

3) It advised the subjects that they would be asked some questions about
how they valued certain trade-offs that resulted from these advantages
and disadvantages. This provided a lead in to the actual bidding

portion of the survey.

B'li' E ':

In order to solicit the bids from the subjects, they were asked how much
they would be willing to pay (WTP) in order to eliminate all of the previously
mentioned problems of the plant. Two aspects of this solicitation of the bids are of

particular note. First, the payment vehicle chosen to solicit the bids was an

41

increase in their monthly utility bills, which was believed to be a very believable
and realistic payment vehicle. Second, mention was made that under the
hypothetical scenario, all people in the area would be paying the increased rates,
not just the individual being interviewed. It was believed that this adequately
addressed the possibility of a free rider phenomenon existing.

Once the subjects were advised of the rules of the bidding game, the bids
were solicited in an iterative manner, using a method employed in previous studies
(Randall, et al, 1974; Brookshire, et al, 1976; Thayer, 1981). The choice of the
iterative method also was based on the finding that it resulted in an approximation
of WTP that most closely approached WTA (Coursey, et al, 1984). A one dollar
per month rate increase was chosen as the starting point since this seemed to be
the lowest reasonable amount to suggest as a rate increase. The subjects were
asked if they would be willing to pay this increased rate if all of the possible
problems associated with the plant could be eliminated. From this point, the bid
was increased until the subject gave a ”no" response. The bid that elicited the last
"yes” response became the bid of record, and this was recorded on the subject
response form (Figure 20 in the Appendix). All subsequent responses and data
were also recorded on this form.

Subjects were then queried as to which of the possible concerns of the plant
mentioned previously were included in their decision process, and these were
ranked from most to least important on the form. Subjects then were asked if their
bid would differ if the plant in the scenario was a waste to energy conversion
facility; this question was posed to determine if people perceived the negative
externalities of this type of facility any differently than traditional FFC facilities.
The last part of the bidding exercise queried any individuals who gave zero bids as

42

to the reason for their bid. This was done to determine if the. zero bid was either a

protest bid or a bid exhibiting strategic behavior.

5' '12:

The survey then solicited socioeconomic data from the subjects for two
reasons. First, this would facilitate comparison with previous research into the
spatial and nonspatial nature of facilities that generate negative externalities (the
vast majority of this research gathered this type of data). Second, as discussed
previously, Papageorgiou (1978) and Dear, et al, (1980, 1982) proposed that the
interaction of the externality surface and the population surface determined the
societal relevance of the externality, perhaps even to the point of determining the
urban structure itself. Gathering this data would therefore allow this researcher to
explore these interactions at length.

Specifically, the following types of data were solicited from the subjects:

1) tenure at their current and previous addresses;

2) number of children, adults and senior citizens in their household;
3) marital status of the subject;

4) level of education that the subject had attained;

5) occupation of the subject;

6) age of the subject; and

7) approximate family income, and whether it was a single or dual income.

These variables were chosen on the basis of previous research, as well as

intuitive notions of this author, as to how different groups of individuals would

43

perceive the negative externalities that result from being proximal to a FF C
facility. The following outcomes were expected as a result of the research (ceten's

pan'bus, or other things being equal):

1) People living at their residence for longer periods of time would be
more acclimated to the effects and would therefore perceive the
negative externalities to a lesser degree, resulting in lower bids.

2) Married people, those with more children, or seniors in the household
would be more sensitive to health concerns, and therefore, would
generate higher bids.

3) People having attained a higher education would be more aware of the
effects associated with the plant and therefore would generate
higher bids.

4) People with occupations in the industrial sector would be more
acclimated to similar effects in their workplace, and therefore, would
generate lower bids.

5) People with jobs outside the industrial sector would have an opposite
situation and generate higher bids.

6) Higher income families would be more sensitive to the effects of the
FFC facility and therefore, would generate higher bids.

While the basis’ for these expectations are not in the form of formal
hypotheses, they were still of interest to the author; therefore, data were gathered

in order to determine if the expectations would be born out in the analyses.

 

The last type of data that was gathered in the interview concerned the
subjects’ level of awareness of legislation designed to protect the environment and
the support of the respondents for such legislation. The literature indicates that
this effect may exist, and this author also had an interest regarding the effect these
variables have on the distance decay function. While these concepts may seem to
be born out of common sense, mention is made of them, and the literature dealing
with them, in order to justify the awareness and support questions included in the
survey.

Van der Plight (1984) examined the factors that influence the acceptance
of nuclear energy facilities and found that a priori views of this energy source
affected the degree to which people would accept the siting of a facility in close
proximity to their residence. This has been termed the NTMBY (Not In My Back
Yard) syndrome, a common description of this perceptual phenomenon (Furuseth,
1989).

Levenson (1974) examined the effect of a priori membership in anti-
pollution organizations on individuals’ perceptions of pollution. Anti-pollution
organization members in general felt that pollution was a greater problem and had
more negative consequences. In addition, they were more active in their expression
of these views.

Sundstrum, et al, (1977) examined the factors that contributed to
acceptance of a proposed nuclear power plant and found that two dimensions
acted on this acceptability, an economic dimension and a hazards/ disruption

dimension. As one would expect, individuals scoring high on the economic

45

dimension were more prone to favor the plant, whereas those high on the
hazards / disruption dimension opposed the plant.

The above mentioned research and this author’s intuitive interest in these
issues drove the formulation of the portion of the survey that dealt with the
Statement of Purpose of the National Environmental Policy Act (NEPA), as
depicted in Figure 21 in the Appendix. In addition, this measure allowed this
researcher to standardize responses within a population of respondents. This item
includes a paraphrased version of the formal statement of purpose of this law

(Public Law 91-190, 1970) as follows. This law seeks:

1) to encourage harmony between humans and the environment,
2) to prevent or eliminate damage to the environment, and

3) to enrich the understanding of the environment.

After this statement are two questions that query the subjects about their a
priori awareness of this law and their support for the 'mtent of this law. A concerted
effort was made to focus the subjects’ attention on the specific issues of, being
mm and in snpmnef, the intent of this law. Perhaps more importantly, care
was taken to avoid confounding their response with any feelings on the genetal
issue of taxation, or how the funding for this law was to be obtained. This author
felt this manner of structuring the question conformed to the guidelines that
Crespi (1972) and Brookshire, et al (1974) proposed for valid questionnaire design.

An additional aspect of this item worth noting is the use of a linear
response scale with the instructions to have the subjects place a mark on the scale

to indicate their response. This method yielded a variable that was truly

46

interval /ratio in nature, and one that could be treated with greater statistical rigor
than traditional Iikert scale questions.

Wu:

The last page of the subject response form allowed the interviewer to
record the sex of the subject, the type of housing unit, the time of day and day of
the week that the interview was performed, and whether or not the subject had
signed the consent form. The first two variables were gathered to determine if
there was an effect of the subject’s sex or type of housing unit on the relationship
between distance to the FF C facility and perception of attendant negative
externalities. The time of day and day of the week variables were gathered to
check for a temporal bias to the data, and the consent form item allowed
verification that this procedure had been followed, if the need to provide

verification arose at a later date.
Wm

Several criteria were included in the process of selecting the study area
where the survey was to administered. The FFC facility used as the focus of the

survey and the surrounding area had to satisfy the following criteria.

1) The FFC facility involved had to generate a large enough output of
pollutants to generate externalities of a nature to be perceived by the

people living at different proximities to it.

47

2) The FFC facility had to be physically large enough to facilitate
identification by the subjects in the study.

3) The FFC facility had to be located in an area that had an adequate
number of people living downwind of it to supply the required

subject population.

The Air Quality Division of the Michigan Department of Natural
Resources was contacted to determine which site or sites fulfilled these criteria. Six
possible plants were found to exist in the Lansing Tri-county area, and these are
listed in Table 5 and depicted on the map in Figure 13. Inspection of the spatial
distribution of these plants and their corresponding output figures revealed that
the Board of Water and Light’s Eckert Station Plant suited the site selection
criteria the best. It generated the largest pollutant output of the six sources, was
the most visible, was the most identifiable (even being known to Lansing residents
by the nicknames of "Winkin’, Blinkin’, and Nod" and the Three Sisters”), and had
the requisite number of people living proximate to it to generate an adequate

survey sample size.

48

To Detroit

.33 >37. Lo £68 an 023. 2 853 33o; 3383 m «o 9&2 .2 eSmE

Evitfi eh

 

 

l

 

 

 

 

>462

'9'“

NICO

   

oe
,. r.?f'f“f7f"'.‘_'e'_'"""

"‘c'

 

 

' to Detroit

a. 8 x

\«o
'
.n
Pu
h

 

 

 

 

 

 

 

ill
.. . , . . .
. ., .7 .
. . . . .. e u . . .. . , ,. . ...
.u. .. ...:
~ . .la . i . e. a
......q a , . . . . . .. . ...
. . . . .n ..n , ... . . . t . u .. . m.
h u.. ... . y . ... I .a 1. tr 1.. ....
. .2 .. .. \ , . a \. . s
a . o . . . o. ... ...
. . r . . ... . . . .. ... . _.
. .4 ... . .. .. . ... .. .. . c ....2.... a
.. ..r.. ..,:.,., . . , ... e . ,.....
.q: . .v .... . . . . ,
. . 1.. . ...s ... . C .. .
T . «3,... . 3 a . .. . . v . ... Z ..
I.-. or. ~ . - ul-wki‘. 1! ' l
: I a .....aliro
. ........J... I .. . , . . r . it, ,1!- III r rrllu IIIIII
. . .... c .. ,. . .. . .y
. _ ,. V... >3:
. n u v .
!

 

 

 

 

 

3(8.0¢n

 

 

 

at

 

 

 

32253. 3 38¢

 

.~ .
8:8 E at“

a c2553

 

...., . 1...: u.
H , .... . 5.2m... ... r. a. .
a... .. .2 . .... .
.‘ . . . ..

. . b

... .. laid . ' W
[Pupfgi fl 2 .

 

 

49

N 6..

Mn?

0 .9 .o_.

m .9. .9.

Mr

0 .9.

03235 8: Sun
...Fm NF 0.. x m.
03935 .0: San
o3m=m>m .0: Emu
am up or x m.o_.

Dhm NF 0.. x vdw
333925

35985 ozmaozomm u 9

8......
ocean to omnozomm .. or
23.96235. ... m
93.26 295 a m
”gagged

oz 1. 8. x 83%

=0 .302 km
300 ... 08.8

02 a: mop x momfi
.500 ... 8h .8

oz 0.. 8. x «8.8
.80 e $32
.80 e 839.

=0 .30: m5
300 ... 8‘35

953%

«co—30 :02: a 302

88 88: u » ”amaze.
“sanding: aqfiuufifl
can: b no EnE gauged

New. 565:5 .00 96.0 :25
38. $82:
now. e an .5... 822m 23.0 in
an... SN
flog e 82 an... .26... 08
ea... $82:
now. e Btu Eu... .96.. 8.2
as... e 5. .5:
now. » $2 566 each .56
sane » Ba
Anew. ... on twp Ear. 5.56 team So
gm Eam—

.§§ . zoz essen— aaao .3 a»... we use
8 33032 no.8 mamas 2: E 35: «membe— xmw .m 033—.

com 35 Snow 2&8

50

War:

With the selection of the target FFC facility and the development of the
survey instrument established, a pretest was administered to several people living
in the sampling area. At the suggestion of Cynthia Wheatley and Eleanor Love of
the Lansing Planning Department, the presidents of the Cherry Hill, Green Oaks,
and East Side Neighborhood Associations were contacted to act as pretest
subjects. These individuals and their spouses provided valuable insight into
administering the survey instrument in their respective neighborhoods.

Two initial instruments were tested that were similar to the final
instrument with the exception of the speg'fie nature of the scenario presented.
Both presented a hypothetical, newly proposed power plant to be placed at various
locations. The subjects were then queried how much they would be willing to pay
(WTP) to either move the plant to a location farther away from their residence
(pretest survey #1), or to prevent the plant from being built at all (pretest survey

#2). Several problems were evident with these scenarios, which included:

1) a difficulty in getting the subjects to understand the rules of the
bidding game;

2) a basic problem of credibility regarding the need for a m plant;

3) a failure by the subjects to understand exactly what they were being
asked to pay to avoid or prevent; and

4) an equity problem dealing with simply pushing the plant and its

attendant problems into ”someone else’s backyard".

51

Based on the comments of the people who participated in the pretest, the
instrument was changed to the form that was administered in the final gathering of
the data.

51.5 ”1.:

Similar to the strategy used by Furuseth (1989), a stratified, random
sampling strategy was used, with a random sub-sample being taken in each of four
zones, stratified by distance to the Eckert Station Plant (Figure 14)., The distances

of the four zones from the plant were as follows:

1) 1\4 to 1\2 mile 3) 1 1\4 to 1 1\2 miles
2) 1\2 to 3\4 mile 4) 1 1\2 to 2 miles

The rational for this type of a sample at these distances was similar to that used by
Dear, et a1 (1980) and reflected a trade-off between three factors.

1) The zones must be large enough and of such a nature to generate
enough people for a reliable sample.

2) Previous research by Dear, et al (1980) and Furuseth (1989) indicated
that the SEF may be fairly compact spatially, reflecting the fact that a
FFC site is not perceived to be as noxious as a nuclear plant.

3) The number of subjects in the sample and the spatial extent of the
sampling must be such that a single interviewer could perform the

interviews in a reasonable period of time.

52

 

—>z
Puolg)
13990

 

Ottawa . Mm,
,.,,:._.,4¥:~: :5; :-: ‘2“ . . .
, . 1'; ' ' ’5‘"
a - an: - I ------ a: r-’
Itiiii1?:"':-:-":5"1‘=' 4.:3'2: . '
‘ re-xafi-i” ',:- .rft.«.c:.'.’e... ..
9‘ 43:?- .. I
. '2' I. .r.

 

Kalamazoo

l-496

 

eluerlplsuued

 

 

 

 

 

 

Figure 14. Sampling areas relative to the power plant location.

Based on this set of spatially stratified zones, a sub-sample size (ni) of
thirty people in each zone was chosen, yielding a total sample size (N) of 120
people in all. This represented about fifteen percent of the approximate 800

households in the four zones, and compares quite favorably with sample sizes in

previous research.
With this scheme established, 800 cover letters (Figure 22 in the Appendix)

were hand-delivered to the prospective subject households in the four areas. The

purpose of this activity was threefold.

53

1) This would notify all potential subject households of the fact that
interviews would be commencing in the near future, increasing the
likelihood that they would elect to participate.

2) This allowed this author to operate more efficiently and safely
in these neighborhoods.

3) This also would allow any condemned or vacant dwellings to be noted so
these could be taken out of the total pool of possible subject residences.

With the strategy established, the cover letters distributed, and any vacant
or condemned dwellings removed from the pool of prospective subjects, the
selection of the subject households was done in a random manner as detailed

below.

1) All prospective subject households within each zone were given a
discrete number.

2) A random number generator was used to produce fifty prospective
households within each zone; this assured a valid sample size of at least
thirty subjects, given that some people would be unwilling to participate
or would not be at home.

3) These fifty prospective subject residence locations were then transferred
to 1 to 100 scale assessor’s maps, which were used in the field to locate

and keep track of the subject residences.

The implementation of the actual survey work in the field took place
during the two week period from March 26, 1990 to April 10, 1990. Following the

54

randomization scheme outlined above, 1 16 usable personal interviews were

administered, with the data being recorded on the subject response forms.
DamAnaheis:

Once the data were gathered, some preliminary coding was performed, and
the data were entered into a Compaq 386 micro computer and analyzed in
SYSTAT (The System for Statistics, Version 3.0, 1986).

Since the data were to be analyzed in a multiple regression framework, the
first step in the analysis was to do a set of diagnostic statistical runs to determine
how closely the data conformed to the assumptions underlying the General Linear
Model (GLM) (Poole and O’Farrell, 1971). Some of the diagnostics were done
before the regression runs occurred, while others were performed in an interactive

manner. These initial analyses included:

1) a review of plots of the variables to determine if the variables
themselves were normally distributed;

2) a review of correlation matrices to determine if multicollinearity
was present;

3) a review of the KS Lilliefors statistic to determine the degree of
normality and homoscedasticity that was present among both the
variables and the residuals or error term;

4) a review of the tolerance statistics that accompanied the simple and
multiple regression runs involving the dependent variable of the bid that
subjects gave for improving air quality (BID) and the main independent

variable of distance from the subjects residence to the

55

FFC facility (DIS);

5) a review of the plots of the residual term to determine normality
of the residuals;

6) a review of plots of the residuals versus DIS to determine the
homoscedasticity of the error term in relation to the independent
variable; and

7) an examination of the leverage values resulting from the regression runs
mentioned above in item number four to determine if there were any

undue leverage values resulting from these runs.

As mentioned above, some of these diagnostics were run a priori, while others were
run interactively, in an ongoing manner with the subsequent simple and multiple
regression runs.

Once all of these analyses were performed, they were reviewed in detail to
determine which of several potential structures being considered to relate distance
to the bid variable yielded the best behavior among the residuals and other
pertinent statistics. It should be noted that a basic philosophy was adhered to in
this stage of the analysis. While the literature (Shaw and Wheeler, 1985; Clark and
Hosldng, 1986) was consulted in detail to help determine the structure being
considered, thought was also given to the underlying processes that were
hypothesized to be acting on the relationship between distance and the subjects’
bid. This author was after a balance between a structure that behaved well
statistically and had a reasonable degree of theoretical basis to it.

The various structures explored included a simple linear form, various

power curves, exponential curves, higher order polynomials, and certain log

56

transformations of these curves. The more promising of these are portrayed in

Figure 15. The log transformations were explored for two reasons.

1) The log structure had a certain amount of theoretical basis to it
concerning how the variables being considered may interact together.

2) The log transformation allowed the regressions to be run in the linear
form that satisfied the ”inherently linear" nonlinear condition of the
GLM (Pindyck and Rubinfeld, 1981).

Again, the intent at this stage of the analysis was to determine a structure that
achieved a balance between theoretical soundness and adequate behavior of the
diagnostics.

Once this structure was determined, an extensive series of runs was
performed that searched for the other bivariate and multivariate relationships that
existed among the variables. The author followed previous research (Dear et al,
1980 and 1982; CEQ, 1980; Levensen, 1974; and Sundstrum, 1977) to guide the
search for these relationships. With this description of the analysis used and the
justification for it in place, discussion will now turn to a formal review of the

results.

 

57

 

 

 

a) y-a-bx b) y..axb

 

 

 

 

. 2
c) y a+bx+cx d) y-ae-bxz

 

 

 

 

O) y-aebX-CXZ f) ygae-bx

Figure 15. Various regression structures explored.

IllI'B 10.:

The best fitting structure for the regression that related the subjects bid for
improving air quality (BID) to the distance at which they were located from the
FFC facility (DIS) was the log transformation of the first order negative
exponential as detailed below. (The transformation is listed first, with the raw form
of the relationship following; an asterisk indicates significant coefficients at an

alpha level of .05).
LOGBID = 2583'- 357’ (DIS) Equation 1A
BID = 13237 e -357 (DIS) Equation 13

Examination of the tolerance values, KS Lilliefors statistics, residuals, and leverage
values indicate that this structural form is relatively well behaved. The relationship
is in the direction that was anticipated, and the coefficients are both significant at
an alpha level of .05. Collinearity of the variables is not a problem, and neither the
KS statistic nor the plots of the residuals indicate a significant departure from the
normality assumption of the GLM. Lastly, the leverage values do not exhibit any

extreme behavior.

58

59

WW:

Papageorgiou’s (1978) suggestion that urban form results from an
interaction of a population surface and an externality surface drove the exploration
of these related research questions. Specifically, variables that were found to effect
perceptions of externalities in previous research were examined; in addition, this
author’s intuitive views exerted a certain amount of influence at this stage of the
analysis.

It should be noted that the log transformation was retained as the basic
structure for these and all other subsequent explorations into the relationships
between the BID variable and related variables. While the underlying rational for
this structure, as regards npn;distanee variables, is not as strong as for the distance
variable, it did yield relationships that had the greatest statistical significance. As
mentioned previously, this analysis was performed with the concept of a balance
between theoretical soundness and statistical validity in mind.

Several bivariate relationships were found to exist between the BID
variable and the socioeconomic variables - i.e., SUPPORT (the subjects
expression of support for the NEPA legislation, HEALTH (the ranking of health
concerns as a consideration in their bid), and SCHOOL (the level of education
that the subject had completed). The variables and the coefficients of these
relevant regression equations (numbers 2, 3, and 4) are detailed in Table 6, along
with the R squared statistics and an indication of significance at the .05 level.

All of the coefficients are significant at the .05 level, with the exception of
the intercept in equation 4, and the relationships are all in the expected direction.

Examination of the diagnostic statistics mentioned above indicates that all of these

6.0

Figure 6. Coefficients, t values, R squared values and indication of

WWWWW
Malues

fl
1A

10

Name

Constant
DIS

Constant

SUPPORT

Constant
HEALTH

Constant
SCHOOL

Constant
DIS
HEALTH

Constant
DIS
SCHOOL

Constant
DIS

SUPPORT

Constant
DIS
INCOME

Constant
DIS
SCHOOL

SUPPORT

Constant
DIS
SCHOOL

SUPPORT

+2.583
-.357

+1 .375
+.008

+1.490
+.766

+.748
+.105

+1 .859
-.396
+.850

+1.081
-.426
+.116

+1 .803
-.357
+.008

+2.392
-.431
+.125

+.780
-.414
+.090
+.006

+.826
-.413
+.093
+.006

+12.121

+1

-2.078

+4.968
+3.188

+4.30?
+2.075

+1 .561
+3.047

+4.973
-2.341
+2.338

+2.231
-2.578
+3.417

+5408
-2.215
+3.143

+9.876
-2.442
+1 .620

+1.562
-2.619
+2.689
+2.610

.442/+.467

-2.580
+2.726
+2.443

.039

.088

.039

.080

.086

.134

.129

.062

.186

.189

significance for the regression equations reported in the text.

Significant

Mistral

Yes
Yes .

Yes
Yes

Yes
Yes

No
Yes

Yes
Yes
Yes

Yes
Yes
Yes

Yes
Yes
Yes
Yes
Yes
No

No
Yes

Yes
NO

Yes
Yes

61

relationships behave relatively well, with no major problems in the distribution of
the error term, no collinearity among the variables themselves, and reasonable
leverage values.

Several multivariate relationships were found to exist among the data,
whereby certain socioeconomic variables acted in conjunction with the DIS
variable in effecting the subjects bid (BID). The first of these regressions (numbers
5, 6, and 7), as indicated in Table 6, are relatively strong relationships; again,
significance of the coefficients is indicated in the table.

All of these coefficients and the overall relationships are significant at the
.05 level, and the relationships are in the direction that was anticipated.
Examination of the diagnostic statistics indicates again that these relationships
behave with no major problems with regards to distribution of the residuals, no
collinearity among the variables, and no extreme behavior of the leverage values.

Three other multivariate relationships (numbers 8, 9, and 10) were found
that were not quite as strong as the above mentioned ones, but still deserve
mention. Most of these coefficients are significant at the .05 level with the
exception of the income coefficient in equation 8, and the intercepts in the last two
equations.

The last two equations relate the effect that level of education (SCHOOL)
, and support for the NEPA legislation (SUPPORT) had on the relationship
between D18 and BID, but there is a minor difference in the coefficients. The last
equation contains a binary choice variable that indicates whether a person in the
subject’s family had a serious breathing ailment. This variable acts as an intercept
dummy variable that generates coefficients with different values than those in

Equation 9. Hence, the presence or absence of a person having a breathing

62

ailment effects the overall relationship among these variables, but the resulting
intercepts are not significant at the .05 level.

In addition to the lack of statistical significance in these latter relationships,
there are some problems with the behavior of the residual term (specifically,
evidence of heteroscedasticity in the distribution of this term) as well as leverage
values considerably higher than the level deemed safe by Wrigley (1983). While
these problems and the marginal level of significance suggest these results are not
as strong as those previously reported, they hint at some underlying processes that
are of considerable interest. Therefore, mention is made of them now, with further
discussion following in the latter portion of this manuscript.

A last group of regressions that is not even formally reported in the tables
or equations will be briefly discussed before moving to the discussion of the results.
In an attempt to examine whether family structure or tenure of residence had any
effect on the BID variable, regressions examining these variables were run both via
the LOG transformation and the linear form, with no significant results.

The variable for tenure that examined how long residents had lived at their
present residence yielded a zero coefficient that was not significant. Perhaps even
more interesting, the coefficients obtained for number of children or seniors in the
household, while not statistically significant at the .05 level, were actually eppesjte
to the direction anticipated. This suggests that these variables may be masking a
negative income effect, whereby households with more children, or seniors on
limited incomes, may not have had the financial resources to provide a higher bid.
Comments noted in the field support this possibility, although this facet of the
regressions was not tested statistically.

The last set of results to be presented concern the difference of the mean

BID among various subgroups of the sample population. This portion of the data

63

set either did not reveal significant results in the above discussed regressions, or
was not suited to treatment in this manner. Accordingly, the mean BID’s for these
variables were examined via the T test to determine if they were statistically
different from each other.

The following subgroups were examined in this manner:

1) the bids that people gave to improve air quality effected by a FF C
facility, as opposed to a hypothetical waste to energy incinerator;

2) the bids given by male as opposed to female respondents;

3) the bids of individuals involved in factory or industrial occupations
versus occupations outside of the industrial sector;

4) the bids of individuals living in single family detached homes versus
apartment dwellers;, and

5) the bids of married versus single individuals.

As indicated in Table 7, the only set of mean bids that were statistieally
different were those of bids related to FFC facilities versus those related to waste
to energy incinerators. While the other pairs of mean bids exhibited minor
differences in values, these differences were not significant at the .05 level.

With this review of the statistical results in place, attention will now turn to
a more extensive discussion of these results and the relevance they have to the
various research questions posed in the statement of the problem and the

hypotheses.

64

Table 7. T test results comparing mean bids for certain sub-groups.

.Q5_lexel

Bid for FFC facility $12.76 4.733 Yes
Bid tor waste to energy Incinerator $14.67

Bid by males $12.60 .729 No
Bid by females $14.71

Bid by industrial occupation workers $11 .00 .451 no
Bid by non-industrial occupation workers $12.95

Bid by single-Iamily dwelling occupants $11.69 .796 no
Bid by multi-tamily dwelling occupants $13.88

Bid by married Individuals $13.86 .1 15 no
Bid by single individuals $13.48

I! . i .
I] I I . B l D . :

As formally noted in the statement of problem and research hypotheses,
the main research question deals with the relationship between the distance at
which people live from a FFC facility that generated negative externalities (DIS)
and their perceived valuation of air quality (BID). As noted in Equation 1, the log
transformation of the first order negative exponential was found to be the best
fitting structure that relates (DIS) to (BID) and indicates that there is a negative
relationship between these variables.

This result supports research hypothesis HlA, and reveals a distance decay
function acting on the perception that individuals living downwind from this FFC
facility have regarding the negative externalities being generated. These findings

are consistent with the majority of research reviewed previously, including that of

65

Furuseth (1989), Van der Plight (1984), Dear, et a1 (1980, 1982), and the wealth of
work done concerning evacuation behavior surrounding nuclear power plants done
by Johnson, et al (1983, 1984, 1985, 1986, 1989) and the USNRC (1978, 1980).

The valuation that people hold for improvements in air quality effected by
the plant is higher near the facility and then decreases via the form specified in
equation 1 as the distance to the source increases. Those individuals living nearest
to the FFC facility perceive the negative externalities to the greatest degree, and
this perception lessens as the distance to the facility increases. Evidently, any
assimilation effect that may exist as a result of people becoming acclimated to the

effects of the plant is superseded by the distance decay function.
Relatedflncstiem:

In addition to the bivariate relationship between DIS and BID, several
other relationships were found to be statistically significant at the .05 level. As
noted in the previous section, SUPPORT, HEALTH, and SCHOOL all affected
the BID variable as hypothesized.

The level of support that an individual had for national environmental
legislation (SUPPORT) was positively related to the evaluation the individual had
for improvements in air quality. As was anticipated, this support translated into a
higher evaluation for this improvement; this is consistent with previous findings by
Van der Plight ( 1984) and Levenson (1974) regarding a priori views of facilities
that generate negative externalities and peoples’ perceptions associated with the
externalities.

An individual’s ranking of health concerns also had the anticipated effect

on the BID variable. People who expressed a greater concern for the health

66

related effects of the externalities associated with the plant apparently were more
sensitive to these effects, and this translated into higher bids.

Lastly, an individual’s level of education was also positively related to their
bids. As anticipated, these individuals tended to be more aware of both the direct
effects of the plant as well as the implications to the human condition; this was
revealed in their bids.

When these variables were examined to determine how they acted in
conjunction with the distance variable, similar results were found. As detailed in
equations 5, 6, and 7, these variables were all significant at the .05 level, and the
relationships found were in the direction expected. The coefficient of DIS
remained negative, as anticipated, whereas the coefficients of SCHOOL,
SUPPORT, and HEALTH all remained positive, as hypothesized.

When several of these ancillary variables were included with the DIS
variable, the statistical validity of the resulting regressions began to erode. As
mentioned previously, Equations 8, 9, and 10 moved towards the margins of
significance, although the direction of the relationships was still as expected.

Perhaps the most interesting of these latter regressions were Equations 9
and 10, which contained the binary chOice variable that dealt with the presence or
absence of a serious breathing ailment in the respondent’s family. While the lack
of statistical validity makes drawing any major conclusions inappropriate, it is
interesting to note that the presence of a serious breathing ailment in the
household has some "effect on the resulting regression.

The last group of findings to be discussed is the results of the T tests that
examined the existence of any differences in the mean bids among certain sub-
groups. As previously indicated, the only statistically significant result of these tests

was the examination of the differences of the mean bids resulting from perceptions

67

of a FFC facility as opposed to the perception of a waste toenergy incinerator. As
indicated in Table 7, the difference of these two mean bids is statistically
significant at the .05 level. Clearly, people in this study perceive the resulting
negative externalities of these facilities differently.

The other T test results do not indicate any differences that are statistically
significant. As indicated in Table 7, while these mean bids do differ slightly, the
differences are not significant; in fact, the bids for individuals living in single family
dwellings was actually lower than for individuals living in multi-family dwellings,
which was the opposite of what was anticipated. Sex of the respondent, marital

status, and job type all failed to generate any significantly different bids.

W11:

FFC facilities play a pre-eminent role in meeting U.S. domestic energy
needs and generate a variety of negative externalities that have significant effects
on the natural and human environment. The manner in which these externalities
interact with the population surface of urban areas may actually define the
structure of these urban areas (Papageorgiou, 1978). Human perceptions of these
processes are a natural result of these interactions and can have a major influence
on the acceptance of existing facilities as well as the ability to site future facilities
(Davis, 1984-1985; Furuseth, 1989). Accordingly, research into peoples’
perceptions of these facilities and how these perceptions vary across space is
warranted.

This research used a survey instrument to implement the contingent
valuation technique to capture peoples’ evaluation of air quality (via a series of
bids for this non-market good) at various sites downwind from a major FFC facility
in the Lansing Michigan area. The main research question that was posed
concerned the nature of the relationship between the distance at which people
lived from the facility in question and the bid that the subjects gave for their
evaluation of air quality affected by the plant.

A two-tailed research hypothesis describing a positive and negative
relationship between these variables was tested in a regression framework. In
addition, several other bivariate and multivariate relationships involving several
variables that described certain facets of the population surface of the study area

also were tested in this manner.

68

69

The results indicate that a distance decay function exists between the
distance at which people live from a FFC facility and the value they place on air
quality; log-linear regression showed the first order negative exponential was the
best fitting structure. In addition, individuals’ level of support for national
environmental legislation, their ranking of health concerns, and their level of
education completed had the expected positive effect on the bid variable. These
variables exerted an effect on the bid variable in a bivariate set of relationships as
well as in a multivariate set of relationships whereby they acted in conjunction with
the distance variable.

Certain sub-groups of the subject population were examined to determine if
there were any difierences in their mean bids for the proposed improvements of
air quality. While there were minor differences in these mean bids, none of the
differences were significant at the .05 level. However, the bids that individuals gave
to mitigate the effects of an existing EECjaeility as opposed to a hypothetical gaste
mnemjaeilin did differ significantly, and this finding indicates that people
perceive the negative externalities of these facilities quite differently.

Q 115' 'fi [151'

As outlined in the introduction of this manuscript, the manner by which
people perceive negative externalities that result from certain energy facilities has
both theoretical and societal relevance. Determination of the spatial dimension of
peoples’ perceptions of these facilities adds to the existing literature concerning
this issue; in this study, this researcher specifically addressed the question of how
the externality surface resulting from these facilities interacts with the population

surface of urban areas to help determine their structure and form.

70

The distance decay function that this study has found to exist supports the
majority of research in this field, and reflects the fact that the effect of distance to
a given interaction or activity does indeed have a dampening effect on the
perception of this interaction or activity. The other relationships that were found
to exist among the variables examined, lend further insight into the manner in
which the "people" facet of urban areas influences the end result of the
externalities resulting from FFC facilities.

This distance decay relationship can be expanded by rotating the curve
indicated in equation 1 around the vertical axis; this generates a cone or born, with
the power plant at the center of the cone. The volume of this horn represents the
financial value of eliminating the negative effects of the plant to the people living
proximate to the facility. As such, it represents the welfare value of such an
improvement, and this value can have utility in the resolution of the issue of
whether to compensate the individuals living proximate to the facility or attempt to
mitigate the effects of the facility.

Given that people living proximate to such a facility have access to valid
information, have the freedom to act upon this information, and are given a choice
as to whether to voluntarily accept the effects of the negative externalities, the
ethical issue of compensation versus mitigation invites discussion. Arguments in
favor of both of these options can be made, and these arguments involve the
different social discount rates that certain individuals have regarding the trade-offs
that exist between retention of environmental quality and economic gain or loss.
Another facet of this question is the tradeoff between efficiency in a Pareto
optimal sense versus a sense of social welfare or equity.

As is so often the case in issues of social welfare, the real question is "who

decides?”; specifically, who decides whether the above mentioned welfare value is

71

utilized to mitigate the environmental disruption or to compensate those who bear
the most direct effects of it? Do policy making bodies of a popularly elected
government make this decision (for the good of those involved) or is it best left up
to the people involved to make the decision directly? Who decides whether these
resources are used to approximate Pareto optimality in resource utilization, or to
attempt to address the social equity issue?

These very questions were posed to this author in the final phase of this
research; summarizing this discussion may help answer the above questions, while
probably posing or leaving open other related issues. Given the assumptions of
people having access to valid information and the freedom to chose, this author
argues that, for the most part, those receiving the direct effects of the above
mentioned externalities should have a considerable voice in deciding the
mitigation versus compensation issue.

Two significant caveats to this view deserve attention. The above set of
assumptions implied that those parties receiving the majority of the effects have
the ability and freedom to evaluate the trade-offs involved and to make a decision;
this implies a sense of maturity of thought and the power to render this decision.
What of those individuals who do not have this decision making power, specifically
children or other people in. a family in a subordinate position? If one expands this
discussion to facilities that may generate more pervasive and damaging
externalities than FFC facilities (i.e., nuclear weapons facilities and toxic or
nuclear waste disposal sites), this issue takes on even greater importance.

This author suggests that some form of social protection may need to be
exercised on behalf of these individuals, particularly in mes of facilities generating
more damaging externalities. The facility involved in this study operates within the

set of laws that regulates air pollution; yet, questions are raised by certain special

72

interest groups as to the efficacy of these laws. If this analysis and discussion is
extended to facilities that either do not operate within relevant laws, or involve
issues where valid laws are yet to be concretely established, the argument for
policy makers to exert a sense of social responsibility takes on greater relevance.
Although this most recent discussion has departed from the specific results of the
research, this author believes the issues raised in this area are some of the most
important facets of how certain negative externalities interact with the human
dimension of our urban built environment.

As Davis (1984-1985) and Furuseth (1989) stated in previous research, the
perceptions that individuals have for these types of facilities may be both a
necessary and sufficient condition for the acceptance of existing facilities, as well
as the ability to site future ones. While the results of this research may not have
direct utility in the planning and policy processes involved with siting and
managing these facilities, this author concurs with the view that Furuseth stated in
the close of his research on hazardous waste facilities (Furuseth, 1989). Optimal
siting and management decisions will require that policy makers and researchers
interested in resource management issues be well aware of how individuals living
in proximity to these facilities perceive them. As this research has answered some
of the questions relating to this perception, this author feels that the effort was

warranted and has added to our knowledge base in this important problem area.
WW:

As is so often the case with a research effort of this nature, even the
suggestions of an experienced research committee and the pretesting of the survey

instrument left some facets of the research question unexplored. While the results

73

reported in this manuscript were significant at the levels cited, the low R squared
values found indicate that there may be other variables involved in the
relationships studied that were not included in the analysis.

Specifically, the effect of income on the regressions examined may warrant
further research. The questionnaire used did query the respondents in a general
manner, via ranges, about their income, but the coarse resolution of the data
generated may have limited its usefulness. (The intent of this technique of using
ranges was to defuse any resistance that the subjects may have had regarding this
question.) Income was found to have the expected positive relationship, with
higher income individuals generating higher bids, but the relationship lacked
statistical significance. Comments by a considerable number of respondents also
suggested that there may have been an unexplored income effect acting on their
bids.

Two possibilities for further exploration of this effect come to mind. First,
determining subjects’ existing utility bills would allow determination of what
percentage increase over their existing expenses their bid represented. Second,
determining the respondents’ precise income levels (i.e., asking for a specific figure
as opposed to ranges) would probably provide a more useful index of their
financial resources. These two enhancements to the survey instrument may help
uncover any hidden income effect(s).

Another idea for future research would be that of a cross-sectional study
examining perceptions of both an existing FFC facility and an mating waste to
energy incinerator. The findings of this research indicated people perceived an
m FFC facility differently than a W facility; it would be interesting
to see if this difference in perception would be born out in a cross-sectional study

that compared two existing facilities.

74

Related to this issue is the possibility of a trans-frontier effect acting on the
perceptions of facilities generating negative externalities that move across a
political border. It has been suggested that the large waste to energy facility that
operates in the greater Detroit, Michigan area is most likely exporting its
externalities to our Canadian neighbors in the Windsor, Ontario area. This seems
a promising study area in which the differential effects of Canadian versus United
States citizenship and the effect of externalities from "Someone Elses’ Backyard"

could be explored.

75

I 131]":

Distance decay function (DDF) - the relationship that describes how the effect or
perception of a given interaction or activity declines as the distance to the given

activity or interaction increases.

Fossil Fuel Combustion (FFC) - the burning of fossil fuels (coal, natural gas, and

petroleum) in order to meet domestic energy needs.

LULU’s - locally unwanted land uses; the group of land uses that generate negative

effects that are usually unwanted by the residents of a given locality.

NTMBY’s - Not In My Back Yard; the usual manner by which people in local areas

react to the above mentioned land uses (LULU’s).

Negative externality - a negative or bad effect of a given activity that tends to occur
outside of the normal market mechanism that is responsible for delivering the

given activity. An example is the pollution resulting from industrial activities that is
not mitigated; the costs of dealing with this pollution that is not included in market

delivery mechanisms is an extension of this initial externality.

Spatial externality field (SEF) - the concept that describes how negative
externalities are distributed spatially over a given geographic region. This consists

of three components:

76

1) the intensity - the total impact of the externality,
2) the extent - the actual spatial dimension of the impact
of the externality, and
3) the rate of distance decay - the rate at which the perception or impact of

the externality diminishes over distance.

WTA (Willingness to Accept) - the amount of money that must be paid to people

to voluntarily accept the effects of negative externalities.

WTP (Willingness to Pay) - the amount of money that pe0ple will pay to avoid

receiving negative externalities.

El . Ell'llll :

BID - the amount of money that respondents were willing to pay to eliminate the
possible effects of the FFC plant in the study.

DIS - the actual distance from the FFC facility to the respondents house.

HEALTH - the respondents ranking of health concerns as a concern that effected
their bid.

INCOME - the income classification that captured the respondent’s income level.

SCHOOL - the highest year of schooling that the respondent completed.

77

SUPPORT - the degree to which respondents supported the statement of purpose
of the NEPA statement.

APPENDIX

78
APPENDIX

I. Min:

As part of my master’s research at Michigan State, I am
studying how people feel about power plants and pollution. The
current issue of where to site certain energy facilities makes
this a matter of some importance to us as consumers, but please
understand that many of my questions will be based on imaginary
events. I do not work for the power company and have no knowledge
of their plans. I will be talking to you about the Eckert Station
power plant. and then asking you some questions about the plant
and some questions about yourself. Please rest assured that .11
your answers will be kept strictly confidential.

Do you have any general questions about the survey that I am
conducting today? If not, we can now move to the consent form and
then on to the questions in the survey.

11. W:

This is a consent form that I need to have you read and sign
before we get into the survey itself. (land the foam to the
person being interviewed.) Can you take a brief moment to read
and sign it? This is on a separate sheet from my notes and will
be stored separately as required by federal regulations.

Thank you.

III. Wm:

The Eckert Station power plant is the largest plant in the
Lansing area and also produces the most pollution of any plant.
You may know it by one of its nicknames. "The Three Sisters“ or
“Hinkin’. Blinkin’. and Nod”. Here is a photograph of the plant
and a map showing where it is located in the Lansing ares. (Show
themapandphototothepersonbeimginterviewed.)Areyou
familiar with this plant?.... Good.

While it uses modern means to reduce this pollution to
levels within the law. it still produces alct of pollution.
Phasihle negative effects from this pollution include:

health risks,

higher noise levels,

negative effects on property values.

more dirt on our homes and cars. and

visual or aesthetic problems with the plant
appearance itself; how it actually looks.

I’ll!”

Figure 16. The script used to guide the interviews.

79

On the positive side, Lansing enjoys the lowest utility
rates of any northern city, and the Eckert plant is a very
important part of the power generating capability of the Board of
Water and Light. Put simply, the operation of the plant has both
plus’s and minus’s.

I am interested in how you feel about the trade-offs that
result from these plus’s and minus’s. I will be asking you some
questions to find out how you value the different costs and
benefits that are mentioned above. Do you have any questions so
far?.... Good.

Suppose that we could eliminate all of the possible negative
effects of the Eckert Station plant; in exchange for this. higher
utility rates would result. Please bear in mind that this
increase in costs would occur through higher monthly utility
bills and would apply to everyone in your area. not Just you
(there will be no free riders).

Figure 16 (cont’d).

80

CONSENT FORM

1. I have freely agreed to take part in a scientific study being
conducted by Christopher Brown, a graduate student in the
Geography Department at Michigan State University, working under
the guidance of Dr. Joanne Westphal. I will be taking a survey
that asks me questions about power plants, pollution, and some
information about myself.

2. The survey has been explained to me, and I understand the
explanation that has been given and what my role in the survey
will be.

3. I understand that I am free to discontinue participating in
the survey at any time. I understand that the expected length of
the survey will be about 20 minutes.

4. I understand that the results of my participation in the
survey will kept in strict confidence, as will those of all other
people participating. In other words, all participants will
remain anonymous in the reporting of results. Within these
restrictions, results of the study will be made available to me
at my request.

5. I understand that, at my request, I can receive additional
explanation of the study after my participation is completed.

Signed:

 

Date:

 

Figure 17. The consent form.

81

 

 

Figure 18. 6 x 8 inch reduction of the original 8 x 10 inch
color print of the Eckert Station Plant.

82

e s e ‘ ~‘Ctfi5‘\.§&

.’

Te

‘70 00mm

.08

     

seam
Rates 25%

[a Detroit

 

.3

Y
I")
'

      

3;. my
use...
.uom

To

...—25.5. 5 zoom

”‘5
Ne

3:: ... ...em

«@2524:

.. n. .. ,u.~.s ...; J. .... ... ..~.. .. .s. ... ....

 

see—d 2: wigs—BE E 353232 2:
Eu 3 tom: 9:: Rimmed 05 we come—60m .3 2&5

Etc? es

 

 

‘1

 

‘gsaaa

 

,s u e
...e. . .7 ... .. I.
...... . . . .

e .enfisn Jul. 9. v. u m . .
. n, s s
.... ...uuzflt. ..
.lrxu. t .

..W‘L... . .

..

 

 

 

see _ e2:

.5: .
a:

 

-»-- .rs.~-.r'

'V

.'m‘—- “ '-I-. -

Ann's-c.-- H.1-

83

I. Was:

A) How much would you be willing to pay to
willing to pay:

Would you be

18 more per month?__________ 108 more per
28 more per month?__________ 158 more per
33 more per month?__________ 203 more per
48 more per month?__________ 25$ more per
58 more per month?__________ 308 more per
Continue until a “no” response is given.

bring this about?

month?
month?
month?
month?
month?

If this is between 58

increments. bid down from the no response until you receive a yes

response; this is the bid to record

 

B) Which of the following concerns associated with the plant were
reasons that influenced your decision to pay the higher rates? Of
those mentioned. can you rank them from most to least important?

Health concerns

Impact on property values

Noise pollution

Dirt getting on houses and cars
Aesthetic or visual impact of the plant

Other

rank
rank
rank
rank
rank_________
rank_________

 

 

 

 

 

 

Are there any other reasons that influenced your decision?

 

C) If the power plant involved in the survey was to be an
incinerator that burned solid waste in order to produce electricity.
would the amount you would be willing to pay be different? If so. what

would the new dollar amount be?

 

Figure 20. Subject response form.

II. I:eatment_gt_lern_fiids= (Query only if sero bids were received).

What was the reason that you were unwilling to pay a higher
utility bill to eliminate the negative effects of pollution associated
with the plant?

a) Did you feel that eliminating the negative effects of the
plant was truly not worth paying more on your utility bill?

b: Did you feel that it was unfair for those that receive
the damages of pollution to pay money to avoid such damage?

c)0ther

 

III. Barnum: .

If you don’t mind. I would now like to ask you some questions
about yourself. Again. please be assured that the responses to these
questions also will be kept strictly confidential. Your responses will
not be associated with your name or address.

A) How long have you lived at your present address?

 

Prior to this, where did you live, and for how long?

Location How Long

B) How many members are there in your household?

Number of children
Number of adults
Any senior citisens?

 

 

 

C) Are you married or single?

D) What is the approximate grade in school that you have
completed? (Circle the response that corresponds to their answer.)

I 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17+
(High School) (College) (Grad school)

E) What is your occupation?

 

Figure 20 (cont’d).

85

F) What is your approximate age?

 

G) Do you or any of your family members have a serious problem
with breathing (asthma. emphysema, or bronchitis)? __________ If yes,
which ailments?

 

B) This is the statement of purpose for the National
Environmental Protection Act that caused the EPA to be formed (Bend
the NEPA form to the subject.); I’d like for us to read it together.
Two questions follow this statement; please place a dot on the
accompanying scale at the place that best describes your answer to the
questions. (Bead with the respondent from their form).

This law establishes a national policy for the environment and
provides for the establishment of a Council on Environmental Quality.
Specifically, this law:

* encourages productive and enjoyable harmony between
humans and their environment.

* promotes efforts which will prevent or eliminate
damage to the environment. and

* enriches the understanding of the natural resource
systems important to the Nation.

Can you place a dot on the accompanying scale at the place that
best describes your answer to the questions.

Thank You. Just 2 more questions.

8) What is your approximate family income?

Less than 810.000
820.000 - 030.000
830.000 - 840.000
840.000 and over

Is this a dual income or a single income?
IV. Conclusion:
Thank you for your time and patience in taking this survey. I

really appreciate your cooperation; the information that you gave me
will help me in completing my research at Michigan State.

 

Figure 20 (cont’d).

86

v. Whats:
A) Sex of the subject:
Male
Female
B) Type of housing unit:
Single family detached
Small Multi-family unit (less than 5 units)
Large apartment building (5 or more units)
C) Time of day \ Day of week:

Time of day

 

Day of the week

 

D) Consent form signed?

 

Figure 20 (cont’d).

87

Statement of Purpose, National Environmental Protection Act
NEPA

This law establishes a national policy for the environment and
provides for the establishment of a Council on Environmental Quality.
Specifically, this law:

* encourages productive and enjoyable harmony between
humans and their environment,

* promotes efforts which will prevent or eliminate
damage to the environment. and

* enriches the understanding of the natural resource
systems important to the nation.

Place a dot on the line that follows the question at the place
that best indicates your response.

How aware were you that such a law existed before reading this
statement?

Very unaware Very aware
I J
I I

How supportive are you of the use of government funds to carry
out the intent of this law?

Very unsupportive Very supportive-
l I
l 1

Figure 21. NEPA question form.

MICHIGAN STATE UNIVERSITY

DEPARTMENT Of GEOGRAPHY

EAST LANSING 0 MICHIGAN 0 48814-1!"
SIS NATURAL SCIENCE

March 21, 1990

Dear Lansing Resident,

As part of my masters degree in geography at Michigan State University, I am
studying how people feel about power plants and pollution. To gather the
information that I need in this study, I will be giving short personal
interviews of the residents that live near the Board of Water and Light
facility located at Eckert Station. I will be doing these interviews over the
next few weeks. Although the study does deal with the Eckert Station, it does
not reflect any actions, views, or plans of the Board of Water and Light.

_L;; of the information that I gather will be treated in the strictest
confidence. Your responses will not be associated with either your name or
your address.

I would appreciate it very much if you could give me about 20 minutes of your
time when I call on you in the coming weeks. Participation is totally
voluntary, of course. I plan to be making my calls on weekday afternoons and
evenings and on weekend days between the hours of 10:00 am and 5:00 pm. I hope
these hours will be the most convenient for you. Should I arrive at an
inconvenient time, please feel free to ask that I return later.

If you wish to confirm that this request is legitimate, please feel free to
contact the following people in the Geography Department at Michigan State
University.

Dr. Judy Olson, Chairperson Dr. Joanne Westphal, Assoc. Professor
Department of Geography, MSU Department of Geography, MSU

315 Natural Science Building 108 UPLA Building

East Lansing, Mi. 48824 East Lansing, Mi. 48824

(517) 355-4651 (517) 353-7880

Thank you for your cooperation in this matter.

Sincerely,

~
Christopher Brown
Graduate Student
Department of Geography
Michigan State University

Figure 22. Coverletter.

BIBLIOGRAPHY

89

BIBLIOGRAPHY

Ajzeal and M Fishbein.1980.llnd9_rsmldmg_Ammdesand_Ertdlcung§ooal
W. Englewood Cliffs, NJ: Prentice Hall, Inc. '

Bowman, C. H. and M. Fishbein. 1978. "Understanding Public Reaction to Energy

Proposals: An plication of the Fishbein Model." WWW
135352110193! 8(4): 19-340.

Brookshire, D. S. B. C. Ives, and W. D. Schulze. 1976 "The Valuation of Aesthetic
Preferences. _, ! _ _ _ - __c.-_-."u’! 3:325-346.

Clark. W A V.. and P L Hosking 1986 W
New York City, NY: John Wiley and Sons.

Con ess of the United States of America. 1970. W
W. Washington D. C. U. S. Government Printing Office.

Council on Environmental Quality. 1980 31W
Washington D. C.: U. S. Government Printing Office.

Coursey, D. L, W. D. Schulze, and J. J. Hovis. 1983. ”A Comparison of Alternative
Valuation Mechanisms for Non-Market Commodities." Unpublished Manuscript,
Department of Economics, University of Wyoming.

Crespi, 1.. 1971. "What Kinds of Attitude Measures are Predictive of Behavior?"
hlbnsflmmenfluaneflx 35: 327-334.

Davis, C. 1984-1985. "Substance and Procedure in Hazardous 1n Hazardous waste

Facility Siting." InumalniEnnmnmentaLsttems “(1)51452

Dear, M. J. and S. M. Taylor. 1982. n r ' ni
Mentallieflm. London. England: Pion. Ltd.

Dear, M. J., S. M. Taylor, and G. B. Hall. 1980. "External Effects of Mental Health
Facilities. . tr . . . . t - t. 70(3) ..342-352

Ener Information Administration. 1990. Wm Washington
epartment of Energy.

Festinger, I... 1957. WWW. Stanford, CA: Stanford

University Press.

 

 

 

90

Flynn, C B.- 1979 .MMSJMILéfihSmmMX Washington, DC: US.
Government Printing Office (NUREG/ -1093).

Foust. B. J . and A De Souza. 1978. WWW
Introdugtign. Columbus, OH: Charles E. Merrill Publishing Company.

Furuseth, O. J. 1989. "Community Sensitivity0 to a Hazardous Waste Facility."
landscamnsulthanmanmng 17(4) 357-

Hill, A. B.. 1965. "The Environment and Diseases. Associations and Causation"
'0.“11' 0._1.'too__.'w\0.u'0°°~lt' 901010~10 '11t‘nu111'

58:272.

Johnson, J. R. 1985. "A Model of Evacuation-Decision Making m a Nuclear
Reactor Emergency. Wm 75(4): 405-418.

Johnson, J. H. Jr. and D. J. Ziegler. 1983. "Distin unguishlngj-Iuman Responses to
Radiological Emergencies." W 59' 38

.1984 "A Spatial Analysis of Evacuation

 

Intentions at the Shoreham NuclearPower Station." In W
M H hn 1 .Edited by Martin J. Pasqualetti and K.

David Pijawka. Boulder, CO: Westv1ew Press.

1.986 "Modeling Evacuation Behavior

During the Three Mile Island Reactor Crisis." WWW
20(3): 165-171

 

.1989 ”Nuclear Protective Action
Advisories: A Policy Analysis and Evaluation."Gr_Q_\1th_and_Change 20(4): 5 1-65.

Keeling, C. D.. 1973. "Industrial Production of Carbon Dioxide from Fossil Fuels
and Limestone. Ignns 25(2): 174-197.

Keeling, C. D., A. Adams, Jr., C. A. Ekdahl, Jr. and P. R. Guenther. 1976a.
"Atmospheric Carbon Dioxide Variations at the South Pole." Iellus 28(6). 552-564.

Keelin C. D., R. B. Bacastow, A. E. Bainbridge, C. A. Ekdahl, J r., P. R.
Guen er, and L. S. Waterman. 1976b. "Atmos heric Carbon Dioxide Variations
at Mauna Loa Observatory, Hawaii." Iellns 28 6): 538-551.

Kellogg, W. W. and R. Schware. 1981. W. Boulder, CO:
Westview Press.

Kneese A V and B Bower 1972 W
W. Baltimore, : Johns Hopkins University Press.

Knetsch J. L. and J .A. Sinden. 1984. "Willin ess to Pay and Compensation
Demanded: Experimental Evidence of an nexpected Disparity 1n Measures of
Value” W forthcoming

Lave, L. B. and E. B. Seskin. 1970. "Air Pollution and Human Health." Sgienge
169: 723-733.

 

91

. 1977. AiLEollutionmdflumanliealth
Baltimore, MD: Johns Hopkins University Press.

Levenson, H.. 1974. "Involvement in Anti llution Activities and Perceived
Nfgbative Consequences from Pollution." WM 38:1105-

Losch, August. 1954. WM. New Haven, CN: Yale
University Press.

Lynn. D A. 1976 WW Reading. MA' Addison-
Welsey.

Marwell, G. and R. E. Ames. 1981. "Economists Free Ride, Does Anyone Else?"
loumalofhlhlicficonomiss 15: 295-310.

Miller. G. L Jr. 1980. W5 Belmont.
CA: Wadsworth Publishing ompany.

Nourse, H. O.. 1967. "The Effects of Air Pollution on House Values." Land
5432119111195 43: 181-189.

Odland, J. and B. Balzer. 1978. "localized Externalities, Conta tagious Processes, and
the Deterioration of Urban Housing: An Empirical Approach. W

Elamungicim 13: 87-93
Papateor u on, G...J 1978. "S

  
 

atial Externalities I and II: Theo and Applications."
- ‘201'h68(4) 4:6.-492

Pigozzi, B. W.. 1990. Personal Conversation. Department of Geography. Michigan
State University, East Lansing, MI.

Pindyck. R S and D L Rubinfeld 1981 EronometrieMmlelsandfsonomic
Em. New York City, NY: McGraw-Hill Book Company.

Poole, M. A. and P. N. O’Farrell. 1971. "The Assumptions of the Linear

Regression Model" WWW 52:145-
158.

Randall, A., B. Ives, and C. Eastman. 1974. "Bidding Games for Valuation of
Aesthetic Environmegéall‘lglprovcmentS” loumalnll Ennmnmcnmlflmomiss
Wm 1' l -

Ridker, R.G.. 1967a. W New York City, NY
Richard Praeger Press.

Ridker, R. G. and J. A. Henning. 1967b. 'The Determination of Residential
Property Values with Special ngfegelé-cgstp Air Pollution (St. Louis, Missouri)"
We

 

 

92

Schulze, W. D., R. C. d’Ar e, and D. S. Brookshire. 1981.
"Valuing Environmental modifies: Some Recent Experiments." Land
Egongmics 57(2): 151-172.

. ShaW. G. and D. Wheeler. 1986.5.tau5ucmmnmquesmficogram1calmalxsn
New York City, NY: John Wiley and Sons.

Sundstrom, E., J. W. Lounsbury, C. R. Schuller, J. R. Fowler, and T. J. Mattingly,
Jr. 1977. "Community Attitudes Toward a Proposed Nuclear Power Generating
E1313 0? a Function of Expected Outcomes. l

Thayer, M. A.. 1981. "Contin unent Valuation Techniques for Assessing

Envnonmental 1m :1sz er Evidence." loumalLEnnmmnentaLEconmcs
andManaztmtm :27-44.

United States Nuclear Regulatory Commission\Environmental Protection Agency.
1978.Pl ninonBifthDvlm f ..wnt owl o-r-vrnmn
'cnioo'lil r 51'11--".11 no 1'1hW.1Pwr
Hams. ashington, D. .:U. G.overnment Printing Office. (NUREG-.0396)

United States Nuclear Regulatory Commission\Federal Emergency Management
Agency. 1980. ri ri i n Ev i f R i l l 11

MW Su pl rt of Nuclea_r_9_e__anlsP w r P
Washington, D.C.: US. Government P nt‘ng Office. (NUREG-0654).

Van der Plight, J., J. R. Eiser, and R. Spears. 1984. "Public Attitudes to Nuclear
Energy." W 12: 302-305.

Vleeming, R. G. 1985. "Factors Affecting Attitudes Toward Nuclear Power 1n the
Netherlands" IheioumalnLSocialehology 125(1)=119-125

Von Thunen, J. H. 1821. l' i ' f
Wow. Hamburg, Germany: Publisher Unknown.

Weiand, K. F..1973."Air Pollution and Property Values: A Study of the St. Louis
Area." Was: 13: 91-95

Wrigley, N. 1983. "Quantitative Methods. On Data and Diagnostics."Eng1;ess_in
W 7(4). 567-577.

Zcrbe. R O 1969 WWW.
Toronto, Canada: Ontario Department of Public Health.

Ziegler D. J. S. Brunn, and J. H. Johnson, Jr. 1981. "Evacuation From a Nuclear
Tec ological Disaster." W710): 1-16.

Ziegler, D. J. and J. H. Johnson, Jr. 1989. "Information Planning and the

Geo a hical Context of Radiolo cal Emergency Management." 1131399331191;
MW 14 350-363