Association for Information Systems
AIS Electronic Library (AISeL)
AMCIS 2006 Proceedings
Americas Conference on Information Systems
(AMCIS)
December 2006
Perceived Health Risk Effects on the Adoption of
3G Cell Phones
Mihail Cocosila
McMaster University- Canada
Ofir Turel
McMaster University- Canada
Norm Archer
McMaster University- Canada
Yufei Yuan
McMaster University- Canada
Follow this and additional works at: http://aisel.aisnet.org/amcis2006
Recommended Citation
Cocosila, Mihail; Turel, Ofir; Archer, Norm; and Yuan, Yufei, "Perceived Health Risk Effects on the Adoption of 3G Cell Phones"
(2006). AMCIS 2006 Proceedings. 347.
http://aisel.aisnet.org/amcis2006/347
This material is brought to you by the Americas Conference on Information Systems (AMCIS) at AIS Electronic Library (AISeL). It has been accepted
for inclusion in AMCIS 2006 Proceedings by an authorized administrator of AIS Electronic Library (AISeL). For more information, please contact
elibrary@aisnet.org.
Cocosila et al.
Perceived Health Risks of 3G Cell Phones
Perceived Health Risk Effects
on the Adoption of 3G Cell Phones
Mihail Cocosila
DeGroote School of Business,
McMaster University, Canada
cocosim@mcmaster.ca
Ofir Turel
DeGroote School of Business,
McMaster University, Canada
turelo@mcmaster.ca
Norm Archer
DeGroote School of Business,
McMaster University, Canada
archer@mcmaster.ca
Yufei Yuan
DeGroote School of Business,
McMaster University, Canada
yuanyuf@mcmaster.ca
ABSTRACT
The possible health hazards of cell phones are a controversial issue that is being debated in various literatures. This study
reports on an empirical investigation of the effect of these hazards on the intention to use third generation (3G) cell phones. A
model explicating the relationships between health risk perception and behavioural intention to use cell phones is developed
and tested using structural equation modeling techniques. Furthermore, the moderating roles of several demographic
variables are examined. To obtain variation in health risk perceptions, the study employs an experimental design in which
two groups receive contradictory information from trusted sources on the possible health hazards of cell phones. Overall, this
study integrates the perceived health risk concept into the technology acceptance literature.
Keywords
Technology adoption, mobile commerce, cell phones, human computer interaction, perceived risk, radiation hazard, health
INTRODUCTION
Mobile phones are the booming communication tools of today. Statistics show that there are over 1.3 billion users worldwide
with 140 million American, 320 million European, and 200 million Chinese subscribers (Tomnay, Pitts and Fairley, 2005)
and the numbers continue to grow. Mobile phones have developed into a phenomenon with considerable multilateral social
consequences, thus becoming more than merely a technical innovation or a social craze.
One of the controversies accompanying the unprecedented growth of cell phone use is the possible health hazard arising from
high levels of radiation. It is known that longer exposure to electromagnetic frequencies might lead to a significant absorption
of energy by the human body and to an increase in body temperature (Repacholi, 2001). Due to the obvious importance of
this issue, numerous studies have attempted to answer the question ‘Does the use of cell phones pose a threat to health?’,
without providing a definite and generally accepted answer thus far. Specific medical study findings have resulted in both
‘yes’ and ‘no’ answers to this question (Ozturan, Erdem, Miman, Kalcioglu and Oncel, 2002, Leszczynski, Joenväärä,
Reivinen and Kuokka, 2002) or were contradictory and inconclusive (Hamblin and Wood, 2002, White, 2004). Furthermore,
most of the studies urged caution in extrapolating their results and called for further research.
In this study, we do not attempt to examine the radiation hazard itself, nor do we take a stance regarding it. Rather, we focus
on the perception of the possible health risk associated with cell phone use. This is important, as perceptions regarding the
hazard, and not the actual hazard, affect user behaviors. Studies investigating the influence of perceived risk on intention to
use an information technology (IT) have mostly considered the influence of perceived risk as a whole, or examined some
salient risk facets which did not include a health component (Featherman and Pavlou, 2003). As such, studies targeting
perceived health hazards associated with the use of IT are rare to date. Generally, such studies have considered eye strain,
fatigue due to posture, and mental stress induced by working extensively at desktop or laptop computers (Yamamoto, 1999).
No research has been identified that investigated scientifically what cell phone users and non-users think of the possible risks
Proceedings of the Twelfth Americas Conference on Information Systems, Acapulco, Mexico August 04th-06th 2006
2843
Cocosila et al.
Perceived Health Risks of 3G Cell Phones
associated with the use of these devices, and how these perceptions influence their behaviors. Such an investigation is
certainly necessary for cell phones, as some medical studies mention the possible risk of cancer from cell phone use.
For this study, we have chosen 3G cell phones as the IT artifact. This new class of phones was selected not because they
might produce different levels of radiation from phones belonging to previous generations, but rather because their newness
increases the uncertainty and, possibly, the risk perception associated with their use (Lin, 2003).
Accordingly, the goal of this paper is to examine the influence of perceived health risk associated with cell phone use on
adoption behavior. An empirical investigation involving 206 respondents was conducted using Partial Least Squares (PLS)
techniques. The rest of this paper is structured as follows: the subsequent two sections depict the theoretical background and
model and hypothesis development. Next, research methodology and results are presented. The final section includes
discussion and conclusions.
THEORETICAL BACKGROUND
Studying how and why individuals adopt new information technologies has been a popular area of information systems (IS)
research for some time. This literature has drawn upon several theories and has identified a number of prominent constructs
that affect adoption (see a review in Venkatesh, Morris, Davis and Davis, 2003). Most of these models examine the user
factors that would, more or less, lead users to accept and, eventually, use a technology. However, recent IS research
acknowledges that, besides the favoring factors, there are also unfavorable factors that may be equally important. Some of
these ‘con’ factors have been encompassed under the term of ‘perceived risk’.
In this study the consumer behavior view of risk is taken. Thus, perceived risk is defined as a subjective expectation of loss,
or a disadvantage usually associated with a purchase (Stone and Grønhaug, 1993). As such, perceived risk reflects an
individual perspective that may not necessarily be in accordance with the objective risk involved in the action at hand.
An increasing stream of consumer behavior research acknowledges that, in order to reduce the effects of perceived risk,
studies must recognize and measure the effects of the various facets of risk perceptions (Lim, 2003). The ‘classical’ facets
include: (1) financial (or economic), (2) performance, (3) social, (4) physical (or health), (5) psychological, and (6) time risk
(Laroche, McDougall, Bergeron and Yang, 2004, Lim, 2003). In addition, there is a perceived overall risk that is a trade-off
measure between the various components (Jacoby and Kaplan, 1972). It has been shown that the above six types of risk
explain a significant part of the overall perceived risk and, consequently, of the intention to acquire a product or service (Jih,
Wong and Chang, 2005). This line of research has consistently found that physical risk was the least important component.
However, it is reasonable to notice that little health risk is usually involved when purchasing most consumer products or
services. This may also be a cause for the absence of the perceived health risk factor in IS studies. Thus, although perceived
risk has become an increasingly popular construct in IS research in recent years, it has been dealing mostly with problems
associated with online shopping: e.g., intangibility of the merchandise (Bielen and Semples, 2004) and the perception of
insufficient security of the e-commerce channel (Kim and Lennon, 2000). Risk perceptions, however, may be important with
regard to technologies that might impose life threatening hazards, such as cancer. As 3G cell phones represent an
embodiment of the latest IT which, according to some studies, might also have serious health consequences, one possible
research question is:
How important is the influence of a perceived health hazard on the intention to use third generation (3G) cell phones?
DEVELOPMENT OF RESEARCH MODEL AND HYPOTHESES
Traditional consumer behavior research has considered perceived health risk (HR) as having an indirect effect (through
overall risk) on the intention to acquire a product or service (Brooker, 1984). Although studies have shown constantly that
health hazards perceived by potential buyers have the smallest influence of the six classical risk components, recent
investigations have also revealed that the relative importance of the risk dimensions vary according to the product or service
category (Laroche et al., 2004). Given the controversy over the level of health risks associated with 3G cell phones, it is
reasonable to assume that perceived health risk is a significant factor affecting the adoption of these technologies.
Recent IS studies have found that perceived overall risk plays an important role in adoption-related behaviors and have
integrated this construct mostly into the technology acceptance model (Featherman and Pavlou, 2003). Specifically,
perceived risk was found to be a significant inhibitor of perceived usefulness (PU) and adoption intention. The same effects
of perceived risk on PU and behavioral intention (BI) regarding e-services were observed in the study of Featherman and
Wells (2004). This view is echoed by Chan and Lu (2004) who found that perceived risk is important for potential users that
tend to see the hazard as decreasing the usefulness of an Internet banking service. Therefore, in parallel with these studies, it
is reasonable to assume that perceived health risk, as a component of the overall risk, would similarly inhibit the usefulness
Proceedings of the Twelfth Americas Conference on Information Systems, Acapulco, Mexico August 04th-06th 2006
2844
Cocosila et al.
Perceived Health Risks of 3G Cell Phones
and behavioral intention to use 3G cell phones. Thus, having as support the findings of both consumer behavior and IS
research studies, the following hypotheses are suggested:
H1: The higher the perceived health risk of 3G cell phones the lower the behavioral intention to use this type of cell phones.
H2: The higher the perceived health risk of 3G cell phones the lower the perceived usefulness of this type of cell phones.
It is widely accepted in the technology adoption literature that increased usefulness (i.e., the degree to which individuals
believe that using a technology will help them to improve performance) has a strong positive influence on the behavioral
intention to use that technology (Venkatesh, Morris, Davis and Davis, 2003). Accordingly, the hypothesis below is
suggested:
H3: The higher the perceived usefulness of 3G cell phones the higher the behavioral intention to use this type of cell phone.
Figure 1 encapsulates the hypothesized relationships:
Figure 1. Model of Perceived Health Risk Influence on User Adoption Intention of 3G Cell Phones
METHODOLOGY AND RESULTS
Instrument Development and Data Collection
The hypotheses were tested through a cross-sectional study, using empirical data gathered through an online survey hosted on
a university website and targeting the business and academic community in Canada. Of the 215 responses, 9 were excluded
from the data analysis as being incomplete. Overall, 206 valid responses were obtained. Information on the number of
individuals who might have browsed through the website, noticed the note on the study, but did not complete the survey, is
not available. As such, response rate is not reported.
Of the 206 responses, 107 participants were women (51.94%) and 99 were men (48.06%). Participants indicated their age on
a nine-point scale ranging from ‘below 26’ to ‘over 60’ years old with increments of five years. Over 48% of all users were
under 26 years old. Participants had an average of 6.76 years of cell phone use, with a median of 10 years (range from 0, i.e.,
non-users, to 20), and an average of 7.51 cell phone calls made and received per day, with a median of 30 (range from 0, i.e.,
non-users, to 60).
The questionnaire was divided into two sections: one elicited demographic information such as age, gender, and cell phone
experience, and the other one contained questions pertaining to the constructs in the theoretical model. BI and the PU of 3G
cell phones were captured with multi-item scales adapted from Venkatesh and Davis (2000) that have been widely used in the
technology adoption literature. The HR construct was measured with a three-item scale adapted from the studies of Stone and
collaborators (Stone and Grønhaug, 1993, Stone and Mason, 1995) which were forerunners in validating multiple item
measures of the overall risk and six most popular risk dimensions in the mid 1990s. Survey questions pertaining to the
theoretical model are depicted in Table 1.
Proceedings of the Twelfth Americas Conference on Information Systems, Acapulco, Mexico August 04th-06th 2006
2845
Cocosila et al.
Perceived Health Risks of 3G Cell Phones
Table 1. Survey Items
The introductory page of the online survey was utilized for manipulating the health risk perceptions of individuals, by
presenting contrasting information from a credible source - the British Broadcasting Corporation (BBC) website - on the
possible health hazards associated with the use of 3G cell phones. The purpose of this manipulation was to add more variance
to user pre-existing perceptions, thus facilitating the analysis of the influence of HR. The research used a “random without
replacement procedure” (Van der Heijden, Ogertschnig and Van der Gaast, 2005) commonly utilized in laboratory
experiments: a participant visiting the introductory page of the online survey was presented information from the BBC
website supporting the idea that cell phones are dangerous for health, the next participant entering was presented other
information from the same BBC website supporting the idea that cell phones do not pose health hazards, and so on in an
alternating manner. After completing the survey, all participants were presented a debriefing page explaining that the BBC
website was presenting, in fact, a balanced view and that there was no clear evidence to date for either of the two sides. After
completing the survey participants were also invited to visit other online information sources supporting this unbiased
position.
Analysis of Variance (ANOVA) was used to examine whether the health hazard manipulation had the expected effect.
Indeed, individuals in the ‘safe’ condition reported lower risk perceptions (mean of 2.67) than individuals in the ‘unsafe’
condition (mean of 3.61). This difference was significant at the 0.0001 level (F(1,203)=15.48).
Model Estimations
A two-step approach was employed for model assessment (Anderson and Gerbing, 1988). Accordingly, tests of the structural
model were conducted only after the measurement model was validated. Both the measurement and structural models were
estimated by using the PLS techniques (Chin, 1998). This approach was chosen since it fits small-sample exploratory
research and poses minimal constraints on the distribution of variables (Thomas, Lu and Cedzynski, 2005). The results of
these analyses are reported in the next subsections.
The Measurement Model
Statistics on the model’s constructs and their corresponding measurement items are presented in Table 2.
Proceedings of the Twelfth Americas Conference on Information Systems, Acapulco, Mexico August 04th-06th 2006
2846
Cocosila et al.
Perceived Health Risks of 3G Cell Phones
Table 2. Measurement Model Statistics
The results demonstrate that the loadings of all items exceed the threshold of 0.7, the item-to-total correlations of all
indicators are greater than 0.35, and the residual variance of all items is fairly small, consistent with the literature
recommendations (Nunnally, 1978). Overall, it was concluded that items have reasonably good psychometric properties and
they share a reasonable proportion of their variance with the latent variables they pertain to. As such, all items were retained
in the measurement model.
Furthermore, Table 2 also demonstrates that the model’s latent variables hold reasonably good psychometric properties. First,
all constructs are highly reliable since their Cronbach alphas exceed 0.91. Second, constructs are very consistent and present
good convergent validity. This is indicated by Fornell and Larcker’s (1981) measures of internal consistency and convergent
validity, which are greater than 0.7 and 0.5 respectively, for all constructs.
Two tests were conducted in order to assess the discriminant and convergent validities of the model’s constructs. First, Table
3 outlines the loadings and cross-loadings of the items on constructs. A visual inspection of the table reveals that items load
highly on the construct they should pertain to, but do not load highly on other constructs. Second, a comparison of the
average variance extracted (AVE) from each construct with its communal variances shared with other constructs was
conducted (Fornell and Larcker, 1981). As shown in Table 4, the square root of the AVE for each construct is higher than the
inter-construct correlations. Given the results of the two tests, there is strong confidence in the discriminant and convergent
validity of the constructs.
Proceedings of the Twelfth Americas Conference on Information Systems, Acapulco, Mexico August 04th-06th 2006
2847
Cocosila et al.
Perceived Health Risks of 3G Cell Phones
Table 3. Loadings and Cross-Loadings
Table 4. AVE and Inter-Construct
Correlations
The Measurement Model
Following the positive results of the analyses, the structural model was examined, with a bootstrapping procedure employing
two hundred re-samples for deriving t-statistics for the structural paths. This number of re-samples is recommended for
obtaining reasonable standard error estimates (Chin, 2001). The structural model, path coefficients, and their p-values are
presented in Figure 2.
Figure 2. The Structural Model
Proceedings of the Twelfth Americas Conference on Information Systems, Acapulco, Mexico August 04th-06th 2006
2848
Cocosila et al.
Perceived Health Risks of 3G Cell Phones
The main conclusion drawn from the structural model analysis depicted in Figure 2 is that two out of the three hypotheses
(H2 and H3) are supported at the .05 level or better. It is clear that health risk perceptions affect the perceived usefulness of
3G cell phones, which in turn, influences user behavioral intention to use these phones. However, the analysis shows that
perceived health risk does not exert a direct effect on behavioral intention to use 3G cell phones (i.e., H1 is not supported).
Furthermore, the very low level of PU variance accounted by HR (R2=0.02) implies that perceived health risk has an almost
negligible explanatory value for the perceived usefulness of 3G cell phones.
The pattern of relationships that emerged suggests that the effect of perceived health risk on behavioral intentions may be
mediated through the assessment of perceived usefulness. This proposition was tested using the Baron and Kenny Procedure
(Baron and Kenny, 1986). To ensure comparability of the tested latent variable models (Kenny, 2006), a visual inspection of
the factor loadings of all models was conducted. This inspection reveals that factor loadings in all three models are almost
identical. As such, it is reasonable to assume that changes in path coefficients stem mostly from structural differences and not
from changes in the measurement model. Again, bootstrapping with 200 re-samples was used for assessing significance
levels. Table 5 outlines the path coefficients and their significance in the three PLS models that were examined.
Table 5. Mediation Tests
Table 5 demonstrates that all of the conditions for full mediation, as outlined by Barron and Kenny (1986), are met. The first
two models show that health risk assessments have a reasonable direct effect on both behavioral intentions to use 3G cell
phones (the outcome variable), and the perceived usefulness of these phones (the mediator variable). The third model then
shows that, once the mediator is introduced, the direct effect of HR on BI becomes almost zero and is not significant. The
significance of the indirect effect (HR PU BI) was assessed using the Sobel test (MacKinnon, Lockwood, Hoffman, West
and Sheets, 2002, Sobel, 1982). A test statistic of 2.06 indicates that the fully mediated path is different from zero at the 0.05
level. As such, there is confidence in the full mediation proposition.
Moderation Analyses
As indicated in the literature review section, it is interesting to look more closely at the implications of HR for some
categories of cell phone users. As a first step into this direction, we investigated whether age and gender influence the
magnitude and significance of this study’s research model relationships. Not all of these effects, however, are well grounded
in theory. Thus, this section is only an exploratory investigation of the moderating role of age and gender.
To analyze gender moderation, the sample was split by gender, and two corresponding PLS models were estimated. To test
age moderation the lower (below 26) and upper (above 31) age quartiles were taken, and two subsequent PLS models were
estimated. Note that directly comparing PLS models from different populations is not always sensible, since differences in
the structural models may result from changes in the measurement models (Carte and Russell, 2003). As such, Chow tests
(Chow, 1960) were conducted on the PLS relationships, using the residuals obtained from multiple regressions. These
regressions utilized composite scores for the latent variables, which were based on the PLS weights of the items (i.e.,
weighted average scores). The results are presented in Table 6.
Proceedings of the Twelfth Americas Conference on Information Systems, Acapulco, Mexico August 04th-06th 2006
2849
Cocosila et al.
Perceived Health Risks of 3G Cell Phones
*
P<0.005
** P<0.01
*** P<0.001
Table 6. Moderation Tests
Several observations can be made based on the above analysis. First, age plays an import role in moderating all of the
model’s relationships. Focusing only on significant coefficients, it is concluded that the effect of perceived health risk on
perceived usefulness is stronger for older people. Perceptions of usefulness, in turn, have a stronger effect on behavioral
intentions in this context, for older people. As such, the indirect effect of HR on BI is stronger for older people than for
younger ones.
Second, gender moderates all of the model’s relationships as well. Focusing only on significant coefficients, it is concluded
that the effect of perceived health risk on PU and the effect of PU on BI are stronger for males than for females. Therefore,
the overall indirect effect of health risk perceptions on behavioral intentions is stronger for males than for females. This
implies that females are less concerned with health risks when evaluating the usefulness of 3G cell phones, and in developing
behavioral intentions to use them.
DISCUSSION AND CONCLUSIONS
The purpose of this study was to investigate the effects of the perceived health hazard associated with 3G cell phones on the
intention to adopt them. An empirical investigation included a manipulation that presented each half of the participants with
opposing views about the safety hazards of 3G cell phones.
First, this study found that perceived health risk has a direct non-significant influence on intention to use cell phones. This is
consistent with consumer behavior studies, which found health risk to be the least important risk facet in a purchase decision
(Brooker, 1984), and with mobile phone studies findings. For instance, a survey of 1,340 secondary-school students in
Teesside, in northeast England, showed that only 9.1% of the respondents indicated “fear that using mobile may damage
health” (Madell and Muncer, 2004) as a reason for not using a cell phone.
Second, the study found that perceived health risk of 3G cell phones negatively affect the perceived usefulness of these
devices. This is similar to findings of other research showing that overall risk diminishes the usefulness of IT services
(Featherman and Pavlou, 2003, Featherman and Wells, 2004). An expected relationship, consistent with the overwhelming
majority of IS studies, is that perceived usefulness positively and strongly influences the intention to use 3G cell phones.
Therefore, although health concerns have some negative impact on the intention to use cell phones, perceived usefulness
remains the dominant factor.
Age moderation on perceived risk is a complex issue in the literature. Some studies show that as people grow older, they tend
to perceive less risk in various activities because of “decreased emotional responsiveness”, “increased emotional control”,
and “psychological immunization” (Jorm, 2000). Other studies, dealing especially with the perceived consequences of
smoking, show that young people do not understand the risk because of failing to see its cumulative nature that occurs due a
repeated exposure to the hazard over time (Slovic, 2000). The findings of our study are consistent with the latter view:
younger individuals (under 26) are less concerned with the possible health risk when evaluating the usefulness of 3G cell
phones and in deciding to adopt this technology compared to mature individuals (above 31).
Proceedings of the Twelfth Americas Conference on Information Systems, Acapulco, Mexico August 04th-06th 2006
2850
Cocosila et al.
Perceived Health Risks of 3G Cell Phones
Regarding gender influence, previous research usually suggests that, in general, women perceive more threats in hazards and,
because of that, manifest more insecurity compared to men (Gutteling and Wiegman, 1993, Finucane, Slovic, Mertz, Flynn
and Satterfield, 2000). However this study did not find that health risk perceived by females was significant.
Overall, this study was a first attempt to integrate the concept of perceived health risk into technology adoption research.
Beyond its academic value, the study may provide practical interest for marketing professionals regarding the identification
of user and non-user categories with higher concerns about the possible health hazards of cell phones. Obviously, further
investigations are necessary into the possible influence of other factors intervening in the relationship between perceived
health hazards of using IT, and intention to use it.
REFERENCES
1.
Anderson, J. C. and Gerbing, D. W. (1988) Structural Equation Modeling in practice: A review and recommended twostep approach, Psychological Bulletin, 103, 3, 411-423.
2.
Baron, R. M. and Kenny, D. A. (1986) The Moderator Mediator Variable Distinction in Social Psychological-Research Conceptual, Strategic, and Statistical Considerations, Journal of Personality and Social Psychology, 51, 6, 1173-1182.
3.
Bielen, F. and Semples, C. (2004) Working Paper 106/04, IAG, Universite Catholique de Louvain, Belgium, Available
at http://www.isys.ucl.ac.be/working_papers/documents/WP106Sempels.pdf, Accessed December 5, 2005.
4.
Brooker, G. (1984) An assessment of an expanded measure of perceived risk, Advances in Consumer Research, 11, 439441.
5.
Carte, T. A. and Russell, C. J. (2003) In pursuit of moderators: Nine common errors and their solutions, MIS Quarterly,
27, 3, 479-501.
6.
Chan, S.-C. and Lu, M. (2004) Understanding Internet banking adoption and use behavior: A Hong Kong Perspective,
Journal of Global Information Management, 12, 3, 21-43.
7.
Chin, W. W. (1998) The Partial Least Squares approach for Structural Equation Modeling, In: Modern methods for
business research(Ed, Marcoulides, A.) Lawrence Erlbaum Associates, Mahwa, N.J., pp. 295-336.
8.
Chin, W. W. (2001) PLS-Graph User’s Guide, Version 3.0, Soft Modelling Inc.
9.
Chow, G. C. (1960) Tests of equality between sets of coefficients in two linear regressions, Econometrica, 28, 3, 591605.
10. Featherman, M. and Pavlou, P. (2003) Predicting e-services adoption: A perceived risk facets perspective, International
Journal of Human Computer Studies, 59, 1, 451-474.
11. Featherman, M. S. and Wells, J. D. (2004) "The Intangibility of e-services: Effects on artificiality, perceived risk, and
adoption", Proceedings of the 37th Hawaii International Conference on System Sciences.
12. Finucane, M. L., Slovic, P., Mertz, C. K., Flynn, J. and Satterfield, T. A. (2000) Gender, race, and perceived risk: The
‘white male' effect, Health, Risk & Society, 2, 2, 159 - 172.
13. Fornell, C. and Larcker, D. F. (1981) Evaluating structural equation models with unobservable variables and
measurement error, Journal of Marketing Research, 18, 1, 39-50.
14. Gutteling, J. M. and Wiegman, O. (1993) Gender-specific reactions to environmental hazards in the Netherlands, Sex
Roles, 28, 7-8, 433 - 447.
15. Hamblin, D. L. and Wood, A. W. (2002) Effects of mobile phone emissions on human brain activity and sleep variables,
International Journal of Radiation Biology, 78, 8, 659 - 669.
16. Jacoby, J. and Kaplan, L. B. (1972) "The components of perceived risk".
17. Jih, W.-J., Wong, S.-Y. and Chang, T.-B. (2005) Effects of perceived risks on adoption of internet banking services: An
empirical investigation in Taiwan, International Journal of E-Business Research, 1, 1, 70-88.
18. Jorm, A. F. (2000) Does old age reduce the risk of anxiety and depression? A review of epidemiological studies across
the adult life span, Psychological Medicine, 30, 1, 11-22.
19. Kenny, D. A. (2006), Mediation. Retrieved Feb. 20, 2006, from http://davidakenny.net/cm/mediate.htm
20. Kim, M. and Lennon, S. J. (2000) Television Shopping for Apparel in the United States: Effects of Perceived Amount of
Information on Perceived Risks and Purchase Intentions, Family and Consumer Sciences Research Journal, 28, 3, 301330.
Proceedings of the Twelfth Americas Conference on Information Systems, Acapulco, Mexico August 04th-06th 2006
2851
Cocosila et al.
Perceived Health Risks of 3G Cell Phones
21. Laroche, M., McDougall, G. H. G., Bergeron, J. and Yang, Z. (2004) Exploring How Intangibility Affects Perceived
Risk, Journal of Service Research, 6, 4, 373-389.
22. Leszczynski, D., Joenväärä, S., Reivinen, J. and Kuokka, R. (2002) Non-thermal activation of the hsp27/p38MAPK
stress pathway by mobile phone radiation in human endothelial cells: Molecular mechanism for cancer- and blood-brain
barrier-related effects, Differentiation, 70, 2-3, 120-129.
23. Lim, N. (2003) Consumers’ perceived risk: sources versus consequences, Electronic Commerce Research and
Applications, 2, 216–228.
24. Lin, J. C. (2003) Health aspects of wireless communication: Personal wireless communication radiation and the eye lens,
Mobile Computing and Communications Review, 7, 3, 1-7.
25. MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G. and Sheets, V. (2002) A comparison of methods to
test mediation and other intervening variable effects, Psychological Methods, 7, 1, 83-104.
26. Madell, D. and Muncer, S. (2004) Back from the beach but hanging on the telephone? English adolescents’ attitudes and
experiences of mobile phones and the Internet, CyberPsychology & Behavior, 7, 3, 359-367.
27. Nunnally, J. C. (1978) Psychometric theory, McGraw-Hill, New York.
28. Ozturan, O., Erdem, T., Miman, M. C., Kalcioglu, M. T. and Oncel, S. h. (2002) Effects of the electromagnetic field of
mobile telephones on hearing, Acta Oto-Laryngologica, 122, 3, 289 - 293.
29. Repacholi, M. H. (2001) Health risks from the use of mobile phones, Toxicology Letters, 120, 323–331.
30. Sobel, M. E. (1982) Asymptotic confidence intervals for indirect effects in structural equation models, In: Sociological
Methodology(Ed, Leinhardt, S.) American Sociological Association, Washington DC, pp. 290-312.
31. Stone, R. N. and Grønhaug, K. (1993) Perceived risk: Further considerations for the marketing discipline, European
Journal of Marketing, 27, 3, 39-50.
32. Stone, R. N. and Mason, B. J. (1995) Attitude and risk: Exploring the relationship, Psychology & Marketing, 12, 2, 135153.
33. Thomas, R. D., Lu, I. R. R. and Cedzynski, M. (2005) Partial Least Squares: A critical review and a potential alternative,
Administrative Sciences Association of Canada (ASAC ) Conference, Toronto, Ontario, Canada.
34. Tomnay, J. E., Pitts, M. K. and Fairley, C. K. (2005) New technology and partner notification - why aren’t we using
them?, International Journal of STD & AIDS, 16, 19–22.
35. Van der Heijden, H., Ogertschnig, M. and Van der Gaast, L. (2005) "Effects of context relevance and perceived risk on
user acceptance of mobile information services", Thirteenth European Conference on Information Systems, Regensburg,
Germany.
36. Venkatesh, V. and Davis, F. (2000) A theoretical extension of the technology acceptance model: Four longitudinal field
studies, Management Science, 46, 2, 186-204.
37. Venkatesh, V., Morris, M. G., Davis, G. B. and Davis, F. D. (2003) User acceptance of information technology: Toward
a unified view, MIS Quarterly, 27, 3, 425-478.
38. White, C. (2004) Risk of cancer from mobile phones is unclear, British Medical Journal, 328, 124-124.
39. Yamamoto, S. (1999) Adverse effects of video display terminals on health, Asian Medical Journal, 42, 6, 245-252.
Proceedings of the Twelfth Americas Conference on Information Systems, Acapulco, Mexico August 04th-06th 2006
2852