m.

{mitt-3::

uw‘m

 

 

 

 

:3...
“Put”, .3... .
. .3... ~

5. I,
I . 34.?!
31.2151: 9. t}
4.... 57—!
X.

 

 

W35

, I
57/1945 ‘75

 

w—v

LIBRARY
Michigan State
University

 

 

 

This is to certify that the

dissertation entitled
Golfers' UV exposure, health beliefs and practices,

and intention to adopt UV protective clothing

presented by

Heewon Sung

has been accepted towards fulfillment
of the requirements for

Ph.D. degreein Human Environment & Design

 

 

‘r

Major professor

Date 3247;- 5% 3003

MS U is an Affirmatiw Action/Equal Opportunity Institution 0-12771

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJClRC/DateDue.p85—p.1 5

GOLFERS’ UV EXPOSURE, HEALTH BELIEFS AND PRACTICES, AND
INTENTION TO ADOPT UV PROTECTIVE CLOTHING

By

Heewon Sung

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Human Environment and Design

2003

ABSTRACT

GOLFERS’ UV EXPOSURE, HEALTH BELIEFS AND PRACTICES, AND
INTENTION TO ADOPT UV PROTECTIVE CLOTHING

By

Heewon Sung

Since the major factor of developing skin cancer is associated with excessive
exposure to ultraviolet radiation (UVR), outdoor enthusiasts, such as golfers, need to be
focused on for skin cancer prevention messages. This study was divided into two phases:
The first phase focused on the cumulative exposure to UVR during golf rounds and
protective nature of clothing, and the standardized images of golfers in magazines and in
the local pro golf shops; the second phase focused on golfers’ health beliefs and practices
regarding sun precautions, and its impact on the intention to wear UV specialized shirts.

In Phase I, polysulphone film badges were used to measure personal and ambient
exposures to UVR. The experiments were conducted in August and in October, and seven
subjects participated in each experiment. Sixteen bandages were placed on the upper
body sites over and under the subject’s shirts. In both experiments, the shoulder and back
sites over the shirts received the greatest UVR exposure, while the back neck and front
neck were the two most exposed areas under the shirts. Results of paired sample t-test
between the absorbance values of badges over and under the clothing indicated golf shirts
significantly reduced UVR exposure in diverse body locations, and summer shirts were

less effective in blocking UVR than the thicker fall shirts or UV specialized shirts.

Market research was conducted as a part of phase I to provide the normative looks
of golfers photographed in the magazines and general information of golf clothing
available in the local pro shops. The most standardized look of golfers found in market
research, was the short-sleeved cotton polo shirts. Only a few UV specialized products,
such as hats, short-sleeved shirts, or sunglasses, were available at local stores, since the
majority of customers was not interested in them.

In phase II, 158 respondents completed a 58-item questionnaire. Ninety-one
percent were male, and their age ranged from 16 to 80 years old. About 30% reported
using sunscreen on the face and 20% used it on the body. A psychosocial model was
proposed based on two theoretical frameworks, the health belief model and Rogers’
attributes of innovations. Among the 1 I health belief model variables, peer’s sun
protective behavior was the most important predictor of sun protective actions, followed
by susceptibility, marital status, psychological barrier, benefit, and having family
members who have skin cancer, explaining 44% of the variance in the regression model.
Triability and relative advantage among five attribute variables were significant
predictors of intention to adopt UV specialized shirts (R2 = .411). The intention to adopt
was also predicted by behavioral barriers, current sun protection behavior, and ethnicity,
increasing the explanatory power to 50% of the variances in the proposed model.

The present study suggests that the health belief model variables play an
important role in explaining sun protective behavior and intention to adopt the preventive
innovation. Suggestions are given for improving UV protection through clothing design

as well as educational program design.

Copyright by
HEEWON SUNG
2003

ACKNOWLEDGMENTS

I thank God for helping me and guiding me to complete my study from the
beginning to this moment. Without the Lord, it would not have been possible.
Appreciation is sincerely expressed to my major professor and dissertation director, Dr.
Ann Slocum for her continued encouragement, commitment, and guidance throughout
my doctoral program. I would like to thank Dr. Sally Helvenston and Dr. Linda Good for
their support and assistance through my program at Michigan State University. I would
also like to thank Dr. Fred Oswald for his willingness and interest to help me as a
committee member. I greatly appreciate my committee members’ advice and assistance.

Special thank-you is extended to Dr. Per Askeland for his technical advice and
assistance in using polysulphone films. I also like to thank Matt Pinter, the Professional
Golf Management program coordinator, and Dr. Marilyn Keigley and Linda Plank at the
department of marketing at Ferris State University for helping me to collect data from
their students. Gratitude is also extended to Sean O’Connor, the superintendent at Forest
Akers Golf Course of Michigan State University for allowing me to collect data.

I would like to thank my husband, Wonjoo Roh, for his encouragement, support,
and understanding during my studies. Thanks are extended to my family: My parents and
parents-in-law, my sister and her family, and my brother, Jungwhan Sung and his family,
who stand by me and support me all the time. Additional thanks are expressed to the
members of the Matthew Sok, and Sojung Lee, Hyunjoo Lee, and Wanjoon Kim, for their

help and support.

TABLE OF CONTENTS

LIST OF TABLES .............................................................................................................. x
LIST OF FIGURES ......................................................................................................... xiii
CHAPTER ONE ................................................................................................................. 1
Introduction ..................................................................................................................... 1
Objectives and Research Questions ............................................................................ 3
Phase I — UV Exposure of Golfers and Market Research of Golf Clothing ........... 3

Phase II — Perceived Beliefs about Health and Attributes toward an Innovation... 4
Rationale for the Study ............................................................................................... 5
CHAPTER TWO: PHASE I — UV EXPOSURE ................................................................ 7
Literature Review ............................................................................................................ 7
Major Types of Skin Cancer ....................................................................................... 7
Ultraviolet Radiation (UVR) ....................................................................................... 8
UVR Received by Occupation and Activity ............................................................... 9
Body-Specific Sites of Skin Cancer Diagnosed ........................................................ l3
Golfers and Skin Cancer ........................................................................................... 15

Sun Protective Clothing ............................................................................................ 17
Methodology ................................................................................................................. 20
Preparation of the Polysulphone Film Dosimeter ..................................................... 20
Pretests of the Polysulphone Films ....................................................................... 21
Experiment I .......................................................................................................... 23
Experiment 11 ........................................................................................................ 25
Sampling Method ...................................................................................................... 25
Data Collection ......................................................................................................... 26
Experiment I .......................................................................................................... 26
Experiment 11 ........................................................................................................ 27

Data Analysis and Discussion ....................................................................................... 29
Experiment I .............................................................................................................. 29
Descriptions of Subjects’ Garments ...................................................................... 29
Absorbance Changes of Personal Dosimeters ...................................................... 29
Absorbance Changes of Ambient Dosimeters ...................................................... 32
Comparison of Personal Dosimeters with Ambient Exposure ............................. 34
Experiment 11 ............................................................................................................ 37
Descriptions of Subjects’ Garments ...................................................................... 37
Absorbance Changes of Personal Dosimeters ...................................................... 37

vi

Absorbance Changes in Ambient Dosimeters ...................................................... 42

Comparison of Personal Dosimeters with Ambient Exposure ............................. 43
Summary ................................................................................................................... 47
Research Question 1 for Objective 1 .................................................................... 47
Research Question 2 for Objective 1 .................................................................... 49
Research Question 1 for Objective 2 .................................................................... 49
Research Question 2 for Objective 2 .................................................................... 50
Limitations ................................................................................................................ 52
CHAPTER THREE: PHASE I — MARKET RESEARCH .............................................. 54
Standardized Appearance from Published Images ....................................................... 55
Methodology ............................................................................................................. 55
Results and Discussion ............................................................................................. 56
Characteristics of Shirts ........................................................................................ 56
Characteristics of Other Garments ........................................................................ 57
Summary ................................................................................................................... 58
Research Question 1 for Objective 3 .................................................................... 58
Market Survey ............................................................................................................... 59
Methodology ............................................................................................................. 59
Results and Discussion ............................................................................................. 60
Interview Question 1 ............................................................................................. 60
Interview Question 2 ............................................................................................. 61
Interview Question 3 ............................................................................................. 62
Interview Question 4 ............................................................................................. 63
Summary ................................................................................................................... 63
Research Question 2 for Objective 3 .................................................................... 63
Research Question 3 for Objective 3 .................................................................... 64
Limitations ................................................................................................................ 64
CHAPTER FOUR: PHASE II — HEALTH BELIEFS AND INNOVATION ................. 65
Literature Review .......................................................................................................... 65
Health Belief Model .................................................................................................. 66
Perceived Susceptibility ........................................................................................ 66
Perceived Severity ................................................................................................ 67
Perceived Benefits ................................................................................................ 67
Behavioral Barriers ............................................................................................... 67
Psychological Barriers .......................................................................................... 68
Modifying Factors ................................................................................................. 68
Relative Importance .............................................................................................. 70
Attributes of Innovation ............................................................................................ 73
Relative Advantage ............................................................................................... 74
Compatibility ........................................................................................................ 75
Complexity ............................................................................................................ 76
Triability ............................................................................................................... 77

vii

Observability ......................................................................................................... 78

Relative Importance .............................................................................................. 79
Attributes of an Innovation and Health Belief Model .............................................. 81
Implications ............................................................................................................... 83
Sun Protective Behaviors .......................................................................................... 84

Methodology ................................................................................................................. 87
Instrument Development ........................................................................................... 87
Sampling ................................................................................................................... 89
Pretest ........................................................................................................................ 89
Data Collection Procedure ........................................................................................ 91
Data Analysis Methods ............................................................................................. 92

Findings and Discussions .............................................................................................. 94
Data Analysis and Findings ...................................................................................... 94

Description of Demographics ............................................................................... 94

Description of Variables in the Health Belief Model ........................................... 94

Description of Variables in the Attributes of an Innovation ............................... 100

Description of the Summated Scale .................................................................... 103

Correlation Analysis ........................................................................................... 104

Regression Analysis ............................................................................................ 106
Discussions and Summary of Research Questions ................................................. 114

Research Question 1 for Objective 4 .................................................................. 114

Research Question 2 for Objective 4 .................................................................. 117

Research Question 1 for Objective 5 .................................................................. 119

Research Question 2 for Objective 5 .................................................................. 120

Research Question 1 for Objective 6 .................................................................. 121

Research Question 2 for Objective 6 .................................................................. 122
Summary ................................................................................................................. 124

CHAPTER FIVE ............................................................................................................ 127

Summary and Conclusions ......................................................................................... 127
Phase I ..................................................................................................................... 127

Field Experiments ............................................................................................... 127

Market Research ................................................................................................. 129
Phase II .................................................................................................................... 129

Health Belief Model ............................................................................................ 130

Attributes of a Preventive Innovation ................................................................. 131

Proposed Research Model ................................................................................... 132
Limitations of the Study .......................................................................................... 132

Phase I ................................................................................................................. 132

Phase II ................................................................................................................ 133
Implications for Practice and Further Research ...................................................... 135

Design Perspective .............................................................................................. 135

Business Perspective ........................................................................................... 138

Research Perspective .......................................................................................... 139

Education Perspective ......................................................................................... 141

viii

APPENDIX A: Experiments and Market Research ....................................................... 145

Consent Form for Experiment .................................................................................... 151
Consent Form for Interview ........................................................................................ 152
APPENDIX B: Questionnaires ....................................................................................... 153
APPENDIX C: Regression Assumptions ....................................................................... 164
Assumption of Regression Model A ........................................................................... 165
Assumption of Regression Model B ........................................................................... 170
Assumption of Regression Model C ........................................................................... 173
REFERENCE .................................................................................................................. 177

ix

LIST OF TABLES

Table 1. Summary of Literature regarding UVR Received by Activity and Body Sites. 12

Table 2. Summary of Literature regarding Body Specific Sites of Skin Cancer .............. 15

Table 3. Site-Specific UV Exposures and Paired Sample T-test between Under and Over
Clothing in Experiment I .................................................................................. 30

Table 4. Comparison between Left and Right Sides of the Body Over Clothing and Paired
Sample T-test in Experiment I .......................................................................... 32

Table 5. Absorbance Change of Ambient Dosimeter by Hours Exposed in Experiment I

........................................................................................................................... 32
Table 6. Z-scores for Ambient Dosimeters by Hours Exposed in Experiment I .............. 33
Table 7. Regression Analysis for Hours Exposed Predicting Absorbance Change on

Ambient Exposure ............................................................................................ 34

Table 8. Mean Proportion of Ambient UVR Received at 16 Body Sites in Experiment 136

Table 9. Site-Specific UV Exposures Under Clothing and Independent Sample T-test
between UV Shirts and Regular Shirts in Experiment 11 .................................. 38

Table 10. Site-Specific UV Exposures and Paired Sample T-test between Under and Over
Clothing in Experiment 11 ................................................................................. 40

Table 11. Comparison between Left and Right Sides of the Body Over Clothing and
Paired Sample T-test in Experiment 11 .............................................................. 41

Table 12.

Table 13.

Table 14.

Table 15.

Table 16.

Table 17.

Table 18.

Table 19.

Table 20.

Table 21.

Table 22.

Table 23.

Table 24.

Table 25.

Table 26.

Absorbance Change of Ambient Dosimeter by Hours Exposed in Experiment 11
........................................................................................................................... 42
Z-scores for Ambient Dosimeters by Hours Exposed in Experiment 11 ........... 43
Regression Analysis for Hours Exposed Predicting Absorbance Change in
Ambient Exposure ............................................................................................ 44
Mean Proportion of Ambient UVR Received at 16 Body Sites in Experiment 11
........................................................................................................................... 46
Summary of Garment Characteristics in Golf Digest ....................................... 57
Sun-Related Activities in a Typical Week in the Summer ............................... 95
Current Sun Protective Behaviors ..................................................................... 95
Perceived Susceptibility and Perceived Severity .............................................. 96
Perceived Benefits of Practicing Sun Protection .............................................. 97
Behavioral Barriers and Psychological Barriers ............................................... 98
Cues to Action of Practicing Sun Protection and Tanning ............................... 99
Perceived Attributes of UV Specialized Shirts ............................................... 101
Intention to Adopt a UV specialized Shirt ...................................................... 102
Descriptive Statistics of Variables in the Proposed Research Model ............. 103
Correlations among Variables in the Research Model ................................... 105

xi

Table 27. Regression Analysis of the Health Belief Model Variables on Sun Protective
Behaviors (Model A) ...................................................................................... 107

Table 28. Regression Analysis of Five Attribute Variables on Intention to Adopt (Model
B) ..................................................................................................................... 110

Table 29. Regression Analysis of All Variables in the Research Model on Intention to
Adopt (Model C) ............................................................................................. 113

Table A1. Record of Clothing Characteristics in Experiment I (August 28, 2002) ........ 146

Table A2. Record of Clothing Characteristics in Experiment 11 (October 11, 2002) ..... 147

Table A3. Frequency of Garment Characteristics Photographed in Golf Digest ........... 148

Table A4. Frequency of Colors of Photographed Images in Golf Digest ....................... 150

Table B1. Items in the Health Belief Model and Attributes of Innovations in Pretest 154

xii

LIST OF FIGURES

Figure 1. Absorbance Changes of Ambient Dosimeters by Hours Exposed in Pretest 22

Figure 2. Absorbance Change in One Hour of Ambient Exposure by Time of Day ........ 23

Figure 3. 16 Body Locations for Film Badges in Experiments ........................................ 24
Figure 4. Scatterplot of Cumulative Ambient Exposure by Hours in Experiment I ......... 33
Figure 5. Box Plot of Cumulative Ambient Exposure by Hours ...................................... 33

Figure 6. Scatterplot of Cumulative Ambient Exposure and Regression Line for
Experiment I .................................................................................................... 35

Figure 7. Scatterplot of Cumulative Ambient Exposure by Hours in Experiment 11 ....... 43

Figure 8. Box Plot of Cumulative Ambient Exposure by Hours ...................................... 43

Figure 9. Scatterplot of Cumulative Ambient Exposure and Regression Line for
Experiment 11 .................................................................................................. 44

Figure 10. Mean Proportions of Ambient UVR Exposure Over Clothing for Experiment I
and II ............................................................................................................... 48

Figure 11. Mean Proportions of Ambient UVR Exposure Under Clothing for Experiment
I and II ............................................................................................................. 50

Figure 12. Mean Proportions of Ambient UVR Exposure Under Different Types of Shirts
for Experiment I and II ................................................................................... 51

Figure 13. A Conceptual Framework of the Health Belief Model ................................... 69

xiii

Figure 14.

Figure 15.

Figure 16.

Figure 17.

Figure B1.

Figure C1.

Figure C2.

Figure C3.

Figure C4.

Figure C5.

Figure C6.

Figure C7.

Figure C8.

A Conceptual Framework of Attributes of UV Specialized Shirts ................. 79
A Proposed Research Model for Adoption of a UV Specialized Shirt ........... 82
Significant Relationships in the Research Model .......................................... 125
Proposed Design of UV Protective Shirt for Golfers .................................... 136
A Black-and-White Photograph for the Pretest ............................................ 157
Partial Regression Scatterplots of 11 Health Belief Variables on Sun
Protective Behaviors ...................................................................... . ............... l 65
Histogram of the Standardized Residual of Sun Protective Behaviors ........ 167
Normal Probability Plot of the Standardized Residual of Sun Protective
Behaviors ...................................................................................................... 167
Scatterplots between Health Belief Variable and Standardized Residual of Sun
Protective Behaviors ..................................................................................... 168
Partial Regression Scatterplots of Five Attributes on Intention to Adopt 170
Histogram of the Standardized Residual of Intention to Adopt .................... 171
Normal Probability Plot of the Standardized Residual of Intention to Adopt1 71
Scatterplots between Attribute Variable and Standardized Residual of
Intention to Adopt ......................................................................................... 172

xiv

Figure C9. Partial Correlation Scatterplots between All Independent Variables and IA 173

Figure C10. Histogram of the Standardized Residual of IA ........................................... 174

Figure C11. Normal Probability Plot of the Standardized Residual of IA ..................... 175

Figure C12. Scatterplots between the Significant Independent Variables and Standardized
Residual of Intention to Adopt ...................................................................... 175

XV

CHAPTER ONE
Introduction

Cancer is defined as “a group of diseases characterized by uncontrolled growth
and spread of abnormal cells” (American Cancer Society, 2002, p. 1). Among several
types of cancers, some types, such as skin cancers are preventable or curable if found in
their early stages. The major factor of developing skin cancer is associated with excessive
exposure to ultraviolet (UV) radiation. In order to reduce the incidence of skin cancer, the
American Cancer Society recommends limiting or avoiding sun exposure during the
midday hours, from 10 am. to 4 pm. Additional recommendations are to cover the body
by wearing a hat, sunglasses, long-sleeved shirt and long pants, and by applying
sunscreen when outdoors (American Academy of Dermatology, 2002c; American Cancer
Society, 2002).

Although skin cancers are “the most easily and successfully treated human
cancers” (Amir, Wright, Kemohan, & Hart, 2000), skin cancer accounts for nearly 50%
of new cancer cases, and it is estimated that more than one million new cases will be
diagnosed in the US. in 2003 (American Academy of Dermatology, 2002a). Due to
melanoma, which is the most serious type of skin cancers, one person dies every hour,
and melanoma is the fifth most common cancer in men and the seventh in women
(American Academy of Dermatology, 2002a). Accordingly, outdoor enthusiasts, such as
runners or golfers, should focus on sun protective practices, because their outdoor
activities could lead to extensive sun exposure during peak sun hours (Gravel, 1997;

Liffrig, 2001).

As Gravel (1997) indicated, golfers are more at risk than the average person
because they often spend more than five hours in the sun while playing a game of golf.
This study pays particular attention to golfers because they usually expose themselves to
the sun during the peak time and because they are free to choose their clothing as
compared with others in outdoor sports that require uniforms such as baseball or soccer.
Moreover, little effort is being made to target outdoor recreational groups regarding the
practice of sun safety with the exception of sunbathing habits (Newman, Woodruff, Agro,
& Mayer, 1996; Vail-Smith, & Felts, 1993). Therefore, analyzing golfer’s perceptions
about skin cancer risks and their current sun protective actions deserves attention.
Moreover, to understand golfers’ attitudes toward covering up their body while playing
golf would be important in developing educational information or improving the design
of golf clothing.

UV specialized shirts increase sun protection, since this garment is designed to
prevent UV radiation transmission to the skin. In addition, wearing UV specialized
clothing could be more convenient than using sunscreen for golfers since individuals
have to reapply sunscreen frequently while playing golf for effective protection during
the prolonged exposure. Actually, most skin cancer prevention campaigns are inclined to
focus on the use of sunscreen (Glanz, Lew, Song, & Cook, 1999), so it is helpful to
determine whether the alternative practices, such as use of sun protective clothing, are
equivalent or more effective. Analyzing golfers’ perceived characteristics of UV
specialized clothing would provide results that may assist product development
specialists to identify problems and to modify garments to accelerate promotion of UV

specialized clothing.

Very little research, however, has focused on the effectiveness of clothing, related
to the amount of UV radiation body-specific areas received, nor has site specific data
collected from the viewpoint of functional apparel design. Armstrong and Kricker (1993)
explain that all body sites are not equally at risk for melanoma. For example, the highest
incidence rate of melanoma for males is on the trunk, whereas for females it is the lower
limbs (Bulliard, 2000). In addition, depending on gender, ethnicity, and geographic
locations, people have different probabilities of developing skin cancer in exposed-body
sites. Identifying differences in the amount of UV radiation received by specific body
sites would be valuable to apparel designers in planning UV protective clothing.

Objectives and Research Questions

This study consists of two phases. In the first phase, field experiments were
conducted to analyze the cumulative UV radiation received by specific body sites. In
addition, market research provided background information about typical golf shirts and
UV specialized products. In the second phase, golfers’ perceptions about skin cancers and
attributes of UV specialized clothing were investigated using quantitative methods. The
two phases are discussed in separate chapters that include literature review, research
methods, and discussion of results. This study is a part of NC-170, “mediating exposure
to environmental hazards through textile systems,” directed by Dr. Slocum and was
financially supported by the Michigan Agricultural Experiment Station.

Phase I — UV Exposure of Golfers and Market Research of Golf Clothing
Objective 1 : To determine the cumulative exposure of different body sites to UV

radiation during golf rounds.

 

 

Research question 1: To what extent do golfers receive UV radiation at 16 upper body
sites?
Research question 2: What parts of the upper body receive the most UV radiation while
playing golf?
Objective 2: To evaluate the effects of clothing in reducing UV radiation
transmissions from sunlight.
Research question 1: To what extent do golf shirts reduce UV radiation, e.g. are there any
differences in the absorbance values between over the shirts and under the shirts?
Research question 2: Do the differences in absorbance values under the shirt vary with
the types of shirts, UV specialized shirts vs. regular golf shirts?
Objective 3: To provide background information of golf clothing available
commercially in the selected market segment.
Research question 1: What is the standardized or normative look of golfers in magazines?
Research question 2: What kinds of traditional clothing are available for golfers in local
pro golf stores?
Research question 3: To what extent are UV specialized products commercially available
in the local stores?
Phase II — Perceived Beliefs about Health and Attributes toward an Innovation
Objective 4: To assess the state of beliefs and current behaviors regarding sun
protection of golfers and examine the relationships between the two.
Research question 1: To what extent do golfers perceive susceptibility to and severity of
skin cancer, benefits of sun protective action, barriers to taking action, and what sun

protective practices do they engage?

 

Research question 2: Which variables in the health belief model contribute most to
explaining their current sun protective behaviors?

Objective 5: To examine whether perceived attributes of a UV specialized shirt
are related to golfers’ intention to adopt it.
Research question 1: To what extent and how is each attribute of UV protective clothing
related to the intention to adopt?
Research question 2: What is the relative importance among attribute variables in
predicting the intention to adopt UV specialized shirts for the potential adopters?

Objective 6: To investigate the relationships among beliefs about sun protection,
current sun protective behavior, perceived attributes of a UV specialized shirt, and the
intention to adopt a UV specialized shirt.
Research question 1: How are beliefs about skin cancer associated with perceived
attributes of a UV specialized shirt as an innovation and the intention to adopt it?
Research question 2: Which variables in the final model contribute most to explaining the
intention to adopt the UV specialized shirt?

Rationale for the Study

This study uses a holistic approach to examine UV radiation exposure of golfers,
normative looks of golfers in the market, golfers’ sun protective beliefs and current
practices, golfers’ perceived attributes of a UV specialized shirts, and the intention to
adopt the UV specialized shirt. Knowing what factors influence golfers’ current
behaviors would be helpful for designing educational messages. Sun safety messages.
could be tailored so as to influence their lifestyle changes. Analyzing perceived

characteristics of the innovation and cumulative exposure to the specific body sites would

provide useful information for apparel designers to develop UV protective clothing.
Overall, the two phases of this study would supplement each other by enriching our

understanding of the general perspective of golfers regarding UV protection and sun

safety issues.

CHAPTER TWO: PHASE I — UV EXPOSURE
Literature Review
Major Types of Skin Cancer

There are three forms of skin cancers. Basal cell carcinomas (BCC) often appear
on the head, neck and hands and are “the most common skin cancers found in fair-
skinned persons” (American Academy of Dermatology, 2002c, 1[ 10). These tumors grow
slowly and do not spread easily to other parts of the body, but can go through the skin to
the bone and cause serious damage to the local area (American Academy of
Dermatology, 2002c; Keesling & Friedman, 1987).

Squamous cell carcinoma (SCC) is “the second most common skin cancer found
in fair-skinned persons” and is typically found on the rim of the ear, the face, the lips and
mouth (American Academy of Dermatology, 2002c, 11 12). This type is more harmful
than BCC because “it can spread easily to internal organs and lymph glands” (Keesling &
Friedman, 1987, p. 478). BCC and SCC were categorized as nonmelanoma skin cancer
(Keesling & Friedman, 1987) and more than 95 percent of them are curable if detected in
the early stages (American Academy of Dermatology, 2002a).

The third type of skin cancer is malignant melanoma, the most serious form of
skin cancer. Melanoma is “characterized by the uncontrolled growth of pigment-
producing tanning cells”; melanoma accounts for 77% of all deaths caused by skin cancer
(American Academy of Dermatology, 2002b, 1} 1). This type of skin cancer is estimated
to be found in about 91,900 new cases in 2003, and “one in 67 Americans have a lifetime
risk of developing invasive melanoma, a 2000% increase from 1930” (11 2). Caucasians

have an incidence rate ten times higher than other ethnic groups diagnosed with

melanoma, but melanoma is usually treatable if found in the early stages (American
Academy of Dermatology, 2002b).

In 2002, it was estimated that more than one million new skin cancer cases would
be diagnosed in the US, comprising about 80% of BCC, 16 % of SCC, and 4% of
melanoma (American Academy of Dermatology, 2002a). Empirical research suggests
that skin cancer is highly associated with cumulative sun exposure (Johnson &
Lookingbill, 1984; Newman, et al., 1996) as well as behavioral factors such as outdoor
work or leisure activities (Carmel, Shani, & Rosenberg, 1994). Garbe and Buettner
(2000) noted that developing malignant melanoma and BCC were related to sun exposure
in youth and high intermittent sun exposure during recreational activities, whereas SCC
was more attributable to cumulative long-term sun exposure. Due to these factors, skin
cancer is one of the most preventable cancers (Marlenga, 1995), and there has been
increasing attention paid to skin cancer prevention promotions associated with sun
exposure habits.

Ultraviolet Radiation (U VR)

Sunlight wavelengths span from 270 to 5,000nm and the major wavelengths in
cutaneous studies were from 270 to 800nm that consisted of visible light (400 to 800 nm),
Ultraviolet A (UVA: 320 to 400nm), Ultraviolet B (UVB: 290 to 320 nm), and
Ultraviolet C (UVC: 200 to 290 nm) (Davis, 2000). Visible light between 400 and 800nm
is “the white light we can see” (Davis, 2000, p.79) and has no harmful effect on skin
(Liffiig, 2001). UVC is mostly absorbed by atmospheric ozone so that it does not reach

human skin (Davis, 2000). Therefore, UVA and UVB are the most significant factors in

damaging skin.

UVA accounts for 95% of all solar UV radiation and penetrates into the skin
deeper than UVB. UVA is the major cause of aging skin and tanning beds use a similar
wavelength (Davis, 2000; Liffrig, 2001 ). UVB is the major cause of skin cancer and
sunburn because it is “highly reactive with macromolecules in the skin” (Liffrig, 2001, p.
196). UVA and UVB are more intense when “the sun is directly overhead” (between 10
am. and 2 p.m.), “at higher altitudes”, and “on hot and humid days” due to lower density
of air (Davis, 2000, p. 81). The depletion of ozone by chemical reactions and the
Antarctic ozone hole also increase the levels of UVR at the earth’s surface, resulting in an
increase in new cases of skin cancer (Diffey, 1992; Herlihy, Gies, Roy, & Jones, 1994).

U VR Received by Occupation and Activity

Many outdoor occupations and leisure activities expose people to the sun while
sunlight is direct and intense. Some empirical studies have identified which occupations
receive higher levels of UVR. Larkd and Diffey (1983) compared UVB doses received by
different types of indoor and outdoor workers in Sweden during two separate months.
They found that people who were engaged in outdoor occupations such as the military or
fishermen received UVB doses about 3.75 to 6 times more than did those in indoor
occupations. However, the relative differences in UVB doses between the two groups
were minimized during the summer season, indicating that indoor workers seemed to
devote a large amount of time in their summer vacations to sun seeking activities
compared with other seasons.

Gies, Roy, Toomey, MacLennan, and Watson (1995) compared the UVR
exposure for two body locations, the top of the shoulder and the center of the chest,

among three occupational groups: 16 physical education teachers, 11 grounds staff, and 8

lifeguards during five consecutive weekdays. Overall, physical education teachers
received the highest UVR exposure while lifeguards received the lowest. According to
subjects’ daily reports of outdoor activities, lifeguards usually stayed in the beach shelters
during their duty times, so they were relatively less exposed to the UVR. In comparison
with the chest site, the shoulder site reached a 1.3 fold to 2 fold greater exposure to the
sun across the three occupations. The authors concluded that the amount of the UVR
exposure to the three occupational groups exceed the standard exposure limits to a great
extent, indicating the importance of educating outdoor workers regarding precautions
against sun exposure.

One recent study compared the amounts of UVR received by various body
locations for two occupations. Vishvakarman, Wong, and Boreham (2001) examined 12
post mail delivery persons (PMDP) and 6 physical education teachers (PET). The order
of body sites, from highest to lowest UVR for PMDP was hand, back, and chest, whereas
for PET, it was t0p of the head, the back, shoulder, and thigh and chest. This study
concluded that the different order and amounts of UVR exposure to the body sites were
due to the nature of their work and amount of time they spent outside across four seasons.
Their findings support the need for obtaining site specific exposure information by
occupation and/or activity.

Exposure also varies with outdoor activities. Diffey, Larko, and Swanbeck (1982)
measured the amounts of UVB radiation received by participants in different activities in
different geographic locations, and the maximum amount of UVB available. Sunbathers
on a beach at Canary Islands (28°N latitude, February, clear and sunny, 11:30-12:30)

received 80% of the available UVB, while sunbathers at Corfu (39°N, September, hazy,

10

 

(IQ

C?

H:

512

12:00-13:00) received 69% of UVB. The proportion of UVB exposure received by skiers
at Gaschurn, Austria (47°N, March, clear and sunny, 8:30-15:30) was 23 %, whereas the
proportion received by sailors at Gothenburg, Sweden (58°N, July, sunny, 9:00-20:00)
was 15%. Their study shows how the UVB doses received by subjects can vary
contingent upon subjects’ activities, geographic location, season, weather condition (clear
and sunny, light cloud, snow storm), or time of a day (8:00-9:00 vs. 12:00-13:00).
Holman, Gibson, Stephenson, and Armstrong (1983) compared UVR received at
five body sites on subjects in nine outdoor activities. The proportion of ambient UVR
received from highest to lowest, averaged across body sites by activity was swimming at
the ocean, boating, sunbathing, hiking, golf, fishing, tennis, pool swimming, cricket, and
gardening. The researchers also reported that the middle of center back (thoracic spine)
and the dorsum of the hand were more exposed than the cheek, anterior surfaces of the
mid-thigh, or posterior surface of mid-calf. The researchers pointed out the different
levels of sun exposure among various outdoor activities. For instance, hiking, golf, and
fishing were associated with relatively more UVR exposure than tennis or gardening,
perhaps due to open areas or proximity to water. That is, environmental factors also
influence the UVR dose received by individuals while they are exposed to the sunlight.
Herlihy, et al., (1994) measured the UV doses received at several anatomical sites on 94
subjects in six outdoor activities in Hobart, Tasmania. Tennis, sailing, and golf were the
three activities that had the highest proportions of ambient UVR, 1.00, 0.87, and 0.67
respectively, while gardening, walking, and swimming had the lowest proportion of
ambient UVR, 0.26, 0.31, and 0.49 respectively. Although the golf course used in

Herlihy, et al.’s study was shaded with trees, compared with other activities, golfers were

11

exposed to relatively higher amount of the UVR doses due to the long period of time

exposed. The researchers also indicated that the time of a day subjects were engaged in

activities affected the amount of UVR. Their findings were different from Holman, et

al.’s study (1983) in the order of the proportion of ambient UVR obtained by outdoor

activities. The differences in the two studies could be due to the location of tennis courts

or topography of golf courses.

Table 1. Summary of Literature regarding UVR Received by Activity and Body Sites

 

 

 

 

 

 

 

Source Comparisons UVR received in order
Larkd & Diffey
Occupations Outdoor > indoor workers
(1983)
Gies, et al., Occupations PET > grounds staff > lifeguards
(1995) Body sites Shoulder > chest
Vishvakarman, PMDP: Hand > back > chest
Body sites
et al., (2001) PET: Top of hat > back > shoulder > thigh > chest
Diffey, et al., Recreations Sunbathing at Canary Islands > at Corfu
(1982) Skiing at Austria > Sailing at Sweden
Recreations Swimming at the ocean > boating > sunbathing >
Holman, et al., hiking > golf > fishing > tennis > pool swimming >
(1983) cricket > gardening
Body sites Back, dorsum of hand > check, thigh, calf
Recreations Tennis > sailing > golf > swimming > walking >
Herlihy, et al.,
gardening
(1994)
Body sites Shoulder > back > thigh > calf > chest > hands > cheek

 

 

12

In Herlihy, et al’s study (1994), the shoulder received the highest prOportions of
ambient UVR (65%), followed by the back (55%), thigh (36%), calf (33%), chest (30%),
hands (26%), and check (13%). In the study by Holman, et al. (1983) only two body sites
of golfers were recorded, and the hand received higher proportions of ambient exposure
(55%) than the check (24%), which was consistent with findings of Herlihy, et al.’s study
(1994) in terms of order, but not in terms of the proportion of ambient UVR obtained.
Table 1 summarizes the previous studies, regarding the extent of UVR received by
subject’s activities and body-specific sites.

Body-Specific Sites of Skin Cancer Diagnosed
A few researchers identified specific body sites of skin cancer. Among 242

Caucasian residents who were diagnosed with BCC in Kauai, Hawaii, the most frequently
diagnosed body sites were the head and neck, followed by the trunk and limbs (Reizner,
Chuang, Elpem, Stone, & Farmer, 1993). The mean age of the patients is 56.5 years, and
males have higher incidence rates (almost 2 times) than do females. The Kauai residents
have higher incidence rates on the trunk, especially the back, than do Minnesota residents
in their study. This was attributed to the warm weather in Kauai that lead to limited skin
cover during most outdoor activities.

Elwood and Gallagher (1998) investigated the relationships between the body
sites and patients’ age. One-thousand-thirty-three melanoma patients in British Columbia
between 1991 and 1992 were divided into two age groups (under age 50 and over 50) to
compare their incidence rates of melanoma tumor per unit area of specific body sites.
Among both males and females under age 50, the back was the highest in the incidence

rates, while the face was the highest in both sexes at ages over 50. In addition, females

13

over 50 showed higher incidence rates on the upper arm and the leg than other groups.
The researchers confirmed the hypothesis that melanoma patients under age 50 had
higher incidence rates on body sites that were exposed to the sun intermittently, while
incidence rates for patients over 50 were more likely to be associated with body sites
receiving continuous sun exposure.

Buettner and Raasch (1998) surveyed 3,536 patients of the three major types of
skin cancer between December 1996 and December 1997 in Australia. BCC was mostly
found on the face and back, while SCC was diagnosed on the face, legs, and upper limbs.
The highest incidence rates of melanoma in males were in the back, neck, and shoulder,
while the highest incidence rates in females, by body sites, were the neck, back, and face.
The researchers highlighted the positive relationships between the highly sun-exposed
body sites, such as the face and the high risk of developing skin cancer. Also, they
pointed out that clothing habits would influence developing skin cancers at different body
sites. For example, males have higher incidence rate than females on the back, while
females have higher incidence rates than males on the legs due to type of clothing worn.

Bulliard (2000) examined 16,117 incidents and 3,150 death records of melanoma
between 1968 and 1999 in New Zealand. Overall, the three body sites with the highest
incidence and death rates of melanoma were trunk, upper limbs and lower limbs. In
comparison with females, males had higher incidence rates on the body sites of trunk,
scalp/neck, and ears, whereas females had higher incidence rates than males on the lower
limbs and upper limbs. However, males had higher death rates than females in all body
sites except for lower limbs. These findings are consistent with previous studies, and

indicate the importance of clothing habits and sun exposure behaviors. Table 2

14

summarizes the previous studies regarding body-specific sites diagnosed with skin

cancer.

Table 2. Summary of Literature regarding Body Specific Sites of Skin Cancer

 

Source

Sample

(Years sample was collected)

Types of skin cancer

& specific body sites diagnosed

 

Reizner, et al.,

242 Caucasian residents in

BCC a: Head & neck, trunk, limbs

 

 

 

(1993) Kauai, Hawaii (1983-87)
Elwood & 1033 melanoma patients in Melanoma 3|: Under 50 - Back
Gallagher (1998) British Columbia (1991-92) Over 50 - Face
BCC a: Face, back
SCC 8: Face, leg, upper limbs
Buettner & 3,536 skin cancer patients in
Melanoma 8: Back
Raasch (1998) Australia (1996-97)
Male-Back, neck, shoulder
Female-Neck, back, face
16,117 incidents and 3,150
Melanoma 8: Trunk, upper limbs,
Bulliard (2000) death records of melanoma in

 

New Zealand (1968-99)

lower limb

 

3 Both males and females

Golfers and Skin Cancer

Recently, the American Academy of Dermatology with the support of the

National Golf Foundation and the Golf Channel national cable network (American

Academy of Dermatology, 2000) has targeted golfers with a skin cancer educational

Campaign. Because of their long time exposure to the sun during peak sun hours,

15

dermatologist MacDougal of Manhattan Beach, California indicated that “it puts golfers
especially at risk” (Gravel, 1997, p. 117). The American Academy of Dermatology also
has worked with the famous professional golfer, Greg Norman as a role model of sun
safety and protection (American Academy of Dermatology, 2001). The campaign focused
on “the odds of getting skin cancer (1 in 5) versus the odds of getting a hole in one (1 in
12,000) during your lifetime” (11 3).

Hanke, Zollinger, O’Brian, and Bianco (1985) stressed that higher risks of
developing skin cancer for professional and amateur golfers were due to the cumulative
exposure during their lifetime. When comparing female professional and amateur golfers
in developing basal cell carcinoma (BCC), 4 of 51 professional golfers (7.8%) and 11 of
142 amateur golfers (7.7%) were diagnosed (Hanke, etal., 1985). Significant factors
related to BCC cases were original hair color (brown and red) and color of eyes (green
and blue). The body sites where BCC was most frequently observed from highest to
lowest were nose, chin; elbow; cheek, eyelid, eyebrow, temple; and thigh. With respect to
use of sunscreen, about 63% of professional golfers used sunscreen regularly while
49.3% of amateur golfers used it, although the differences were not statistically
significant. With respect to the mean hours of golf played per day, professional golfers
spent much more time, 6.2 hours under the sun, than amateur golfers did, 2.1 hours. In
comparison with professional golfers, amateur golfers might ignore their intermittent sun
exposure during their golf rounds, resulting in less use of sunscreen, which would lead to
similar percentage of diagnosing BCC on amateur golfers.

In summary, previous studies suggest that different occupations or activities yield

different amounts of UVR exposure to various body sites (see Table 1). Also, different

16

types of skin cancer were diagnosed at diverse body locations (see Table 2). Perhaps
several confounding variables affect the UVR received by individuals: geographic
location, topography of the place, time of year, time of a day, period of exposure, and
differences in personal habits. In order to offer better protection from UVR, it is
necessary to analyze which body sites are exposed to UVR the most when traditional
clothing is worn while playing golf. Results of this analysis will assist in generating the
design criteria for developing UV protective garments.

Sun Protective Clothing

Use of sunscreen sometimes leads to inadequate sun protection to the body.
People use an improper amount of sunscreen, use sunscreen with an insufficient sun
protection factor (SPF), fail to cover exposed body sites, and do not sufficiently reapply
sunscreen during prolonged exposure (Garbe & Buettner, 2000). Thus, wearing sun
protective clothing would be a much simpler and more effective way to practice sun
safety than using sunscreen (Garnbichler, Altrneyer, & Hoffrnann, 2002).

Sun protective clothing indicates garments that provide adequate sun protection
for all day exposure and have a SPF of 15 or higher for fabrics (Diffey, 2000; Liffrig,
2001). Protective quality of fabrics varies with the specific characteristics of fabrics:
Fabric porosity, fiber types (cotton, wool, polyester, etc), color, weight, and thickness
(mass per unit area).

Fabric porosity is related to the fabric construction (woven vs. knit). Woven
fabrics generally provide better sun protection than knit fabrics, since knit fabric has
larger spaces between yarns, resulting in greater UVR transmission than woven fabric

(Capjack, et al., 1994). Fabric stretch also provides less protection by increasing fabric

17

porosity (Gambichler, et al., 2002). In terms of fiber types, polyester fabrics provided 3 to
4 times higher SPF values than cottons did if they were similar in construction, although
cotton fabrics were much more comfortable to wear in summer (Davis, Capjack, Kerr, &
Fedosejevs, 1997). However, polyester fabric was less effective in preventing UVA than
UVB transmission.

The color of a fabric could enhance SPF, since “the absorption band for some
dyes extends into the UVR spectral region so that such dyes act as UVR absorbers”
(Curiskis & Pailthorpe, 1996, p. 16). Dark and bold colors provide better sun protection
than white, light, or pastel colors (Osterwalder & Rohwer, 2002). Fabric weight or
thickness is a significant factor in influencing SPF: The higher the weight per unit area,
the better UV protection (Capjack, et al., 1994). Other factors such as bleach and wetness
also influence SPF (Gambichler, et al., 2002). Unbleached cotton or silk provides better
protection from UV radiation than their bleached counterparts “due to UV-absorbing
natural pigments and other impurities” (p. 18). Wet fabric reduces the values of SPF
significantly compared with a dry condition, especially for cotton fabrics, because the
presence of water in the spaces between yarns scatters light, increasing UV transmission.
In addition, repeated laundry with a detergent containing optical brightening agent would
increase SPF of some types of fabrics, such as cotton or cotton-and-polyester blend
fabrics (Gambichler, et al., 2002).

For better UV protection, UVR absorbers have been incorporated in the
construction process of fiber or in the finishing process of fabrics (Capjack, etal., 1994).
Clothing with UVR absorbers provides a 30 or higher SPF and is advertised as medical

devices (Liffrig, 2001). However, as Capjack et al. mentioned (1994), the usefulness of

18

UVR absorbers in increasing sun protection values in comparison to other “normal”
fabrics have not been consistent, although the price of UVR-absorber specialized
garments may be higher than normal garments. Most regular clothing provides some
basic level of sun protection. About 85% of 5000 fabrics tested at Australian Radiation
Laboratory provided SPF of more than 20, and most summer fabrics provided SPF of 10
or higher (Diffey, 2000).

For this study, a UV specialized garment is defined as a garment that is designed
to maximize UV protection by using fabrics that are specialized with UVR absorbers.
Very little research has focused on the effectiveness of UV specialized garments in terms
of UV radiation received by body-specific areas, compared with other normal garments.
Hence, this study examined the cumulative exposure of different body sites and effects of

different types of fabrics in preventing UV radiation transmissions during golf rounds.

19

Methodology
Preparation of the Polysulphone Film Dosimeter

Ultraviolet B (UVB), ranging from 280 nm to 315 nm is a significant factor in
causing skin cancer (Liffring, 2001). A number of researchers (Diffey, 1989; Gies, et al.,
1995; Herlihy, et al., 1994; Larkd & Diffey, 1983) have used polysulphone films to
measure personal exposure to UVB after Davis, Deane, and Diffey introduced it in 1976.
When this film is exposed to UVB, it degrades (depolymerizes) with a change in its
absorbance that is measurable at a wavelength of 330nm (Holman, et al., 1983;
Rosenthal, Lew, Rouleau, & Thomson, 1990).

The Composite Materials and Structures Center, Michigan State University made
the film used in this study. Polysulphone crystals were dissolved in CHzClz
(Dichloromethane). The solution was poured onto a flat glass plate (approximately 11
inch x 8.5 inch) and spread evenly by rolling a glass bar over the liquid several times.
When the solvent vaporized in approximately three minutes, the film was removed from
the glass plate. The film is colorless, transparent, very thin and fragile. The process was
repeated to produce the quantity needed.

The change of UVB absorbance (AA) was measured at 330nm. The change was
determined by measuring each film badge before and after exposure to UV radiation
(UVR) in the UV spectrophotometer. For this study, a Perkin Elmer Lambda 900 UV
spectrophotometer with a beam of 32mm X 13mm with a central aperture of 18mm x
7mm was used. This beam size is larger than previous studies (Larkd & Diffey, 1983;
Rosenthal, et al., 1990; Vishvakarman, et al., 2001), so the size of film badges was

increased. The film was cut into 35mm X 13mm rectangles for the badges.

20

 

To measure absorbance, first, each film badge was cleaned with isopropyl alcohol
to remove the contamination from the surface which might increase the absorbance value
(Rosenthal, et al., 1990). Second, the film badge was stood against the beam by matching
the center of the film with the center of the beam aperture. Finally, in the
spectrophotometer, a light passed through the film, and the absorbance value of an
unexposed film badge was recorded at 330nm. The film was measured again after
exposed to UVB. Then the change in absorbance (AA) of each film badge was calculated
by subtracting the value of the unexposed film from the value of the exposed.

The film badges for each subject and ambient exposure were stored in separate
envelopes which were put into separate small boxes. Those small boxes were put again in
a large shoe box for the purpose of blocking UVR during storage.

Pretests of the Polysulphone Films

After preparing the films, the ambient level of radiation during six hours was
measured on August 10, 2002 in order to test the measurement capacity of the film
badges to determine if they were practical. Twelve film badges were prepared and placed
on papers on the top of a car that was located in an unshaded, open area. It was a sunny
day, the temperature was over 80 °F, and the UV index was 8. Between 10:30 am. and
4:30 pm, two film badges were removed from the car every hour. Four out of 12 badges
were damaged, so four of six time intervals are represented by only one sample. Figure 1
shows the distribution of the absorbance changes in the first pretest. Afier four hours of
exposure, the absorbance changes of the film badges were decreased. Because only one
sample represented between four hours and six hours exposure, the researcher decided to

increase the numbers of badges in the second pretest. In addition, because the films were

21

made by hand, the absorbance values of unexposed film badges varied, which might
cause inconsistent distributions of the data in Figure 1. Hence, the range of variability in
absorbance values of unexposed film badges was limited to 0.035 variation from
minimum to maximum.

Figure 1. Absorbance Changes of Ambient Dosimeters by Hours Exposed in Pretest

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I T
0.3
m .
,3; 0.25 t I i
5 . i
8 0.2 I +
§ 0.15 j @
g 0.1 I
0.05
0 4v 1 7 g I 9
i Hours of Exposure
1
I I0 Samp|e1 - Sample2 I

 

 

A second pretest was conducted on August 18, 2002, to examine the absorbance
changes by hour on a very sunny day with a temperature of about 80 °F, and a UV index
of 7. Twenty four film badges were prepared. The absorbance values of unexposed film
badges between 0.08 and 0.105 were selected. Beginning at 8 am, four badges were
exposed to the sun for an hour. They were then removed and four more badges were
exposed for an hour. Badges were changed each hour until 2 pm. Figure 2 shows that
the values of the absorbance increased steadily over time. Hence, the researcher decided

to use the polysulphone films as a UVR dosimeter for this study.

22

Figure 2. Absorbance Change in One Hour of Ambient Exposure by Time of Day

 

0.1600
0.1400 .1

0.1200 L L—
0.1000 4 '
0.0800 f —~ -~— — __..
0.0600 , i —— ————-———» -—- —~. ‘
0.0400 ,( ——- ‘ I
0.0200 4‘ ‘

0.0000 . - + . .
3.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00

1
I
l
1 Time
1

 

 

 

 

 

 

 

 

Absorbance Change

 

 

 

 

 

 

 

[0 Sample 1 I Sample 2 A Sanpje 3 X Sanple 41

 

 

Experiment I

For experiment I, a total of 196 film bandages were constructed to measure UVB
doses: 84 bandages for 12 body locations under clothing and 112 bandages for 16
locations over clothing or areas directly exposed to the sun for seven subjects. In
addition, 30 film badges were prepared to measure the ambient level while subjects
played golf for six hours. To minimize the variation among film badges, films with
absorbance values outside the 0.035 range were excluded.

To prepare a personal dosimeter, both sides of a film badge were attached onto an
adhesive bandage by using double sided tape and/or transparent tape. The aperture of the
film badge was kept at 23mm x 13mm. Then the bandage was labeled with an
abbreviation for a subject and a specific body location. A code identified films placed
over and under clothing. For instance, the film bandage attached at the front neck over

the clothing for subject A was labeled as AlNF whereas the film bandage attached at the

23

same location but under the clothing was as AONF. The sixteen locations of the upper

body included neck (front & back), shoulder (lefi & right), chest (left & right), back (left

& right), anterior arm (left & right), posterior arm (left & right), anterior forearm (left &

right), and posterior forearm (left & right). Specific sites are marked in Figure 3.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘ I X
i

AAR « AAL PAL

CR CL
AFR , i 1 ‘PFL

AFL
I

chi:

FRONT BACK

 

 

 

 

NF
NB
CR
CL
SR
SL
BR
BL

PAR

PAL

AAR

AAL

PFR

PFL

AFR

AFL

 

Neck-Front
Neck-Back
Chest-R
Chest-L
Shoulder-R
Shoulder-L
Back-R
Back-L
Posterior
Arm-R
Posterior
Arm-L
Anterior
Arm-R
Anterior
Arm-L
Posterior

F orearm-R
Posterior
Forearm-L
Anterior
Forearm -R
Anterior
Forearm -L

 

 

 

Figure 3. 16 Body Locations for Film Badges in Experiments
(Modified from Buettner & Raasch, 1998, p. 588)

24

 

Experiment 11

The Polysulphone films for personal dosimeters and for ambient tests were
prepared in the same way as Experiment I. To control the variance, the unexposed films
with absorbance values only between 0.073 and 0.103 at 330nm were included. A total
of 254 film bandages were constructed for Experiment 11: 224 bandages for 16 body
locations under and over clothing for seven subjects (16 x 2 x 7) and 30 film badges for
ambient tests (5 x 6 hr) with the badges labeled and stored in the same way as in
experiment one.

Sampling Method

The population of this study is golfers who are 18 years of age or older, living in
Michigan. From this population, 10 volunteers participated in the experiments. The
sample was obtained by snowball sampling method. The researcher contacted three male
golfers by phone and asked them to participate in this study after explaining the study
purpose, and elements of informed consent. Also, the researcher explained the incentive
that participants would play 18 holes of golf without charge on the day of the experiment
and receive a gift certificate to play another day. When the subject voluntarily agreed to
participate, the researcher scheduled a date for the experiment and asked them to
recommend others with whom they played golf. Although participants in experiments
might be diverse in height, weight, body size, or age, it was assumed that those variations

would not bias the data for UV exposure to the body.

25

Data Collection
Experiment I

The first experiment was conducted at Forest Akers Golf Course East, at
Michigan State University on August 28, 2002. The day was mostly sunny with a
temperature in the mid 705 and a UV index of 7.

The researcher met subjects at the golf course 30 minutes before tee time. She
explained the experimental procedure, and obtained their written informed consent. In
addition, specific features of subjects’ clothing such as fiber content, fabric structure, hue,
shade of hue, and style of shirt were recorded on Table A1 in Appendix A because these
characteristics influence protection from UVR transmission (Davis, et al., 1997;
Gambichler, etal., 2002). Based on Figure 3, twelve bandages were placed on the skin of
each subject on 12 body locations under the shirts and 12 bandages were placed on the
surface of the shirts close to, but not directly over the badge that was under the clothing.
Four bandages were attached to the skin of both right and left forearms that were directly
exposed to the sun, because all subjects wore short-sleeve shirts. The subjects were
exposed to the sun for approximately four and half hours: The first team of four subjects
was exposed to the sun between 10:20 am. and 3 pm, and the second team of three
subjects was exposed from 11:10 am. to 3:20 pm. After they finished playing golf, the
film badges were retrieved and restored in double boxes as previously described.

To measure ambient exposure, 30 film badges were placed on papers on the roof
of a car. The car was located in an open, unshaded area of the parking lot of the same
golf course. Thirty badges were initially exposed when the subjects started playing golf.

Five were removed each hour during the time participants were playing. Although the

26

play was finished within five hours, the researcher obtained a total of six hours of
exposure for the maximum length of play. Again, ambient films were stored in double
boxes after retrieval.

Of the total 196 film badges for the subjects, nine were lost (4.6%) and 44 were
damaged or torn (22.4%). Some loss and damage was probably due to the fact that the
subjects had to carry the golf bags on the shoulder and the motion of the arm caused
rubbing against film surfaces. Two out of 30 films for ambient measurement (6.7%)
were damaged. The absorbance value of retrieved film badges was measured by
spectrophotometer between 24 hours and 36 hours after exposure, since exposed film
continued to degrade even though it was stored in UVR black boxes (Diffey, 1989).
Experiment 11

The second experiment was conducted at the same golf course of Michigan State
University on October 11, 2002. Since the subjects had to wear long-sleeved UV
specialized shirts on the date of the second experiment, the researcher waited until the
weather was cooler to avoid any potential problem with heat stress. Four of seven in the
first experiment agreed to participate in the second experiment and to wear UV
specialized shirts.

The day was mostly sunny with the temperature in the mid 50 °F in the morning
and about 70 °F in the afiemoon; the UV index was 4. The researcher met the subjects at
East golf course 30 minutes before the tee time. After explaining the experiment purpose
and process and obtaining subjects’ written informed consent, the researcher placed 16
bandages under a subject’s shirt and 16 bandages on the surface of the shirt on the basis

of Figure 3. Four subjects who participated in Experiment I wore UV specialized shirts,

27

 

which were constructed to block approximately 97% of UV radiation. Three wore their
own shirts. All subjects wore long-sleeved shirts, but one subject rolled up his sleeves
during the experiment. Thus, one set of film badges on his forearms was directly
exposed to the sun. The researcher recorded the absorbance values of those four film
badges for this subject as being over clothing because those films were directly exposed
to the sun although they were attached on the skin. Then, the front and back sides of a
subject were photographed, and clothing features were recorded on Table A2, in
Appendix A. The first team of three subjects was exposed to the sun between 10 am. and
3 pm, and the second team of four subjects was exposed from 11 am. to 3:40 pm. The
ambient levels were measured in the same manner as the first experiment. In order to
reduce damage to the film badges on the shoulder and back sites, each participant used a
pull cart to carry the golf bag.

When the film badges were retrieved after the experiment, 22 (10%) were
damaged or torn and 2 (0.9%) were lost among 220 film badges for personal dosimeters.
Among 30 film badges for the ambient test, two were damaged (6.7%). The absorbance
value of exposed film badges were measured within 24 to 36 hours after film badges
were retrieved. Eventually, the data obtained from the experiments would help to
determine which body sites were exposed most among upper parts of the body and where

golfers need additional protection.

28

Data Analysis and Discussion
Experiment I

Descriptions of Subjects ’ Garments

All seven subjects wore short sleeved knit shirts (see Table A1 in Appendix A).
Four of them wore polo knit shirts (two or three buttons with knit collar), and one wore
wool-and-nylon blend knit tee in black color. The other two wore cotton tees, and one of
them wore a cotton vest over the tee. For this subject, the researcher attached film badges i
for under clothing measurement under the vest but over the tee. Six subjects wore 100% ‘
cotton shirts; three were white and three were of dark colors. All participants wore hats;
five wore baseball caps and two wore visors. Only two of seven subjects wore sunglasses
while playing golf. Six subjects were right handed, so they wore gloves on the left hands.
Absorbance Changes of Personal Dosimeters

Table 3 presents the mean and standard deviation of the absorbance changes (AA)
among different body locations and results of paired sample t-test between over and
under the shirts. The means of absorbance changes at each body location varied from
0.008 to 0.123 under clothing and from 0.062 to 0.254 over clothing. The amount of
exposure at different body sites was influenced by the individual’s movement, posture
and orientation of the body to the sun as they were playing golf. A comparison of the
mean scores shows that the most exposed site of the upper body over the shirts was the
right shoulder, followed by the right back, the left back, and the left shoulder. The back
of the neck and the front of the neck were the two most exposed areas under the shirts.
This probably indicates that the badges were exposed directly to the sun even though they

were intended to be covered by clothing. Although four of seven subjects wore t-shirts

29

with collars, the collars did not prevent exposure from UVB due to golfers’ stance while

playing. Bending at the waist, resulted in high absorbance mean scores at the back of the

neck both under and over the clothing (AAU = .123; AAQ = .159). For this reason, the

back of the neck had the highest absorbance value among the body locations under

clothing. The front neck under the clothing also had a high mean score, because subjects

unbuttoned the shirts exposing the badge under the shirt to the sun. The effects of

clothing cannot be compared for forearm sites since all subjects wore short sleeve shirts.

Table 3. Site-Specific UV Exposures and Paired Sample T-test between Under and Over

 

 

 

 

Clothing in Experiment I
Under Clothing Over Clothing
Body Locations (AAU) (AAO) t-(zegrte P
Mean (N )a SD Mean (N )a SD
NCCk'Front .072 (7) .043 .094 (7) .057 1.140 (6) .298
NCCk‘BaCk .123 (4) .059 .159 (7) .072 0.951 (3) .412
Chest-R° .019 (7) .025 .135 (5) .040 5.877 (4) .004**
Chest-Ld .008 (6) .002 .107 (4) .022 9.276 (3) .003**
ShOUIder'R .012 (6) .013 .254 (4) .046 11.662 (3) .001 **
Shoulder-L .010 (6) .009 .183 (4) .037 1 1.529 (3) .001M
BaCk'R .010 (5) .008 .205 (4) .062 4.067 (2) .055
3301914 .011 (4) .010 .185 (5) .041 2.629 (I) .231
POSterior Arm'R .006 (6) .008 .1 13 (6) .029 7.692 (4) .002'”
P05131101" Arm'L .018 (5) .027 .132 (5) .018 10.104 (2) .010*
Anterior Arm-R .012 (3) .019 .062 (3) .045 n.a.c n.a.
Amerior Arm'L .016 (4) .026 .105 (6) .027 2.766 (2) .1 10
Posterior For earm-R n.a. n.a. .088 (4) .028 n.a. n.a.
Posterior Forearm-L n.a. n.a. .091 (6) .052 n.a. n.a.
Anterior Forearm 'R n.a. n.a. .125 (7) .026 n.a. n.a.
Anterior Forearm 'L n.a. n.a. .107 (3) .035 n.a. n.a.
TOTAL .029 (45) .044 .143 (45) .064 9.452 (44) .000 ***

 

 

 

 

 

(N) ': Numbers of samples. (dfT’: Degree of freedom of t-test. R‘: Right. L d: Left. n.a.‘: cannot analde

*p < .05, "p < .01, *"p < .001

30

Paired sample t-test was used to analyze the differences between the absorbance
values of dosimeters placed over and under the clothing since the two values at the same
body sites of a subject were obtained under the same conditions and were dependent on
subjects’ movement. The mean difference between absorbance values over and under
clothing across all body sites was significant at the .001 level (t = 9.452, df = 44, p =
.000). Specifically, significant differences in absorbance changes between over and under
clothing were found at the right and left of the chest, the right and left of the shoulder,
and the right and left of posterior arm in Table 3 (p < .05). This finding indicated that the
regular golf t-shirts that were worn by the subjects reduced the UVR transmission to
some extent although the characteristics of t-shirts varied.

Interestingly, absorbance values for the back site were not significant statistically
either at the right or the left side although it approached significance for the right side of
the body. This may be due to the limited number of paired samples. The anterior arm was
not significantly different in the absorbance values between over and under clothing
while the posterior arm was significant. This finding might imply that swing movement
is less influential than movement from walking. Anterior arm is mostly exposed to the
sun while swinging the golf club, whereas the posterior arms as well as the shoulder sites
are always exposed to the sun during golf rounds.

In comparison of the absorbance values between left and right sides of the same
body location in Table 4, paired sample t-test showed that only the shoulder site over the
clothing was significant. The explanation of this difference is not readily apparent. It

might be attributable to carrying a golf bag on the shoulder or to measurement errors.

31

These findings might be useful information for golf wear designers, indicating that
golfers could be exposed to different amounts of UVR in relation to body sites.

Table 4. Comparison between Left and Right Sides of the Body Over Clothing and Paired
Sample T-test in Experiment I

 

 

 

, Left Side (AAO) Right Side (AAO) Maine
Body Location 1, p
Mean (N)" SD Mean (N)a SD (d9
Shoulder .186 (2) .033 .267 (2) .031 53.995 (1) 012*

 

 

 

 

 

(N) ': Numbers of samples. (dOb: Degree of Freedom of t-test.

‘p < .05.

A bsorbance Changes of Ambient Dosimeters

Table 5 presents the absorbance changes (AA) of five film badges exposed to ambient
conditions for each time interval, and Figure 4 shows the scatterplot of these data. The
absorbance values increased from one to four hours of exposure and then dropped for the
5th and 6th hours. The reduced value after four hours indicates that the polysulphone film
when exposed to UVB was saturated after four hours (Gies, et al., 1995; Herlihy, etal.,

1994). In addition, a few outliers were found as shown in Figure 4 and Figure 5.

Table 5. Absorbance Change of Ambient Dosimeter by Hours Exposed in Experiment I

 

 

Time Film 1 Film 2 Film 3 Film 4 Film 5 Mean SD
10:20-1 1 :20 .1184 .1158 .0961 .0968 .0929 .104 .012
10:20-12:20 .1917 X .1368 .1737 .2384 .185 .042
10:20-13:20 .3370 .3183 .3208 .1891 .2850 .290 .060
10:20-14:20 .3853 .4081 .3677 .4083 .3792 .390 .018
10:20-15:20 .4564 .3777 .3803 .3740 .2905 .376 .059
10:20-16:20 .1715 .3747 X .2476 .2815 .269 .084

 

 

 

 

32

"-4 .m-fl

Figure 4. Scatterplot of Cumulative Ambient Figure 5. Box Plot of Cumulative

 

 

 

     

 

 

 

 

 

 

Exposure by Hours in Experiment 1 Ambient Exposure by Hours
5
*
.44
g I o
.3, ° 0 '3” i *
D O 0
8 .21 D a 3’ ‘2‘
r: o a (U
6‘? ., 6
g a 8 1
.1 ‘ I g
E e
O
3
0.0 . . . f . . < °-° . . r . . .
o 1 2 3 4 5 6 7 1 2 3 4 5 6
Hours of Exposure Hours of Exposure

Z-Scores reported in Table 6 were calculated in order to find out “how many
standard deviations the observed score lies from the mean” (Shavelson, 1995 , p. 125).
The formula of z-score is:

X—X
2:
SD

 

where: X = variable X
)7 = mean of a set of scores on variable X

SD = standard deviation

Table 6. Z-scores for Ambient Dosimeters by Hours Exposed in Experiment I

 

 

10:20-12:20 .1547 X —l.1453 -.2708 1.2614
10:20-13:20 .7887 .4743 .5175 -l.6965 -.0841
10:20-14:20 -.2470 1.0228 -1.2223 1.0292 -.5827
10:20-15:20 1.3720 .0330 .0777 -.0302 -1.4525
10:20-16:20 -1.1553 1.2566 X -.2521 .1507

 

 

33

Based on z-scores and the box plot (Figure 5), three values (2 _>_ 1.3, bold, Table 6) were
deleted as outliers.
Comparison of Personal Dosimeters with Ambient Exposure

In order to compare the absorbance change of each body location with the
ambient level, the regression line of the ambient data was necessary. AS Holman, et a1.
(1983) indicated after some period of time a linear relationship did not exist in the
absorbance changes of ambient exposure, and this is consistent with the data in Table 6.
It happened because the polysulphone films were saturated at some point as UV radiation
dosages increased. That is, it could be considered that the absorbance changes of film
badges would increase as time increases if there were no saturation limits of the
polysulphone film. In Figure 4, the relationship was approximately linear between the
absorbance changes and one- to four-hour exposure. The researcher included only this
four-hour interval and cleaned the data by deleting three outliers in Table 6, and excluded
the data for the 5th and 6th hour due to the nonlinearity. The linear regression equation
was obtained by regression analysis. The result is reported in Table 7.

Y = 0.098X + 0.003119

where: Y = total amount of ambient exposure received by a subject

X = total amount of hours exposed to UVR

Table 7. Regression Analysis for Hours Exposed Predicting Absorbance Change on

 

 

 

 

Ambient Exposure
Variable B SE B B p
Constant .003 .015 .833
Hours exposed .098 .005 .978 ,00()* * *
r"*p < .001

34

An R2 of .956 (Adj. R2 = .953) indicated that the regression line was fitted well to
the sample data. The scatter plot and the fitted line of the absorbance changes of the
ambient UVB exposure between 10:20 am. and 2:20 pm. are presented in Figure 6.

Figure 6. Scatterplot of Cumulative Ambient Exposure and Regression Line for
Experiment I

 

 

 

 

8, .2«

c

m 0

.c

0 o

8 I

c 1

to

E

O

8

< 0.0 j ' . I qu=0.9557
O 1 2 3 4 5

Hours of Exposure

From this equation, it is possible to obtain the total corresponding amount of UVR
exposure received by golfers during play. The proportion of total ambient exposure to
each body site for each subject was obtained for comparisons among different body
locations. The two teams of golfers were exposed to the sun for different amounts of
time, so the total amount of the absorbance changes of ambient exposure was calculated
separately as follow:

Team 1: Y = 0.098 x (4.67 hr) + 0.003119 = 0.460779 2 0.461

Team 2: Y = 0.098 x (4.17 hr) + 0.003119 = 0.411779 2 0.412

Then the fraction of total ambient UVB received at each body site was obtained by
dividing the absorbance change of each location of a subject by the total ambient

exposure during the subject’s golf rounds.

35

Table 8 presents the mean proportion of the total ambient exposure for each body
location received by seven golfers during Experiment I. The data show that the amounts
of UVB radiation received under the clothing during playing golf ranged from 1.3% to
29% of the total ambient UVB, while subjects received 14.7% to 59% of ambient UVB
directly over the clothing. Since ambient UVB was measured at the same golf course in
an open, horizontal, unshaded area, the proportions indicate that golfers were not exposed
fully to UVB. Exposure was influenced by golfer’s position and orientation to the sun
while playing, golfer’s clothing characteristics, weather, and topography of the golf
course.

Table 8. Mean Proportion of Ambient UVR Received at 16 Body Sites in Experiment I

 

 

 

. Under Clothing Over Clothing
Body Locatrons Mean :6 SD sample Mean i SD sa‘T‘Ple
L srze srze
Neck-Front .166 :1: .097 7 .210 :L- .122 7
Neck-Back .289 i .152 4 .357 i .153 7
Chest-Ra .042 d: .053 7 .309 :1: .106 5
Chest-Lb .018 :1: .006 6 .238 :t .042 4
Shoulder-R .028 :t .031 6 .586 i: .119 4
Shoulder-L .023 :l: .020 6 .421 :1: .090 4
Back-R .022 d: .018 5 .469 i .128 4
Back-L .025 :1: .022 4 .421 :1: .098 5
Posterior Arm-R .013 i .018 6 .262 i .078 6
Posterior Arm-L .040 :1: .058 5 .301 :t .048 5
Anterior Arm-R .026 :t .041 3 .147 :1: .111 3
Anterior Arm-L .037 :L- .062 4 .237 :L- .065 6
Posterior Forearm-R n.a.° .196 i .059 4
Posterior Forearm—L n.a. .204 :h .112 6
Anterior Forearm -R n.a. .285 :1: .061 7
Anterior Forearm -L n.a. .241 :1: .071 3

 

 

 

 

 

R‘: Right. Lb: Left. n.a.c cannot analyzed

36

 

The lower proportion of ambient UVB under the clothing compared with the
proportion over the clothing indicated how important wearing a shirt would be to reduce
UVB exposure while exercising outside. In addition, body sites over the clothing that
were directly exposed to the sun were diverse in the proportion absorbed as well. This
can be explained by the effect of Shadows from body posture, movement of the golfer, or
individual’s different techniques. These data would be useful to compare with the data
collected under a different ambient condition such as the different UV index.

Experiment 11
Descriptions of Subjects ’ Garments

On the date of the second experiment, all subjects wore long-sleeved shirts. The
description of subjects’ garment characteristics are provided in Table A2 in Appendix A.
Four subjects were asked to wear beige UV specialized woven shirts that were provided
by the researcher. The shirts were purchased at Sun Precautions, Inc. They were made of
100% nylon, constructed to block approximately 97% of UVA and UVB radiation. Mesh
panels were inserted under the arms in the sleeve and side line, and in the back yoke of
the Shirt for ventilation. The other three participants wore their own shirts. Two wore
cotton knit sweatshirts in white and in navy color, and the other wore a brown corduroy
woven shirt with a collar. All subjects wore hats: six wore baseball caps and one wore a
visor. No one wore sunglasses during their play. Six subjects were right handed, so they
wore gloves on the left hands.

Absorbance Changes of Personal Dosimeters
Comparison between different types of shirts. For personal exposure

measurements, first, the researcher compared the effect of clothing on the absorbance

37

changes of film badges in the same locations under the clothing. Table 9 shows the
comparison of the mean differences between the absorbance changes under UV
protective Shirts and under subject’s own shirts.

Table 9. Site-Specific UV Exposures Under Clothing and Independent Sample T—test
between UV Shirts and Regular Shirts in Experiment 11

 

 

 

 

 

 

 

UV Shirts Regular Shirts
, Under Clothing Under Clothing
Body Locations ( A Auuv) ( A Aur) t-value p
Mean (N )a SD Mean (N )a SD
Neck-Front .0471(4) .019 .0476 (3) .041 .19 .99
Neck-Back .0535 (3) .018 .0000 (1) n.a.b -2.60 .12
Chest-Rc .0000 (4) .000 .0046 (3) .008 1.00 .42
Chest-Ld .0038 (3) .005 .0018 (3) .003 -.64 .56
Shoulder-R .0001 (4) .000 .0010 (2) .002 .89 .54
Shoulder-L .0014 (4) .003 .0000 (3) .000 -.85 .44
Back-R .0018 (4) .003 .0000 (3) .000 -.94 .39
Back-L .0011 (4) .002 .0000 (2) .000 -.67 .54
Posterior Arm-R .0010 (3) .002 .0000 (2) .000 -.78 .50
Posterior Arm-L .0027 (4) .005 .0000 (3) .000 -.85 .44
Anterior Arm-R .0037 (4) .006 .0014 (3) .003 -.65 .55
Anterior Arm-L .0008 (3) .001 .0000 (2) .000 -.78 .45
Posterior Forearm-R .0227 (4) .029 .0000 (2) .000 -1.03 .36
Posterior Forearm-L .0098 (3) .006 .0000 (2) .000 -2.39 .01*
Anterior Forearm -R .0114 (4) .008 .0000 (1) n.a. -1.25 .30
Anterior Forearm -L .0128 (2) .018 .0000 (2) .000 -1.01 .42

 

(N)': Numbers of samples. n.a.b: cannot analyzed R‘: Right. Ld: Left.
*p < .05.

There was no Significant difference in amount of UVB absorbed through two
different types of shirts, except for the left posterior forearm. The mean absorbance value
of the left posterior forearm was higher in UV Specialized shirts (AAUUV = .0098) than

regular shirts (AAUT = .0000). This difference could be attributed to subject’s clothing

38

habits (unfastening the sleeve cuffs), clothing characteristics (a slit in the sleeve), or a
measurement error. Table 9 suggests that UV specialized Shirts do not block UV radiation
more effectively than the regular fall Shirts. It might be due to the thickness and weight of
subject’s own shirts: The sweatshirts and a corduroy shirt were thicker than UV
Specialized shirts as well as summer knit golf shirts, thus they offered more sun
protection.

Comparison between under and over clothing. Table 10 presents the means and
standard deviations of absorbance changes in each body location. In addition, it includes
the results of paired sample t-test, comparing the absorbance values of the dosimeters in
the same location between under and over clothing. Since there were no significant
differences in the absorbance changes between two types of shirts except for the posterior
forearm (Table 9), the data of under UV specialized shirts were combined with the data
under regular shirts to increase the sample size. Significant differences of the mean
absorbance values between under clothing and over clothing were found in all body
locations except for the front neck area. It is noteworthy that wearing UV specialized
shirts or fall shirts would prevent UVR transmission to a great extent. The only body Site
where the mean exposure over clothing was not significantly greater than the
corresponding position under clothing was at the front neck. This might be due to the fact
that the film bandage location of the front neck over clothing was inconsistent across
subjects, because five of six subjects unbuttoned their Shirts, and one wore a shirt without
a collar.

Under clothing. Among the body Sites under clothing, the most exposed area was

the front neck (AAU = .047), followed by the back of the neck (AAU = .040). Because

39

most subjects unbuttoned their Shirts when they were playing, it might have led to greater
exposure of the front neck to the sun than the other sites. Although the difference
between under and over clothing at the back of the neck was significant because shirts
with collars could generate Shade, the neck was highly exposed to the sun due to the body
posture required to play golf. The least exposed areas under clothing were the shoulder

and back, perhaps because these areas were covered with two layers of fabric of the

 

 

 

 

bodice back yoke.
Table 10. Site-Specific UV Exposures and Paired Sample T-test between Under and Over
Clothing in Experiment 11
Under Clothing Over Clothing
Body Locations (AAU) (AAO) “$52" p
Mean (N )1’ SD Mean (N)a SD

Neck-Front .0473 (7) .027 .0558 (5) .033 .891 (4) .423
Neck-Back .0401 (4) .030 .0954 (7) .027 12.327 (3) .001“
Chest-R° .0020 (7) .005 .0904 (6) .016 13.644 (5) .000***
Chest-Ld .0028 (6) .004 .0893 (5) .028 5.386(3) .013*
Shoulder-R .0004 (6) .001 .1237 (6) .012 26.023 (5) .000***
Shoulder-L .0008 (7) .002 .1480 (6) .020 17.287 (5) .000***
Back-R .0010 (7) .003 .1148 (7) .014 25.245 (6) .000***
Back-L .0007 (6) .002 .0996 (7) .015 17.990 (5) 000*"
Posterior Arm-R .0006 (5) .001 .0664 (7) .027 5.218 (4) .006M
Posterior Arm-L .0015 (7) .004 .0762 (7) .011 13.953 (6) .000***
Anterior Arm-R .0027 (7) .004 .0694 (7) .029 5.786 (6) .001"
Anterior Arm-L .0005 (5) .001 .0562 (6) .025 3.766 (3) .033*
Posterior Forearm-R .0151 (6) .026 .0682 (6) .017 3.575 (4) .023*
Posterior Forearm-L .0059 (5) .007 .0664 (7) .025 10.147 (4) .001”
Anterior Forearm -R .0091 (5) .009 .0497 (6) .015 7.334 (3) .005"
Anterior Forearm -L .0064 (4) .013 .0437 (7) .015 5.403 (3) .012*

TOTAL .0074 (85) .016 .0853 (85) .034 18.569 (84) .000***

 

 

 

 

 

(N)': Numbers of samples. (dOb: Degree of freedom of t-test. R“: Right. L ‘1: Left.

*p < .05, Mp < .01, *"p < .001

40

Over clothing. The most highly exposed areas over the clothing were the left
shoulder (AAU = .148), followed by the right shoulder (AAU = .124), the right back (AAU
= .115), and the left back (AAU = .10), consistent with the findings of Experiment I.
When golfers play, they have to bend at the waist, causing greater UVR exposure for
these sites than the chest sites. Also, the back Sites might be more directly exposed to the
sun when golfers walked on the golf course during play. The least exposed areas among
the upper body sites were the anterior forearm areas and front neck.

Comparison between left and right sides. When comparing the absorbance
changes between left and right sites of the upper body, a Significant difference was found
only in the back over clothing at 01 = .05 level (Table 11). There were no statistical
differences for the other sites, indicating that left and right sides of the upper body under
clothing or over clothing, except for the shoulder area, were exposed to the sun in a
similar way while playing golf. In comparison to Table 4 in Experiment I, the difference
between left and right sides of body was revealed on the Shoulder during the summer
season and on the back during the fall. This might be related to the angle of the sun and
golfer’s orientation to the sun.

Table 11. Comparison between Left and Right Sides of the Body Over Clothing and
Paired Sample T-test in Experiment 11

 

. Lefi Side (AAO) Right Side (AAQ) t-value
Body Locatron d b p
Mean (N)a SD Mean (N)a SD ( 0

 

 

 

 

 

 

Back .0996 (7) .015 .1148 (7) .014 3.027 (Q .023*

 

(N)': Numbers of sample data. (dflb: Degree of Freedom of t-test.

*p < .05.

41

it

Absorbance Changes in Ambient Dosimeters

Table 12 shows the absorbance changes (AA) of ambient exposure measurements
for each time interval and provides the means, and standard deviations. When comparing
the mean of each time interval, the absorbance change increased steadily from one hour
to four hours of exposure, but did not increase after four-hour exposure. The absorbance
value at six hours dropped. The variation of the absorbance values after five hours
exposure might be due to exceeding the saturation limit of the polysulphone films with
the increase of UV radiation dosage.

Table 12. Absorbance Change of Ambient Dosimeter by Hours Exposed in Experiment 11

 

 

 

 

Time Film 1 Film 2 Film 3 Film 4 Film 5 Mean SD
10:00-11:00 .0318 .0440 .0431 .0179 .0221 .032 .012
10:00-12:00 .0643 .0630 .0830 .0364 .0616 .062 .017
10:00-13:00 .1195 .0778 .1080 .1062 .0904 .100 .016
10:00-14:00 .0797 .1353 .1359 .1439 .1076 .121 .027
10:00-15:00 .1337 .1260 .1356 X .1272 .131 .005
10:00-16:00 .1304 .1007 .1094 .1636 X .126 .028

 

 

Figure 7 is the scatterplot of the absorbance changes of ambient exposure in
Experiment 11 and Figure 8 is the box plot of the same data, demonstrating the location of
the mean and the potential outliers. The data that were far away from the mean would
result in large standard deviation.

To check the outlier, the researchers obtained the z-scores in Table 13 in the same
manner as experiment I and four values of z-scores (bold) that were greater than the
absolute value of 1.3 (1.3 standard deviation away from the mean) were considered as

outliers.

42

Figure 7. Scatterplot of Cumulative Ambient Figure 8. Box Plot of Cumulative
Exposure by Hours in Experiment 11 Ambient Exposure by Hours

13 m

 

 

.18 ° .16.

.141 : .14.
5 ° -
12‘ O 121
a ° °
10* o .10.
II
.08 ° , a .081 *

l ! *

0 1 i 5 i 5 6 7 1 2 5 i 5 6

 

 

 

 

 

 

 

 

Absorbance Change
2
Absorbance Change
2

Hours of Exposure Hours of Exposure

Table 13. Z-scores for Ambient Dosimeters by Hours Exposed in Experiment 11

 

 

Time Film 1 Film 2 Film 3 Film 4 Film 5
=I=10:00-1 1:00 .0063 1.0320 .9484 -1.1694 -.8173
lO:00-12:00 .1593 .0822 1.2851 -1.5220 -.0047
10:00-13:00 1.1701 -1.3826 .4682 .3557 -.6113
10:00-14:00 -l.53l7 .5572 .5796 .8792 -.4843
10200-15z00 .6495 -.9855 1.0485 X -.7124
10:00-16200 .1559 -.9060 -.5928 1.3429 X

 

 

Comparison of Personal Dosimeters with Ambient Exposure

In order to standardize the absorbance values of personal exposure to the ambient
UVB, the researcher needed to determine the equivalent absorbance changes in the
ambient level as in Experiment I. As Herlihy, et a1. (1994) emphasized that the
absorbance change of the film badges “Should be kept within the linear region” (p. 289),
the researcher included only the data between one hour and four hours of exposure and
ignored the data for the 5th and 6th hour due to the nonlinear relationships. The data from
one to four hours of exposure was comparable with the first experiment. A regression

equation was obtained by a regression analysis as follows:

43

Y = 0.0336X — 0.00003
where: Y = total amount of ambient exposure received by a subject
X = total amount of hours exposed to UVR

Table 14. Regression Analysis for Hours Exposed Predicting Absorbance Change in

 

 

 

 

Ambient Exposure
Variable B SE B B p
Constant -.0030 .007 .997
Hours exposed .0336 .003 .958 .000***
"*p < .001

From the equation, it is predicted that for every additional hour of exposure, the
absorbance value will increase by .0336. R2 is .918 (adjusted R2 = .912), suggesting that
the regression line is fitted well to the sample data. Figure 9 shows the scatterplot and the

regression line of the ambient exposure during four hour intervals.

Figure 9. Scatterplot of Cumulative Ambient Exposure and Regression Line for
Experiment 11

 

 

 

 

Absorbance Change

 

o 1 2 :1 4 5

Hours of Exposure

However, if X = 0 (zero hours exposed to the sun), the calculated absorbance

value is -.00003, which is very small value and illogical because the absorbance change

44

has to start at zero when the film badges are not exposed to the sun. Accordingly, the
equation of Y = 0.0336X was used for Experiment 11. From this equation, the total
corresponding amount of UVR exposure received by golfers during play was obtained.
The first team was exposed between 10 am. and 3 pm, and the second team was
exposed between 11 am. and 3:40 pm. The total amount of ambient exposure for each
team was calculated separately as follows:

Teamle = 0.0336 x (5 hr) = 0.168

Team 2: Y = 0.0336 x (4.67 hr) = 0.156912 z 0.157

Then the amount of UV radiation each body site received was expressed as a proportion
of the total ambient exposure by dividing the absorbance change of each body location by
the total amount of ambient exposure of each team.

Table 15 presents the proportion of the total ambient exposure for each body
location received by golfers during Experiment 11. The proportions of ambient UVB
under the shirts (first column) ranged from 0.3% (right shoulder & anterior left arm) to
29% (front neck), whereas the proportions over the clothing (last column) ranged from
27.2% (anterior left forearm) to 92.5% (left shoulder). The most exposed Sites over
clothing after the left shoulder, in order were the right shoulder (76%), the right back
(71%), the left back (62%), the back of the neck (59%), and the right (56%) and left
(54%) of the chest sites. Under the clothing, the body sites, such as the shoulder, back,
posterior arm, and the anterior left arm received less than 1% of ambient UVB across all
types of shirts. That is, the tnmks of golfers was exposed to UVB to a great extent, but
UVB penetrated the shirts less than 1% of the ambient while playing golf when golfers

wore UV Specialized shirts or heavy fall shirts.

45

Table 15. Mean Proportion of Ambient UVR Received at 16 Body Sites in Experiment 11

 

 

I Under Clothing l Over Clothing

Body Locations I Mean i SD Mean iSD Mean 3: SD I Mean i SD
(Total)' (UV Shrrt)‘ (Regular Shirt)‘ (Total)'

Neck-p,0m .290 e .165 (7) .288 i: .107 (4) .293 e .253 (3) .349 i .209 (5)
Neck_Back .252 r .195 (4) .336 e .121 (3) n.a. (1) .592 r .176 (7)
(31,65pr .012 :1: .031 (7) 0 (4) .027 a. .047 (3) .562 r .094 (6)
(3,68ch .017 r .022 (6) .023 r 027(3) .011 e .018 (3) .543 r. .162 (5)
Shoulder-R .003 .1.— .005 (6) .001 :h .001 (4) .006 3: .009 (2) .763 :r .091 (6)
Shoulder-L .005 e .013 (7) .009 i .017 (4) 0(3) .925 r .144 (6)
Back_R .006 :r .016 (7) .011 e .021 (4) 0(3) .712 :t .102 (7)
Back_L .005 r .011 (6) .007 :r .014 (4) 0 (2) .617 r .093 (7)
posterior Am,_R .004 .1. .008 (5) .006 :1: .010 (3) 0(2) .411 :1: .170 (7)
posterior AmL .009 r .024 (7) .016 r .032 (4) 0(3) .473 i .079 (7)
Anterior Arm-R .016 :1: .026 (7) .022 r. .033 (4) .008 :1: .015 (3) .430 :t .180 (7)
Anterior Arm-L .003 r .007 (5) .005 r .009 (3) 0(2) .347 :1: .158 (6)
£32323, .095 :1: .163 (6) .143 r .187 (4) 0(2) .420 e .103 (6)
2:21:11 .036 i .040 (5) .059 i .032 (3) 0 (2) .410 e .144 (7)
fig}: _R .056 i .053 (5) .071 e .049 (4) n.a (1) .307 :1: .098 (6)
£353: _L .038 e .076 (4) .076 e .106 (2) 0 (2) .272 r .094 (7)

 

 

 

 

 

(N)': Numbers of samples. Rb: Right. L °: Left.

Specifically, the highest proportions of the ambient UVB under the UV

specialized Shirts (second column), if the front and back of the neck were excluded due to

the inconsistent measurement, was the posterior right forearms. This Site received about

14% of ambient UVB. The other forearm sites also received higher proportions of UVB

compared with the other body sites. This might be because the film badges under the UV

shirts were exposed through the unfastened cuffs of the sleeves. The proportions of

ambient UVB among the body sites under the regular shirts ranged between 0% and 3%

46

except the front neck. Most of body Sites under the typical fall shirts did not register UVB

penetration through the fabrics. That is, if golfers wore sweatshirts or corduroy shirts,

which were much heavier and thicker than UV specialized shirts while they played, those

shirts were more effective in preventing UVB exposure than UV specialized Shirts.
Summary

The objectives of this chapter are to determine the cumulative exposure of
different body Sites to UVB during golf rounds and to evaluate the effects of clothing to
prevent UV transmissions from the sunlight.

Research Question 1 for Objective 1
- To What Extent do Golfers Receive UV Radiation with respect to 16 Upper Body Sites?

Figure 10 presents the mean proportion of ambient UVB at each body site over
the clothing for Experiment I and Experiment 11. The proportions over the clothing
indicate the amount of UV radiation received directly by the different body sites with
respect to the ambient conditions of each experiment date. It indicates how much a golfer
is exposed to the sun relative to the available ambient exposure. The proportion allows
direct comparison in two different ambient conditions.

In both experiments, each body location received different amounts of UVR, but
there was a pattern for the data for body locations. Although the graphs do not match
exactly in both experiments, the shoulder and back sites receive the greatest UVR
exposure, followed by the back of the neck, chest, posterior arms, and anterior forearms.
These findings are comparable to the study of Elwood and Gallagher (1998) that
melanoma was more likely to be diagnosed at the back and the upper arms of males under

ages 50 years old. The other sites such as anterior arm (AAR & AAL) and posterior

47

 

forearm (PFR & PF L) were not consistent for absorbance proportions, which might be

due to individual’s movements.

Figure 10. Mean Proportions of Ambient UVR Exposure Over Clothing for Experiment I

 

 

 

 

 

 

and II

' * .—' — #' "1 r" *1

1: Emma! I tfixgalmsm IL'
1.00 —- ——~ —-~— _- ’
E 0.90 -—— ———— —-— ——~ w ———— —«:‘
50.80«~—- — —— —A
E 0.70 +—~ —— —— ~— A w ___ 7*
0.60 - —~ — — f — .I
E 0.50 -L. ,- —— ———: — - e — '
.9 0.40 -,_ - —— — — —-——4,i
£0.30 ,2 e — ~ ~41
._ 0.20- e— A ~— —— «1‘
‘L 0.10 A- — ~ — —— ————— 11‘
0.00 . . - . '1
mmtzu mutt») quzui
l
2200 mm§§§§3§g§§§.
Body Locations .

 

 

 

NF
NB
CR
CL
SR
SL
BR
BL
PAR
PAL
AAR
AAL

PFR

PFL

AFR

AFL

 

Neck-Front
Neck-Back
Chest-R
Chest-L
Shoulder-R
Shoulder-L
Back-R

Back-L
Posterior Arm-R
Posterior Arm-L
Anterior Arm-R
Anterior Arm-L
Posterior

F orearm-R
Posterior
Forearm-L
Anterior
Forearm -R
Anterior
Forearm -L

 

 

The graph of Experiment 11 in Figure 10 Shows higher proportions of ambient

48

UVR than that of Experiment I. The proportion was obtained by dividing the absorbance
change of each body location by the total amount of ambient exposure, and the total
amount of the ambient exposure received by golfers in Experiment 11 (AAambn = .168 for
team 1 & .157 for team 2) were much lower than those in Experiment I (AAambI = .461 for
team 1 & .412 for team 2). The amount of UVR during both experiments would be
influenced by the ambient conditions, such as UVR intensity. The UV indexes which
indicate “UV intensity levels on a scale of 0 to 10+” (U.S. Environment Protection
Agency, 2002) are different on the two experiment days. The day of the first experiment

(summer), the UV index was 7; the day of the second experiment (fall), the UV index

was 4. Hence, the difference in the proportions shown in Figure 10 may reflect the
different intensity of the UV radiation.

Research Question 2 for Objective 1

- What Parts of the Upper Body Receive the Most UV Radiation while Playing Golf?

When comparing the mean absorbance changes across body-Specific sites in
Table 3 and in Table 10, the shoulder and back on both right and left sides had the
highest mean scores among all body sites, followed by the back of the neck, chest, and
posterior arms. These findings are consistent with the findings of Gie, et al.’s study
(1995) and Herlihy, et al.’s (1994). The back sides of golfers seem to be more exposed to
the sun than the front sides of their bodies while they play golf. The anterior arm, the
anterior forearms, and the front neck received relatively lower mean scores than other
sites, although the right and left Sides of those areas differed for both experiments.
Research Question 1 for Objective 2
- To What Extent do Golf Shirts Reduce UV Radiation?

Results of paired sample t-test in Table 3 and Table 10 Show Significant
differences in the mean absorbance changes between over and under clothing. While the
chest, shoulder, and posterior arm in Experiment I were significant in the effects of golf t-
shirts to reduce UVB transmission, all body sites except for the front neck were
Significant in Experiment 11.

Figure 11 presents the mean proportion of ambient UVB under clothing for
Experiment I and II. The absorbance proportions in Experiment I were higher than those
in Experiment 11 across all body sites. The mean proportions of Experiment I are obtained

from summer thin, short-sleeved knit Shirts, whereas the data of Experiment 11 are based

49

 

on the sum of two types of clothing, the UV specialized Shirts and subjects’ typical fall
Shirts. Figure 11 indicates that summer shirts are less effective in blocking UV radiation
than the thicker fall shirts or UV specialized Shirts. In Figure 10, the shoulder and the
back over clothing were exposed to UVR the most, but these locations had lower
proportion under clothing in Figure 11. The two layers of back yoke of UV specialized
Shirts and corduroy shirts would contribute to producing an effective barrier to UVR.

Figure 11. Mean Proportions of Ambient UVR Exposure Under Clothing for Experiment

 

 

 

 

 

 

 

 

 

 

  

 

            

I and II
AT T— _ — — — Ti 7 7 7— A _ NF Neck-Front
i NB Neck-Back
1 I —--——— f" a 1 CR Chest-R
i" Ejfir'ment 1“ t Eminent H1 1 cr Chest-L
{ SR Shoulder-R
0'12 T" — T “m m " 7 ‘ SL Shoulder-L
ac) 0.1 * BR Back-R
16 BL Back-L
E 0.08 « PAR Posterior Arm-R
‘5 0 O 6 - _ fl ‘_ . . PAL Posterior Arm-L
g ' AAR Anterior Arm-R
E 0.04 : AAL Anterior Arm-L
Posterior
1 5 0-02 ' PFR Forearm-R
* . . . , first.
at ~J z -I it u a: u z ~J q: A; A: ~) ,
C.) C.) 02 ”J m m it it :5 V: in in. f; f? I AFR Anterror
_ 1 Forearm -R
Body Locations 1 AFL Anterror
1_ _ __-_,, __ - ._- ._ - ,- A--.lfi__ , 7 _ __1 Forearm -L

 

 

 

 

Research Question 2 for Objective 2
- Do the Differences in Absorbance Values Under the Shirt Vary with the Types of Shirts,
UV Specialized Shirts and Regular Golf Shirts?

Specific comparisons in the two experiments under clothing were possible if the
values of the neck sites were excluded because of great differences between the neck Sites
and the other upper body parts. The data of forearm sites were not included also, because

every subject in Experiment I wore short Sleeve Shirts so the data under clothing were not

50

obtained. Figure 12 presents only the trunk parts and the upper arms of the body and
compares the proportions of ambient UVB received by those body Sites between
Experiment I and Experiment 11.

Figure 12. Mean Proportions of Ambient UVR Exposure Under Different Types of Shirts
for Experiment I and II

1 _ A

1ft§wnmer knit shirts -“tuyipséiéliadfle :Ravla fa“ shins

 

 

 

 

1
l
1 0.045 e- ~-—~—e I
' 1
1

 

 

 

 

 

 

 

 

i g 0.040 r
l :5 0.035 t

5 0.030 «
t F 0'0” " CR ChestR

E 3:33? 4” f '= CL Chest-L
1 §_ 0 010 ,_ . 1 SR Shoulder-R
j E ' SL Shoulder-L
1 0-005 “ 1 BR Back-R
1 0-000 ‘ i BL Back-L

<1~ \2 4r \2 Q- s» 51- V 8— \2 PAR Posterior Arm-R

; Q 0 ‘° 2 ’2’ 3’ sf 3" ‘23" if PAL Posterior Arm-L
1 Body Locations ’ AAR Anterior Arm-R
17—4___ _ fi__ __ AAL Anterior Arm-L

 

 

 

 

In Figure 12, the summer Shirts of Experiment I had the highest proportion of
ambient UV radiation absorbed under clothing at most body locations. Especially, the
right of the chest (CR), posterior left arm (PAL), and anterior left arm (AAL) received
higher proportions of ambient UVB compared with other types of Shirts. In Experiment I,
all seven subjects wore short-Sleeved knit Shirts. The film badges of the left arm under the
short Sleeves can be exposed to the sun when the golfer swings the club, leading to higher
absorbance of UVB doses. Both types of Shirts in Experiment II gave more protection
than the shirts worn in Experiment 1. Regular fall shirts provided better protection from
UV transmission than UV specialized shirts, but were not acceptable during the summer

season.

51

In order to examine the Significant differences in the mean proportions of ambient
UVB among three types of shirts, the ANOVA test was used. No significant difference
was found at each body location. Although summer shirts absorbed the higher
proportions of ambient UVB than the other two types of shirts in Figure 12, the
differences were not significant statistically, perhaps because of the small absorbance
values. However, summer polo shirts offer similar UV protection in Sheer’s article
(1999). She reported that commercially available UV specialized shirts provided SPF of
30 or higher according to Consumer Reports, whereas the typical polo shirts from Land’s
End or Hanes provided SPF of 15 to 20, which was an acceptable level of SPF for sun
protective clothing.

Limitations

The field experiments have several limitations. Several confounding variables
might affect the amount of UVB received by the subject: for instance, height of the sun in
the sky, the UV index on the date of experiments, topography of golf course (reflection
from water or sand), and subject’s characteristics, such as different types of clothing,
body stance and posture, or golf skills. Those factors probably caused the variances of
absorbance changes in this study. Also, the data are based on a small number of
participants, and film badges on some of body locations were lost, resulting in lower
reliability of the findings. The film provided for this study became saturated after four
hours of exposure and could not accurately measure exposure beyond that time. Thus it
was impossible to measure exposure for the test hours participants played golf. For these
reasons, interpretation of the findings and generalization to the population should be

made with caution.

52

This study focused on only one activity, golf at the same geographic location,
repeated twice in order to reduce the variances, while a number of previous studies have
investigated different types of activities at the different locations (Herlihy, et al., 1994;
Holman, et al., 1983; Vishvakarman, et al., 2001). Because the objectives of this study
were to identify the amount of UVB received by golfers and to examine the effect of the
typical shirts to protection from UVB, the findings of this study would fulfill these

purposes at least.

53

CHAPTER THREE: PHASE I — MARKET RESEARCH

It would be valuable to investigate what types of garments are worn and available
in the selected market segment as background information for understanding consumer
needs. Market research is defined as “the process of collecting, manipulating, retaining,
and using information for the purpose of decision-making in all areas of the marketing
fimction” (Kurtz, 1969, P. 2). Product research is “the most important area” for providing
products that meet consumer tastes and needs, which is directly related to a company’s
future success (p. 11). For this study, product research, focusing on Shirts for golfers was
conducted, using two approaches: Observation of published images and a market survey.

First, the standardized look of golfers in current magazines was explored in order
to understand the norms that would influence golfers’ views about a new product
(innovation). If the new item were not compatible with their values or beliefs about the
images of golfers, the new product would be unacceptable. For instance, if the UV
specialized garments are thought to be inappropriate for the special social settings, such
as golf games with peers, the potential adopters would be less likely to consider it as an
alternative, although it might offer more protection from UVB. Therefore, finding out the
Standardized appearance of golfers in recent years will be helpful in developing UV
protective shirts.

In addition, it is necessary to have knowledge of the related products, available in
the selected market in order to evaluate the innovation. The market survey provides the
general scope for UV Specialized golf shirts as well as typical golf shirts. It focuses on
the general interests and awareness of current consumers regarding UV protective

clothing, but not the specific product dimensions, such as color, quality, size, or price.

54

Standardized Appearance from Published Images
Methodology

In order to determine what the normative appearance or standardized look of
golfers is, the researcher examined the golf magazine, Golf Digest between January 2001
and December 2002. The unit of analysis was men’s golf garments, so images of all male
adults who were playing, posing, or practicing on the golf courses, or who were
instructing anyone were examined. Criteria for inclusion were: The target model should
be in color, hip-level up to head should be visible, and the photograph of the model
Should be larger than three centimeters because if the picture is smaller than this size, it is
difficult to determine the garment characteristics. In addition, front or side views of a
golfer were included; pictures from behind were excluded due to the difficulty of
clarifying the garment features. Advertisements, drawings/sketches, or black and white
photographs were also excluded from the data.

All garments of the target model were observed and recorded in Table A3 in
Appendix A. First, the features of t-shirts were examined: Sleeve length (long or short),
the existence of sleeve band (yes or no), the existence of a collar (yes or no), and color
(Specific color, pattern or stripe). Also, other types of clothing were examined. Outerwear
was divided into jacket/sweater or vest. If a target model wore a long sleeved jacket or a
sweater over the shirt, then it was recorded only as outerwear. If the vest was worn over
Shirts, the features of the shirt and the vest were marked separately. The length of pants
(long or short), types of hats (cap, visor, brimmed, or none), and use of sunglasses (yes or
no) were recorded if those were observable. Use of gloves was not included because most

golfers wore gloves when the full-swing was necessary, and did not wear them for

55

putting. That is, wearing gloves seemed to be more contingent on a golfer’s
position/action at each hole than on sun protective purposes. If the target model wore
different styles or colors of garments, each style was recorded separately. If the garment
characteristics were unobservable or unclassifiable, it was counted in the column U,
which Stands for unobservable.
Results and Discussion

Table A3 in Appendix A presents the detailed results of clothing features, and
Table 16 shows a yearly summary. One issue of Golf Digest was missing, so a total of 23
issues were examined: 270 observations from 12 issues in 2001 and 170 from 11 issues
of 2002. Unobservable categories ranged from 0.9% (Sleeve length) to 8% (pants length).
Characteristics of Shirts

About 82% of golfers in 2001 and 88% in 2002 wore Short-Sleeved Shirts with
collars (89.6%, and 90.5%). That is, the most typical style of golf shirt was a Short-
sleeved polo shirt. Sleeve band was optional. About 55% of the shirts had sleeve bands in
the two consecutive years. The hems of Short sleeves were not tight even if they included
sleeve bands, so the width was Similar with sleeves without sleeve bands. Golfers might
prefer to wear loose and short-sleeved (length to elbow) Shirts instead of long- or fitted-
Sleeve shirts. If the golf shirts were long sleeved, most included sleeve bands and neck
collars. Shirts colors varied: white (14.1), grey (8.6%), navy or black (16.6%),
beige/yellow tone (16.3%), blue tone (18.5%), red tone (21.4%), and green tone (4.5%).
Black and navy colors were categorized as one group because in some pictures it was

hard to distinguish these two colors. There was no specific color pattern according to

56

seasons. More Specific categories of colors and the frequencies are provided in Table A4
in Appendix A.

Table 16. Summary of Garment Characteristics in Golf Digest

 

 

 

 

 

 

 

 

 

 

Garments Characteristics T032? (1% ) To2th)(2°/o) 1‘33"
81 Long 17.0 11.8 15.0
Leg: Short 82.2 88.2 84.5
Unobservable 0.8 0 0.5
Sleeve Yes 55.2 59.4 56.8
. Band No 41.5 38.2 40.2
T-sh1rts Unobservable 3.3 2.4 3.0
Yes 89.6 91.8 90.5
Collar No 5.2 3.5 4.5
Unobservable 5.2 4.7 5.0
Color 70.4 72.3 71.1
Color Stripe 23.7 20.6 22.5
Pattern 5.9 7.1 6.4
Yes J acket/ Sweater 1 1.1 8.8 10.2
Outwear Vest 5 .2 4.1 4.8
None 83.7 87.1 85.0
Long 85.9 94.1 89.1
Pants Short 3 .7 1.8 3 .0
Unobservable 10.4 4.1 7.9
Cap 50.4 57.6 53.2
Hats Yes Visor 7.0 1 1.2 8.6
Brrmmed 5.6 2.4 4.3
None 37.0 28.8 33.9
Sunglasses Yes 8.5 3.5 6.6
No 87.4 93.5 89.8
Unobservable 4.1 3 .0 3 .6
Total
Observation 270 1 70 440

 

 

Characteristics of Other Garments

In all the photographs of golfers, only 10% wore jackets (i.e. wind-shirts) or
sweaters and about 5% wore vests (Table 16). In most cases, golfers wore jackets or
sweaters over polo-type shirts, so the collar was shown over outerwear. About 89% of

golfers wore long pants, and only 3% wore shorts. The Shorts were usually worn in public

57

ranges, rather than golf courses. The types of hats worn by golfers consisted of caps,
53%, visors, 8.6%, brimmed hats, 4.3%, and non users, 34%. Wearing a hat seemed to be
a matter of fashion. Hill and Rassaby (1984) found Significant correlations between the
intention and barriers to wearing a hat among men. Those barriers were the problem of
playing the sport (r = -.43), being self-conscious in a hat (r = -.41), a nuisance on windy
days (r = -.3 7), inconvenient (r = -.34), and getting a sweaty head (r = -.32). These
barriers might be reflected in golfers in the published magazines. Finally, most golfers
(90%) did not wear sunglasses while playing golf. It seems to be comparable with
reasons for not using a hat that affects play.

Summary
Research Question 1 for Objective 3
- What is the Standardized or Normative Look of Golfers in Magazines?

From the results, the standardized image of current golfers is apparent. Most
golfers wore short-sleeved polo shirts that included wide width of sleeve hem and collar,
matched with long pants. There appears to be no color preferences among photographed
golfers’ Shirts. Golfers wore long-sleeved Shirts or jackets occasionally, probably for
cooler weather conditions. A baseball cap was the most preferred head covering if one
was worn. Most golfers did not use sunglasses.

Most target models in magazines were professional golfers and were
photographed at their professional leagues or during instructional settings. Hence, the
results are more reflective of professional and formal images of golfers than images of

the general public. Professional golfers seem to select garments that facilitate their

58

movements over other features of garments. Hence, the wide hems of short sleeves or
long sleeves with sleeve bands were designed for the golfers’ practical performances.

Only two male golfers wore sleeveless shirts and thirteen golfers wore shorts
among 440 observations. Although informal observation indicates that these looks are
common in public golf courses, it constitutes only a small percentage of the published
images. All of the subjects in experiments for this study also wore Short-Sleeved Shirts
with shorts when the experiment was conducted during August, and long-sleeved shirts
with long pants for the October experiment. This suggests that weather conditions affect
their garment selection. Hill and Rassaby (1984) reported that the most significant barrier
to wearing a long-sleeve shirt for men was the discomfort from heat.

Market Survey
Methodology

To learn the availability of traditional golf clothing and UV specialized clothing, a
survey was conducted of selected local stores. The results would be helpful in
understanding the clothing preferences of amateur golfers. The researcher visited 4 of 11
local specialty golf stores listed in the 2002 yellow pages of the Greater Lansing Area
telephone directory. First, the researcher contacted the local golf stores listed and asked
whether they carried golf clothing because some stores only supplied golf equipment.
Only four stores in the Lansing and East Lansing area carried clothing.

After the researcher explained the purpose of the study to a salesperson or a
manager in the selected store, the researcher asked whether he or she would be willing to
be interviewed. All subjects agreed to the interview and were willing to participate at that

time, so the researcher did not need to arrange another meeting. Three stores carry both

59

men’s and women’s clothing, and one store carries only women’s. The researcher asked
questions regarding the typical styles of golf shirts, customer’s interests in and
availability of UV specialized products in the local stores. The interviews took
approximately 10 minutes. The interview questions and protocol were standardized and
the questions are listed with the results.

Results and Discussion

Two of the four interviewees were males, and three of them were store managers.

‘3‘ J's-'1' 15““

I?

Interviewees’ years of experience in sales span from five to 22 years (5, 8, 15, and 22
years). Interviewees were between 25 and 49 years of age.

Interview Question I

- What are the Most Popular Styles to Golfers Who Shop in Your Store?

Common and unique styles of shirts. The most popular and typical style to golfers
was a short-sleeved polo, made of 100% cotton. Popular colors were soft-tone, solids
such as white or grey. Green was another popular color, a symbol of MSU. One retailer
indicated that Michigan residents were conservative rather than fashion conscious such as
residents on the West Coast area, so they preferred typical styles within soft color ranges.
The popular style for women was Sleeveless shirts in summer time. The interviewees
described wicking away moisture shirts or long-sleeved pique rugby shirts (cotton, twill
collar, rubber buttons, rib trim at cuff) as the unique styles in their stores.

Common and unique styles of hats. The most common style of hats for golfers
was the baseball cap (80 to 90%). However, women sometimes looked for a brimmed hat
or visor. The wide brimmed hat or straw hat in summer season was a unique style for a

few male golfers.

60

Seasonal differences. All retailers mentioned there were no seasonal differences
in customer’s search for clothing. Golfers usually looked for short-Sleeved shirts. Golfers
wear Short-sleeved polo shirts in most seasons and they use vests or wind-shirts over
short-sleeved shirts in cool weather. Another reason for no seasonal differences in
customer’s interests was that they shopped and bought short-sleeved shirts during the
winter season for vacation use in warmer places.

Search for long-sleeve shirts in summer. Most golfers, especially males would not
look for long-sleeved shirts in summer. However, a few female customers bought long-
Sleeved shirts in summer to protect or cover their skin.

Interview Question 2
- Are Golfers Interested in or Looking for U V Specialized Clothing?

Who and how often. Three interviewees in local stores that carried both men’s
women’s wear said they were rarely asked about UV specialized products. One manager
said that he had never been asked, and two said one or two customers in a year asked
about UV Specialized products. The interviewees mentioned that the purchasing criteria
for their customers were design of garments, then the price. Hence, after customers select
the design of a Shirt, if the shirt provides UV protection, then it is an additional benefit,
but not a required option.

However, the manager at a women’s golf specialty store mentioned that some
female customers in her store were interested in UV specialized products. Those
customers were concerned about the UV protective frmction of clothing products. Prices

were not important for them, although UV specialized products were more expensive

61

IT“
q‘

than other typical products. They were aware of the UV specialized products to protect
their bodies from the sun.

Types of clothing. The UV specialized product of most interest during the summer
season was a hat, followed by sunglasses and short-sleeved shirts. Customers at the
women’s golf specialty store were interested in long- or Short-sleeved shirts, hats, and
sunglasses.

Interview Question 3
- Does Your Store Carry UV Specialized Products?

Types of clothing. F our retail stores carried a small selections of UV specialized
products. One store carries only UV protective sunglasses. Two stores carry UV
specialized short-Sleeved shirts and hats, and one of these carried UV protective
sunglasses. The women’s specialty store carried the most items, such as shirts, outerwear,
sunglasses, and gloves. The manager at the women’s golf store explained the importance
of gloves for golfers and gardeners. Customers who were interested in UV specialized
products at this store were more concerned about the function of UV protection than style
or price of the products. However, customers at the other three stores did not care much
about the function of UV protection, and the style of garments was the most significant
factor in their purchasing decision. These consumers might not be conscious of the risk of
skin cancer or aging skin.

Because few customers were interested in UV specialized products, the preferred
features of UV specialized products were not discovered. In addition, since the UV

specialized products that were available at local stores were not distinguishable in terms

62

of styles or colors, customers could select the garment without information on the UV
protective function in advance.

Interview Question 4

- Are Your Suppliers Aware of the Needs of UV Specialized Clothing?

Three retailers mentioned that some suppliers have been aware of the importance
of UV Specialized products. One retailer mentioned that they had six lines of men’s wear
and two of them provided UV specialized clothing. However, because their customers
were rarely interested in UV specialized products, it was not necessary to make those
products available to their customers.

Summary
Research Question 2 for Objective 3
- What Kinds of Traditional Clothing are Available for Golfers in Local Pro Golf Stores?

Based on a market survey in which four salespeople in local pro golf shops were
interviewed, the most typical style of golf shirts available and purchased was a short-
sleeved polo, made of cotton in soft colors without seasonal differences. This style
included a knit collar, and rib trim at cuff or turnback hem. The traditional style of golf
shirt was comparable with the findings of the review of published images. A baseball cap

was the most common head covering for customers at local golf stores, which was also

similar with the findings of published images.

63

WAT—fl
" .4

Research Question 3 for Objective 3
- To What Extent are UV Specialized Products Commercially Available in the Local
Stores?

The majority of customers in the four local stores surveyed were not interested in
UV specialized products. Only a small portion of customers, mainly women, looked for
these products, mostly during summer season. Hats, Short-sleeve Shirts, or sunglasses
were the major items for which those customers looked. Accordingly, local stores made
available small amounts of UV specialized products in these limited items of interest to
customers. To most customers, design and price were more important factors than UV
protective functions. Golfers in this selected market seem to be aware of UV protective
clothing, but do not acknowledge the importance of the UV protective function of
clothing. Golf wear suppliers seem to be becoming aware of the importance of UV
protective function for golf clothing.

Limitations

This market survey has limitations. The data were obtained from a small number
of retailers. The retailers had to recall their experience when they responded to the
questions. In addition, few of their customers looked for UV specialized clothing so it
was difficult to characterize customers. However, results from this small study suggest
that UV specialized products could be disseminated more quickly into the target market if
the design of UV specialized products were appropriate or more attractive than other
typical golf shirts at the proper price range, because most male golfers select garments

based on design features and prices.

64

CHAPTER FOUR: PHASE II — HEALTH BELIEFS AND INNOVATION
Literature Review

New cases of melanoma are increasing annually. The American Academy of
Dermatology (2002a) has estimated that 1 in 39 Americans could develop melanoma in
2003. Sun protective actions, hence, have been recommended to reduce the risk of skin
cancer. Even though males have higher rates of incidence and death from skin cancer
than do females, men are less likely to practice sun protective behaviors (U .S.
Department of Health and Human Services, 2000). Several factors could account for the
different sun protective practices between men and women. As Liffiig (2001) pointed
out, “most programs target women since their traditional family role makes them
generally more likely than men to purchase and apply sunscreen products” (p. 199).
Children and youth are another Significant target audience of educational programs due to
the cumulative aspect of UVR during childhood (Buller, Loescher, & Buller, 1994;
Cockbum, Hennrikus, & Sanson-Fisher, 1989; Labat, etal., 1996).

The present study focused on outdoor enthusiasts, especially golfers. Although
increasing attention is being paid to golfers regarding sun safety promotions, golfers’
perceived beliefs about skin cancer risk, attitudes toward UV protective clothing, and sun
protective behaviors have been insufficiently studied. Since the incidences of skin cancer
are associated with unprotected exposure to UVR (Saraiya, Hall, & Uhler, 2002), it is
important to analyze sun protective actions practiced by golfers when outdoors.
Specifically, whether golfers use sunscreen, wear sun protective clothing, or seek the

Shade while they play could be significant determinants of skin cancer risk. For the

65

ill/l.

flSK' g

second phase, two theoretical frameworks, the health belief model and attributes of
innovation were used.
Health Belief Model

The health belief model (HBM) is a useful framework for understanding an
individual’s health beliefs, attitudes and intentions to comply with recommended
protective actions. This model was originally developed by a group of researchers in the
Public Health Service as an attempt to explain why peOple did not comply with
preventive health-related recommendations between 1950 and 1960 (Rosenstock, 1974).
The basic principles of this model were derived from value-expectancy theory, which
“describes behavior or decision-making under conditions of uncertainty wherein behavior
is predicted from both the individual’s valuation of an outcome and the expectation that a
specific action will result in that outcome” (Becker, Maiman, Kirscht, Haefner, &
Drachman, 1977, p. 349). That is, an individual would take an action to avoid or prevent
disease (expectation to the outcome) for the purpose of maintaining one’s health (one’s
value). To do this, several components were necessary: susceptibility, severity, benefits,
barriers, cues to actions, and demographic and socio-psychological variables.
Perceived Susceptibility

Perceived susceptibility is defined as “one’s subjective perception of the risk of
contracting a health condition” (Rosenstock, 1990, p. 43). Since it is a perceived
probability of getting skin cancer in oneself, the range of susceptibility is from a
minimum of denying any possibility of occurrence to a maximum of accepting the real

risk as a possible personal occurrence to the subject (Rosenstock, 1974).

66

P.

ab

01

id
‘17:?

tIISCOr

Perceived Severity

Perceived severity is one’s subjective assessment of the disease potential “for
causing physical harm and interfering with social functioning” (Becker, et al., 1977, p.
349). Physical harm includes a person’s evaluation of death, disability, or pain. Social
functioning includes the consequences of the conditions on work, family life, or social
relations, influenced by disease. Perceived severity is an important part of the HBM
because “even when an individual recognizes personal susceptibility, action will not
occur unless he or she also believes that becoming ill would bring serious organic and/or
social repercussions” (Becker & Maiman, 1975, p.14).

Perceived Benefits

Perceived benefits of taking health-related action are one’s perceived beliefs
about the effectiveness, feasibility or values of a behavior in reducing the health threat
(Becker, et al., 1977; Rosenstock, 1990). For example, an individual’s perceived benefit
of using a sunscreen is for reducing the chance of getting skin cancer, or for increasing a
feeling of control over a health-related problem.

Behavioral Barriers

Behavioral barriers refer to one’s perceived “potential negative aspects of a
particular health action” (Rosenstock, 1990, p. 43) such as physical (safety, side effects),
psychological (discomfort, unpleasant, complexity, time-consuming), or financial costs
(monetary expense) and situational factors. In the present study, behavioral barriers are
based on barriers to practicing the recommended sun protective actions, for instance, the
cost of sun protective products, unfashionable appearance of protective clothing, or

discomfort of the wearing long-Sleeved shirts or wide-brimmed hats.

67

Psychological Barriers

For this study, the perceived benefit of tanning is included as a psychological
barrier to taking sun protective actions. Sunbathing was introduced in the US. in 1914 as
a medical treatment, called “heliotherapy.” Tan skin was perceived as a Sign of good
health as a result of the success of the therapy (Keesling & Friedman, 1987). Also, after
the industrial revolution, darker skin color became a symbol of higher status because
wealthier people had time for outdoor sports and leisure activities (Keesling & Friedman,
1987). Fashion designer, Coco Chanel, also stimulated this trend in the late 1920s with
her bronzed tanning “as a Sign of affluence” (Roach, 1997, p. 116).

The most common motivation for getting a suntan was to look and feel attractive,
healthy, and active (Amir, et al., 2000; Arthey & Clarke, 1995; Broadstock, Borland, &
Gason, 1992; Keesling & Friedman, 1987). Johnson and Lookingbill (1984) found that
attitudes toward suntan varied with age. Among 489 respondents who were outpatients of
dermatology and internal medicine clinics, those younger than 30 years of age had much
stronger beliefs that a suntan looked healthy. Previous studies revealed that the general
propensities of tanners included being more likely to spend time in the sun, to take a risk
for the benefit of a suntan, less likely to have knowledge of skin cancer, and less likely to
use sunscreen (Arthey & Clarke, 1995; Keesling & Friedman, 1987; Miller, Ashton,
McHoskey, & Gimbel, 1990). Accordingly, one’s perceived positive aspects of beliefs
about tanning would be the major barriers to taking sun protective actions.

Modijying Factors
Cue to action is some instigating event that occurs “to set the process in motion”

(Rosenstock, 1974, p. 332) and consists of internal cues (perception of body states or

68

symptoms) and external cues (interpersonal interaction, the impact of media). External
cues include knowing a person who has skin cancer or having a friend who practices sun
protection or who enjoys a suntan (Arthey & Clarke, 1995).

Other variables included in modifying factors are demographic variables (age,
sex, ethnicity), socio-psychological variables (social class, personality, peer and reference
group pressure), and structural variables (knowledge about the disease, prior
contact/experience with the disease). These variables influence perceived benefits and
barriers as well as perceived susceptibility and severity (Rosenstock, 1974).

Figure 13 presents the variables that contribute to explaining sun protective

behaviors for this study based on the existing literature.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Demographic Factors Perceived Benefits of
Gender Ft preventive action - -
Age Practrcrng
Education Recommended
Race Perceived Barriers to Sun Protective
Marital status preventive action Behaviors
. — 0 Behavioral barriers F— ” U33 sunscreen

Cue to Act")? , D Psychological barriers ° Wear long sleeved
Interpersonal 1nteract1on Shirt and long
0 Frrhend s protective pants

be avior . 0W h

. , . Percerved ear a at

a Friend stannrng Susceptibility to skin 0 Stay 1n the shade

perceptlon and caner

 

 

behavior

 

 

Perceived Severity of

Illness of family member ,
skrn cancer

or friend

 

 

 

 

 

 

Figure 13. A Conceptual Framework of the Health Belief Model
(Modified from Rosenstock, 1974, p. 334)

69

Relative Importance

The health belief model (HBM) has played an important part in explaining and
predicting compliance with professionals’ health-related recommendations. AS mentioned
earlier, the HBM was originally employed to explain preventive behaviors, but many
researchers also adapted the HBM to other health-related behaviors.

J anz and Becker (1984) reviewed 29 published studies related to the HBM
between 1974 and 1984 with respect to research design and 4 dimensions of the HBM
(susceptibility, severity, benefit, and behavioral barrier). These studies were categorized
into 3 groups: 13 studies were about compliance with recommendations for preventive
health behaviors such as influenza vaccination, screening examination, and avoiding
risky behaviors (i.e. seat belt use, drunk driving, smoking behavior, exercise regimen).
Thirteen other studies were about compliance with treatment regimens when people were
ill. Treatments were for hypertension, diabetes, or mothers’ compliance with a regimen
for their child’s illness. The final category included 3 studies about clinic utilization for
various reasons. When analyzing the significant findings fi'om 29 studies, the behavioral
barrier dimension was the most significant in explaining compliance with health-related
recommendations, followed by benefits, susceptibility, and severity, in that order.
Though perceived severity was the poorest variable in explaining compliance with
preventive health suggestions, it was the second most important variable in explaining
compliance with prescribed treatment for the sick, because “this HBM dimension is more
meaningful to individuals diagnosed as ill and/or experiencing symptoms” (J anz &
Becker, 1984, p. 37). The susceptibility dimension was more significant in preventive

behavioral studies than studies about compliance with recommended treatments for the

70

 

(ti
“Va
tin
Wt.

0608

Pitta.
t 1‘11ng p.

 

sick, whereas dimension of benefits revealed the reverse results. In conclusion, the
studies confirmed the significance of the HBM as a framework in health-related research
and all dimensions of the HBM contributed to understanding individuals’ health—related
behaviors.

With respect to Skin cancer preventive behaviors, a number of researchers have
adapted the HBM. Glanz, et al., (1999) investigated the predictors that influenced sun
protective practices of children (6-8 years old), their parents, and outdoor recreation
staffs. The researchers found that behavioral barriers and benefits were the most
significant predictors for explaining parents’ preventive behaviors, whereas program
policies of sun protection were the most significant to children’s and staffs’ behaviors.

Behavioral barriers were also the most significant variable among the HBM
dimensions to predict skin cancer prevention practices of Australian college students
(Cody & Lee, 1989). Jackson and Aiken (2000) explored the causal relationships among
HBM variables to predict the intention to practice sun protection among 202 female
Caucasian college students. The significant predictors were norms for sun protection,
advantage of sunbathing (psychological barrier in the present study), susceptibility, and
barriers in order. Severity was not directly related to this dependent variable. In their
study, norms for sun protection were set by peer’s sun protective behavior, which is
considered as cues to action in the present study. 7

In a study of health beliefs among 202 Wisconsin male farmers, approximately
three quarters of them believed that they were more susceptible to Skin cancer than the
average person, that skin cancer was a serious disease, and that sun protective practices

would decrease the probability of developing skin cancer (Marlenga, 1995). However,

71

they did not carry out sun protective behaviors in real life. Almost 70% of the
respondents never or rarely wore sunscreen, long-sleeve shirts, or wide brimmed hats.
The major reason of not wearing long-sleeved shirts was because they were too hot to
wear. Also, respondents did not use sunscreen because they forgot to wear it and/or it
took too much time. Their responses indicated that the barriers to wearing sunscreen,
long-sleeved shirts or hats seemed to override the other HBM dimensions, since they did
not practice sun protective behaviors.

Shem (1986) explored the HBM to predict the likelihood of pesticide applicators
to adopting protective clothing as the preventive behavior. She found that risk takers who
did not take preventive actions had a lower level of perceived susceptibility to and
perceived severity of pesticide-related illness and thought benefits of pesticide use
overweighed the risks. Risk avoiders who complied with protective actions possessed
opposite perceptions.

In conclusion, most dimensions of the HBM were associated with the decision
whether or not to take compliant behavior, although the extent of the significance varied
with the research designs. Significant predictors varied with prevention versus treatment
status of respondents, types of health-related practices, and demographics of the sample,
i.e. gender, age (children, adults, or senior citizen). However, behavioral barriers to
practicing protective actions are relatively significant across studies in predicting whether
the recommended health actions are followed. Psychological barriers and cues to action
are significant as determinants of sun protective behaviors. Overall, analyzing the relative
importance among dimensions of the HBM for the specific sub-population group would

be necessary for preparing more appropriate information.

72

Attributes of Innovation

Another theoretical framework employed for this study is Rogers’ diffusion
theory (1995). According to Rogers, diffusion of an innovation is defined as “the process
by which an innovation is communicated through certain channels over time among the
members of a social system” (p. 5), and an innovation is defined as “an idea, practice, or
object that is perceived as new by an individual or other unit of adoption” (p. 11). A new
style of dress can be explained with respect to diffusion of innovation since dress is an
important communication cue between source and receiver who encode and decode its
meaning at a given time and place. The innovation for this study is a UV specialized shirt
that is specially designed to maximize protection from UV radiation. Adoption of this
innovation is related to skin cancer prevention practices. An innovation, like UV
specialized shirts adopted for preventive purposes, is defined as a preventive innovation
“that an individual adopts at one point in time in order to lower the probability that some
future unwanted event will occur” (Rogers, 1995, p. 70).

An innovation can be described by five attributes that influence the rate of
adoption of the innovation (Rogers, 1995). Studying attributes of innovation is important
due to “its predictive potential to assist in actual diffusion and to increase our theoretical
understanding of diffusion” (Dearing, Meyer, & Kazmiereczak, 1994, p. 18). Yet,
attributes of an innovation in diffusion of clothing has been insufficiently studied.
Understanding the potential adopters’ perceptions prior to diffusion would be useful in
modifying and developing the innovation to match better with their tastes. Five attributes

of innovations are relative advantage, compatibility, complexity, triability, and

Observability.

73

Relative Advantage

Relative advantage is defined as “the degree to which an innovation is perceived
as being better than the idea it supersedes” and different kinds of advantages such as
economic, social prestige, or other noneconomic benefits that could be considered
depending on the nature of an innovation (Rogers, 1995, p. 212). The more potential
adopters perceive that the innovation possesses the advantages of price, image, or quality,
the greater the likelihood that the innovation will be adopted.

Relative advantage then could be illustrated through several dimensions. Since
relative advantage “has been defined rather broadly but operationalized more narrowly,”
Dearing, et a1. (1994, p. 20) disaggregated relative advantage into two mutually exclusive
attributes, economic (monetary benefit) and noneconomic advantage (effectiveness).
Ostlund (1974) disaggregated relative advantage into three dimensions, time saving,
effort saving, and monetary value, and showed different rankings of importance as the
predictors in discriminating between buyers and non-buyers of two new consumer
products. Moore and Benbasat (1991) and Agarwal and Prasad (1997) disaggregated
relative advantage into two dimensions, image (social status) and relative advantage.

For this study, relative advantage was subdivided into economic advantage, social
prestige (image), and overall usefulness. Economic advantage represents cost
effectiveness by adopting the innovation. Since price is an important criterion in clothing
purchase decisions (Eckman, Damhorst, & Kadolph, 1990; Lee & Burns, 1993), it is
crucial to incorporate the economic aspect of relative advantage.

Rogers (1995) emphasized social prestige as a predictor especially in clothing

diffusion. Agarwal and Prasad (1997) used the term, image instead of social prestige,

74

 

since “image captures the perception that using an innovation will contribute to
enhancing the social status of a potential adopter” (p. 562). Image (social prestige) was
also an important factor that influenced diffusion of innovation from the perspective of
structural equivalence (Moore & Benbasat, 1991).

Overall usefulness of an innovation is another dimension of relative advantage,
defined as “the extent to which a potential adopter views the innovation as offering an
advantage over previous ways of performing the same task” (Agarwal & Prasad, 1997,
p.562). This dimension can be applied in this study by determining to what extent the
potential users would perceive UV specialized clothing as better than typical clothing in
terms of quality, performance, fashionability and effectiveness of UVR prevention.
Compatibility

Compatibility is defined as “the degree to which an innovation is perceived as
consistent with the existing values, past experiences, and needs of potential adopters”
(Rogers, 1995, p. 224). Compatibility is positively associated with the rate of adoption:
The more the innovation is compatible with potential consumers’ needs or values, the
more likely they will adopt the innovation. For instance, if potential adopters put small
value on their susceptibility to develop skin cancer, they would have a lower probability
of adopting UV specialized clothing because they do not have sufficient need for the
innovation. However, if they are highly concerned about health regarding skin cancer,
they are more likely to adopt the innovation because it is compatible with their felt needs.

Rogers (1995) also stressed the importance of compatibility with past experience.
People do not accept an innovation if it is too radical or too different from their previous

experience with comparable products. Thus compatibility could be viewed in terms of

75

compatibility with the existing wardrobe as Littrell and Miller (2001) used for their study
of clothing diffusion. This notion is comparable with the applicability in Dearing, et al.’s
study (1994) which was defined as “the degree to which an innovation is communicated
as having more than one use or having use in more than one context” (p. 23). Since
versatility (“adaptability of the garment to various end users and mix-and-match
potential”) and matching (“suitable with some other garment or garments they owned”)
are important criteria when purchasing clothing (Eckman, et al., 1990, p. 17), it would be
worthwhile to identify the degree to which the innovation is perceived as compatible
(matching) with their own garments.

Compatibility with social norms can inhibit or encourage the likelihood of
adoption. Social (interpersonal) or situational factors are important criteria in purchasing
dress (Chang, Burns, & Noel, 1996). If the innovation is consistent with or appropriate to
the potential adopters’ social culture, it is more likely to be adopted. That is, people tend
to wear garments similar to their peers’ because of a desire for conformity or membership
in their social groups (Engel, Blackwell, & Miniard, 1990). Thus, although a preventive
innovation (UV specialized clothing) is perceived superior to traditional golf wear, if it is
not compatible with the group norm of the potential adopters, the probability to adopt
would be lower.

Complexity

Complexity is defined as “the degree to which an innovation is perceived as
relatively difficult to understand and use” (Rogers, 1995, p. 242), and negatively
associated with the probability of adoption. Agarwal and Prasad (1997) used the term,

ease of use for complexity, meaning “relatively free of effort” and “easier to use and less

76

complex” (p. 562). Littrell and Miller (2001) argued that complexity of clothing would
be high when additional learning of garment construction, fabric, silhouette, or fasteners
were required due to lack of familiarity with the garment. For the present study,
complexity was focused on two aspects: ease of wearing and difficulty in understanding
the function of UV prevention with respect to the innovation. UV specialized clothing is
designed to reduce exposure to UV radiation, but the appearance of clothing is similar to
regular garments. Hence, it is expected that there would be little difficulty in wearing or
caring for UV specialized clothing compared with other typical golf garments, but
potential adopters may not understand the UV protective function.
Triability

Triability is defined as “the extent to which potential adopters perceive that they
have an opportunity to experiment with the innovation prior to committing to its usage”
(Agarwal & Prasad, 1997, p. 562). Availability of a small sample or a trial period with a
30-day return policy among commercial products could encourage a higher level of
triability. If potential users have an opportunity to try or use the innovation before
actually adopting, it would reduce uncertainty about the product. Triability will be a more
important attribute for highly involved products, such as clothing, computers or
automobiles because those product categories have higher economic risk or performance
risk, and are more associated with self image than lower-involved products (Assael,
1998). Triability is positively related with the rate of adoption: The higher the degree of

triability of the innovation, the more likely consumers adopt the innovation.

77

Observability

Rogers defines Observability as “the degree to which the results of an innovation
are visible to others” (1995, p. 244). As Rogers mentioned, the preventive innovation
possesses relatively lower Observability because it is difficult to identify a causal
relationship between the adoption of the innovation and the probability of an unwanted
event occurring. For instance, the person who adopts this innovation could develop Skin
cancer due to cumulative effects of UV radiation that occurred before adopting it. Since
the role of a preventive innovation, such as a UV protective shirt, is to prevent a future
unwanted event, results of wearing it are invisible and intangible.

Agarwal and Prasad (1997) separated Observability into two attributes: result
demonstrability and visibility. Result demonstrability refers to the communicability and
tangibility of the results of an innovation, while visibility refers to the degree to which
potential adopters are able to observe the use of the innovation in their situation or social
context. If outcomes of an innovation are more likely to be demonstrated and observed in
a social context, people are more likely to adopt it. Therefore, Observability would be
positively associated with the probability of adoption.

Figure 14 was developed as a conceptual framework to describe the attributes of
an innovation and the intention to adopt. Intention to adopt UV specialized Shirts was
defined in the present study as “a subjective probability... a belief held by the subject
about how likely or unlikely it is that they will act” (Hill & Rassaby, 1984). Thus, having
an intention to adopt UV specialized shirts indicates that respondents are planning to

purchase or wear the innovation.

78

Perceived Attributes of Innovations Intention to Adopt

 

Relative Advantage
Economic
Prestige
Usefulness

 

 

 

Compatibility

Past experience
Existing value

Current needs

Intention to
Adopt UV
specialized Shirt

 

 

 

 

Complexity

 

 

 

Triability

 

 

Observability
Communicability
Visibility

 

 

 

Figure 14. A Conceptual Framework of Attributes of UV Specialized Shirts
Relative Importance

Attributes of innovations were widely employed in the area of information
technology. According to Johnson, Meyer, Woodworth, Ethington, and Stengle (1998),
perceived attributes of an innovation were significantly different contingent on the nature
of innovations in communication of cancer information within one organization. They
examined respondents’ perceived attributes of four different innovations. Respondents
perceived four innovations in a similar way with higher degrees of triability, adaptability,
and acceptance (the mean is within 8.0 to 9.1), but lower degrees of complexity and
compatibility (mean is within 3.5 to 5.4). In another study, compatibility, visibility and
triability were significant predictors of current usage of the World Wide Web, whereas
relative advantage and result demonstrability were significant predictors of the intention

to continue to use the World Wide Web (Agarwal & Prasad, 1997). This study suggests

79

that emphasizing different sets of attributes of an innovation are necessary for different
stages of the innovation-decision process, 6. g. initial adoption and intention to continue to
use.

There are limited studies in the clothing and textiles area with respect to attributes
of innovations. Littrell and Miller (2001) examined the attributes of India-inspired
garments to predict consumers’ acceptance and intention to purchase the innovation. The
perceived characteristics of the innovations consisted of complexity, familiarity with
overall garment features (defined as awareness-knowledge, including relative advantage,
compatibility with existing value system, and Observability by Rogers’ definition) and
compatibility (matching) with existing wardrobe. Consumers’ acceptance and intention to
purchase were positively and significantly related to familiarity and compatibility.
Perceived complexity of garment features such as construction or Silhouette did not have
a significant influence on those two dependent variables. In addition, consumers’
acceptance of the innovation was a significant predictor of intention to purchase.

Researchers have employed attributes of an innovation to discriminate between
adopters and non-adopters of innovations. Shim and Kotsiopulos (1994) explored the
influence of perceived characteristics, individual characteristics, and organizational
characteristics on retailers’ technology innovativeness. They classified apparel and gift
retailers into five adopter categories and compared the differences in the perceived
characteristics of new retail technology. The results showed significant differences in
perceived relative advantage, compatibility, and Observability among adopter categories:
The innovators or early adopters were more likely to perceive those three attributes

positively than late adopters.

80

.AA

The relative importance of attributes of innovations is different based on the
nature of an innovation. Analyzing attributes of innovation would help to increase the
rate of adoption by modifying negative attributes and strengthening the positive attributes
of the innovation. However, very little research has focused on examining attributes of
the preventive innovation. Identifying attributes of UV specialized clothing would
provide a baseline for implementing strategies to increase sun protection practices by
golfers.

Attributes of an Innovation and Health Belief Model

Potential adopters’ intention to adopt an innovation would be influenced by the
previous experiences or beliefs about a certain situation. According to the innovation-
decision process, attitudes toward an innovation are formed after knowledge of the
innovation is obtained or received (Rogers, 1995). Knowledge is shaped and influenced
by consumers’ previous practices, felt needs, and influences of Significant others, because
they have received information that is relevant to their interests, beliefs, or needs.
Analyzing consumers’ perceived attributes of an innovation is a diagnostic tool to
understand their attitudes toward the innovation, so is an indicator of identifying their
intention to adopt it. Relative advantage, compatibility, triability, and Observability, as
perceived by members of a social system, are positively related to the rate of adoption,
while complexity is negatively related to the rate of adoption (Rogers, 1995).

In terms of adopting UV Specialized shirts, if an individual has a strong felt need
to prevent skin cancer so practices sun protection, s/he is more knowledgeable about UV
specialized Shirts, which influences her/his perception of characteristics of the preventive

innovation. This individual would have more a positive perception of compatibility or

81

relative advantage toward UV specialized shirts than a person who does not have a need
for sun protection. Then, the individual’s positive attitudes toward the UV specialized
shirt would lead to a higher intention to adopt it.

Since researchers have done little work to explain attributes of a preventive
innovation related to perceivers’ actual health-related practices and beliefs, the
associations between these two theoretical frameworks would be worthy of a diffusion
study. On the basis of Figure 13 and Figure 14, the two theoretical frameworks are

combined to pr0pose a psychosocial model for this study, Figure 15.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Modifying Beliefs of Current Sun Attitudes toward
Factors Sun Protection Safety Practices Innovation
Perceived
Demographic — Benefits —
Factors
0 Gender
0 Age Behavioral
0 Education _ Barriers Sun Protective
a Race Behaviors .
0 Marital status Attributes 0f
Psychological i—— 0 Use sunscreen Innovatrons
_ Barriers a Wear lon D Relatrve
C A ti 7" l d E. ‘H advantage
ue to c on s eeve S m D Compatibility
and long pants a C .
omplexrty
Interpersonal _ 0 Wear a hat a Triabili
interaction Percerved. . 0 Stay in the a Ob th’l'
D Peer’s protective Susceptrbrhty to shade serva ‘ 1ty
behavior ,skrn caner
0 Peer’s tanning .
perception and Perceived .
behavior Seventy 0f skrn Intention to
cancer Adopt UV
Illness of family specialized Shirt
member or friend

 

 

 

 

 

 

Figure 15. A Proposed Research Model for adoption of a UV Specialized Shirt

82

Implications

The UV specialized shirt is employed as an innovation for this study, Since it is
relatively new to golfers based on market research. Although a few UV specialized
garments are available in the marketplace, most consumers seem unaware of its
availability or importance. Hence, potential adopters seem to be at the knowledge stage in
the innovation-decision process. Evaluating their perceived attributes of the innovation
would be helpful in understanding their attitudes toward it, thereby influencing their
intention to adopt UV Specialized shirts. Information about perceived health beliefs
regarding sun protection would assist in understanding how attributes of the UV
protective shirts are formed.

This study was conducted to learn about the health beliefs and behaviors of the
potential adopters. Findings would be useful to sun safety campaign designers who target
outdoor enthusiasts by identifying the influential factors in altering their lifestyles and
practices. This study would also contribute to our knowledge of diffusion of clothing
fashion. Most diffusion studies in the area of clothing and textiles are focused on
innovativeness of adopters (fashion opinion leadership or innovators) (Goldsmith, Flynn,
& Moore, 1996; Huddleston, Ford, & Bickle, 1993; Shim & Kotsiopulos, 1994;
Workman & Johnson, 1993), but research about perceived characteristics of an
innovation from the perspective of potential adopters has received little attention. This
study would examine golfers’ perceived characteristics of a preventive innovation based
on their health-related beliefs, and would be beneficial for people in product development
by assessing the potential adopters’ reactions that could lead to modifying the garment at

a pre-diffusion stage.

83

Sun Protective Behaviors

Several factors were associated with sun protective behaviors. Among 10,048 US.
white respondents in the 1992 National Health Interview Survey, 53% practiced at least
one of three types of sun protective behaviors: 32% reported using sunscreen/sunblock,
28% wearing protective clothing, and 30% staying in the Shade (Hall, May, Lew, Koh, &
Nadel, 1997). Being old, being Single, having sensitive skin, having repeated sunburn
experiences, and having a history of Skin cancer or Skin cancer exams were significant
factors for respondents engaging in one or more sun protective behaviors.

In a study comparing two national telephone surveys conducted in 1986 and in
1996 by the American Academy of Dermatology, use of sunscreen increased from 35%
in 1986 to 54% in 1996, while regular use of a tanning booth also increased from 2% to
6% (Robinson, Rigel, & Amonette, 1997). When the duration of weekend outdoor
exposure was compared, white males working indoors, with higher incomes tended to
have longer exposure than their opposites. With respect to awareness of skin cancer
prevention information, being white, female, old, and educated were more positively
associated than their opposites. They emphasized that analyzing sun protective behaviors
of specific subsets of the Caucasian population was important to develop adequate
information to the target market. Although the proportion of the population using
sunscreen or having an awareness of skin cancer risks had increased, each subset was still
associated with some skin cancer risk-creating behaviors. For instance, male adults and
those 18-25 years of age tended to experience sunburn more frequently from occupational
and recreational exposure, while women 18-25 years of age used tanning booths more

frequently than the other subsets.

84

Sunburn, experienced in childhood or adulthood, has been related to an increased
risk of melanoma and basal cell carcinoma (Saraiya, et al., 2002). Purdue, Marrett, Peters,
and Rivers (2001) examined the predictors of sunburn among Canadian adults, and found
that being young, being male, and having light Skin color and non-black hair color were
Significantly related to sunburn experiences. Furthermore, spending more leisure time in
the sun, seeking a tan, working outdoors in the summer, and forgetting about protecting
oneself from the sun were significant predictors of sunburn.

Stepanski and Mayer (1998) explained sun protective behaviors among US.
outdoor workers. Of 312 outdoor workers, 50.4% adequately protected themselves from
the sun, but the face and lower arms were the body sites the least protected from the sun.
When the researchers surveyed 240 outdoor workers, California transportation workers
(Caltrans) tended to practice sun protective behaviors most, followed by construction
workers and postal carriers. The researchers pointed out that the clothing policies for each
occupation generated different levels of sun protective behaviors. For instance, Caltrans
and construction workers were required to wear long pants, resulting in higher mean
scores of wearing protective clothing than postal carriers. Across the occupational
groups, wearing sunglasses was the most frequently reported sun protective behavior
(mean = 3.96, with the maximum possible point of 5), followed by wearing a hat (mean =
3.56), wearing other protective clothing (mean = 2.99), and staying in the shade (mean =
2.54). Using sunscreen with SPF of 15 or higher and limiting mid-day sun exposure were
the least frequently reported behaviors.

Vail-Smith and Felts (1993) examined 296 Caucasian college students in the

southeastern US. The majority (61%) were young women 18 years of age. The college

85

students were concerned about the negative effects of sun exposure, such as getting skin
cancer (57.3%) and aging skin prematurely (63.2%). However, only 18.5% of women

and 14.7% of men used sunscreen with at least the SPF of 15 when they went to the
beach and rarely use it at any other time. The proportion of respondents using sunscreen
was much smaller than the proportion of outdoor workers. Half of respondents also
enjoyed sunbathing. Female students were likely to sunbathe more frequently than were
male students. Students who sunbathed frequently (2 or 3 times per week) tended to have
fewer self-perceived risk factors and to have strong beliefs associated with the advantages
of a suntan.

Age was a Significant predictor of perceived susceptibility to Skin cancer in
Grubbs and Tabano’s study (2000): The older the respondents, the more likely they
perceived susceptibility. Grubbs and Tabano stressed the importance of targeting the
younger age group regarding Skin cancer risks and sun precautions.

In summary, males, whites, and indoor workers were more likely to enjoy outdoor
activities. Women, elderly groups, or those having sun sensitive skin or a history of skin
cancer were more likely to take at least one sun protective behavior. Outdoor workers
tended to be more dependent on wearing protective clothing such as a brimmed hat or
long-sleeved shirt than using sunscreen or staying in the Shade. Younger age groups and

women were more likely to have a suntan.

86

Methodology

This section describes the research methods for the second phase. Topics include
instrument development, sampling, data collection procedures, results of a pretest, and
data analysis methods.

Instrument Development

A self-administered questionnaire was developed because standardized
instruments to measure the health belief model or attributes of innovation did not exist.
Each item was modified from previous studies. The instrument consists of three sections:
1) beliefs about skin cancer risk and tanning, and current sun protective behaviors; 2)
perceived attributes of UV specialized Shirts and the intention to purchase the shirt; and
3) demographics. Every continuous variable was measured using a Likert scale, since
“Likert scaling is widely used in instruments measuring opinions, beliefs, and attitudes”
(DeVellis, 1991, p. 69).

The first section was based on the health belief model with 36 items modified
from previous studies (Arthey & Clarke, 1005; Carmel, et al., 1994; Grubbs & Tabano,
2000; Jackson, 1997; Jackson & Aiken, 2000; Vail-Smith & Felts, 1993). Each item and
its source can be found in Table B1 in Appendix B.

The first two items in the first section are regarding exposed hours of outdoor
activities and sunbathing. The next five items relate to current sun protective behaviors,
including use of sunscreen on the face and body, use of long-sleeved shirts or pants to
cover the body, use of a hat, or use of the shade to avoid exposure. Those five items were

measured using five-point Likert-type scales ranging from “never” (1) to “all the time”

(5).

87

Susceptibility to and severity of Skin cancer are measured by four items each.
Three items measure the benefits of taking sun precautions. Six items measure the
behavioral barriers to taking sun protective behaviors, and five items measure the
psychological barriers to sun protection (perceived advantages of tanning). Cues to action
are measured by 7 items: two items for friend’s sun protective behaviors, two for illness
of family/friend and influence of media, and three for friend’s suntan perception and
behavior. Subjects answered each item on five-point Likert-type scales ranging from
“strongly disagree” (1) to “strongly agree” (5).

In the second section, a total of 26 items were modified from Jackson and Aiken,
(2000), Moore and Benbasat, (1991), Shim and Kotsiopulos (1994), and Strutton,
Lumpkin, and Vitell, (1994). There are few scales to measure attributes of innovations
with respect to apparel products, so the researcher needed to extensively modify the
instrument previously used to measure adoption of information technology.

Eight items measure relative advantage: two items for economic advantage, three
for image/prestige, and three for general usefulness. Compatibility and observability each
are measured by four items, and complexity and triability by three items each. The
intention to adopt UV specialized shirts is measured by four items. All items employed
five—point Likert-type scales ranging from “strongly disagree” (1) to “strongly agree” (5).
Each item and source of literature is provided in Table B1 in Appendix B.

For the second section, the researcher provided a black-and-white picture for the
pretest (Figure Bl , in Appendix B) and a color picture for the final test so
respondents gave rating on each attribute item based on the same product. The

photograph, from a commercial catalog supplying sun protective clothing included a male

88

model wearing a long-sleeved polo shirt with dark long pants. Shirts in different colors
and price information were included in order to provide a variety to respondents. The
UV specialized Shirts in the catalog are designed to offer a SPF (sun protection factor) of
30 or more. The description of the products in the picture was provided to help
respondents in understanding the general function of UV specialized shirts.

The last section includes the demographic information: gender, age, education,
marital status, and ethnicity. In addition, the questionnaire asked how often and when the
respondents usually play golf. These two questions would indicate how enthusiastic the
respondents were about golf. The pretest included a total of 69 items.

Sampling

A convenience sampling method was used to collect the data. For this study,
golfer is defined as a person who has played golf. Because it is difficult to figure out who
has played golf in the general population, the researcher visited the golf courses to select
a sample among golfers who finished their plays for the day. Also, students of a
professional golf management (PGM) program at Ferris State University were recruited
to participate.

Pretest

The pretest was conducted at Forest Akers East Golf Course during September
2002. Of the 21 useful responses, 14 respondents were male and seven were female
golfers, ranging in age from 24 to 42 years, with a mean of 30 years.

The pretest was conducted in order to verify the internal consistency among items
of each variable and to check for ambiguity. Items were deleted and modified based on

the results of internal consistency using Cronbach’s alpha coefficients. The alpha for each

89

variable is reported in Table B1 in Appendix B. In addition, three professionals examined
the questionnaire for content validity before the final survey was conducted.

As a result of the pretest responses, sun-related activities, current sun protective
behaviors (SPB), benefits, and psychological barriers were not changed. However, one
item on susceptibility (#10) and one item on severity (#15) were deleted to increase the
alpha and to reduce the number of total items. Two of six items in the behavioral barrier
measurement (#23 & #24) were deleted for the same reason. One item was added
regarding whether wearing a long-Sleeved shirt restricted golfer’s movement. Thus, the
behavioral barrier was focused on the barriers to wearing protective clothing while
playing golf. Among the items of cues to action, the fifth and sixth items (#34 & #35)
were reverse scored, since these two items were negatively correlated with the other
items. Two items in cues to action were removed: The third item (#32) was moved into
the demographic section to be asked as binary responses (yes or no), and the fourth item
(#33) was deleted due to low inter-item correlations. After deleting two items in the
variable cues to action, the alpha increased to 0.68, the minimally acceptable level of
alpha (DeVellis, 1991). After the pretest, 31 items in the first section were used in the
final questionnaire to measure the beliefs about skin cancer and sun protective behaviors.

In the second section, seven items were deleted in order to increase the alpha
scores and reduce the total numbers of items: two items of relative advantage (#1 & #7),
one item of compatibility (#9), two items of complexity (#14 & #15), one item of
triability (#17), and one item of intention to adopt (#26). One item of observability (#20)
was moved to measure complexity, since respondents considered it difficult to understand

the UV protective function rather than difficult to communicate the advantages. A total of

90

19 items was employed to measure the attributes of an innovation and the intention to
adopt it.

In section three, an open-ended question covered the frequency of golf rounds per
each month. Also, the format for measuring the time of a day respondents play golf was
changed. Respondents were instructed to check all that applied, because some
respondents mentioned that they were more likely to play golf any time without specific
patterns and they played more than 3-4 hours for a round of golf, so 2 hour intervals were
not appropriate to the time categories. As mentioned earlier, one item for cues to action,
having family members or fiiends who have skin cancer was included in the demographic
section. The final questionnaire included 58 items and was presented with the cover letter
(Appendix B).

Data Collection Procedure

A self-administered questionnaire was distributed to golf players between
September, 2002 and January, 2003. The superintendent of Forest Akers Golf Courses at
Michigan State University was contacted through letter and phone in order to get
permission to distribute the survey to golfers at the golf courses or at the official events.
After obtaining permission, the researcher visited the two golf courses of Michigan State
University on different days of the week between 2 pm. and 7 pm. and on the dates of
scheduled events between September and October, 2002. The potential subjects who
were heading to parking lots after playing golf were randomly approached. Then the
researcher explained the purpose of the study and gave out the questionnaire to them so
that they had a chance to read the cover letter, since it contained specific information

about the purpose of the study and the elements of informed consent. If a subject agreed

91

to participate voluntarily, the researcher considered it as a verbal consent. If a person
refused to participate, the researcher approached another person. In addition, the
researcher visited two golf practice ranges in Lansing area, and obtained the data in the
same way at the golf courses. There is no control of subject’s gender, age, or ethnicity.

The researcher also contacted the instructors of professional golf management
(PGM) program at Ferris State University to obtain their permission to distribute the
questionnaire to their students. The PGM program, authorized by the Professional
Golfers’ Association of America, trains and educates future golf professionals for the
needs of the golf industry. The use of college-aged sample would be helpful to
understand this subset of the population who are less likely to take sun protective actions
and more likely to have positive attitudes toward sun exposure (Hall, et al., 1997;
Robinson, et al., 1997). In addition, analyzing young people’s attitudes toward the
innovation would be challenging for designers, since this age group is probably less
concerned about health and more interested in appearance. The data at Ferris State
University were collected between November 2002 and January 2003.

Data Analysis Methods

The summated rating scale was developed based on Cronbach’s alpha coefficient
as the measure of internal consistency. The items to measure each variable were summed.
For instance, to measure the variable sun protective behavior (SPB), the respondent’s
rating on each of five items was summed, generating a summated score for SPB. The
advantage of the summated rating scale is to reduce the errors of measurement by
combining multiple items, producing a more reliable measurement of a variable (Spector,

1992).

92

Three types of data analysis methods were used: descriptive statistics of all the
independent and dependent variables in the present study, bivariate relationships among
the dependent and independent variables by Pearson correlation coefficients, and multiple

regression analyses on the two dependent variables.

93

Findings and Discussions

This section presents the data analysis and research findings for the second phase
of the study. The first part describes the findings of data analysis. Next, the discussion of
findings and a summary for each research question are provided.

Data Analysis and Findings

Description of Demographics

A total of 158 useful questionnaires were collected: 106 from golf courses and 52
from the PGM program of Ferris State University. Among eligible participants, the
response rate was 68.1%; 21.6% refused to participate and 10.3% did not complete the
questionnaire. Ninety-one percent of the respondents was male (n = 143) and the
respondents’ age ranged from 16 to 80 years (mean age = 30 years). Although the
researcher tried to approach both sexes at the same rates, male golfers were dominant in
the selected golf courses. About 74% were Caucasians (n = 117) and 65% were single (11
= 103). Thirty-five of 158 respondents (22.2%) answered that they have family members
or friends who have skin cancer.
Description of Variables in the Health Belief Model

Sun-related activities. Frequency descriptions of sun-related activities are
presented in Table 17. Regarding outdoor activities or work in a typical week during the
summer, 40% of the respondents spend more than 25 hours in the sun in a week.
Approximately 54% did not sunbathe in the summer.

Sun protective behaviors. Table 18 presents the distribution of sun protective
behaviors (SPB). About 31% of the respondents often to all the time wore sunscreen on

the face and 20% used it on the body when outdoors, while more than 80% never or

94

Table 17. Sun-Related Activities in a Typical Week in the Summer

 

1-2 3-4 5-7 8-10 11-15 16-20 21-25 25 or
0 hr hrs hrs hrs hrs hrs hrs hrs More Total
N N N N N N N N N N

(W (%) (°/o) (°/o) (%) (%) MD M) (W (°/o)

 

 

 

01.190.“ 3 6 16 22 17 19 12 63 158

actrvrtres/ 0

work (1.9) (3.8) (10.1) (13.9) (10.8) (12.0) (7.6) (39.9) (100)

Sunbathe 85 38 18 5 6 1 1 1 0 155
(53.8) (24.1) (11.4) (3.2) (3.8) (0.6) (0.6) (0.6) (98.1)

 

rarely wear a long-Sleeved Shirt to cover the body from the sun. About 47% of the
respondents wore a hat when they were in the sun. This might be attributed to the high
proportion of males in the sample, since men are more likely to wear a hat (Garbe &
Buettner, 2000). Also, hats are frequently seen in published images of golf players: About
67% of male golfers were wearing hats. Finally, 26% often to all the time tried to stay in
the shade when outdoors. Based on the description in Table 18, approximately a quarter
of respondents practiced one of the recommended behaviors except for wearing sun
protective clothing often to all the times when they were outdoors.

Table 18. Current Sun Protective Behaviors

 

Sometimes Often All the time

 

 

 

 

 

 

SPB Never Rarely Mean
(Alpha = .63) N N N N N SD
(°/o) (%) (°/o) (%) (%)
Sunscreen 36 25 48 35 14 2.78
on the face (22.8) (15.8) (30.4) (22.2) (8.9) 1.27
Sunscreen on 41 34 51 25 6 2.50
the body (26.1) (21.7) (32.5) (15.9) (3.8) 1.15
Wear 79 51 21 7 0 1.72
Clothing (50.0) (32.3) (13.3) (4.4) 0.86
Wear a hat 15 24 45 43 31 3.32
(9.5) (15.2) (28.5) (27.2) (19.6) 1.22
Stay in the 21 41 54 31 10 2.80
Shade (13.4) (26.1) (34.4) (L97) (6.4) 1.10

 

95

In order to find out respondents’ overall sun protective behaviors, the summated
scale of SPB was obtained by the sum of the scores of all five items. The scores of the
SPB variable ranged from 5 to 23, with a mean of 13.1. Higher scores indicate more sun
protective behaviors. The scale of SPB was used as a dependent variable in testing the
health belief model, and as an independent variable in testing the proposed research
model for this study. Cronbach’s alpha of all five items was .63.

Susceptibility and severity. In Table 19, about 30% to 50% of the respondents
agreed or strongly agreed on being susceptible to getting skin cancer as indicated by three
items, whereas 68% to 87% of the respondents agreed or strongly agreed on the two
items for severity of having skin cancer. In addition, about 72% were concerned about
aging due to sun exposure.

Table 19. Perceived Susceptibility and Perceived Severity

 

 

 

 

 

 

 

 

. . . = Strongly . . Strongly Mean
Suscept1brl1ty (Alpha .75 ) Disaeree Drsagree Neither Agree Agree SD
1 am more likely than the average person 2.84
to get skin cancer. 21 4O 52 34 1] 1.12
The possibility of skin cancer worries 19 40 44 45 10 2.92
me. 1.13
Whenever I hear of a friend/relative (or 3 36
public figure) getting skin cancer, it 8 25 42 66 16 1'03
makes me realize I could get it too. '
Severity (Alpha = .71 )
Getting skin cancer would severely affect 5 20 25 60 48 3.80
my life. 1.10
It would be terrible to have skin cancer. 0 6 15 53 83 3:?
It would be terrible to look older than I 3 .92
really am due to too much sun-exposure. 3 l l 3] 63 50 0,98

 

The summated scale for the two variables was also obtained so that the sum of the

items for susceptibility and severity would represent each variable in the regression

96

equations. The higher scores indicated more agreement that they were susceptible to skin
cancer and getting skin cancer was serious to them. Cronbach’s alphas of susceptibility
and severity were .75 and .71 respectively.

Benefits of taking sun protective behaviors. About 80% of the respondents agreed
or strongly agreed on the benefits of covering up their body to reduce the chances of
getting skin cancer or aging their skins (Table 20). Fewer people strongly agreed with the
second item that linked skin cancer with sun exposure compared with the other two items
of benefit: Only half of respondents agreed or strongly agreed that sun avoidance was
related to skin cancer prevention. Because respondents were enthusiastic about outdoor
sports, especially playing golf, they seemed not to connect the outdoor activity with the
chance of developing skin cancer, although they believed that taking sun precautions
would reduce skin cancer risks.

Scores for the three items relating to benefits were summed, with higher scores
indicating more perceived benefits of taking sun protective actions. Cronbach’s alpha for
the three items of benefits yielded .80.

Table 20. Perceived Benefits of Practicing Sun Protection

 

 

 

 

_ Strongly . . Strongly Mean

Benefits (Alpha .80) Disagree Disagree Neither Agree Agree SD
Wearing a long-sleeve shirt, 3 brimmed
hat, or sunscreen when I am in the sun 4.15
would reduce my chances of getting skin 4 4 23 6O 67 0,94
cancer.
Whether or not a person develops skin
cancer is related to how frequently they 3.52
avoid exposure while spending time in the 4 18 51 62 23 0,96
sun.
If more people cover up their bodies when 3 82
they are in the sun, they would reduce the 6 13 28 68 43 1'05

chances of getting age spots and wrinkles.

 

97

Behavioral barriers and psychological barriers. Most respondents perceived
barriers to practicing sun protective behaviors. Table 21 describes behavioral barriers and
psychological barriers (perceived advantages of tanning). More than 80% of the
respondents answered that they would not wear a long-sleeved shirt or a wide-brimmed
hat while playing golf if they were uncomfortable. Regarding the items that represent
psychological barriers, about 49% to 54% believed that tanning made persons look
healthier and more attractive than people without a tan.

Table 21. Behavioral Barriers and Psychological Barriers

 

 

 

 

 

 

 

 

 

 

 

 

Behavioral Barriers Strongly . . Strongly Mean
(Alpha = .60 ) Disagree Disagree Neither Agree Agree SD)
If a long-sleeve shirt feels uncomfortable, 4.26
I would not wear it for golf. 5 9 12 46 86 1.04
If a wide brirnmed-hat feels 4 20
uncomfortable, I would not wear it for 4 9 18 48 79 '

.5 01f. 1.02
The cost of UV specialized clothing (8 3 59
6000-3 80.00) could keep me from 1 l 16 45 41 45 1°20
purchasing it. '
The style/appearance of UV specialized 3.70
clothing could keep me from wearing it. 4 18 45 46 45 1,08
Wearing a long-sleeve shirt while playing 3.47
golf would restrict my movement. 6 29 4O 51 32 1,12

Psychological Barriers
SAdvantage of Tanning, Alpha = .91)
I feel more attractive with a tan. 19 21 23 59 34 3'44
1.30
A tan makes me feel better about myself. 25 33 36 44 18 %33
. . 2.96

I feel healthier With a tan. 25 29 45 46 13 1 20
A sun-tanned person looks healthier than 3.30
someone without a tan. 15 19 47 57 20 1.13
A sun-tanned person looks more attractive 341
than someone who is not tanned. 13 29 3O 5 3 33 1,24

 

The summated scale of behavioral barriers was created by summing the five

items. The five items had a lower internal consistency than other scales ((1 = .60) because

98

each item measured a different aspect of sun protective clothing. The sum of the five
items, however, would imply the sum of the barriers to wearing sun protective clothing.
Higher scores indicated that respondents perceived more barriers to wearing a long-
sleeved shirt or a hat. Perceived psychological barrier is another type of barrier. The five
items were summed, and the higher the scores, the more advantageous is a tan.
Cronbach’s alpha for psychological barriers was .91.

Cues to action. Cues to action (CA) were divided into two dimensions (Table 22).
CA-SP indicates whether respondents have fi'iends who tend to practice sun protective
actions when outdoors. Only 12.6% of the respondents agreed or strongly agreed that
their golf friends usually wore long-sleeved shirts, while 23.4% answered that their golf
friends used sunscreen when outdoors. CA-Tan represents whether respondents have
friends who are more likely to enjoy a suntan and perceive a suntan positively. More than
50% of respondents agreed or strongly agreed that most of their friends had a suntan and
perceived a suntan positively.

Table 22. Cues to Action of Practicing Sun Protection and Tanning

 

 

 

 

 

 

 

Cue to Act (Alpha = .60) Strongly . . Strongly Mean
CA _ SP Diflree Disagree Neither Agree flee SD
People with whom I play outdoor sports 2 21
usually wear a long-sleeve shirt or hat, 44 61 33 16 4 1'04
when they are in the sun. '
People with whom I play outdoor sports 2 79
usually use sunscreen with an SPF of 15 16 45 60 30 7 1'0]
or more when they are in the sun. '
CA — Tan
' . 3.41
Most of my friends have a suntan. 5 24 44 69 15 0 97
Most of my friends feel that a suntan is a 3.37
good thing. 5 25 46 70 12 0.95
My friends disapprove of people who 2.23
have a dark tan.a 45 54 4] 14 4 1.04

 

’ Deleted item in the scale construction

99

The summated scale of CA-SP was composed of two items. The higher the
scores, the more friends who take sun protective actions respondents have. The other
scale of CA-Tan was composed of two items of the peer’s tanning perception and
behavior and excluded the last item due to low item-total correlation. High scores
indicated that respondents’ fiiends were more likely to enjoy a tan. Cronbach’s alpha for
the four items of cues to action yielded .60.

Description of Variables in the Attributes of an Innovation

Relative advantage. Table 23 shows the frequency distribution on five items of
attributes of UV specialized shirts in the second section of the questionnaire. Although
69% of the respondents agree or strongly agreed on the relative advantage of the UVR
protective function over other garments (6th item), only a small proportion of
respondents perceived the advantage of monetary value, image, or style of UV
specialized shirts positively. The summated scale of relative advantage was created by
the sum of the scores of five items, and excluded the fourth item, to increase the internal
consistency, yielding Cronbach’s alpha of .67. A higher score indicates the UV
specialized shirt has more advantages.

Compatibility. More than 27% of the respondents answered that the UV
specialized shirts in the photograph would coordinate with their garments and be
appropriate for their golf games, but only 18% of the respondents agreed that the product
was compatible with their current needs. The scale of compatibility was developed by
summing three items, with higher scores indicating higher compatibility with felt needs

or values of respondents. Cronbach’s alpha of compatibility was .77.

100

Table 23. Perceived Attributes of UV Specialized Shirts

 

 

 

 

 

 

 

 

 

 

 

Relative Advantage Strongly . . Strongly Mean
(Alpha = .67) Disagree Disagree Neither Agree Agree SD
This product would maximize the quality 2.82
for the money that I spend. 9 46 73 23 6 0,89
Golfers who have this product would have 2.13
more prestige than those who do not. 48 60 31 19 0 0,98
Wearing this product would improve my 2.06
image among my golfing friends. 53 56 37 8 3 0,98
1 would be self-conscious about wearing 2.92
this product.3 12 49 49 33 14 1.09
This product would be more fashionable 2 38
or unique in style than other similar types 32 56 49 20 1 0'97
of shirts. '
This product would provide better UV 3.80
protection than the other garments l have. 4 17 28 66 43 1,04
Compatibility (Alpha = .77)

This product would coordinate well with 2.78
the other garments I have. 26 28 61 4O 3 1.06
This product would be compatible with 2.54
my current needs for a new shirt. 25 53 51 28 1 0.98
This product would be appropriate for my 22 39 49 42 5 2.80

golf games. 103

 

Complexity (Alpha = .64)

 

 

 

 

 

It would be difficult to understand how 2 88
this product works to prevent UV 12 53 48 32 13 1.08
radiation. '

1 would have difficulty explaining why 2.79
wearing this product is beneficial. 13 58 49 25 13 1.07

Triability

It would be easy to try out this product 3.46
without a big commitment. a 6 23 39 7O 19 1.01
If this product, made especially for 3 05
golfers, is available in stores where I buy 20 27 44 59 8 '

my golf clothes, I would try it on. 1-12

Observability (Alpha = .64)

 

 

 

I have seen people wearing this product 1.65
when I am outdoors. ’ 92 4O 17 7 2 0.93
I could communicate the advantages of 3.00
wearing this product to others. 16 26 64 46 6 1.01
Ifl use this prOdUCt, the 3 02
advantages/disadvantages would be 1 l 35 60 44 8 '

readily apparent to me. 0.99

“ Deleted item in the scale construction

101

Complexity and triability. About 40% of the respondents reported a low level of
complexity in understanding the UV protective function, and a high level of triability of
UV specialized shirts. The summated scale of complexity consisted of the sum of the
two items, with higher scores indicating higher levels of complexity in understanding its
function. Triability was measured by an independent item (second one) and excluded the
first item due to the ambiguity and less appropriateness to an apparel product.

Observability. Although only 5.7% of the respondents have seen people wearing
UV specialized shirts when outdoors, about 33% agreed or strongly agreed that they
could perceive or communicate the advantage of UV specialized shirt. The summated
scale of observability was built by the sum of the two items; first item was excluded due
to the lower item-total correlation. Cronbach’s alpha of observability was .64.

Intention to adopt UV specialized shirts. Regarding the Intention to Adopt (IA)
UV specialized shirts, approximately 70% of respondents answered that they did not plan
to buy or wear UV specialized shirts. The summated scale of IA was composed of the
sum of the three items in Table 24. Cronbach’s alpha of the three items was 0.83. The
higher the scores, the greater the intention to wear or purchase UV specialized long-
sleeved shirts. This scale is a dependent variable in attributes of innovations (Figure 14)
and in the proposed model (Figure 15), with the possible range of 3 to 15.

Table 24. Intention to Adopt a UV specialized Shirt

 

 

 

 

Intention to Adopt Strongly . . Strongly Mean
(Alpha = .83) Disagree Disagree Neither Agree Agree (SD)
1 plan to wear a UV specialized shirt for 2.09
most outdoor activities. 45 68 32 12 1 (0.92)
I intend to wear a long-sleeve shirt while 60 54 2 6 1 6 2 2.03
playing golf. (1.03)
I intend to purchase a UV specialized shirt 2.12
within the next few years. 50 59 31 16 2 (1.01)

 

102

Although Cronbach’s alphas of variables in the present study were relatively
lower than the alphas in the pretest (Table Bl, in Appendix B), all of the alphas were
above .60, the minimal acceptable level (DeVellis, 1991).

Description of the Summated Scale

The mean, standard deviation, and minimum and maximum values of the
summated scales used in the regression equations are presented in Table 25. SPB was
positively skewed with a mean of 13.1, which was lower than the midpoint of 15 within
the possible range. Severity, benefits, and behavioral barriers showed higher mean scores
compared with the midpoint of the possible ranges, indicating that respondents perceived
the severity of skin cancer, and benefits of and barriers to taking sun protective behaviors
to some extent.

Table 25. Descriptive Statistics of Variables in the Proposed Research Model

 

 

 

 

 

 

 

 

 

Scales of Health Mean Minimum Scales of Mean Minimum
Belief Model (SD) Maximum Attributes (SD) Maximum
Sun Protective 13.09 5 Intention to Adopt 6.23 3
Behavior (SPB) (3.58) 25 (1A) (2.56) 15
Susce tibili 9.09 3 Relative 13.16 5
P W (2.69) 15 Advantage (3.20) 25
. 12.05 3 . . . 8.11 3
Severity (2.38) 1 5 Compatibility (2.60) 15
1 1.49 3 . 5.67 2
Benefit (2.48) 1 5 Complexity (1.84) 10
. . 19.21 5 . . . 3.05 1
Behavroral Barriers (3 .23) 2 5 Triability (1 .12) 5
Psychological 16.00 5 . . 6.02 2
Barriers (5.30) 25 Obsewab‘l‘w (1.72) 10
Cue to Act-Sun 5.00 2
Protect (CA-SP) (1.72) 10
Cue to Act—Tanning 6.77 2
(CA-Tan) (l .69) 1 0

 

 

The right column in Table 25 presents the mean scores of summated scales of

variables in the attributes of UV specialized shirts. The mean of IA was 6.23 within the

103

possible ranges of 3 to 15, indicating lower intention to adopt. The mean of relative
advantage and compatibility showed lower than the midpoint of 15 and 9 respectively
within the possible range, indicating that respondents did not appreciate these attributes
of UV specialized shirts positively.

Correlation Analysis

Before conducting regression analyses on the two dependent variables (SPB and
IA), Pearson correlation coefficients among independent variables and dependent
variables were examined (Table 26). The dependent variable of SPB (sun protective
behavior) was significantly and positively related to susceptibility (r = .254), benefits (r =
.329), and cues to sun protection (CA-SP) (r = .380), and negatively correlated with
psychological barriers (r = -.233). That is, respondents who were more susceptible to skin
cancer risks and perceived more benefits of sun precautions, were more likely to practice
sun protective actions. Also, if respondents have more fiiends who practice sun protective
actions (CA-SP), the respondents were more likely to take sun protective behaviors. The
negative correlation indicates that respondents who perceive more advantages of tanning
are less likely to practice sun protective behaviors.

In terms of attributes of innovations, four of five attributes were significantly and
positively correlated with IA (intention to adopt); the correlation for complexity was not
significant. Relative advantage (r = .524), compatibility (r = .557), and triability (r =
.569) present moderate relationships with IA, indicating that the more respondents
perceive the relative advantage, compatibility, or triability positively, the greater their

intention to purchase or wear the UV specialized shirts.

104

E
\refivfiy: {but}: at: - - r,-
c€ E u:

.\.v\\.

cams-s :52 23 as a 25322...“ a sausage ..

 

 

2: 2:. .38. ER. 32 LE. LAN. N3- .22. EN.- 22..- .52“. So. .42. Ammmveassam
0>mfl$fl°hn~ 55m

2: :3. .58. a2..- .5. ram. .53.. ..E. 3:.- .a33 2:. :2- 28. aé<e§zwfiw

2: .23. 5.- .5. .23. 52. me. S2- «8.- .22. m2. 2:. 32:530.":

2: 42.- :3. 2:. ME.- ..wmm. :2- ..er n.a. ”2. LE. £225.:

2: 32 so. :2. MS- 82 2:. 25.- 87 22- £53585:

2: *5. .42.- EN. 82- .28- E. :2 E. 5:52.555

so: So; «omm. Nmo. awomr .32. _oo. 3;. $353.32 Pia—om.”

. . . . . . . “5:5:- .-
2: L: - amom *NE N8. N: .36— ~u< 3 8:0.»
. . . . . . U89..— ==m I
oo _ m2 - mmo - om _ moo cmc ~u< 8 8:0 .9
”5:5;

2: 3%. 82 .23. $2 a massage

2: w:. vfi. mm: flotsam—Eegaaomé

 

2: 02.. .28. 3:25.».

2: #2. brain

2: 9.53385 ._

gm 5 a: : S a a b a n v m N : a325,

 

 

 

382 5333. 05 a 83%; 28:5 223350 .8 282

105

With respect to the relationships among variables in the proposed research model,
Figure 15, the correlation coefficients among the health belief model and the intention to
adopt UV specialized shirts were examined. Behavioral barriers (r = -.339),
psychological barriers (r = -.176), cues to sun protection (CA-SP, r = .274) and cues to
tanning (CA-Tan, r = -.250) were significantly related with IA. Specifically, if
respondents believe that there are barriers to wearing sun protective clothing, tanning is
beneficial, and their friends enjoy a suntan, respondents are less likely to intend to adopt
UV specialized shirts. In contrast, if respondents have more friends who practice sun
protective behaviors, they are more likely to intend to adopt UV specialized shirts.
Respondents’ current sun protective behaviors are significantly and positively correlated
with the intention to adopt (r = .384).
Regression Analysis

The summated scales of variables were used for the multiple regression analysis.
Linear regression models were developed based on the conceptual frameworks on Figure
13, Figure 14, and Figure 15. The normality of the residuals of the dependent variable in
the each model was assessed by the Kolmogorov-Smimov test, and four outliers were
deleted for violation of normality of distributions. Each regression model is discussed.

Health belief model in Figure 13. The contribution of independent variables in the
health belief model to the dependent variable, sun protective behaviors (SPB) was
examined. Since demographic characteristics help to explain the variance of SPB
according to Figure 13, three demographic variables, age, ethnicity, and marital status
were used as the independent variables for this study: Ethnicity (1 = Caucasian, 0 =

others) and marital status (1 = single, 0 = others) were recoded as dummy variables, and

106

age was used as a continuous variable. Since 90% of respondents were male golfers,
gender was not used as a variable. The information, years of education completed was
also excluded, since this variable was highly correlated with age, and more than 97% of
the sample completed at least high school education.

The seven summated variables were entered with three demographic variables. In
addition, one variable, the part of cues to act, having family members or friends who have
skin cancer, was included as a dummy variable (1 = Yes, 0 = No). The variables and
results of the regression analysis are presented in Table 27.

Table 27. Regression Analysis of the Health Belief Model Variables on Sun Protective

 

 

Behaviors (Model A)

Independent Variables B B Adjfilsted
Constant 8.266 -001
Age -.038 -.151 .123
Marital Status (Single) -1.958 -.267 ~021 *
Race (Caucasian) -.631 -.079 .358
Susceptibility .359 .277 000*”
Severity .020 .014 .837
Benefits .323 .223 .001" 398
Behavioral Barriers -.080 -.075 277
Psychological Barriers -. 160 -.242 001"
Cues to Act - Sun Protect .662 .318 000*"
Cues to Act — Tanning .172 .082 284

Cues to Act — Family Skin
Cancer
*p < .05, "p < .01, *"‘*p < .001

1.187 .142 .041*

Six of 11 independent variables were significant at the .05 level of probability, F
(11, 140) = 10.07, p = .000. Peer’s sun protective behaviors (CA-SP) was the most

important predictor of SPB in comparison among the values of standardized coefficient

107

beta (6 = .318), followed by susceptibility (,B = .277), marital status (being single, 6 = -
.267), psychological barriers (,6 = -.242), benefits of taking sun protective behaviors (fl =
.228), and knowing family members who have skin cancers (6 = .142). R2 value for this
model was .442, indicating that 44.2% of the variance of the dependent variable, SPB
was explained by the six predictors, with an adjusted R2 value of .398.

According to regression assumption, the relationship between independent
variables (X8) and the dependent variable (Y) has to be linear (Lewis-Beck, 1980). These
linear relationships can be examined by the correlation coefficients in Table 26 as
reported in correlation analysis section. In addition, multiple regression analysis
generated the partial regression plots, which should present linear relationships between
each independent variable and the dependent variable “if the assumption of linearity is
met” (Norusis, 1999, p. 501). All of the scatterplots present a linear pattern between X
and Y (see Figure Cla — Clk, in Appendix C). Multicollinearity was also examined by
the correlation coefficients: The highest correlation coefficient among independent
variables in the health belief model was .395 between CA-Tan and psychological barriers
(advantages of tanning) that was far smaller than .85, the level Munro (2001)
recommends to decide multicollinearity. Thus, it was assumed that there was no problem
with multicollinearity.

The assumption regarding the residuals in the regression model was examined in
order to test the linear model. First, the normal distribution of the residual of the
dependent variable was assessed. The histogram in Figure C2 in Appendix C showed that
the residual followed the normal distribution. The normal probability plot indicated that

the distributed points clustered around the given straight line (Figure C3 in Appendix C).

108

In addition, the significant value of the Kolmogorov-Smirnov test indicates the
distribution of the residuals does not differ significantly from normal (p = .200).

Homoskedasticity refers to whether “the variance in prediction errors is more or
less constant across the values of X” (Lewis-Beck, 1980, p. 28). Scatterplots between
eleven independent variables and the standardized residuals of the dependent variable
were obtained to assess homoskedasticity (Figure C4a to Figure C4k in Appendix C).
Distributed values in each stactterplot were similar above and below the regression line
across the values of an independent variable to some extent although a few outliers
existed. In addition, it was assumed that there were no relationships between independent
variables and residuals, since the graphs did not show defined shapes between
independent variables and residuals. Also, the horizontal line showing the relationship
between independent variables and residuals indicated no relationship existed in all
eleven scatterplots. Thus, it was assumed that there were no significant violations in the

residual assumptions.

Attributes of innovation in Figure 14. With respect to attributes of innovations,
five attributes were entered as independent variables to explain the dependent variable,
intention to adopt an innovation (IA). R2 value for this model was .41 l, with an adjusted
R2 value of .391, F (5, 148) = 20.676, p = .000. The beta and standardized coefficients
beta are presented in Table 28. Two of the five attribute variables were significant at .05
levels. Triability contributed the most to IA (,6 = .324), followed by relative advantage (,6
= .23 7). Complexity and observability were insignificant predictors of IA in this model.

Regression analysis generated the partial regression plot between five

independent variables and the dependent variable separately, and showed the linear

109

relationships (Figure C5a — C5e, in Appendix C). Multicollinearity was examined by the
correlation coefficients and tolerance value. Tolerance of a variable measured the
collinearity, meaning “the proportion of the variance in a variable that is not accounted
for by the other independent variables” (Munro, 2001, p. 272). The range of tolerance
among five independent variables ranged from 0.42 (compatibility) to 0.91 (complexity).
The highest correlation coefficient value in Table 29 was 0.647 between compatibility
and relative advantage, which was relatively lower than 0.85, so it was assumed that
multicollinearity was not significant in this regression model.

Table 28. Regression Analysis of Five Attribute Variables on Intention to Adopt (Model

 

 

B)
Independent Variables B B p Adjusted R2
Constant .140 .882
Relative Advantage .188 .237 .007* *
Compatibility .171 .174 .075 . 3 9 1
Complexity -.049 -.036 .592
Triability .742 .324 .000***
Observability .039 .026 .716

 

*p < .05, "p < .01, ***p < .001

The normality of residual was examined by a histogram, a normal probability
plot, and the Kolmogorov-Smimov normal test (p = .097). The graphs showed that
residuals were distributed to the normal curve, although there was a peak at .00 and at .25
above the mean (Figure C6, Appendix C). The distributed points fall close to the line in
the normal probability plot although values around .75 tend to have lower expected
values than observed values (Figure C7, Appendix C).

Homoskedasticity and no correlation between the independent variables and

residual were examined by the scatterplot between standardized residuals and each

110

independent variable. The scatterplots (Figure C8a to Figure C8e, in Appendix C)
indicated that the scatters of the residuals above and below the regression line were
similar, and there was no particular pattern across the scatterplots. Hence, it was assumed
that the variance of the error term was constant for all values of the independent
variables, and there were no relationship between independent variables and residuals.

Final research model in Figure 15. In order to examine the research model in
Figure 15, the final regression equation was developed. All variables in the proposed
research model were entered in the multiple regression equations to explain the
dependent variable of Intention to Adopt (IA) in hierarchical order. Since the research
model combined the two theoretical frameworks, it was necessary to force the order of
entry of variables. Based on the research model, the independent variables were entered
in three blocks: the demographic variables and health belief variables were entered in the
first block, sun protective behavior (SPB) as an independent variable was entered in the
second, and attribute variables were in the third.

The first block was the same regression equation model as in Table 27. Since not
all variables in model A were significant in predicting the dependent variable, as shown
in Table 27, a stepwise solution was appropriate as the entry method for each block of
variables in order to keep only significant variables in the regression equations. In the
second block, SPB was entered based on the research model in Figure 15. According to
order of entry of variables in the equation, significant variables in the first and the second
blocks were retained in the final model with the significant variables in the third block. In

the third block, attribute variables were entered. The third block was the same model as

111

in Table 28, which presented the results of multiple regression equations of five attribute
variables on IA.

The results of multiple regression analysis for the proposed model are shown in
Table 29. R2 value for the final regression model was .501 , with an adjusted R2 value of
.477, F (7, 144) = 20.641, p = .000. With five independent variables, 50.1% of the
variance of IA was explained. Only ethnicity among demographic variables was
significant. The Beta of ethnicity (B = -.851) indicated the mean difference between
Caucasians and other ethnic groups. Caucasians were less likely to adopt the UV
specialized shirts than other ethnic groups when all other predictors are held constant.

Behavioral barriers among the health belief model variables was significant (B = -
.138). The slope of behavioral barriers indicated an decrease in the predicted intention to
adopt UV specialized shirts for one unit of increase in behavioral barriers with all other
variables held constant. In the second block, SPB was the significant variable in
predicting IA. Triability and relative advantages were the significant variables among
five attributes of innovation in the third bock.

Triability had the largest standardized coefficient of beta (,6 = .338) among
significant variables retained in the final model, so this variable was the most important
predictor for IA. Relative advantage was the second most important predictor in this
model (,8 = .250), followed by SPB (,8 = .217), behavioral barriers (,6 = -.177), and
ethnicity (B = -.147). Susceptibility and CA — SP (cues to sun protection) were retained in
the final model, since these two variables were significant in the first block, but

insignificant when the variables in the third block were added.

112

Table 29. Regression Analysis of All Variables in the Research Model on Intention to

 

 

 

 

 

 

Adopt (Model C)
Independent Variables B [3 P Aghliigied:
Constant 2.531 .061
Ethnicity -.851 -.147 .038*
Behavioral Barriers -.138 -. l 77 .006* *
CA-SP -.019 -.013 .858
Susceptibility -.001 -.001 .986 .232
SPB .158 .217 .003" .281
Relative Advantage .199 .250 .000* * 1‘
Triability .780 .338 .000* * * .477

 

*p < .05, **p < .01, ***p < .001

Statistical assumptions of the regression analysis had been examined. Since the
regression model C used the stepwise method, only significant predictors in this model
were reported in the assumption examination in Appendix C. The two insignificant
variables, CA-SP and susceptibility were included in the assumption examination of the
Model C, since they were retained in the final model. These two variables, however,
show no relationships with the dependent variable, IA (Figure C9f and C9g, in Appendix
C). Partial regression plots (Figure C9a — C9e, in Appendix C) and correlation
coefficients in Table 26 suggest that linear relationships exist between predictors and the
dependent variable. The histogram (Figure C10, in Appendix C) and the normal
probability plot (Figure C11, in Appendix C) show the normality of residuals. Residuals
followed the normal curve, and the plotted values were close to the given straight line in
normal probability plot. The Kolmogorov-Smimov test shows the normal distribution of

the residuals (p = .200).

113

In order to check whether the error terms are independent of all predictors (the
correlation between errors and predictor equals zero), scatterplots between the significant
independent variables and residual of the dependent variable were obtained (Figure C12a
— C12g, in Appendix C). The regression line in each scatterplot showed a zero correlation.
Also, the scatters were plotted similarly above and below the regression line across the all
predicted values, so homoskedasticity was satisfied.

Discussions and Summary of Research Questions

The majority of respondents in the present study was characterized as male
(91%), Caucasian (74%), and relatively young (69% under 30). These characteristics are
comparable with other studies of outdoor enthusiasts’ characteristics (Robinson, et al.,
1997). The sample of the present study also has characteristics similar with the predictors
of sunburn found in Purdue, et al.’s study (2001): younger age, males, light skin color
(Caucasians), longer leisure time in the sun, and seeking a tan. Respondents in this study
are enthusiastic about golf, a game that requires some degree of sun exposure
intermittently and cumulatively, leading them to a higher at risk of developing skin
cancer.

Research Question 1 for Objective 4

- To What Extent do Golfers Perceive Susceptibility to and Severity of Skin Cancer,
Benefits of Sun Protective Action, Barriers to Taking Action, and What Sun Protective
Practices do They Engage?

Based on the descriptions of responses between Table 17 and Table 22,
respondents in the present study agreed or strongly agreed on severity of skin cancer,

benefits, and behavioral barriers, but were less likely to use sun protective behaviors.

114

Susceptibility and severity. About 30 to 35% of the respondents agreed or strongly
agreed that they were susceptible to getting skin cancer. More than 50% agreed or
strongly agreed that they were aware of the possibility of getting skin cancer, whenever
they heard of a friend/relative getting skin cancer (Table 22). This result suggested that
respondents generally did not consider the possibility of developing skin cancer which
could be attributable to the age of respondents since younger people tended to think less
about the susceptibility to getting skin cancer than older groups (Grubbs & Tabano,
2000). About 57% of the respondents in this study were 16 to 24 years of age.
Approximately three quarters believed that getting skin cancer would affect their life.

Benefits and barriers. About 80% agreed that sun precaution practices would help
to reduce the possibility of getting skin cancer (Table 20). However, they also agreed that
they would not practice sun protective actions if doing so were uncomfortable (Table 21).
Cost, style, and movement restriction seemed to be less important as a barrier to wearing
UV specialized shirts than discomfort. Another banier, psychological barriers, was also
reported as positive. About 60% agreed that a tan was attractive and 37% thought a tan
was healthy (Table 21). In addition, about half of the respondents believed that a
suntanned person looked more attractive and healthier than someone without a tan. Their
positive perceptions about a tan were significantly and positively correlated with
sunbathing behaviors (r = .325). About 44% of respondents reported that they sunbathe 1
hour up to 25 hours per week during the summer. Positive perception about a suntan
significantly and negatively related to the intention to practice sun protection in the study

of Jackson and Aiken (2000).

115

Cues to actions. Respondents in the present study were more likely to have
friends who enjoyed a tan, but less likely to have friends who performed sun protective
behaviors (Table 22). Friends’ behaviors or norms play an important part in respondents’
sun protective behaviors or sunbathing behaviors. Arthey and Clarke (1995) stressed that
“the beliefs that a suntan makes you look more attractive and healthy suggests that the
perceived opinions of others are having an important effect on people’s sun tanning, and
consequently sun protection, behaviour” (p. 267). Although the present study did not
examine the degree of the impact of significant others, the variable of cues to action
would indicate its importance in predicting the sun protective behaviors.

Outdoor activities. About 70% of the respondents reported that they spent time in
the sun for more than 10 hours in a typical week during the summer (Table 17). Their
outdoor activities could be associated with golf play. Approximately 52% of the
respondents answered that they played at least 10 rounds of golf per month of golfing
season. That means they play, on average, two rounds of golf per week resulting in sun
exposure of about 8 to 10 hours per week. However, 92 of 158 respondents (58.2%)
reported that they played golf in the mid morning and/or early afternoon. About 20.5% of
the respondents mentioned that they played any time of a day if possible. This finding
indicates that respondents in this study are exposed to the sun during peak sun hours,
while recommended sun protective action suggests that outdoor activities should be
minimized between 10 am. and 4 pm. “when sun’s rays are most direct” (Liflring, 2001,
p.197)

Sun protective behaviors. The most reported sun protective behavior taken by

respondents was wearing a hat when outdoors. About 47% of the respondents ofien or all

116

the time wore a hat. Next in frequency were use of sunscreen on the face (31.1%), staying
in the shade (26.1%), and use of sunscreen on the body (19.7%). Only 4.4% wore long
sleeved shirts or long pants when outdoors. From the results, it was assumed that their
bodies were not properly protected from the sun. Based on market research, most golfers
wore short-sleeved shirts while playing. However, they reported in the survey
questionnaires that they neither use sunscreen on the body nor wear long-sleeved shirts.
Findings were consistent with sun protective behaviors of outdoor workers in Stepanski
and Mayer’s study (1998). From observation data, they found that face and lower arms
were the least protected areas of the body. Hence, it would be essential to educate
outdoor occupational or recreational groups about appropriate sun protective practices.

Wearing a hat was the most frequently taken action when outdoors in the present
study and Stepanski and Mayer’s study (1998). However, hats seemed to be worn for
reasons other than sun protection. Respondents in the present study were more dependent
on using sunscreen than wearing long-sleeved shirts, contrary to the outdoor workers in
Stepanski and Mayer’s study (1998). Respondents may think wearing long-sleeved shirts
as sun protection is not appropriate for golfers if it is comfortable while they are playing.
Research Question 2 for Objective 4
- Which Variables in the Health Belief Model Contribute Most to Explaining their
Current Sun Protective Behaviors?

Among eleven variables in the health belief model in regression model A, Table
27, cues to sun protection (CA-SP), susceptibility, marital status, psychological barriers,
benefits, and knowing a person who has skin cancer were significant, in that order, in

CXplaining 44.2% of the variance of sun protective behaviors (SPB). CA-SP, friend’s sun

117

protective behaviors, contributed most. That is, the more respondents have friends who
take sun protective actions, the greater the likelihood that respondents take sun protective
behaviors. Psychological barriers (perceived advantages of tanning) and impact of peers’
sun protection were also significant predictors of sun protective behaviors in Jackson and
Aiken’s study (2000).

Susceptibility was the second most important variable in explaining sun protective
behaviors. The more respondents perceived they were susceptible to developing skin
cancer, the more likely they were to take sun protective behaviors. Severity was not a
significant predictor in the present study. This is consistent with the findings of Jackson
and Aiken’s study (2000). As Janz and Becker (1984) indicated, severity seemed not to
be meaningful to subjects in explaining preventive behaviors. If respondents were
persons who have been diagnosed with skin cancer, severity might help to explain sun
protective behaviors as was true in the studies of compliance with the treatment for the
sick.

Behavioral barrier was the most significant variable in predicting the
recommended health-related behaviors in the 29 studies (J anz & Becker, 1984), but this
factor was not significant in the present study. Rather, the psychological barrier,
perceived advantage of tanning, was a significant predictor, indicating that perceived
body image was more influential in explaining respondents’ sun protective behaviors
than behavioral barriers. Educational messages that influence the perception of
appearance would be necessary, especially for young adults in order to increase sun

protective actions.

118

Being single is another significant predictor which is in contrast to the finding of
Hall, et a1. (1997). In the present study, singles are less likely to take sun protective
behaviors than married people. It might be because the respondents’ marital status was
associated with their age or with their lifestyles. Having family members or friends who
had skin cancer was a significant predictor of sun protective behaviors, supported by the
previous studies. Benefits of taking sun precautions were also significant in explaining
the variance of sun protective behaviors, meaning that the more the benefit perceived, the
more likely respondents practice sun protective behaviors.

Research Question 1 for Objective 5
- To What Extent and How is Each Attribute of UV Protective Clothing Related to the
Intention to Adopt?

Less than 50% of the respondents perceived UV specialized shirts to have a
relative advantage (Table 23). Although 69% agreed that they could have better
protection with UV specialized shirts, respondents disagreed with the superiority of the
image (69%) or fashionability (56%). About 30% of the respondents agreed that the
innovation photographed was compatible with their garments and their situation, and that
the benefits of UV protection were observable. More than 40% of the respondents
perceived that the difficulty of understanding the UV protective function was low, and
triability was high. However, about 70% did not intend to purchase UV specialized shirts.

In order to find out the specific association between five attribute variables and
the dependent variable, intention to adopt (IA), correlation coefficients were examined
(Table26). All variables except complexity were significantly correlated with IA. The

ranges of significant coefficients were from .317 (Observability) to .569 (Triability).

119

Complexity was also an insignificant variable in predicting the likelihood of purchasing
India-inspired garments for US. female consumers (Littrell & Miller, 2001).

Research Question 2 for Objective 5

- What is the Relative Importance among Attribute Variables in Predicting the Intention
to Adopt UV Specialized Shirts for the Potential Adopters?

Although four attribute variables were significantly correlated with the intention
to adopt, only two variables were significant in regression model B in Table 28. R2
indicated that 41% of the variance in the dependent variable, IA, was accounted for by
triability and relative advantages, and triability was the most important predictor of the
intention to adopt UV specialized shirts. That is, the more respondents perceived that the
UV specialized shirt could be tried in the local stores and was better than existing shirts
in style, price, or UV protective function, the greater the likelihood that they would
intend to wear or purchase it. Since UV specialized shirts represented a preventive
innovation and golfers seemed not to be familiar with it, trying it on and advantages of
the shirts were directly and significantly associated with their intentions to wear it.

In the study by Agarwal and Prasad (1997), triability was a significant predictor
of current usage of World Wide Web, while relative advantage was a predictor of the
intention to continue to use it. Their study supported how important it is to emphasize
the different attributes of an innovation in accordance with the perceivers’ states of the
decision-making process. The p-value of compatibility was .075, which was close to the
Significance level of alpha. Complexity and observability did not contribute significantly
to explaining the variance. Since the innovation is relatively new, the effectiveness of UV

protection would not yet be apparent to respondents. Complexity may be an inadequate

120

attribute to describe the clothing innovation that is consistent with the study of Littrell
and Miller (2001).

Research Question 1 for Objective 6

- How are Beliefls about Skin Cancer Associated with Perceived Attributes of a UV
Specialized Shirt as an Innovation and the Intention to Adopt it?

In the Table 26 of correlation coefficients, susceptibility was significantly and
positively correlated with triability (r = .241), indicating that individuals who believe
they have higher levels of susceptibility to skin cancer than the typical person are more
likely to try the UV protective shirt. The perceived benefits of practicing sun protective
behaviors were significantly and positively related to relative advantages (r = .175),
triability (r = .224), and observability (r = .258). That is, if individuals believe that sun
protective behaviors would reduce the chance of getting skin cancer, they perceive the
UV specialized garment as more advantageous, more triable, and more observable than
individuals who perceive fewer benefits. On the contrary, behavioral barriers were
negatively correlated with relative advantages (r = -.208), compatibility (r = -.290), and
triability (r = -.207). The more individuals believe that there are barriers to taking sun
protective actions, the less likely they are to perceive relative advantages, compatibility,
and triability.

Cues to sun protection (CA-SP) were positively and significantly correlated with
relative advantage (r = .230), compatibility (r = .217), and triability (r = .238). If
individuals have more friends who practice sun protective behaviors, they perceive the
innovation as more advantageous, compatible, and triable. However, compatibility was

Significantly and negatively related with cues to tanning (CA-Tan), indicating that

121

respondents tend to perceive UV specialized shirts negatively and incompatible with their
social values, if they have more friends who enjoy a suntan. This is not surprising since
significant others’ opinions play key roles in sun tanning and sun protective behaviors
(Arthey & Clarke, 1995); innovations need to be compatible with what is acceptable to
peers. Complexity was significantly and positively related only with cues to tan. That is,
if individuals’ friends are more likely to enjoy a tan and perceive having a tan positively,
they tend to agree that it would be difficult to understand and explain the UV protective
function of the innovation. Individuals in this category might not attempt to understand
the firnction of UV protection.

Current sun protective behaviors (SPB) were significantly and positively
correlated with relative advantages (r = .234), compatibility (r = .170), triability (r =
.270), observability (r = .203), and intention to adopt (r = .384). The positive correlations
indicate that the more respondents practice sun protective behaviors, the more likely they
perceived UV specialized shirts as advantageous, compatible, triable, and observable, and
the more likely they intended to adopt the shirts.

Findings suggest that beliefs or knowledge about a certain condition (skin cancer
risks) influence the formation of attitudes toward the related object (UV specialized
shirts), affecting their intention to adopt the innovation.
Research Question 2 for Objective 6
- Which Variables in the Final Model Contribute Most to Explaining the Intention to
Adopt the UV Specialized Shirt?

About 50% of the variance of the Intention to Adopt (IA) was explained by five

Variables in the final model in Table 29. Triability contributed most in explaining IA,

122

followed by relative advantages, sun protective behaviors (SPB), behavioral barriers, and
being Caucasians (ethnicity). In the first block of the final model, the significant
contribution of behavioral barriers was reasonable, since this variable focused on barriers
to wearing sun protective clothing. Caucasians would have a lower level of intention to
purchase or wear the UV specialized shirts than other ethnic groups. Though outdoor
workers were more likely to wear long-sleeved shirts for sun protection than to use
sunscreen (Stepanski & Mayer, 1998), Caucasian golfers in this study did not prefer to
wear long-sleeved shirts. The differences could be attributed to the activity in which they
are engaged. Since golfers’ body movements are directly related to their golf scores,
uncomfortableness could be a very significant factor to them. This could explain why
golfers in the present study are more likely to use sunscreen than to wear long-sleeved
shirts in the sun (Table 18).

In the second block, adding one variable, SPB increases the explanatory power of
the IA from .232 to .281 of adjusted R2 in Table 29. This indicates that current sun
protective behavior is a significant predictor of the intention to adopt or purchase UV
specialized shirts, and implies that promoting sun protective behaviors would be helpful
to increase the rate of adoption.

By adding the five attribute variables in the third block, the adjusted R2 increased
to .477. This shows that the attribute variables had some impact on the intention to adopt.
In fact, triability and relative advantage contributed the most, leading to a high increase in
R2 value. In comparison with model B in Table 28, model C is improved overall by
adding health belief variables, demographic variables, and the variable of sun protective

behaviors (SPB).

123

However, the R2 indicated that about 50% of the variability in the dependent
variable (IA) was explained by five variables, indicating that there are other significant
predictors that were not included in explaining the intention to adopt UV specialized
shirts. The imbalanced proportion of demographic variables such as age or gender in the
present sample would contribute to lack of significance in predicting the dependent
variables, reducing the explanatory power.

Summary

The average respondents in this study were not inclined to practice appropriate
sun protective behaviors. Many respondents spent more than 10 hours in the week during
the summer, and 42% sunbathed one to 10 hours a week in the sun during the summer.
The significant relationships in the proposed model are shown in Figure 16. This figure
was developed based on Figure 15, and shows only the significant predictors of the
dependent variables, sun protective behaviors (SPB) and intention to adopt UV
specialized shirts (IA) in three regression models.

In regression model A, six variables among the 11 health belief model variables
were significant in explaining 44.2% of the variance of sun protective behaviors: marital
status (single), susceptibility, benefits, psychological barriers, friends’ sun protective
actions (CA-SP), and having family member or friend who has skin cancer. Marital
status and perceived psychological barriers had negative impacts on the dependent
variable.

In predicting the contributions of five attribute variables on the intention to adopt
UV specialized shirts, only relative advantage and triability were significant predictors of

IA in regression model B, although relative advantages, compatibility, triability, and

124

observability were significantly correlated with IA. About 41% of the variability in 1A
was explained by two variables.

Additional variables in regression model C significantly increased the explained
variability of IA (R2 = .50). IA was explained by ethnicity, behavioral barriers, sun
protective behaviors in the health belief model and relative advantages and triability

among five attributes of innovations. Ethnicity and behavioral barriers were negatively

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

 

 

 

associated with IA.
Behavioral (“l
Barriers
Ethnicity (—)
(Caucasians)
Attributes of
Innovations
0 Relative
Marital Status Advantage
(Single) 0 Triability
v v
Perceived
susceptibility Sun Protective Intention ‘0
Behaviors AdathV
Perceived 0 Use sunscreen specialized
Benefits 0 Wear protective Shrrt
clothing
Psychological ° Wear a hat
Barriers ‘3 Stay in the
shade
Cue to Action (+

 

 

 

0 Friend’s Sun
Protective Behavior

0 Family Member or
Friend has Skin
Cancer

 

 

 

Figure 16. Significant Relationships in the Research Model

125

The findings support the proposed research model partially, and indicate that
variables were excluded from the study that might help to explain the unexplained
variance of the intention to purchase or wear UV specialized shirts. In addition, adoption
of UV specialized clothing was not described with the five attributes. Only two attributes
were influential in predicting IA. Thus, discovering the significant variables for the
regression model A, B, or C would be very worthwhile for future studies. For instance,
investigating the information sources (media or reference group influences), information
process, involvement with the innovation, or motivation toward health behaviors might
contribute. In addition, the health belief variables which were significant predictors of
sun protective behaviors can be related to IA indirectly, so investigating causal
relationships would be useful to understand the directions among these variables in a

future study.

126

CHAPTER FIVE
Summary and Conclusions

This chapter includes a summary of the findings of the study, the limitations of
the study, and the implications for future studies and practice.

The primary purposes of this study were to investigate to what extent golfers were
exposed to UV radiation while they play and to understand how they perceived the risk of
sun exposure and its impact on the intention to wear UV specialized shirts. This study
was divided into two phases and used a variety of data collection methods. A brief
summary of the findings with respect to the objectives of each phase is presented.

Phase I
Field Experiments

The objectives of the first phase were to investigate the cumulative exposure of
different body sites to UV radiation during golf rounds and to evaluate the effects of
clothing in reducing UV radiation transmissions. Polysulphone film badges were used to
measure personal exposure to UVB. When the film is exposed to UVB, it depolymerizes
with a change in its absorbance that is measurable at a wavelength of 330 nm. The
absorbance change (AA) of each dosimeter was calculated by subtracting the value of the
unexposed film from the value of the exposed. The experiments were conducted in
August and in October at Forest Akers Golf Course East at Michigan State University.

Seven subjects participated in each experiment. Sixteen bandages were placed on
upper body sites over and under the clothing. In Experiment 1, seven subjects wore their
own summer golf shirts, while in Experiment 2, four subjects wore UV specialized shirts

that were provided by the researcher and three wore their own fall shirts. While subjects

127

were exposed, ambient exposure was also measured with 30 badges that were placed in
an open, unshaded area. The absorbance values of the personal dosimeters were
standardized to the equivalent absorbance values in the ambient level in order to compare
the data between Experiment 1 and Experiment 2.

The results from both experiments indicates that the upper back of golfers, such as
_the shoulder, back, the back of the neck, and the posterior arms, are more exposed to UV
radiation than the front of the body such as the chest. Under clothing, the back neck and
front of the neck were the two most exposed areas, which indicated that the badges were
exposed directly to the sun even though they were intended to be covered by clothing.
The difference between exposure of the neck and other sites may be attributed to golfers’
stance during play (bending at the waist) and their clothing habits (unbuttoning the
shirts). However, the other body sites under the clothing received a limited amount of UV
radiation, suggesting the importance of clothing in reducing UV radiation transmission.
In addition, while the shoulder and the back over clothing were exposed to UV radiation
the most, these locations were least exposed under clothing, perhaps because two layers
of back yoke in UV specialized shirts and corduroy shirts contributed to better protection.

Comparison of exposure under different clothing indicates the summer cotton
shirts allowed more UV radiation across most body sites than did the UV specialized
shirts or thick fall shirts (Figure 12). Summer golf shirts are less effective in blocking
UV radiation than the thicker fall shirts or UV specialized shirts. However, the
differences in the absorbance values under the three types of shirts were not statistically
significant at each body location, perhaps due to the small absorbance values as measured

by films in this study.

128

Market Research

Market research focused on a survey of standardized images of golfers from
photographs in magazines and interviews of golf store personnel to gain general
information of golf clothing available. Photographs of golfers in Golf Digest in two years
indicated standardized images of golfers, wearing short-sleeved polo shirts that included
wide width of sleeve hem and collar, matched with long pants. The images reflect
professional golfers and may not represent public or amateur norms.

The interviews with four salespeople in local golf stores, however, supported the
findings of published images. The most typical styles of golf shirts available and
purchased were the short-sleeved cotton polo, in soft colors across seasons. A baseball
cap was the most common head covering for customers at local golf stores, which was
also consistent with the findings of published images. Few UV specialized products, such
as hats, short-sleeved shirts, or sunglasses, were available at local stores, because the
majority of customers was not interested in them. Only a small portion of customers,
mainly women, looked for these products. Design and price of shirts were more
important factors than UV protective functions when golfers made purchasing decisions.
It seemed that most customers in the local pro golf shops were aware of UV specialized
products, but had not yet formed attitudes toward those products. That is, golfers in the
selected market are in the knowledge stage in the innovation-decision process.

Phase II
The purposes of the second phase were to assess golfers’ health beliefs and
current sun protective behaviors and to examine their attitudes toward a UV specialized

shirt with relation to the intention to adopt it. In addition, this study proposed a

129

psychosocial model, based on the health belief model and Rogers’ attributes of
innovations, to investigate the relationships between the two theoretical frameworks.

A self-administered questionnaire was developed and distributed at public golf
courses and golf-related programs in a Midwestern area by convenience sampling
intercept method. One-hundred-fifty-eight respondents completed a 58-item
questionnaire. Beliefs, sun protective behaviors, and attribute variables were measured
using five-point Likert-type scales. Ninety-one percent were male, and their age ranged
from 16 to 80 years old (mean age = 30 years). About 74% were Caucasians and 65%
were single. Respondents’ personal characteristics were comparable with those who were
more likely to be exposed to the sun for long periods of time (Robinson, et al., 1997) and
to have repeated sunburn experiences (Purdue, et al., 2001).

Health Belief Model

The majority of respondents agreed or strongly agreed on the severity of skin
cancer, benefits of taking sun precautions, and barriers to wearing sun protective
clothing. With respect to protecting the face, about 50% of the respondents wore a hat,
and 30% used sunscreen on the face frequently or all the time when outdoors. With
respect to protecting the body, only 20% of the respondents used sunscreen on the body,
and 4% wore protective clothing often or all the time when outdoors. The findings
indicate that the majority of respondents do not protect themselves adequately, despite
being exposed to the sun to a great degree while playing golf.

The health belief variable that contributed most to predicting sun protective
behaviors was cue to sun protection (CA-SP), e.g., a friend’s sun protective behavior.

This finding supports Arthey and Clarke’s argument (1995) that the influence of peers on

130

sun protection would make their knowledge and beliefs about sun precautions transfer
into actions. Susceptibility, marital status (single), psychological barriers, benefits, and
having family members who have skin cancer were also significant predictors of sun
protective behaviors in that order (Table 27). The more likely respondents believe they
are susceptible to getting skin cancer, believe practicing sun precautions are beneficial,
and have family members or friends who have skin cancer, the more likely they are to
practice sun protective behaviors. Being single and psychological barriers had negative
impacts on sun protective behaviors.

Attributes of a Preventive Innovation

According to Rogers (1995), an innovation can be described by five attributes:
relative advantage, compatibility, complexity, triability, and observability. About 70% of
the respondents in this study agreed that the photographed UV specialized shirt had
superior UV protective qualities, compared to other typical shirts. However, they did not
agree with the advantages of style, prestigious image, or economic value of the shirts.
Only 30% of the respondents agreed that the innovation was compatible with their
garments and that the benefits of UV protection were observable. About 70% reported
that they did not intend to purchase or wear a UV specialized shirt.

The intention to adopt UV specialized shirts (IA) was predicted by triability and
relative advantages among five attributes of the innovation in the regression model in
Table 28. This finding is compatible with market survey results. As salespeople
indicated, the important criteria used when purchasing golf clothing in local stores are
design and price of a garment. Consequently, the innovation needs to be perceived by

consumers as superior in style and price as well as the UV protective function in

131

comparison to other existing garments in order to be adopted. It also needs to be available
to try on in the store.
Proposed Research Model

Combining the health belief model with Rogers’ attributes of innovations
increased the explanatory power of the variance of intention to adopt (IA) in Table 29.
Triability and relative advantages among five attributes were the most significant
predictors in the pr0posed model. Current sun protective behaviors were a significant
indicator of IA. The behavioral barrier variable in the health belief model was also a
significant predictor of IA and was associated negatively. That is, in order to increase the
adoption of UV specialized shirts, the shirts should have few barriers to being worn: UV
specialized shirts need to be perceived as comfortable, inexpensive, and/or attractive.
Caucasians have a lower intention rate to adopt a UV specialized shirt than do other
ethnic groups, although they have higher incidence rates of skin cancer than other groups
have (Arthey & Clarke, 1995).

Limitations of the Study

Several limitations in this study should be considered when interpreting the
research findings.
Phase I

The polysulphone films used as personal dosimeters have limitations. This film
reached the saturation point before the maximum length of the golf game, so the
researcher generated the total amount of UVB exposure of golfers based on the
regression line. In addition, some personal dosimeters were lost, especially in Experiment

I, presumably due to the golfers’ sweat under the shirts and to their carrying the golf bags

132

on the shoulders over the shirts where the film badges were placed. Field experiments
typically involve a small numbers of subjects and data are subject to confounding
variables, such as height of the sun in the sky, season, the UV index of the experiment
dates, topography of golf course, orientation of golfers to the sun, and subjects’
characteristics. Although this study reported the characteristics of clothing worn by the
participants while they played golf, the specific factors affecting the sun protectiveness of
fabric, such as fabric count (the number of yams/cm for tightness of the weave or knit),
mass per unit area, wetness, stretch effect, color effect, or washing effect, were not
controlled or studied. Future studies could examine the contribution of these variables to
the effectiveness of golf shirts in blocking UV radiation transmissions.

In the review of published images, the researcher reviewed one publication, Golf
Digest, which focused on the activities of professional golf players, so the observed
appearances represented only formal images of golfers. It would be useful to observe the
appearance or look of the typical golfers at public golf courses across different seasons.
With respect to personal interviews with the salespeople in the pro shOps, the researcher
visited four stores in the local area that sold clothing. The interviewees were asked to
recall previous sales experiences. They responded based on their knowledge or beliefs
about sun protection. Another study could involve interviewing customers or reviewing
sales data.

Phase II

There were also limitations in the survey portion of the study. The sample size

and sampling methods could be improved. The participants were recruited through a

convenience sampling method. Non respondents who refirsed to fill out the questionnaire

133

or who did not complete the questionnaire might be less likely to be interested in sun
protection. The sample was not balanced with respect to age, gender, marital status, or
ethnicity and that might influence the results obtained, so caution is necessary in
interpreting the findings.

There were few standardized measurements of the health belief model or the
attributes of fashion products as innovations. The instrument for the present study was
modified and developed based on previous studies. Although the instrument was
examined for reliability and validity during a pretest, the reliability scores of some of the
variables in the instrument were less than the desired level of .70 (Spector, 1992), which
might affect the significances in the regression models.

This study combined the different types of sun protective behaviors as one
dimension, and this might produce findings that are different from the literature. Hall, et
a1. (1997) found that different indicators were associated with the different types of sun
protective behaviors: Age and sunburn experience were significantly related to wearing
protective clothing, and a history of skin cancer was related with use of sunscreen.
Another limitation could be that people adopt health-related actions for non-health
reasons. For instance, an individual could wear a hat for social acceptance or for fashion.
Thus, it would be interesting to investigate golfers’ sun protective behaviors and beliefs
with a qualitative research approach.

The male model in the photograph of the questionnaire might influence positive
or negative attitudes toward UV specialized clothing. The model’s physical
characteristics such as body type, facial attractiveness, gender, or age could impact the

judgment of attractiveness of the clothing that the model wore.

134

 

IE
W

he

p2

SU

1'6
512

sh

ex
fal

m:

lhc‘

eff

Implications for Practice and Further Research
Design Perspective

Data from the experiments indicate which body sites should be more protected to
reduce UV radiation transmission for golfers. In addition, summer cotton shirts, typically
worn by golfers, were less effective in protecting the human skin from UV radiation than
heavier fall shirts or UV specialized shirts. The American Academy of Dermatology
(2002c) recommends wearing tightly woven clothing such as a long-sleeved shirt and
pants and a wide-brimmed hat (3-inch brim) that covers the face and ears, along with
sunglasses with UV protection. Most golf consumers, however, would hesitate to follow
these practices, especially during the summer season due to discomfort from heat or
restriction of movement to play. In addition, the findings in the market research show the
standardized images of golfers: The most general look is to wear a short-sleeved polo knit
shirt, and few wear long-sleeved shirts. Hence, promoting long-sleeved golf shirts during
the summer season will not be compatible with golfers’ social norms.

Cotton, linen, or rayon fabrics are the most popular in the summer, since they are
excellent in moisture absorbency, but inferior in sun protection factors to polyester
fabrics (Davis, et al., 1997). The most common fabric for golfers is cotton, based on
market research and clothing features of subjects in experiments (Table A1 and A2, in
Appendix A). Thus, it is crucial to retain cotton fabrics for a golf shirt in order to increase
the adoption of sun protective clothing.

The findings in the field experiments can be put into practice in designing an

improved golf shirt. Since the two layers of cloth in the back yoke of shirts were more

effective in reducing UV radiation than one layer of fabric, this approach can be applied

135

to the excessively exposed body sites such as the shoulder, back, or posterior arm. This
will also take care of any differences in the UV exposure to the lefi and right side of the
shoulder (Table 4) and the back (Table 11). Two layers of fabrics in the yoke area can be
made up of two layers of cotton fabrics or UV specialized fabric over cotton fabric.
Increasing yoke size will provide more protection to the back side of the body. Slightly
longer sleeves to cover the elbow would also add protection. In addition, a collar that
stands high would reduce the exposure of the back of the neck to the sun. Figure 17

presents the UV protective shirt designed on the basis of the implications of this study.

Two layers

/ of fabrics

    

 

 

 

 

iz—“Q--~--_-..--..5J L- ...... it

Figure 17. Proposed Design of UV Protective Shirt for Golfers

The front neck is one of the most exposed areas under the clothing, suggesting
that different clothing habits, like buttoning the shirt, would reduce the body exposure at
this site. Also, the forearm sites were more exposed to the sun than the upper arm sites in

Experiment 11 (Table 15) although the subjects wore long-sleeved shirts, mainly caused

136

 

b:

C(

d:

W

ax

PC

by unfastened cuffs of the UV specialized shirt sleeves. Hence, it would be necessary to
educate the consumers about appropriate clothing habits for sun protection.

Based on findings in the phase II, behavioral barrier, relative advantage, and
triability were significant in predicting the intention to adopt UV specialized shirts. That
is, in order to promote sun protective clothing, barriers to wearing clothing, such as
discomfort or movement restriction should be considered in designing clothing. In
addition, the designers need to pay attention to style, image, and economic values in
designing UV specialized shirts and make them available in the local stores in order to
increase consumer’s intention to adopt. In fact, about 35% to 70 % of the respondents in
the second phase reported that they did not notice relative advantages of economic
values, prestigious image, or fashionable style of the photographed shirts in the
instrument. In order to overcome these limitations of UV specialized shirts, designers
need to analyze the target consumers’ needs with respect to aesthetic needs (design and
color), psychosocial needs (social norm), physical needs (comfort), and technical needs
(movement). The designers must reflect on consumers’ needs and the standardized
images of current golfers. Observation of average golfers at public golf courses during
the summer season would be helpful for identifying the standardized images in the
process of designing UV protective clothing. Triability was the most important predictor
of the intention to adopt UV specialized shirts in regression model B and C. That is, in
order to increase the adoption of sun protective clothing, those products should be
available in the local store. Market survey in the present study indicated that only a small

portion of UV specialized products were available in the selected local store.

137

A recent study reports that the use of UV absorbers as a new laundry additive
during household laundering resulted in a significant improvement of UV protection of
fabrics (Osterwalder & Rohwer, 2002). Since summer cotton shirts were less effective in
protecting golfers from sunlight than thick fall shirts or UV specialized shirts, based on
the field experiments, effective UV absorbers as a laundry additive would be beneficial
and feasible for consruners if they work properly. Hence, testing the effects of UV
absorbers after laundry process, compared with the normal clothing in terms of the
duration and the amount of UV radiation prevention, would be necessary.

Business Perspective

For retailers of golf clothing, UV specialized short-sleeved shirts can be supplied
with sunscreen for using it on forearm sites to promote sun protection. As a commercial
package, a short-sleeved golf shirt can be purchased only with sunscreen, or sunscreen
can be provided as a gift when golfers purchase golf clothing. In addition, educational
messages targeting golfers need to be supplied with sunscreen in the pro golf stores.

F lyers, including information about the proper sun protective actions and the data
regarding UVR exposure at different body sites or the effects of clothing, such as the
differences in the absorbance values of UVR between over and under the shirts, could
stimulate consumers to comply with the recommended sun protection.

For golf course managers, the UV index card can be provided to golfers, or the
UV index of each day can be posted in the golf course, so golfers become aware of the
UV index. Also, providing information about the definition of the UV index, the amounts
of UV radiation received by golfers during the different seasons, and the risk of

developing skin cancer when outdoors will be useful to increase sun protective practices.

138

Research Perspective

Although increasing attention is being given to golfers regarding sun protective
practices due to their prolonged exposure when the sun is the strongest, little is known
about golfers’ beliefs and prevention practices. This study makes a small contribution to
that body of literature. Additional research is necessary to confirm the findings. By using
a random or stratified sampling method, examining the general population of golfers
without geographic limitation is recommended in order to generalize golfers’ beliefs and
behaviors regarding sun precautions. A list of public golf course users could be a possible
sampling frame. Comparing age or gender differences of golf players would provide
specific and appropriate information to the different subpopulation groups. Replication
of the instrument across different sample groups could be helpful to establishing a stable
instrument.

In addition, researchers in future studies need to explore and test variables to
account for the unexplained variance of the final research model, which combined the
health belief model with attributes of the preventive innovation. For example, it would
be important to examine social influences on sun protection, such as, to what extent, to
whom, and in which stage of the decision process significant others influence the
potential adopters’ practices. Investigating the structure and strength of the potential
adopters’ social network would be a related study, since an innovation is communicated
through social networks (Rogers, 1995). Furthermore, sunbathing and sun protecting
behaviors are affected by the perceiver’s willingness to belong to his/her social network

(Keesling & Friedman, 1987). This would contribute to developing a more complete

139

model. The present study analyzed the data by using multiple regressions. It would be
valuable to use path analysis to specify the causal relationship among variables.

In addition, discovering consumers’ perceived attributes of an innovation will be
useful, before diffusion is attempted. For the clothing innovation, only two attributes
contributed to explain 41% of the variances of the intention to adopt it. A number of the
clothing studies have provided clothing evaluative criteria with respect to price, quality,
performance, usefulness (matching, appropriateness, utility), and aesthetic criteria (color,
styling, fabric) (Dickson & Littrell, 1997; Eckman, et al., 1990; Lee & Burns, 1993).
Most of these criteria were examined as the relative advantages in the present study,
which might lead to a low level of explanation of the variance. It will be worthwhile to
attempt to use these criteria as an individual attribute of the clothing innovation. Also,
other extrinsic cues, such as brand name or store images, can be examined as additional
factors that influence the consumers’ intention to adopt clothing. Consumers’ fashion or
shopping involvement can be associated as well.

A longitudinal study is recommended for future research. Since this study
measured the golfer’s intention to adopt or purchase UV specialized shirts, it would be
important to examine whether the innovation has actually been adopted. Also, it will be
useful to compare the outcomes of self-reports with direct observations of respondents’
sun protective behaviors.

Conducting market survey of golf clothing in other geographic locations would be
necessary. In areas warmer than Michigan, potential adopters might be more
knowledgeable about UV specialized shirts since they are exposed to the sun all year

round. On the contrary, they might not consider it as sun protection, since wearing a shirt

140

would be uncomfortable due to heat stress. This can be a part of the analysis process in
designing functional clothing.
Education Perspective

Findings in the second phase would guide the direction of educational messages
to promote skin cancer prevention. Outdoor enthusiasts have a higher risk of skin cancer
than the average person since they have been associated with high cumulative UVR
exposure. The present study found that the majority of respondents did not protect their
body properly while they played golf. However, staying out of the sun during the midday
would not be a suitable message for outdoor enthusiasts. Promotion of wearing long-
sleeved shirts in the summer as sun precaution would not be appropriate to golfers, as
market research described. Golfers would wear short-sleeved shirts if the weather is not
cool. Survey data indicate that respondents are more apt to use sunscreen than wear long-
sleeved shirts. If so, educating golfers regarding use of sunscreen on the exposed body
sites, such as forearm and legs, as well as wearing shirts when outdoors, would be
important. Actually, forearm sites of golfers were the least exposed areas directly to the
sun based on experiments. Thus, if a golfer feels heat stress in wearing a long-sleeved
shirt while playing golf, it will be appropriate to use sunscreen on the forearm sites. Golf
course staffs or instructors will take important roles to promote use of sunscreen. They
can remind golfers of the importance of using sunscreen before starting the golf games.

The educational messages could be designed differently according to various age
groups. As Mahler, F itzpartick, Parker, and Lapin (1997) found, appearance-based
messages, presenting photoaging or skin cancer slides associated with lack of sun

protection were more effective with the older age group (26-43 years of age) than with

141

the younger group (18-25 years of age) in increasing the intention to use of sunscreen.
For the young adult group, acceptance from their peers is significant, so attractive
practice guides that are comparable with their social norms are appropriate. Messages
that would influence their basic concepts of beauty or fashion would be necessary. The
proper role models or opinion leaders of the young age group will be influential in
changing their beliefs, leading to changing their behaviors.

This study shows how each dimension of the health belief model played
significant roles in explaining sun protective behaviors. Findings can assist practitioners
in designing sun protection programs. Focus on the significant belief variables, such as
cues to sun protection, susceptibility, perceived benefits, or psychological barriers
(perceived advantages of tanning) in designing messages to golfers is necessary in order
to effectively influence or change golfers’ sun protective behaviors. For instance, it is
necessary to explain the possibility of developing skin cancer especially to young adults
in the near future if they do not practice sun protection. According to the American
Academy of Dermatology (2002a), melanoma, the most serious type of skin cancer is
more common than any other type of cancer among women between 25 and 29 years old.
This fact might be associated with women’s positive perceptions about tanning. Hence,
the educational messages focusing on the association between sunbathing/artificial
tanning and the risk of developing skin cancer photoaging of skin are required. In
addition, targeting young stars who do not have a tan would be effective in influencing
the young adults’ perception about beauty.

However, it would be important to control the level of threat of the message for

practitioners. Becker and Maiman mentioned that “very low levels of perceived severity

142

are not sufficiently motivating, while very high levels of perceived seriousness are
inhibiting” (1975, p. 14). That is, insufficient level of threat would make the audience
reluctant to accept the recommendations. Excessive fear would cause negative results, as
well, causing avoidance of the recommended sun protective actions and/or denial of the
possibility of their developing skin cancer. Therefore, it will be important to analyze the
extent to which the targeted group feels susceptible or believes the disease is severe from
the educational messages they received in order to enhance compliance.

To increase compliance, campaign designers need to provide specific action
guides and encourage self-confidence to perform actions, which is effective in increasing
compliant behaviors (Mattson, 1999; Rodrigue, 1996). Those variables would also help to
reduce the perceived barriers to taking actions. After designing and implementing the
educational messages, it would be necessary to conduct longitudinal studies to find how
behavioral changes occur after the target consumers are exposed to the messages.

The findings in this study suggest that golfers are at the knowledge stage in the
innovation-decision process, and they have not formed favorable or unfavorable attitudes
toward the preventive innovation. Golfers’ intention to adopt a preventive innovation is
predicted by attributes of the innovation, health beliefs, and their current behaviors.
Triability and relative advantages of UV specialized shirts, the barriers to wearing sun
protective clothing, and sun protective behaviors are the significant predictors to increase
the intention to adopt. This study suggests implications for apparel designers and
business practitioners to reduce UV radiation exposure by golfers. However, the skin
cancer incidence rates could be reduced by changing the individuals’ health-related

beliefs and practices of sun precautions.

143

APPENDICES

144

APPENDIX A

Experiments and Market Research

145

 

 

 

 

 

 

 

 

 

 

 

 

 

 
    

 

 

 

 

 

 

 

 

 

 

 
 

     

 

  

 

 

 

 

 

 

 

 

 

S S S E o E c S o E S S. E E E o e 38.
V.226
2.2: o------------..-: 3% see >
> > > > 2:: > > >
8500 R22 0 8035
2.8: oil--2 92... see >
> > > > 828 2:3 > > > 58.35
2.2: a ................ a 928.80 >
> > > > 880 823 > > > m 28.3%
2.3 832 a ................ a £5 588
> > > > 2.22283 > v.85 > > > 98.35 >
o ................ 92m
> > > > 2.2: a n 323 > > > see: .588 >
5:00
o 28.3%
95— .595
2.2: a ................ a 223 . r >
> > > > > > > + as, 888
cos-80 + >20 m Sufism
2.2: a ................ .2 2.2... see
> > > > c980 >>az > > > < Sufism >
82> 81m $0 2053 :5— 3:00:00 02 mo> oZ mo> team weed $2820
A m 02 mo». 02 23035 Loni x30 25 3:8 3mm Emce— uuofizm BEE
a; o: 28 ea 2.25
. can x82 33% goo—w
82% as» Em as: baa-p

 

 

 

 

 

 

22.8 .mm ease; _ aoaeooxm 2 85:38an 2:520 .8 283: .3. 222

146

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Q: to Q» o o 5: o to Cm EN 5: to o tn o E \- .88-
.. ............... 9:5
> > > > 2.2: a .— 552: > > > 28288 >
8:80
98.35
8:830: Aooooc
> > > > O ................ A 823 > > > :_ 3:: >
2.9.8
232 3 E8839
.. e m 28.3%
> > > > $2: D ................ a ow_om > > > 92m >3 >
832 m 80.3%
> > > > 2.2: a ................ -_ own: > > > €52: >
822 : soozem
---------------- 92:85
> > > > 2.2: a n 22.2 > > > 888 >
:280
828.33.
> > > > 2.2: a ................ -: owe: > > > :28 >8 >
822 m 28.35
> > > > $2: 0 ................ 4 090m > > > 93m >30
.522 58.35 >
82> Etm 9:0 :08? :EM :2 mo> 02 no». tozm wfii
o 8 o 3:880 :5 2w: 0: 3:875 80
A M Z > Z 8 08885 :onE V. .A E 8:00 gem £w:o_ 80.3.6 :3
> 288m 880 we 032$ :82 0.62m 9.62m
€288: was 8m m8: Ham-..-

 

 

 

 

 

 

 

288 .: 889 = 85:82.: a 88:22.38 8:88 .8 E8: .2 2%.:

147

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

148

 

A _HZV 98:87. ova—05 wZ .ANHZV 829603 28.8:— n :Z .02 ”oz 80> ” 3 83635 ”L .tonm ” m .w:oq H...—
—.v v.5” md cfin 9m 9.6 vém VA: Wm afiw Ema N.m —.: ad WMN v.2. N.m N.m edw QM m4? N.mm ad N.Nm h— Ax: o\o
: cMN MN 2: 2 E 0: MN 2 NMN ONN 3 GM ©~ v0 ca~ z 3 NVN a N: El N NNN 3. CNN _SOH
~ ~ _ w m M o _ _ _ ~ ~ N _ _ M — h c N ~ _ M _ ooD
M~ N M _ _ 2 cN E _ N M 2 N E N w o— fl M— _ CN >02
NM N Mfi v _ 2 M _M 0N w v v 9N vM _ 2 MN _ MN A: VM :00
OM M N VN v OM MN m _ N. 0N _ N OM N _ NN 0N w QM new
_ fl N v a N _ A: _ _ _ ~ _ m h _ fl _ ~ w h _ _ N M~ ws<
cN 2 N _ S N 5N MN M M N c— t _ MN m— v_ VN m MN 33-
N _ _ v _ N h — _ N _ M _ _ ~ V v c E _ _ M M _ _ 3 0:3.
ON N. b _ ~ o v MN NN _ N a 2 M _ MN _ w E MN v RN .3:
VM N c v m 1N m _ CM cN v o N. MN M M OM v E w_ MN m cM a<
c _ v o _ _ _ v v N — E v N N z N v c _ b a :N _ v ©~ .32
_ _ _ c _ m ~ — — o— ~ ~ 0 c N— h m ~ — ~ N~ no”—
_N _ 0 fl m S m _ ©~ _N _ m S _ N 2 _ h z _N _ NN :3.
firm 82> 80 60> WNW—WWW :8qu &:m L oz .5 L oz 3 o— ..m .4
oz 3 oz o: em .4 Z 880 5:02
3 3 52:: 8.80 x82 98m 262m 533 262m ”MW-w.—
mNMWO Pg mPZ<m figmmH-DO FEW-H _ “SN

 

 

 

 

 

:8me .200 :_ 82388823 moum8888n0 0:02:80 .8 8:288: .m< 83:..-

 

 

CHZO 93:3? 252: ”$2 ANHZO $238? ova—05” 2 team ”pm mac; ”D 03835 ”L 02 “oz mo> ” o>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

149

 

9M WOO 96 O.MM WV 0.” Ndm OS O.M —.Ow 9mm QV NA: #6 m.NN ~46 O.m m6. WOO O.M NOV wdm m6 mi” O.m— GA: nun-”W
O.M m.MO m.m de YN N.: efim —.v a; —.VO ufiw "6 ad fir WON MN“. b6 m.M «JO v.N Ndm vdm O Ndw m4 — Ax: .X.
m O2 0 Ov v O_ MO N. M of w: h 2 N~ MM MN_ m e 02 v we _O~ O O2 ON O: 130%
c L m L L O _ _ v _ v _ N _ v _ _ L — v _ w m V— N c _ 00Q
2 N _ N O_ _ 3 N_ L N 2 N M_ m O~ M~ N 2 >02
_N m N E N O_ O— N N m 3 _ ON a ML w— M _N woo
O_ _ c _ M O— M t t _ N L N t _ O- h M_ M: N ON mom
w=<
N M_ m N m ~ 3 ML N v : M L : N m m N— M 2 33.
v _ _ v _ L O m _ M L _ _ _ H M _ N M _ m O _ v _ _ m L 05:.
_ h m m w a N w w _ _ c a w 32
N o_ M m m : _N f _ N N m ‘1 _ ON L N. ML KL v ~N .a<
O L O _ _ O N. N _ M _ c O _ M N. O _ O L .82
o _ _ h N w h _ O _ L O b O — E O m 3 _ h — pom
N _ v _ N. N _ _ _ _ _ v N. N_ N. m _ _ _ N _ :3.
firm 8m; gnu 50> WWW—WWW Fauna Ebw L “.2 3 L uZ 3 L A.m .4
L ”Z 3 oz .5 L pm .4 oz .3 Egan 860 3:8 x82 Eam 052m Swan— o>oo_m WM.“ 5:22
mwm/MMWO PE HZ<L M<m3 AFDC Hgmuh. NOON

 

 

 

 

 

 

 

325280 .. “mama .290 E wonmfiwouocm 85380820 808.80 mo 3:252”— .m< 2an

 

H28 ~89sz mo 83 O0 :52me ”a

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

OO_ NOV 9w WE WM: 9v . 259::—
OOO NOV ON TV ms Om Ym ON md ON m.O m.O ON 252$
Sm _2 BN 2 MN 3 S 3 ON O H _ O 830
E8520 xoflm Liz 03m 52% SEE 580
xoflm 03?: 5.80 gm x80 ®>:O
A<HOH E=m R20 .0 >>mZ x3e \oim BEL E080 Ems ZOAOU
>mmU MU<Am .8 >><Z Mia—m ZHHMU
w. _ m MAN v.3 O: . 23.—om
w. H m 9O 0.0 ON N.O~ w: 5.5 «6 ON. :2 Evy—om
N3 N M N NM m N M: O 3 830
55m ii 6.03 LESMSm com owsfio 32.0% ammom BEBOOO BE? mow—OD
Gm:— HU—Hm ELLE

 

 

 

 

 

 

“8me :00 E momma: Baafimofif go 9200 mo 55:35 é< 2an

150

Consent Form for Experiment

The purpose of this study is to measure the amount of ultraviolet radiation (UVR)
received at different sites of upper body and to identify which areas are exposed the most
during play. Golfers tend to be at higher risk of developing skin cancer than the average
person due to the extensive exposure during periods of sun peak time since cumulative
UV exposure has strongly associated with skin cancer. The ultimate goal will be to
develop clothing that offers more UV protection service.

You are invited to play a free game of golf (18 holes) at East course of MSU during a day
when you will be asked to wear personal dosimeters at several body locations over or
under the golf clothing. The dosimeter (film badge) will be held by adhesive bandage
which is unnoticeable by others. Record will be made based on fiber content, fabric
structure, hue, shade of color, and style of clothing that you choose on the day of
experiment.

During these processes, you have the right to refuse to participate or withdraw this study
at any time. Your name or identification will not be associated with results. Your privacy
will be protected to the maximum extent allowable by law. All data recorded will be
accessible only to a few of permitted researchers. Data will be kept confidential and will
be aggregated for any reports. Your participation is entirely voluntary.

If you have any questions about this study, please contact Dr. Ann Slocum (phone: 355-
3779, ascolcum@msu.edu). If you have questions or concerns regarding your rights as a
study participant, or are dissatisfied at any time with any aspect of this study, you may
contact Ashir Kumar, M.D., Chair of the University Committee on Research involving
Human Subjects by phone: (517) 355-2180, email: ucrihs@msu.edu, or regular mail: 202
Olds Hall, East Lansing, M148824.

Consent for Participants

I have read the above description and voluntarily agree to participate in today’s
experiment for this study.

Subject’s Name:

 

Subject’s Signature:

 

Date:

 

Please indicate whether you consent to following category:
Yes No
I consent to having my picture taken for educational and research

Eli—moses.—

151

Consent Form for Interview

The purpose of this study is to investigate golf clothing currently available in stores. In
order to design clothing that offers more effective UV protection, it is necessary to
examine what types and features of golf clothing commercially available and generally
worn by players. The data obtained from this study will be helpful to identify the design
problems and to appreciate the scope of UV specialized clothing.

Your store was randomly selected from 12 local golf clothing suppliers/retailers listed in
2002 Yellow book in Greater Lansing Area. You will be asked a few questions regarding
your products commercially available and your customers’ preferences. The interview
will be taken approximately 10 minutes and can be tape-recorded. In addition, description
of clothing in your store will be recorded, for example, styles, price, fiber content, fabric
structure, and color of golf t-shirts.

Your participation is entirely voluntary. You have the right to withdraw from this study at
any time. Your name will not be associated with the results. Your privacy will be
protected to the maximum extent allowable by law. All data recorded will be accessible
only to a few permitted researchers. Data will be kept confidential and will be aggregated
for any reports. The audio tape will be destroyed after the data reported.

If you have any questions about this study, please contact Dr. Ann Slocum (phone: 355-
3779, ascolcum@msu.edu) or Heewon Sung (phone: 381-1510, sungheew@msu.edu). If
you have questions or concerns regarding your rights as a study participant, or are
dissatisfied at any time with any aspect of this study, you may contact Ashir Kumar,
M.D., Chair of the University Committee on Research involving Human Subjects by
phone: (517) 355-2180, email: ucrihs@msu.edu, or regular mail: 202 Olds Hall, East
Lansing, MI 48824.

Consent for Participants
I have read the above description and voluntarily agree to interview for this study.

Subject’s Name:

 

Subject’s Signature:

 

Date:

 

YES No
I consent to be tape-recorded during the interview.

 

Subject’s Signature:

 

152

APPENDIX B

Questionnaires

 

153

Table B1. Items in the Health Belief Model and Attributes of Innovations in Pretest

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Sample Item Source Reliability
Health Belief Model
1. In the typical week. approximately how many hours do n.a.
Sun-Related you spend in the sun? Jackson,
Activities 2. In the typical week. approximately how many hours do (1997)
you sunbathe?
3. 1 use sunscreen with a sun protection factor (SPF) of at 0.82 (5)
least 15 on my face when I am in the sun.
4. I use sunscreen with SPF of at least 15 on my body Carmel et al.,
Current Sun when I am in the sun. (1994)
Protective 5. I wear protective clothing to cover my body like a long
Behavior sleeved shirt and long pants when I am in the sun. Jackson.
6. I wear a hat when I am in the sun. (1997)
7. I try to stay in the shade as much as possible when I am
outdoors.
8. 1 am more likely than the average person to get skin 0.70
cancer.
9. The possibility of skin cancer worries me. Jackson, (1:13:13 3
Susceptibility 10. 2:10;] 5::211: .worry about getting skin cancer until I (1997) a = 0.83 (3)
l 1. Whenever I hear of a friend/relative (or public figure) Gmbbs &
getting skin cancer, it makes me realize I could get it Tabano (2000)
too.
12. Getting skin cancer would severely affect my life. 0.80
13. It would be terrible to have skin cancer. ““50“,
. 14. It would be terrible to look older than 1 really am due to (1997) 9 Item 8
Seventy too much sun-exposure deleth’
. ' , _ Jackson & a = 0.86 (3)
15. Skin cancer can be taken care of pretty easrly With Aiken (2000)
medication in the future.
16. Wearing a long-sleeve shirt, a brimmed hat, or 0.78 (3)
sunscreen when I am in the sun would reduce my
chances of getting skin cancer. Jackson &
17. Whether or not a person develops skin cancer is related Aiken (2000)
Benefits to. how frequently they avoid exposure while spending
time in the sun. Carmel ct al.,
18. If more people cover up their bodies when they are in (1994)
the sun, they would reduce the chances of getting age
spots and wrinkles.
19. Ii is likely that a long-sleeve shirt feels uncomfortable 0.74 (6)
would keep me from wearing it.
20. It is likely that a wide brimmed-hat feels uncomfortable '9 Items ‘6
would keep me from wearing it. Jackson, 8‘37 dcmcd’
B . 21. It is likely that the cost of UV specialized clothing (1997) a - 0'81 (4)
ehavnoral . .
Barriers would keep me from usrng it. . . . .
22. It 15 11ker that the style/appearance of UV specralized Vail-Smith &
clothing would keep me from wearing it. Felts (1993)
23. Having to reapply sunscreen while spending some time
in the sun bothers me.
24. I don’t like to use sunscreen because it feels unpleasant.
25. I feel more attractive with a tan. 13‘3““, 0'79 (5)
Psychological 26. A tan makes me feel better about myself. (1997)
Barriers 27. I feel healthier with a tan. . Carmel et al.,
28. A sun-tanned person looks healthier than someone (1994)

 

154

 

 

 

 

 

 

 

 

Variables Sample Item Source Reliability
without a tan.
29. A sun-tanned person looks more attractive than Vail-Smith &
someone who is not tanned. 125130993)
30. My friends usually wear protective clothing, like a long- 0.52 (7)
sleeve shirt or hat, when in the sun.
31. My friends always use sunscreen with a sun protection '3‘ Items 25
factor (SPF) of at least 15 when they are in the sun. JaCkson’ 8‘36 deletcd’
Cue to Act 32. I have family members or friends who have skin cancer. (1997) a — 0'68 (5)
33. I have noticed more news announcements and articles Arthey &
about risks of sun exposure. Clarke, (1995)
34. Most of my friends are tan.
35. Most of my friends feel that a suntan is a good thing.
36. My friends disapprove of people who have a dark tan.
Attributes of an Innovation
[Economic Advantage] 0.7 8 (8)
1. I cannot get the best value for the money by using this
product. -> Item 1 &
2. This product would maximize the quality for the money 7 deleted,
that I spend. a = 0.83 (6)
[Image]
3. Golfers who have this product would have more Moore 8‘
prestige than those who do not. agaisat
Relative 4. Wearing this product would improve my image among
advantage my golfing friends. Strutton et
5. I would be self-conscious about wearing this product. al., (199,4)
[Overall Usefulness]
6. This product is more fashionable or unique in style than
other similar types of shirts.
7. This product would be better in performance (i.e. fit,
comfort) than other similar types of shirts.
8. This product would provide better UV protection than
the other garments I have.
9. This product does not differ from what I have worn in 0.66 (4)
the past. Strutton, et
10. This product would be coordinated with the other al., (1994) -) Item 9
Compatibility garments I have. deleted.
1 1. This product would be compatible with my current need I;::;: a = 0-73 (3)
ofa new shirt. “991)
12. This product would be appropriate for my golf games.
Strutton, et 0.49
13. It would be difficult to understand how this product 3]" (1994) -) Item l4 &
Complexity works to prevent UV radiation. 15 deleted,
14. Overall, this product would be easy to use. Moore & and add “cm
15. I believe that this product is cumbersome to use. Benbasat 20’ 0 = 0-66
(1991) (2)
16. It would be easy to try out this product without a big 0.89
commitment. -) Item 17
Triabili 17. Before deciding whether to purchase clothing, I prefer to Strutton, et deleted
ty try it on. al., (1994) a - 0-83 (2)
18. If this product, made specially for golfers, is available in

 

stores where I buy my golf clothes, I would try it on.

 

 

 

155

 

 

 

 

 

Variables Sample Item Source Reliability
Attributes of an Innovation
[Visibility] 0.71
19. I have seen people wearing this product when I am 9 "cm 20
outdoors. Moore & ducted,
20. If I use this product, the advantages/disadvantages Benbasat a = 0'82 (3)
. . would be readily apparent to me. (1991)
Obsewab‘l‘w [Result Demonstrability]
21. I would have difficulty explaining why wearing this Strutton, et
product is beneficial. al-i (1994)
22. I would communicate the advantages of wearing this
product to others.
0.86
23. I plan to wear a UV specialized shirt for most outdoor Jackson & $11131 26
. activities. Aiken (2000) c e .
rim?“ to 24. I intend to wear a long-sleeve shirt while playing golf. a = 0'87 (3)
p 25. I intend to purchase a UV specialized shirt within the Shim &
next few years. Drake, (1990)

 

26. Ijrobably won’t purchase a UV specialized shirt.

 

 

 

156

 

.O B A “Em “8.53.55
.55.“ 3% E390
$55 .23 @8358
52529.5 55. "Em
+om Logic can 9:5 ace
<>D 59: 5:53.. >2
we {a :2: 29: 433
ears—w...— wzceofira >3

 

 

 

 

 

.85 33 3.05 no 65:88:“ .533 Eco
8 cosmic—u £3009. mm 8355 $5. .953 comics.— ~o_o_>ab_: 8 .5898 3.2: 295 Show
a .8 x8: 2: Lo x23 can Joan 52:27. :2: 958 33 : .EoEton—xo uoE 05 :o 3QO

  

Figure B1. A Black-and-White Photograph for the Pretest

157

Questionnaire

158

    

COLLEGE or
HUMAN ECOLOGY

Department at llama
Environment and Deal"

Michigan State University
204 Hunan Ecology Building
East Lanslno. Mlchiom
48824403)

(517) 3557712
me (517) 432-1053

MSUIsmatbm-dm.
WW.

 

MICHIGAN STATE

UNIVERSITY
October2002

 

Dear Golfer:

Outdoor enthusiasts such as golfers have paid limited attention to sun safety practices,
although their activities lead to extensive sun exposure during peak sun hours. You are
invited to participate in a survey, conducted by Michigan State University Clothing and
Textile researchers. The goal of the project is to design a shirt that offers more effective
UV protection than currently available golf shirts.

The purpose of this survey is to learn about your beliefs about sun exposure, and your
perCeptions of a prototype shirt that offers UV protection. Your answers to the following
questions will aid us in designing an acceptable golf shirt.

Please read each question carefully and answer as accurately as possible. The
questionnaire has three sections and it will take you approximately 20 minutes to
complete. You have the right to refuse to answer or to withdraw from this study at any
time. Your responses will be anonymous. Data will be kept confidential and will be
aggregated for any reports. All data recorded will be accessible only to the principal
investigators, and the data will be destroyed after the findings are reported. By
completing and returning this questionnaire, you indicate your voluntary agreement to
participate.

If you have any questions about this study, please contact Dr. Ann Slocum (phone: 355-
3779, W or Heewon Sung (phone: 381-1510, W. If
you have questions or concerns regarding your rights as a study participant, or are
dissatisfied at any time with any aspect of this study, you may contact Ashir Kumar,
M.D., Chair of the University Committee on Research involving Human Subjects by
phone: (517) 355-2180, email: 391W or regular mail: 202 Olds Hall, East
Lansing, MI 48824.

Thank you for your time and cooperation in answering our questions. The information
will be very helpful.

Sincerely yours,

Ann C. Slocum Heewon Sung

Associate Professor Doctoral Student
Michigan State University Michigan State University

159

 

QUESTIONNAIRE

 

Section I: This section asks about your health beliefs and your
behavior in the sun.

 

The following questions ask about your current practices when you are outdoors. Please
respond to each statement by circling the appropriate number on the scale provided.

I. In the mica] week in the summer, approximately how many hours do you spend in the sun?
(1)0 hrs/week (2) 1-2hrs/week (3) 3-4hrs/week (4) 5-7hrs/week (5) 8-10hrs/week
(6) ll-lShrs/week (7) 16- 20 hrs/week (8) 21-25hrs/week (9) more than 25hrs/week

2. In the gpical week in the summer, approximately how many hours do you sunbathe?
(1)0 hrs/week (2) l-2hrs/week (3) 3-4hrs/week (4) 5-7hrs/week (5) 8-10hrs/week

 

(6) ”-1 Shrs/week (7) 16- 20 hrs/week (8) 21-25hrs/week (9) more than 25hrs/week Li“!
Some All
Never times the Time
3. I use sunscreen with a sun protection factor (SPF) of at least
15 on my face when I am in the sun. 1 2 3 4 5
4. I use sunscreen with SPF of at least 15 on my body when I l 2 3 4 5

am in the sun.

5. I wear clothing to cover my body like a long sleeved shirt 1 2 3 4 5
and long pants when I am in the sun.

6. I wear a hat when I am in the sun. 1 2 3 4 5
7. I try to stay in the shade as much as possible when I am 1 2 3 4 5
outdoors.

Below are some beliefs people have about sun protection. Please indicate whether you
disagree or agree with each statement by circling the appropriate number on the scale
provided.

Strongly Strongly
Disagree Agree
8. Wearing a long-sleeve shirt, 3 brimmed hat, or sunscreen
when I am in the sun would reduce my chances of getting 1 2 3 4 5
skin cancer.
9. Whether or not a person develops skin cancer is related to
how frequently they avoid exposure while spending time in l 2 3 4 5
the sun.
10.1f more people cover up their bodies when they are in the
sun, they would reduce the chances of getting age spots and l 2 3 4 5

wrinkles.

160

The following questions are regarding your beliefs about tanning. Please indicate whether
you disagree or agree with each statement by circling the appropriate number on the scale
provided.

Strongly Strongly
Disagree Agree
11. I feel more attractive with a tan. 1 2 3 4 5
12. A tan makes me feel better about myself. 1 2 3 4 5
13. I feel healthier with a tan. 1 2 3 4 5
14. A sun-tanned person looks healthier than someone
without a tan. 1 2 3 4 5
15. A sun-tanned person looks more attractive than someone
who is not tanned. l 2 3 4 5

Please indicate whether you disagree or agree with each statement by circling the
appropriate number on the scale provided.

Strongly Strongly
Disagree Agree
16. People with whom I play outdoor sports usually wear a
long-sleeve shirt or hat, when they are in the sun. 1 2 3 4 5
17. People with whom I play outdoor sports usually use
sunscreen with an SPF of 15 or more when they are in the l 2 3 4 5
sun.
18. Most of my friends have a suntan. l 2 3 4 5
19. Most of my friends feel that a suntan is a good thing. 1 2 3 4 5
20. My friends disapprove of people who have a dark tan. 1 2 3 4 5

Below are some factors that may keep you from practicing sun protection. Please indicate
whether you disagree or agree with each statement by circling the appropriate number on
the scale provided.

Strongly Strongly
Disagree Agree
21. If a long-sleeve shirt feels uncomfortable, I would not wear
it for golf. 1 2 3 4 5
22. If a wide brimmed-hat feels uncomfortable, I would not
wear it for golf. 1 2 3 4 5
23. The cost of UV specialized clothing (3 6000-3 80.00)
could keep me from purchasing it. 1 2 3 4 S
24. The style/appearance of UV specialized clothing could
keep me from wearing it. 1 2 3 4 5
25. Wearing a long-sleeve shirt while playing golf would
restrict my movement. 1 2 3 4 5

The following questions are your beliefs about skin cancer risk. Please indicate whether you
disagree or agree with each statement by circling the appropriate number on the scale.

Strongly Strongly
Disagree Agree
26. I am more likely than the average person to get skin
cancer. 1 2 3 4 5
27 . The possibility of skin cancer worries me. 1 2 3 4 5
28. Whenever I hear of a friend/relative (or public figure)
getting skin cancer, it makes me realize I could get it too. 1 2 3 4 5
29. Getting skin cancer would severely affect my life. 1 2 3 4 5
30. It would be terrible to have skin cancer. 1 2 3 4 5
31. It would be terrible to look older than I really am due to too
much sun-exposure. 1 2 3 4 5

I61

 

better UV protection during playing golf. Assume that you can purchase it
the regular golf stores. The price is approximately $60.00-S80.00. Please
examine the sample shirt and answer in a way that shows how you feel.

 

Section II: This section asks about your perception and opinion about a UV
specialized shirt (see attached color picture). This product is designed to offer

in

 

 

Please indicate whether you disagree or agree with each statement by circling the
appropriate number on the scale provided.

Strongly Strongly
Disagree Agree

1. This product would maximize the quality for the money

that I spend. 1 2 3 4 5
2. Golfers who have this product would have more prestige

than those who do not. 1 2 3 4 5
3. Wearing this product would improve my image among my

golfing friends. 1 2 3 4 5
4. I would be self-conscious about wearing this product. 1 2 3 4 5
5. This product would be more fashionable or unique in style

than other similar types of shirts. I 2 3 4 5
6. This product would provide better UV protection than the

other garments I have. 1 2 3 4 5
7. This product would coordinate well with the other

garments I have. 1 2 3 4 5
8. This product would be compatible with my current needs

for a new shirt. 1 2 3 4 5
9. This product would be appropriate for my golf games. 1 2 3 4 5
10. It would be difficult to understand how this product works t1

prevent UV radiation. 1 2 3 4 5
l l. I would have difficulty explaining why wearing this

product is beneficial. l 2 3 4 5
12. It would be easy to try out this product without a big

commitment. l 2 3 4 5
13. If this product, made specially for golfers, is available in

stores where I buy my golf clothes, I would try it on. 1 2 3 4 5
l4. 1 have seen people wearing this product when I am

outdoors. l 2 3 4 5
15. I could communicate the advantages of wearing this

product to others. 1 2 3 4 5
16. If I use this product, the advantages/disadvantages would

be readily apparent to me. 1 2 3 4 5

Please indicate the likelihood of your future behavior by circling the appropriate number

on the scale provided.

Strongly Strongly
Disagree Agree
17. I plan to wear a UV specialized shirt for most outdoor
activities. 1 2 3 4 5
18. I intend to wear a long-sleeve shirt while playing golf. 1 2 3 4 5
19. I intend to purchase a UV specialized shirt within the next
few years. 1 2 3 4 5

162

 

Section III: This section asks about your demographic information. ]

 

Please respond to each statement by marking X next to the most appropriate response or fill
in the blank in the space provided.

1. Gender? [ ]Male [ ]Female

2. What is you age? years old

 

3. What is the highest level of education you have completed?
] Eight Grade
] High school degree

I
[
[ ] 2-year degree, Some college, or Associate degree
[ ] 4-year college degree

[

] Graduate School or Advanced degree

4. What is your marital status?

[ ] Single
[ ] Married -—> If married, do you have children younger than 18 who live in your
[ ] Others /' household? [ ] Yes [ ]No

5. What is your race/ethnicity?

 

[ ] Caucasian [ ] African American
[ ] Asian [ ] Hispanic
[ ] Native American [ ] Others
6. How many rounds of golf do you play each month of the golfing season? rounds/month

7. What time of a day do you usually play golf? (Check all that applied)
] Early morning

] Mid morning

[

[

[ ] Early afternoon

[ ] Mid to late afiemoon
[

] Evening

8. Do you have family members or friends who have skin cancer? [ ] Yes [ ] No

Thanks you for your participation!

163

APPENDIX C

Regression Assumptions

I64

Assumption of Regression Model A

Regression model A: T 0 Examine the contributions of] 1 variables in the health belief

model to sun protective behaviors (SPB).

X = Age, marital status, ethnicity, susceptibility, severity, benefits, behavioral barriers,
advantages of tan, cues to sun protection (CA-SP), cues to tanning (CA-Tan), &
cues to family having skin cancer

Y = Sun protective behaviors (SPB)

Figure C1. Partial Regression Scatterplots of 11 Health Belief Variables on Sun
Protective Behaviors

Figure Cla. Age and SPB Figure Clb. Marital Status and SPB

 

 

   

 

Sun Protective Behaviors

 

 

 

 

Sun Protective Behaviors
L 0
0°

 

Age Marital status

Figure Clc. Ethnicity and SPB Figure Cld. Susceptibility and SPB

 

 

   

 

 

 

 

Sun Protective Behavrors
a
0
Sun Protective Behaviors
a

 

 

Ethnicity Susceptibility

I65

Figure Cle. Severity and SPB Figure le. Perceived Benefits and SPB

 

 

 

 

 

 

 

 

 

 

O U
s
61 o a 3. e a
on ° ° 0 e o
O
4i o o o 4‘ Q a n
a on O f
o a
2 °. :o..:°°w:s§°s °
0 O D an“ &0 no 9
a D an a O
a .. ° ”z “i,- ~. .hv‘e" a
a 2 ° ' t.‘ "' ° It;
a
E ‘2‘ a o o 9:: o no not: do" g
a ° 0 o
o a ° ° 9 °°e an a1
2 4 on 2
s .. °
§ ,. 1‘:
‘9' o 8 0 a e n
D h 4
‘2 ii . ° 2 ‘ ° ..
a o
8 -e . . 0:) -a -
-8 -6 -4 2 0 2 4 6 -10 8 -6 4 2 O 2 4 6
Severity Perceived Benefits

Figure Clg. Behavioral Barriers and SPB Figure Clh. Psychological Barriers and
SPB

 

 

 

 

 

 

 

 

 

8 8
° 9
5 a 61 c
a a c 0 ° 0
9° 1: o e
‘ 1 a no I a
, . .. , .. . g
a '0 '° ° ° ° ° m
2‘ 0" a o o O S a} :9 3 g
a I o s
o $ 0 0° 0"
e 0 4 o o a an 0 a rfi o a g
Q ° ° 1: s c 0 . o c Q, .
a a OH O O flfi on 0 g
c 0
g 2 ° ‘; . a e a. o o no i '
u D O . c
9 ° .. . o 3
§ ’4 . a o 0 CD
9 a
'9' .
‘2 ‘5‘ ° .
o
0:) a -a f
10 0 10 -20 10 O 10
Behavioral Barriers Psychological Barriers

Figure Cli. CA-SP and SPB Figure C1 j. CA-Tan and SPB

 

 

 

 

 

 

 

 

 

a a
O 6 a
er 1. a , q, o
4 ° 0 no 0 °
5 ° , .
0 00°
2 o a o °095 “é. :
0 o 3°:00‘32" ”0° 0 9
g g 0 a Do 0; “Diana 0 a
Z S 2 Q; count: 0 c o
8 ' 8 0 ‘Po" ° 0
a o ° . .° “N".
on
E a. 9‘ . ° 0
C c o
3
(3 -a o a) -a
4 2 o 2 4 a 4 2 o 2 4 6
Cues to Sun Protection Cues to Tanning

I66

Figure C 1k.Family-Skin Cancer and SPB

 

 

 

 

Sun Protective Behaviors
r.

 

having family/friends who have skin cancer

Figure C2. Histogram of the Standardized Residual of Sun Protective Behaviors

30

 

 

 

 

Frequency

‘53 ‘12.. "~» "~ >5 Re to a 9.. vs a

Standardized Residual

Figure C3. Normal Probability Plot of the Standardized Residual of Sun Protective
Behaviors

1.00

 

75. /

 

 

Expected Cumulative Probabilities
\

ooo . .
000 25 so 75 100

 

Observed Cumulative Probabilities

Kolmogorov-Smirnov Normal Test: 2 = .057, df = 152, p = .200

I67

Figure C4. Scatterplots between Health Belief Variable and Standardized Residual of Sun
Protective Behaviors

Figure C4a. Age and Standardized Residuals

Standardized Residual

 

 

 

 

 

10

ab 30 56 60 70 ab 90

Age

Figure C4c. Ethnicity and Standardized

 

 

 

 

 

 

Resrduals
3
21 a I
° 8
1‘ l
o
o
I
_ O 3
W e
3
'0
.a a
O .1.
m
'8 o
.E o
a .
'U o o
c
N
a .3 ' V V r W V
- 2 0.0 .2 .4 .6 .5 1.0 1.2

Race (0=other. 1=Caucasians)

Figure C4e. Severity and Standardized

Standardized Residual

 

 

 

 

 

Residuals
3
2‘ a
. l
" - . 2 E I l :
° - 3 r 2 r
0 a ° 3 3 8 ° I E
I g I E s l
e z B '
-1. . : = 9 ° I
8
-2. .
I
-3 , , . . o f
4 6 8 10 12 14 16
Severity

168

Standardized Residual

Standardized Residual

Figure C4b. Marital Status and

 

 

 

 

 

 

 

 

 

 

Standardized Residuals
3
2. j ;
a 5
1‘ 3 a
l
0 l I
° r
, I
1 4 . !
- r
B
.24 .
-3 T Y Y ‘7 v
-.2 0.0 .2 .4 .6 .e 1.'o 1.2
Marital status (O=other, 1=single)
Figure C4d. Susceptibility and
Standardized Residuals
3
8
2‘ .
° 9
1‘ ' Z s i I : 2
g ‘ I I o o i
o 3 _ I I ' I ' l .
a I: E z I .
-1 . E . 3 o E I e
a .
B
-21 a
.3 v V 1 0 V I V
2 4 6 8 1O 12 14 16
Susceptibility

Figure C4f. Perceived Benefits and

Standardized Residual

Standardized Residuals

 

 

 

 

 

3
2 1 8 a
8 o
o a o
‘ O
1 I a I . .
l - '
a , g I
o - a Q i ' B
r I
o ' B u I . o
0 ° 0 u o
I
8 ' ° 9
-1 . 8 . 3 u a
D
.2 4
-3 V fi V
2 4 6 B 10 12 14 16
Benefits

and Standardized Residuals

Figure C4h. Psychological barriers

Standardized Residuals

Figure C4g. Behavioral Barriers and

 

 

o a o
o 80
on o
g.
ul ocean
I dead a
I one I
D O
a o c
on o
a o o
o I o

 

 

 

 

 

 

 

 

eeooflu o
as
.- olooelona 98
e o a o I on a
a duel. one
o onllloeo o o 0
ocean Us Ila e
o in;' one o
a Bee all I
I. a colon a
a :5 a
0 loo
0 a on
on
n
e

339.com 859.0333

16

2'4

13

{a

14

1'2

1b

Psychological Barriers

Behavioral Barriers

Figure C4j. CA-Tan and Standardized
Residuals

Figure C4i. CA-SP and Standardized
Residuals

 

 

 

 

 

"M
o a on o .w
o a on on In a
o n III-all 'loo 0 o .8
dead! I'I'e RBI 0 o a
o o a Ieaoqunn cell! a 16
e e 08.... can do
a o no on do a o 14
e an a
e 12
< l 4 t 0
2 1 o 1 9.. 3

.2663. BuEaucSm

 

 

 

 

 

all.
no vaol
a
on o I u n o o 18
a a an. al no a o
o o a '8' an Ion .— 0 r6
0 noel IIIeoa III-o o o o
o n aglvllall a o a v4
0 e I III a 08
on. can on a .2
< < J 4 O
2 1 o 4 9.. 1..

_naHuom gaucaucflm

Cues to Tanning

Cues to Sun Protection

Figure C4k. Family-Skin Cancer and Standardized Residuals

 

ofo

 

 

 

I one 'Bnn I an O o D o w
B Ioeol'llo CI
2 1 o 4 2.. 1..

 

.338". Seascape

12

1.0

Having family/friends who have skin cancer (0=no. 1=yes)

169

Assumption of Regression Model B

Regression Model B: T 0 Examine the Contributions of Five Attributes Variables to

Intention to Adopt UV Specialized Shirts (IA)

X = Relative advantage, compatibility, complexity, triability, observability

Y = Intention to Adopt (IA)

Figure C5. Partial Regression Scatterplots of Five Attributes on Intention to Adopt

Figure C5a. Relative Advantage and IA

 

6<

Intention to Adopt

 

 

 

 

Relative Advantage

Figure C5c. Complexity and IA

 

 

 

 

 

4i 0 no
an o
°n
2 . o O a
a "e $° 1P 0
o 8 a ll), 0 o 8 “a on On 0:30
E‘s— m ' -- n
o‘ g C n V v 435%“
II a one a a
B o o ‘2" ° °

.- @ ° 0
8' 2 . "In, ° 0
‘O
< 3.0
9 o
C 4‘
_O
a
C
93
E -6

-6 -4 2 O 2 6

Complexity

170

Intention to Adopt

Intention to Adopt

 

Figure C5b. Compatibility and IA

 

 

 

 

Compatability

Figure C5d. Triability and IA

 

 

 

 

 

(pl
Ni
a
0'1
.a
N
U

Triability

Figure CSe. Observability and IA

 

 

 

 

 

 

 

 

41 o 3 ea
5
2. ,, .° : . .
2 a ‘2'. °
8 u
no fig 5 9.0 9 20°: 90
0 _i 3 an n a, g .. 1? n
on “00:": k .
a 2 35‘s? °° °
0 e
g a o
E .41 a
g
C
Q
s e
-6 4 2 6 2 4 6
Observability
Figure C6. Histogram of the Standardized Residual of Intention to Adopt
30
20«
10«
6‘
5 Std. Dev: .98
3 Mean = 0,00
3% 0 N = 154.00

 

%%@@%%%%%%%
Standardized Residual
Figure C7. Normal Probability Plot of the Standardized Residual of Intention to Adopt

1.00 /
,0

.75 -

 

 

 

 

Expected Cumulative Probabilities

 

0.00 .55 .5'0 7'5 100

Observed Cumulative Probabilities

Kolmogorov-Smirnov Normal Test: 2 = .066, df = 154, p = .097

I71

Figure C8. Scatterplots between Attribute Variable and Standardized Residual of

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Intention to Adopt
Figure C8a. Relative Advantage and Figure C8b. Compatibility and
Standardized Residuals Standardized Residuals
3 3
2 a a 2i
1‘ 2 g . B a a 6 ° ° H g ' a 2
I g a 3 o I g E 3 ° 3
, ,;-.:E;j$ , i : - P. j I .2. , .
s o°: Ea . -a '3' g . l i
1:3, .=i:°a‘ '3: '=§a..e
m e o e m o o
3 ° ., a 3 . g
.5 .5 e
g .21 3 -2t
0 10 20 30 2 4 6 6 1O 12
Relative Advantage Compatability
Figure C8c. Complexity and Figure C8d. Triability and
Standardized Residuals Standardized Residuals
3 u 3
2 3 . 2 i
' ° 2 s E E
1< E E 9 1 ° 2 o E
. i a I ' 2 i t i I
_ 0 I l e i 4v i 0 e g I g
g . a E i 2 2 g 3 ° 3 I
Z 4 i i = - ' é ° 9 2 a
s 5 ° 8 i 3
g .24 g -2l 5
O 2 4 6 6 10 12 O 1 2 3 4 5
Complexity Triability
Figure C8c. Observability and Standardized Residuals
3
2 5 a
a 3 i
1 . i - l
a : I i I .
- 0 V I : = ' :
é a i l I :
g H ‘ g a
E E '
2 -21
§
5': -3 , Y Y , .
0 2 4 6 6 10 12
Observability

172

Assumption of Regression Model C

Model C .° T 0 Examine the Contributions of Variables in the Proposed Model to Intention

to Adopt (IA )

X = Age, marital status, ethnicity, susceptibility, severity, benefits, behavioral barriers,
advantages of tan, cues to sun protection (CA-SP), cues to tanning (CA-Tan), cues
to family having skin cancer; Sun protective behavior; Relative advantage,
compatibility, complexity, triability, observability

Y = Intention to Adopt

Figure C9. Partial Correlation Scatterplots between All Independent Variables and IA

Figure C9a. Ethnicity and IA

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EL '2‘ a
2
.9.
,5. “
g.
E -e f t

.1_5 1 0 5 0.0 5 1 0

Ethnicity
Figure C9b. Perceived Barriers and IA Figure C9c. Sun Protective Behaviors and
IA
6 6
u 4
I I . I g
I, % I I
Bi- 2 '8’ I: o 5 . 8- .2T a a J
2 ° '" é .°
2 2 .3" °
8 4+ 5 .41
= a
10 0 10 -6 -6 -4 2 0 2 4 6 6
Behavioral Barriers Sun Protective Behaviors

173

Figure C9d. Relative Advantages and IA

Figure C9e. Triability and IA

 

 

 

 

.. ..
n. _ a
8 8
< <
2 9.
C I:
o _o
'5 a
C C
2 3
E '5 E -6
~45 -6 -4 -2 0 2 4 6 13 -3 2 -1 O 1 2 “

 

 

 

 

 

 

Relative Advantages

Figure C9f. CA-SP and IA

Triability

Figure C9g. Susceptibility and SPB

 

 

 

 

 

 

 

 

 

 

I! O a I I
n
4 a n a t . 9..
on u a ..
u ..
3 a I: a D D n
2 " a all u a a u 2 n fl “ l a. I a
n n a n a "’ :"u" " I °
a. "a n E : n "a ° an n n' u a u a“ on ‘ n “a 9". .
a II an n a n n u a
.. .. ‘h: ° an I :2 ° ’ " n ° " ° ° i.” “n a“
v a u n r n u a I u u n
n n 3‘ U “in“. a 9 0 a n
D a D n D“ n a a E .5 n “no I I g In a a D
a 2 n e a I u "an i: I: l' o '5. 2 n a u a .‘h n I on g a a
'8 ' :1 ° (.1 B '1 8 . a gal u a u
< " ° . < . = °
0 '1 In a a 9 I a “a
a
E 4 o . c 4 u
o o
a " a ,
2 E
5 .5 , I_: -6 ,
d 2 l o 1 2 4 45 .4 .2 o 2 4 e
Cues to Sun Protection Susceptibility

Figure C10. Histogram of the Standardized Residual of IA

on
s

 

 

Frequency

 

Std Dev = .97
Mean = .00
N = 153.00

 

%%%%%%%%%%%

Standardized Residual

l74

Figure C11. Normal Probability Plot of the Standardized Residual of IA

 

 

 

 

100
f
I}:
at?“
75 ‘
8 x”
E r
s r“
5 so I
a)
2
E
E
3 .l I
O ,25 /
U
9.
o
in
3 J}
u.) 000
0.00 25 .50 {75 1,00

Observed Cumulative Probabilities

Kolmogorov-Smirnov Normal Test: 2 = .033, df = 153, p = .200

Figure C12. Scatterplots between the Significant Independent Variables and Standardized
Residual of Intention to Adopt

Figure C12a. Ethnicity and Standardized Residuals

3

 

24

 

 

_a
com on and

 

 

Standardized Residual

 

-.2 0'0 .'2 .3 E, .a 1,0 1.2

Race (0=others, 1=Caucasians)

175

Figure C12c. Sun Protective Behaviors and
Standardized Residuals

Standardized Residuals

Figure C12b. Behavioral Barriers and

 

 

 

 

 

_aaEmom u0N_EIuc5w

 

 

a II Ilia ooBooo

one:

1.i§

:

 

 

 

o 1 9.. 1..

_uauaom noseuufiw

1?)

2'4

25

2'0

1':

{6

1'4

12

16

Sun Protective Behaviors

Behavioral Barriers

Figure C12c. Triability and
Standardized Residuals

Standardized Residuals

Figure C12d. Relative Advantage and

 

 

 

 

 

 

 

 

 

 

o o o o o o
on Io Illanllelele
o e 'olIflL‘eeIa III.
e eucalni-ooa
' as 0
m u o m. m. 1..
.328”. 852223
e
e e
e
a I o o o
In 600 n a- u
eon on!
9' o I Be 0
8000905 no I. o
0" llnl one on
I Erlle e u
on uncle l
I I III a
o o 0' e
I
o o e
.
9,. r o m m. 1..

.2680”. 85322.55

2b

1b

Triability

Relative Advantage

Standardized Residuals

Figure C12g. Susceptibility and

Standardized Residuals

Figure C 12f. CA-SP and

 

 

 

 

 

 

 

 

 

 

w
o o o o v“
o I e I
no In I a 7mm
e I II I Irena .-
e e ' III-.0 ea 7W
I all In .1 II I In
no .889 e o lo '8
lo ell—o .- I
o e e a! 6
e o o. no a
i '4
e
4 l 2
el 0 e.l 2 3
.2633. 859353
R
o .W
a cage 5 .8
on on e I II
o lei- Been-aloe +5
0 0 filled. a
09030 Icon. all use 14
g 08.0 a
In! EU a e .2
1 o
1 0 «_i 2 3

632.com nanuucSm

Susceptibility

Cues to Sun Protection

176

REFERENCE

Agarwal, R., & Prasad, J. (1997). The role of innovation characteristics and perceived
voluntariness in the acceptance of information technologies. Decision Sciences,
28(3), 557-582.

American Academy of Dermatology. (April, 2000). New golf partners join hale Irwin
and the American Academy of Dermatology in Driving Home Sun Protection
Message. Retrieved March 12, 2003, from
http://www.aad.org/PressReleases/golfmessagehtml

American Academy of Dermatology. (April 25, 2001). Greg Norman joins skin cancer
awareness campaign. Retrieved March 12, 2003, from
http://www.aad.org/PressReleases/GNorman.htrnl

American Academy of Dermatology. (2002a). 2003 Skin cancer fact sheet. Retrieved
March 12, 2003, from
http://www.aad.org/SkinCancerNews/WhatIsSkinCancer/SCancerFacts.html

American Academy of Dermatology. (2002b). Malignant2003 Melanoma fact sheet.
Retrieved March 12, 2003, from http://www.aad.org/magmel.html

American Academy of Dermatology. (2002c). Patient information: Skin cancer.
Retrieved March 12, 2003, from http://www.aad.org/pamphlets/skincan.html

American Cancer Society (2002). Cancer facts & figures 2002. Retrieved March 12,

2003, from
http://www.cancer.org/downloads/STT/CancerFacts&Figure32002TM.pdf

Amir, 2., Wright, A., Kemohan, E. E. M., & Hart, G. (2000). Attitudes, beliefs and
behaviour regarding the use of sunbeds among healthcare workers in Bradford.
European Journal of Cancer Care, 9, 76-79.

177

Armstrong, B. K., & Kricker, A. (1993). How much melanoma is caused by sun
exposure? Melanoma Research, 3, 3945-401.

Arthey, S., & Clarke, V. A. (1995). Suntanning and sun protection: A review of the
psychological literature._Social Science and Medicine, 40 (2), 265-274.

Assael, H. (1998). Consumer behavior and marketing action (6th ed.). Cincinnati, OH:
South-Western College Publishing.

Becker, M. H., & Maiman, L. A. (1975). Sociobehavioral determinants of compliance
with health and medical care recommendations. Medical Care, 13 (1), 10-24.

Becker, M. H., Maiman, L. A., Kirscht, J. P., Haefner, D. P., & Drachman, R. H. (1977).
The health belief model and prediction of dietary compliance: A field
experiment. Journal of Health and Social Behavior, 18 (4), 348-366.

Broadstock, M., Borland, R., & Gason, R. (1992). Effects of suntan on judgments of
healthiness and attractiveness by adolescents. Journal of Applied Social
Psychology, 22 (2), 157-172.

Buettner, P. G., & Raasch, B. A. (1998). Incidence rates of skin cancer in Townsville,
Australia. International Journal of Cancer, 78 (5), 587-593.

Buller, M. K., Loescher, L. J ., & Buller, D. B. (1994). Sunshine and skin health: A
curriculum for skin cancer prevention education. Journal of Cancer Education, 9
(3), 155-162.

Bulliard, J. (2000). Site-specific risk of cutaneous malignant melanoma and pattern of
sun exposure in New Zealand. International Journal of Cancer, 85 (5), 627-632.

Capjack, L., Kerr, N., Davis, D., Fedosejevs, R., Hatch, K.L., & Markee, N. L. (1994).
Protection of humans from ultraviolet radiation through the use of textiles: A
review. Family and Consumer Sciences Research Journal, 23 (2), 198-218.

178

Carmel, S., Shani, E., & Rosenberg, L. (1994). The role of age and an expanded health
belief model in predicting skin cancer protective behavior. Health Education
Research, 9(4), 433-447.

Chang, Y., Burns, L. D., & Noel, C. J. (1996). Attitudinal versus normative influence in
the purchase of brand-name casual apparel. Family and Consumer Sciences
Research Journal, 25 (1), p. 76-109.

Cockburn, J. D., Hennrikus, R. S., & Sanson-Fisher, R. (August, 1989). Adolescent use
of sun-protection measures. The Medical Journal of A ustralia, 151, 136-140.

Cody, R., & Lee, C. (1989). Behaviors, beliefs, and intentions in skin cancer prevention.
Journal of Behavioral Medicine, 13 (4), 373-389.

Curiskis, J., & Pailthorpe, M. (1996). Apparel textiles and sun protection. Textiles
Magazine, 25 (4), 13-17.

Davis, J. L. (2000). Sun and active patients: Preventing acute and cumulative skin
damage. The Physician and Sportsmedicine, 28 (7), p. 79-85.

Davis, 8., Capjack, L., Kerr, N., & Fedosejevs, R. (1997). Clothing as protection from
ultraviolet radiation: Which fabric is most effective? International Journal of
Dermatology, 36, 374-379.

Davis, A., Deane, G. H. W., & Diffey, B. L, (1976). Possible dosimeter for ultraviolet
radiation. Nature, 261, 169-170.

Dearing, J. W., Meyer, G., & Kazmiereczak, J. (1994). Portraying the new:
Communication between university innovators and potential users. Science
Communication, 16 (1), 11-42.

DeVellis, R. F. (1991). Scale development: Theory and applications. Newbury Park, CA:
Sage Publications, Inc.

I79

Dickson, M. A., & Littrell, M. A. (1997). Consumers of clothing from alternative trading
organizations: Societal attitudes and purchase evaluative criteria. Clothing and
Textiles Research Journal, 15 (1), 20-33.

Diffey, B. L. (1989). Ultraviolet radiation dosimetry with polysulphone film. In Diffey,
B.L. (Ed), Radiation Measurement in Photobiology (135-159). San Diego:
Academic Press Inc.

Diffey, B. L. (1992). Stratospheric ozone depletion and the risk of non-melanoma skin
cancer in a British population. Physics in Medicine and Biology, 3 7 (12), 2267-
2279.

Diffey, B. L. (2000). Has the sun protection factor had its day? British Medical Journal,
320, 176-177.

Diffey, B. L., Larko, 0., & Swanbeck, G. (1982). UV-B doses received during different
outdoor activities and UV-B treatment of psoriasis. British Journal of
Dermatology, 106, 33-41.

Eckman, M., Damhorst, . L., & Kadolph, S. J. (1990). Toward a model of the in-store
purchase decision process: Consumer use of criteria for evaluating women's
apparel. Clothing and Textiles Research Journal, 8 (2), l3-22.

Elwood, J. M., & Gallagher, R. P. (1998). Body site distribution of cutaneous malignant
melanoma in relationship to patterns of sun exposure. International Journal of
Cancer, 78 (3), 276-280.

Engel, J. F., Blackwell, R. D., & Miniard, P. W. (1990). Consumer Behavior (6th ed).
Orlando, FL: Dryden Press.

Gambichler, T., Altrneyer, P., & Hoffinann, K. (2002). Role of clothes in sun protection.
Recent Results in Cancer Research, 160, 15-25.

Garbe, C., & Buettner, P. G. (2000). Predictors of the use of sunscreen in dermatological
patients in Central Europe. Preventive Medicine, 31, 134-139.

180

Gies, H.P., Roy, C.R., Toomey, S., MacLennan, R., & Watson, M. (1995). Solar UVR
exposures of three groups of outdoor workers on the sunshine coast, Queensland.
Phtochemistry and Photobiology, 62 (6), 1015-1021.

Glanz, K., Lew, R. A., Song, V., & Cook, V. A. (1999). Factors associated with skin
cancer prevention practices in a multiethnic population. Health Education &
Behavior, 26 (3), 344-359.

Goldsmith, R. E., Flynn, L. R., & Moore, M. A. (1996). The self-concept of fashion
leaders. Clothing and Textiles Research Journal, 14 (4), 242-247.

Gravel, T. (1997). Skin cancer. Australian and New Zealand Journal of Public Health, 22
(3),116-117, & 301.

Grubbs, L. M., & Tabano, M. (2000). Use of sunscreen in health care professionals: The
health belief model. Cancer Nursing, 23 (1), 164-167.

Hall, H. I., May, D. S., Lew, R. A., Koh, H. K., & Nadel, M. (1997). Sun protection
behaviors of the US. white population. Preventive Medicine, 26 (4), 401-407.

Hanke, C. W., Zollinger, T. W., O’Brian, J. J ., & Bianco, L. ( 1985). Skin cancer in
professional and amateur female golfers. The Physician and Sportsmedicine, 13
(8), 51-52, 61-63, 66-68.

Herlihy, E., Gies, P. H, Roy, C. R., & Jones, M. (1994). Personal dosimetry of solar UV
radiation for different outdoor activities. Phtochemistry and Photobiology, 60 (3),
288-294.

Hill, D., & Rassaby, J. (1984). Determinants of intentions to take precautions against skin
cancer. Community Health Studies, 8 (1), 33-44.

Holman, C. D. J ., Gibson, 1. M., Stephenson, M., & Armstrong, B. K. (1983). Ultraviolet
irradiation of human body sites in relation to occupation and outdoor activity:
Field studies using personal UVR dosimeters. Clinical and Experimental
Dermatology, 8, 269-277.

181

Huddleston, R, Ford, I., & Bickle, M. C. (1993) Demographic and lifestyle
characteristics as predictors of fashion opinion leadership among mature
consumers. Clothing and Textiles Research Journal, 11 (4), 26-31.

Jackson, K. M. (1997). Psychosocial model and intervention to encourage sun protective
behavior (Doctoral dissertation, Arizona State University, 1997). Dissertation
Abstracts International, 58, 3368.

Jackson, K. M., & Aiken, L. S. (2000). A psychological model of sun protection and
sunbathing in young women: The impact of health beliefs, attitudes, norms, and
self-efficacy for sun protection. Health Psychology, 19(5), 469-478.

J anz, N. K., & Becker, M. H. (1984). The health belief model: A decade later. Health
Education Quarterly, 11 (1), 1-47.

Johnson, E. Y., & Lookingbill, DP. (1984). Sunscreen use and sun exposure. Archives
of Dermatology, 120 (6), 727-731.

Johnson, J. D., Meyer, M., Woodworth, M., Ethington, C., & Stengle, W. (1998).
Information technologies within the cancer information service: Factors related to
innovation adoption. Preventive Medicine, 19, 1-13.

Keesling, B., & Friedman, H. S. (1987). Psychosocial factors in sunbathing and
sunscreen use. Health Psychology, 6 (5), 477-493.

Kurtz, R. (1969). Market research: Strategy and techniques. Braintree, MA: D. H. Mark
Publishing Co.

Labat, K., DeLong, M. R., Gahring, 8., Getting, J ., Amir-Fazli, H., & Lee, M. (1996).
Evaluation of a skin cancer intervention program for youth. Journal of Family
and Consumer Sciences, 88, 3-10.

Larko, 0., & Diffey, B. L. (1983). Natural UV-B radiation received by people with
outdoor, indoor, and mixed occupations and UV-B treatment of psoriasis. Clinical
and Experimental Dermatology, 8, 279-285.

182

Lee, M., & Burns, L. D. (1993). Self-consciousness and clothing purchase criteria of
Korean and United States college women. Clothing and Textiles Research
Journal, 11 (4), 32-40.

Lewis-Beck, M. S. (1980). Applied regression: An introduction. Newbury Park, CA:
Sage Publications, Inc.

Liffri g, J. R. (2001). Phototrauma prevention. Wilderness and Environmental Medicine,
12(3), 195-200.

Littrell, M. A., & Miller, N. J. (2001). Marketing across cultures: Consumers’ perceptions
of product complexity, familiarity, and compatibility. Journal of Global
Marketing, 15 (1), 67-86.

Mahler, H. I. M., F itzpartick, B., Parker, R, & Lapin, A. (1997). The relative effects of a
health-based versus an appearance-based intervention designed to increase
sunscreen use. American Journal of Health Promotion, 1 1 (6), 426-429.

Marlenga, B. (1995). The health beliefs and skin cancer prevention practices of
Wisconsin dairy farmers. Oncology Nursing Forum, 22(4), 681-686.

Mattson, M. (1999). Toward a reconceptualization of communication cues to action in
the health belief model: HIV test counseling. Communication Monographs, 66
(3), 240-265.

Miller, A. G., Ashton, W. A., McHoskey, J. W., & Gimbel, J. (1990). What price
attractiveness? Stereotype and risk factors in suntanning behavior. Journal of
Applied Social Psychology, 20 (15), 1272-1300.

Moore, G. C., & Benbasat, I. (1991). Development of an instrument to measure the
perceptions of adopting an information technology innovation. Information
System Research, 2 (3), 192-217.

Munro, B. H. (2001). Statistical methods for health care research (4th ed.). Philadelphia,
PA: Lippincott Williams & Wilkins.

183

Newman, W. G., Woodruff, S. I., Agro, A. D., & Mayer, J. A. (1996). A survey of
recreational sun exposure of residents of San Diego, California. American Journal
of Preventive Medicine, 12 (3), 186-194.

Norusis, M. J. (1999). SPSS 9.0 guide to data analysis. Upper Saddle River, NJ: Prentice-
Hall Inc.

Osterwalder, U., & Rohwer, H. (2002). Improving UV protection by clothing - recent
development. Recent Results Cancer Research, 160, p. 62-9.

Ostlund, L. E. (1974). Perceived innovation attributes as predictors of innovativeness.
Journal of Consumer Research, 1 (2), 23-29.

Purdue, M. P., Marrett, L. D., Peters, L., & Rivers, J. K. (2001). Predictors of sunburn
among Canadian Adults. Preventive Medicine, 33, 305-312.

Reizner, G. T., Chuang, T., Elpem, D. J ., Stone, J. L., & Farmer, E. R. (1993). Basal cell
carcinoma in Kauai, Hawaii: The highest documented incidence in the United
States. Journal of American Academy of Dermatology, 29, 184-189.

Roach, M. (April, 1997). Give your skin a break. Health, 114-118.

Robinson, J. K., Rigel, D. S., & Amonette, R. A. (1997). Trends in sun exposure
knowledge, attitudes, and behaviors: 1986 to 1996. Journal of American Academy
of Dermatology, 3 7 (2), 179-186.

Rodrigue, J. R. (1996). Promoting healthier behaviors, attitudes, and beliefs toward sun
exposure in parents of young children. Journal of Consulting and Clinical
Psychology, 64 (6), 1431-1436.

Rogers, E. M. (1995). Diffusion of innovations (4th ed.). New York, NY: The Free Press.

Rosenstock, I. M. (1974). Historical origins of the health belief model. Health Education
Monographs, 2 (4), 328-335.

184

Rosenstock, I. M. (1990). The health belief model: Explaining health behavior through
expectancies. In K. Glanz, F. M. Lewis, & B. K. Rimer (Eds), Health behavior
and health education: Theory, research, and practice @p. 39-62). San Francisco,
CA: Jossey-Bass Inc.

Rosenthal, F. S., Lew, R. A., Rouleau, L. J ., & Thomson, M. (1990). Ultraviolet exposure
to children from sunlight: A study using personal dosimetry. Photodermatology,
Photoimmunology, and Photomedicine, 7, 77-81.

Saraiya, M., Hall, H. I., & Uhler, R. J. (2002). Sunburn prevalence among adults in the
United States, 1999. American Journal of Preventive Medicine, 23 (2), 91-97.

Shavelson, R. J. (1995). Statistical reasoning for the behavioral sciences (3rd ed.).
Needlham Heights, MA: A Simon & Schuster Co.

Sheer, B. (1999). Issues in summer safety: A call for sun protection. Pediatric Nursing,
25 (3), 319-320, 323-325.

Shem, L. C. (1986). Protective clothing actions of Michigan corn and apple growers in
regard to pesticides (Master thesis, Michigan State University, 1986). Maters
Abstracts International, 24, 390.

Shim, S., & Kotsiopulos, A. (1994). Technology innovativeness and adopter categories of
apparel/Gift retailers: From the diffusion of innovation perspective. Clothing and
Textiles Research Journal, 12 (2), 46-57.

Spector, P. E. (1992). Summated rating scale construction: An introduction. Newbury
Park, CA: Sage Publications, Inc.

Stepanski, B. M. & Mayer, J. A. (1998). Solar protection behaviors among outdoor
workers. Journal of Occupational and Environmental Medicine, 40 (1), 43-48.

Strutton, H. D., Lumpkin, J. R., & Vitell, S. (1994). An applied investigation of Rogers
and Shoemaker’s perceived innovation attribute typology when marketing to
elderly consumers. Journal of Applied Business Research, 10 (1), 1 18-131.

185

US. Department of Health and Human Services (November, 2000). Healthy People
2010, Volume I: 3. Cancer (2nd ed.). Retrieved March 12, 2003, from
http://www.healthypeople.gov/Document/pdf/Volume1/03Cancer.pdf

U.S. Environment Protection Agency. (2002). What is the UV index? Retrieved March
12, 2003, from http://www.epa.gov/sunwise/uvwhat.html

Vail-Smith, K., & Felts, W. M. (1993). Sunbathing: College students’ knowledge,
attitudes, and perceptions of risks. Journal of American College Health, 42 (1),
21-26.

Vishvakarman, D., Wong, J. C., & Boreham, B. W. (2001). Annual occupational
exposure to ultraviolet radiation in central Queensland. Health Physics, 81 (5),
536-544.

Workman, J. E., & Johnson, K. K. P. (1993). Fashion opinion leadership, fashion
innovativeness, and need for variety. Clothing and Textiles Research Journal, 11
(3), 60-64.

186