The Influence of Personal Psychological States and Perceived Device
Characteristics on the Intention to Use Smartwatches
Perceived accuracy, privacy concerns and health consciousness under investigation
Maria-Elisabeth Berger, 12500259
Master’s Thesis
Graduate School of Communication Master’s program Communication Science
Eline Smit, 2020, June 26
Abstract
Recently smartwatches experienced an increase in demand. With functions for tracking and monitoring fitness-related data, they enable consumers to more autonomously take part in managing their health, favorably leading to a general improvement thereof. To test whether individual perceptions such as the accuracy of a smartwatch or privacy concerns stand in relation to usage intentions a cross-sectional online survey was conducted. Additionally, it was tested whether the psychological state of health consciousness accounted for a
moderation effect between privacy concerns and usage intentions. Results indeed revealed a moderation effect, even though not the one anticipated, to be apparent. That is, high health consciousness stood in relation to a stronger negative effect of privacy concerns on usage intentions, as compared to lower usage intentions among those scoring low on the health consciousness scale. However, there was no statistical significance for the main effects of perceived accuracy and privacy concerns on the usage intentions of smartwatches. Overall, this study provided insights into the trending, but still nascent field of smartwatches and leads the way to future studies building up on these findings.
By augmenting human performance, AI has the potential to markedly improve productivity, efficiency, workflow, accuracy and speed, both for doctors and for patients giving more charge and control to consumers through algorithmic support of their data.
– Eric Topol MD, How Artificial Intelligence Could Transform Medicine, 2019 McCarthy (2007) sees the now seemingly ubiquitous term of Artificial Intelligence as “the computational part of the ability to achieve goals in the world.” In recent years technology surrounding AI surged and so did its relevance, as can be inferred by its wide application throughout various fields. In the segment of healthcare, AI technologies aid to reduce medical errors, and improve productivity (Jiang et al., 2017; Johnson et al., 2018). Recently, medical-related AI also found its way into the consumer segment, in the form of so-called consumer wearables, like for example the Apple Watch, and the Samsung, Garmin, or Fitbit
smartwatch. In recent years the adoption of wearable devices increased rapidly from 13% in 2015 to 33% in 2019, according to a survey from Rock Health (2019) on digital health adoption. Moreover, the general smartwatch market is projected to grow at a compound annual growth rate of 14.5% between 2020 and 2025 (Mordor Intelligence, 2019). Through this technology, of which a predominant majority has fitness trackers and health monitoring apps included (Eysenbach et al., 2018), a healthy lifestyle could be communicated with the potential of improving consumers’ quality of life (Mordor Intelligence, 2019), as users are presented with visuals like for example a comparison between their current and their aimed-at activity-level (Helbostad et al., 2017; Trommler, Attig & Franke, 2018). Having the features of fitness trackers and health monitoring represents an important motivation for users to buy smartwatches (Adapa et al., 2018; Jovanov, 2015; Wu, Sum & Nathan-Roberts, 2016). Moreover, adoption was found to be driven by customers believing that through these
features, smartwatches help them with reaching specific goals, such as losing weight, running a marathon or other thematically related goals (Chuah et al., 2016). Highlighting these fitness features’ usability, a qualitative study found activity monitoring to indeed be the second most frequently observed task for which smartwatches are used, after checking for the time (Chun, Dey, Lee & Kim, 2018).
Individuals expect mobile technology to be accurate and work without errors (Spreer & Rauschnabel, 2016). Consequently, devices having functional limitations and being inaccurate do not meet people’s expectations and are therefore considered not fulfilling their purpose, hence useless. Previous research in the context of e-banking already tied a device’s perceived usefulness to its accuracy (Liao & Cheung, 2002). Incorrect and inconsistent measurement of body parameter and activity levels render fitness tracker not useful to their bearer and affect continuous usage negatively (Pal, Funilkul & Vanijja, 2018).
Besides the concept of accuracy influencing device usage, also privacy concerns are regularly brought up for discussion regarding technology adoption. Privacy policies and
disclosure thereof, are relevant to various online behaviors across industries (Boerman et al., 2017; Zarouali et al., 2017). To address rising privacy concerns, privacy laws, requiring companies to be transparent about their data processing practices, were passed (EU General Data Protection Regulation, 2016). However, privacy and legal issues are still prevalent and identified as potential barriers to technology adoption (Datta, Namin & Chatterjee 2018). As they collect, store and might even share data that is often considered being sensitive, such as geographical location, financial information and personal information (e.g. health, habits), this finding is anticipated to also hold true for the segment of smartwatches.
However, while Spreer and Rauschnabel (2016) found technological hurdles to be adoption barriers, their correlation with resistance was not significant. The authors explain this by people possibly “forgiving” certain issues, if other functions are valued higher. Taking up the idea of forgiving some drawbacks for other perceived benefits, within this study this possible scenario is ascribed to also occur among privacy concerns, backing it up with the theory of privacy calculus, suggesting consumers to make mental tradeoffs between perceived benefits and perceived privacy risks, in the process of decision making (Dinev & Hart, 2006). As personal health consciousness is linked to the striving towards a healthy lifestyle, which, as discussed above, might be facilitated by means of smartwatches, I assume this concept eligible to account for such a tradeoff being tolerated. Chau et al. (2019) found people to be more motivated to adopt healthcare technology, if they identify their current behavior to harm their overall health, as they assist in the improvement of their health status. This heightened consciousness towards their personal health might lead to individuals rating the benefits derived out of smartwatch-usage as surpassing perceived privacy risks, hence result in higher usage intentions.
As further growth within the smartwatch segment can be expected, its future
importance and subsequently this study´s societal relevance is strengthened. It is especially important for devices to be precise, as participants stated to have switched to another device
upon encountering faulty measurement with their current device (Chuah et al., 2016). It can therefore be deduced, that people perceiving a certain device to be inaccurate either opt for another product of a competitor or even refrain from buying such technology in the first place. Knowing of inhibitors that keep people from using the technology investigated, may help to further design marketing strategies, dispersing possible doubts. In case of privacy concerns for example, a unary communication can be adopted, assuring people of a certain standards and mechanisms within the company to ensure security over consumer´s data. In order to do so effectively, however, it seems to be paramount to additionally know of possible
moderators, weakening the relationship between inhibitors of adoption and technology usage. Thus, based on the findings considering the relationship between perceived device accuracy, privacy concerns, and usage intentions, communications can be developed specifically at overcoming possible concerns and worries and highlighting desired device characteristics. Moreover, having knowledge of certain personal attitudes moderating possible doubts regarding technology adoption makes marketing towards consumers easier and allows for communication efforts to be more targeted. Subsequently, an increased amount of people might be using WFTs, and by doing so also might be inclined towards improving their general fitness and health. This effect might even counter-act widespread demographic trends such as the rise in obesity or chronic diseases. A healthier society further supports the cutting of costs within the health-sector, by lessening expensive treatments, connected to illnesses such as diabetes or cardiovascular diseases (Levy, 2014).
Although the concepts of technology adoption have been used extensively to measure end-user perceptions, research in the field of wearables is nascent. Up to this point, only Pal, Funilkul and Vanijja (2018) conducted research on the influences of perceived accuracy on smartwatches, finding it to have the maximum impact on continuous usage out of all the other variables investigated.
The novelty of this current work can be judged by the following aspect. Assuming a moderating role of health consciousness on privacy concerns and usage intention is a novel approach, that to my knowledge and based on extensive literature review never was
investigated before. Being informed about possible moderators enables us to better understand the underlying motivations and factors influencing the adoption of smartwatches. Clearly, a research gap exists, which is aimed to be bridge within this paper, by answering the following research questions.
RQ1: Does perceived accuracy of smartwatches stand in relation to the usage intentions thereof?
RQ 2: Do privacy concerns stand in relation to usage intentions of smartwatches? Is this relationship moderated by people´s health consciousness?
Conceptual Model Figure 1 Conceptual Model Perceived accuracy of smartwatch Usage intentions Health Consciousness Privacy concerns
Theoretical framework
Perceived accuracy
Consumers generally expect the technology they are using to be accurate (Spreer & Rauschnabel, 2016). However, insights gained from semi-structured interviews within the settings of an ethnographic study, revealed several smartwatch-users to have encountered instances, in which their device was obviously mal-functioning by providing them with inaccurate data (Pal, Funilkul & Vanijja, 2018). The authors argued that such technical flaws made users unsatisfied prompting them to discontinue their usage. In other words, whether the technology indeed measures what it is supposed to measure, as well as the perceived accuracy of the respective data provided, renders the device (not) valuable to people, resulting in its (dis)continued usage. This estimation is confirmed by research evidence for perceived accuracy of smartwatches to be a strong predictor for continuous usage thereof (Pal, Funilkul & Vanijja, 2018).
The Technology Acceptance Model (TAM) designed by Davis (1989) found vast application in the field of the adoption of novel technology. It has been shown to be of immense robustness throughout its application across a broad range of technologies
(Yousafzai, Foxall, & Pallister, 2007). The model suggests the concepts of perceived ease of use and perceived usefulness as the two main drivers for the adoption of the technology under consideration. As the ease of use heavily relies on prior user-experience (Brown, 2002; Taylor & Todd, 1995), this first of the two concepts is supposed to be of less importance, as our society becomes increasingly tech savvy (Eshet-Alkalai & Chajut, 2009) and thus more at ease with using and applying new technology. Moreover, also McQuail (2010) argued that in the 1960s, along with the terms’ origin, our society entered the stage of the “information society” era, wherein information is produced and distributed mainly by means of computer-based technology, implying our growing acquaintance with and knowledge of modern technology.
The second concept within the TAM, perceived usefulness, is defined as “the degree to which a person believes using a particular system could enhance his or her job
performance” (Davis, 1989, p. 320) Regarding the context of this study an adapted and in prior research (Pal, Funilkul & Vanijja, 2018) used version of this definition was leaned on. It states perceived usefulness of smartwatches to be a belief about the extent that they are able to increase a consumers’ personal efficiency in a way that they are more productive and
organized (Chuah et al., 2016). Enhancing people’s fitness performance is one of the desired outcomes to the use of smartwatches and their fitness tracking functions. This betterment of an individuals’ status quo regarding health and fitness is reached by smartwatches tracking their data, analyzing it, finding instances for improvement and finally report these back to the consumers. Analyzing the data output after a workout for example might give users the opportunity to spot room for enhancing their efforts, when seeing post-hoc, that their heart rate during a certain exercise to have been below their set goal. Therefore, as with these functions smartwatches describe a means to facilitate reaching an individual’s fitness and health related goals, accuracy of their measurement seems to be of paramount importance. In other words, to guarantee improvement, transparent and true data from which deductions about specific betterment can be made, are needed.
Within the settings of a qualitative interview study, sport enthusiasts for example were found to have purchased fitness trackers due to their specific functionalities like the step and calorie counter, or the heart rate monitor. Further, the users relied on them being accurate and abandoned them, when they found them not to work properly or display inaccurate data (Canhoto & Arp, 2017). Concluding from these findings, the concept of perceived accuracy seems to be of immense importance, when considering the adoption of wearable devices such as smartwatches. Furthermore, the measurement of such sensible data like the heart rate being precise is especially important, as a faulty monitoring and subsequent communication thereof, might deceive the user. Thus, being notified about an allegedly irregular heartbeat-rate, a
service the Apple Watch for example offers, might lead to unnecessary stress scenarios, and result in avoidable and long doctor´s visits, in case of a faulty measurement.
Prior qualitative research into consumer’s experiences with WFTs, revealed concerns about their smartwatches, feeling that they were tracking inaccurate data and not working smoothly. Several examples were mentioned by various participants, such as one respondent stating to have received a notification of having reached 10,000 steps, while driving a car and another one complaining that the device never recognized her treadmill movements (Pal, Funilkul & Vanijja, 2018). Indeed, several studies found smartwatches to be less accurate with increasing intensity of physical activity (Bai et al., 2018; Khushhal et al., 2018). Bai et al. (2018), for example showed Fitbits Charge HRs mean absolute percentage error to be higher in light activity (10.1%) as compared to sedentary behaviors (7.2%). Similarly, Khushhal et al. (2017) showed decreasing validity of the heart rate sensor among Apple Watches with increasing intensity of exercise, recording lower values than the actual heart rate. Displaying an incorrect heartbeat-rate, well below the actual one, when working out, might lead to overestimation of one self’s capabilities and subsequent over-performance. A heartrate being too high during physical activity bears the risk of a health threat, if it occurs continuously. Consequences for exceeded heart rates reach from poor recovery after
exercising to more severe implications, such as increased cardiac risks (Atwal, Porter & Macdonald, 2002). Thus, as most consumers naturally seek to avoid such stressful situations and possible health hazards, people´s intention towards using smartwatches are hypothesized to stand in relationship to the perceived accuracy thereof. Moreover, according to prospect theory (Tversky & Kahnemann, 1981) people tend to avoid uncertain scenarios, thus favoring perceived exact and trustworthy data over perceived uncertain and non-reliable data.
Following this line of argumentation, the first hypothesis is worded as follows:
H1: There is a positive relationship between perceived accuracy of smartwatch-measurement
Privacy concerns
The global number of internet users amounted to 4.13 billion in 2019, describing an increase of about 0.21 billion users compared to the previous year (Clement, 2020). With the rise of new media and the advent of the seeming pervasiveness of the internet, people’s concerns about their personal data being collected also grew steadily and still are growing. As of February 2019, 39% of internet users in Europe stated to be much or somewhat more concerned regarding their online privacy as compared to one year ago. Globally, 53% of internet users agreed with this statement (Clement, 2019). Considering that wearables collect sensible data about people’s health, it seems especially likely for privacy concerns to arise and play an important role when it comes to these devices’ usage. An in-depth interview revealed identity theft to be the most striking privacy concern among smartwatch users, followed by the leaking of financial and health information (Udoh & Alkharashi, 2016). Moreover, while the continuous collection, transmission and storage of personal data posits without doubt some benefits to the end-users, by for example enhancing motivation through regular reminders to work out or drawing comparisons to the achievements of connected friends, it also highlights novel challenges considering their privacy. Indeed, Lidynia, Schomakers and Ziefle (2019) found privacy concerns to be one of the major barriers to the use of wearables. In accordance with this, Pal, Funilkul & Vanijja (2018) demonstrated high privacy concerns to result in lower usage intentions of smartwatches. On the other hand though, Udoh and Alkharashi (2016) conducted interviews among active smartwatch users as well as those being merely familiar with this technology, gaining contradicting insights. Upon being asked about the awareness of privacy risks, one participant stated, that he was not aware, but would nevertheless refrain from using his watch any differently, as he already thinks about this whenever he uses his computer or phone anyways. In general, even though most of their participants claimed to be concerned about privacy and acknowledged the risks, an equal amount maintained an “I don´t care” attitude at the end of the interviews and stated that they
had nothing to hide. This study, however, was only small-scale and included participants in a younger age-group only. Therefore this current study takes up on the findings of the first research mentioned and hypothesizes that people high in privacy concerns to also have lower intentions on using smartwatches. This is even more so relevant, as the sample for this research will constitute of respondents across all age groups, other than in the study by Udoh and Alkharashi (2016), in which only students between the age of 19 and 26 were engaged. As individuals were found to be more anxious about the disclosure of their private data with rising age (Goldfarb & Tucker, 2012), thus less willing to use devices that infringe their privacy, this reasoning seems plausible. Based on this argumentation the second hypothesis can be deduced:
H2: High privacy concerns among participants stand in relationship to lower usage intentions
of smartwatches. Health consciousness
The concept of health consciousness was defined in various ways throughout literature and years. Kraft and Goodell (1993) for example, expressed health consciousness as the extent to which individuals are concerned with nutrient-intake, physical activities and a healthy living environment, describing it as a “well-being” lifestyle, thus taking a behavioral perspective to the term. Many other prior definitions also relied on the measurement of tangible behaviors directed towards health. Hong (2009), however, re-conceptualized the term from a
psychological perspective and incorporated personal aspects determining the extent to which people are conscious about their status of health. Consequently, health consciousness is seen as a thorough mental orientation toward health, describing a psychological state. It loads on the three distinctive dimensions of self-health-awareness, personal responsibility for one’s individual health as well as health motivation (Hong, 2009). In this study Hong’s (2009) definition was used, as the research clearly suggests the concept of health consciousness to be more powerful in the prediction of other various health behaviors, when it is measured as a
holistic psychological state towards health, as opposed to when only actual and specific behaviors are observed and measured. Indeed, the author found health consciousness to have implications not only on the processing of health-related information but also the subsequent intention to act in a healthy manner (Hong 2009). Similarly, Claussen et al. (2015) found individuals following a health-conscious lifestyle to be more likely to engage in preventive health behaviors, such as engaging in physical activity or eating healthy foods, as compared to those not being health conscious. Based on these findings, people having a high sense of health consciousness seemingly attribute more value to their health as opposed to those
scoring low on the scale. Smartwatches, with their underlying purpose of providing users with health-related information, as well as motivation to engage in healthy activities, describe a means to facilitate these overarching goals of health-conscious people. In addition, fitness apps and trackers on smartwatches can further even help to increase personal value related to one’s health (Adapa et al., 2018), meaning that individuals might be likely to gain even higher health consciousness, once they interact with smartwatches, as the topic of health and fitness becomes more salient. Within this context, health consciousness is assumed to be a possible subject to change if an intervention is undertaken, as the re-conceptualization as defined by Hong (2009) portrays a psychological state, which is described by prior research to vary with alterations in resources or environment changes (Murrell & Norris, 1983). Individuals who already have a high sense of health consciousness may find communicated health messages of smartwatches to be even more personally relevant, as they are more likely to memorize health contents and incorporate them into their future behavior (Dutta-Bergman, 2006).
Thus, as highly health conscious people should, based on the argumentation above, attach great value to their personal health and facilitators thereof (i.e. smartwatches), this personal orientation towards health is anticipated to weaken the negative relationship between privacy concerns and intentions to use the facilitating device. Findings of people claiming to be concerned about their privacy, yet still trading it for small incentives, such as the attention
of peers when posting online, was coined “privacy paradox” in literature (Barnes, 2006). Attempting to find possible explanations for this phenomenon of privacy concerns being overshadowed by small incentives, this study draws upon findings of privacy calculus theory (Dinev, Hart, 2006), which argues that in the context of internet transactions people weigh between the expected benefits and the expected losses of disclosing personal information, thus performing a mental calculus. Laufer and Wolfe (1977) claimed information disclosure to be acceptable, as long as it assures certain benefits to the individual consumer, and as long as only a moderate level thereof is apparent. Therefore, individuals are similarly supposed to perform a calculus between the expected loss of privacy from adopting smartwatches, and the potential worthy outcomes of information disclosure.
Dinev and Hart (2008) additionally claimed high behavioral intention to be preceded by higher allurement beliefs than general privacy concerns. In the context of this research, this means that the more tremendously people value the health-aspects of smartwatches, hence scoring higher on health consciousness, the more they are likely to trade personal information against the benefits derived out of wearables usage, as the beliefs about the positive outcomes outweigh the privacy risk beliefs. Following this stream of argumentation, high health
consciousness is assumed to weaken the relationship between privacy concerns stemming from smartwatch usage and usage intentions, as the perceived potential gains of easier reachable health goals by means of the adoption of smartwatches surpasses these expected losses. Consequently, the following hypothesis emerges:
H3: High health consciousness among participants will weaken the negative relationship
between privacy concerns and usage intentions of smartwatches. Methods
The conceptual model (see Figure 1) was tested empirically using data that was collected by means of a survey that included items for the respective constructs specified in the model.
Sample
Ethical approval for the study was obtained from the review board of the Amsterdam School of Communication Research (ASCoR). As wearables like the Apple Watch or Fitbit are designed to appeal to a non-specific customer segment, thus are designed to be used by virtually everyone, the sample recruited for the purpose of this study had the only demand of men and women being above the age of 18 and below the age of 80. The age restriction of 80 derives from an insight of a report on the demographics among wearable users. According to it users over the age of 65 amounted for only 4.6% of users (Statista, 2016). Moreover, most of the people above the age of 80 were anticipated to not use the devices’ fitness tracking functions, as with growing age, physical activity decreases (Milanović et al., 2013). Adding to my point is the claim that elderly users (65-75 years) not only encounter problems with the navigation through the menu of a smartwatch but also have troubles reading off the screen due to a perceived poor contrast and the small screen size of smartwatches (Zotz et al., 2018). These circumstances very likely factor into the minuscule adoption rate among this respective population. As the outcome variable will be measured in behavioral intention, it was not necessary for each respondent to own a smartwatch, although it was asked for in the
questionnaire. In total 221 people from 21 different nations took part in the survey, of which 35 participants did not finish it, resulting in a total sample size of 186 individuals. 61.3% (n = 114) of respondents were female, and ages ranged from 18 to 72 with an average age of 32 (M = 32,31, SD = 13.79). With 54.8%, the majority of respondents (n = 102) was from Austria, followed by the Netherlands (18.3%) and Italy (2.7%). The level of education was rather balanced, with 36.6% (n = 186) indicating to own a bachelor’s degree, 22.0% (n = 68) having a master’s degree and 22.0% (n= 41) of respondents being High School graduates or their country’s respective equivalent.
Among the respondents only 24.2% (n = 45) indicated to currently use a smartwatch, out of which 73.3% (n = 33) used it for tracking fitness and health related data. 48,9% (n =
22) reported to have encountered some functional limitations to their device, such as the smartwatch counting steps, while they were driving the car et cetera.
Design and Procedure
To test the hypotheses, a cross-sectional survey was conducted. It was carried out during a time frame of one month, between the 9th of April and the 12th of May 2020. The link to the
questionnaire, designed with the online tool Qualtrics, was accessible through means of laptops and computers, as well as – given their global penetration rate of 41.5% (O’Dea, 2020) – smartphones, in order to reach a large, dispersed and diverse audience. The language of the survey was English. Easy non-scientific terminology was used in an attempt to ensure that the questionnaire was understandable to laymen. This was checked for by having a non-native English and middle-aged person with no scientific background read the questions and report back. To recruit participants, the questionnaire was shared via personal channels, such as e-mail and WhatsApp, as well as on social media platforms. Specifically, having about 900 followers on Instagram and over 400 friends on Facebook, the first author’s private accounts were used to share the survey by means of a link, as the accounts reached on these networks is considerable. Each respondent was asked to fill out the survey by themselves and
individually, thus providing self-reported data.
First of all, people were informed about the study’s general topic, the author conducting the research and the institution standing behind the project, as well as contact details in case of queries or complaints. After obtaining informed consent of the participants, the online survey started with general demographic questions such as the respondent’s age or country of residence. The questionnaire proceeded with a short introduction to smartwatches and the first set of questions concerning perceived accuracy thereof. The next block was displaying questions about privacy concerns, followed by a statement battery about people´s health consciousness. The survey concluded with a 3-item set of questions about participant´s
intention to use a smartwatch. Finally, respondents were thanked for their cooperation and led to the end of the survey.
Measurements
In order to promote content validity, already validated measures from past studies were used and slightly adapted and reworded so as to specifically relate to the context of this study.
All items were measured using a seven-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). Factor validity was measured by means of a principal axis factor analysis with oblique rotation, conducted for each scale with the IBM program SPSS Statistics 25. The Cronbach’s alpha value for each construct was calculated with a reliability analysis, after which one new variable for each latent construct was computed.
Perceived accuracy
As the scale for perceived accuracy and functional limitations (= PAFL) used in Pal, Funilkul and Vaniija’s (2018) study heavily correlates with the background of the current study, a slightly adapted version (see Apendix A) thereof was applied. Only three of its originally four items were used, as the fourth (“I prefer my smartwatch to be shock, dust, and water proof” ) was due to its focus on functional limitations of the hardware, not relevant to this specific context, hence eliminating the part about functional limitations of the scale. Items for measurement included for example “I doubt whether the fitness data collected by a smartwatch is accurate.”
The results of the component correlation matrix confirmed the assumptions of a correlation between the factors. One factor was found to have an eigenvalue above 1 explaining a variance of 63.7%. This suggests that the scale items are unidimensional. A subsequent reliability analysis revealed a Cronbach’s Alpha of .71, representing an acceptable internal reliability (Ursachi et al., 2015). None of the items needed to be deleted in order to improve the scale. This allowed to compute the new predictor variable “Perceived Accuracy”. After employing
descriptive analysis, it was apparent that perceived accuracy among respondents (n = 186) was on average quite balanced (M = 3.78, SD = 1.21).
Privacy concerns
To measure privacy concerns a three-item scale adapted from Pal, Funilkul and Vanijja (2018) (see Appendix A), was applied. The scale included items such as “Smartwatches can use my personal information without my knowledge violating my privacy.”.
Factor analysis revealed the three items to load on one factor, explaining 61.4% of the variance. The scale’s internal reliability was reasonable with a Cronbach’s Alpha value of .68. Out of the three items used, the second independent variable “Privacy Concerns” was
computed showing to be moderately high (M = 4.44, SD = 1.22) among participants (n = 186). Health consciousness
As Hong’s (2009) definition and re-conceptualization of health consciousness already have been adopted throughout this work, so was her 11 items scale (see Appendix A), constructed to measure the psychological state of health consciousness. In her study she found health consciousness to load on three factors, therefore having three major dimensions, which are: self-health awareness, personal responsibility and health motivation. Items applied within this study are for example statements like “I´m generally attentive to my inner feelings about my health.”, “I take responsibility for the state of my health.” or “Living life without disease and illness is very important to me.”.
Out of the 11 items measuring health consciousness, one item was reverse coded, as it was the only one question in the scale being worded positively. Cronbach’s Alpha showed very good internal reliability with a value of .81 for all items together. A new moderator-variable, named “Health Consciousness” was computed. Descriptive analysis showed participants to have a rather high consciousness regarding their health (M = 5.35, SD = .72).
The outcome variable of intention to use a smartwatch was measured by directly asking participants to self-report their intention to use a smartwatch, following Ajzen’s (1991), Theory of Planned Behavior. Additionally, one item from Davis (1989) was applied, as well as one item from Kuo and Yen (2009) resulting in a total of three items measuring behavioral intention. The statements were worded in a manner for people to assume they owned a smartwatch in case they did not already have one.
The three items measuring usage intention loaded on one factor, accounting for 84.6% of the variance. The scale was found to have very good internal reliability with a Cronbach’s Alpha of .91. Out of these three items the new outcome variable “Usage Intention” was computed. The average usage intention within the sample was moderately high (M = 4.89, SD = 1.49).
Results
For the analysis of the direct relationship between the two predictor variables of perceived accuracy and privacy concerns and the outcome variable of behavioral intention to use a smartwatch, a Multiple Regression Analysis was conducted, using the statistical analytics software SPSS 25.
First, the prerequisite of normality was checked by looking at the regression
histogram, which, however, was ambiguous, thus giving no definite conclusion as to whether the outcome variable was normally distributed. Subsequently a Q-Q plot was created,
showing normality of the residuals, as the points were clustered around the diagonal line (see Appendix B). Afterwards linearity of the data was checked by analyzing the scatterplot of the outcome value against the residuals and applying a Loess Curve. Since the residuals seemed to approximately evenly scatter around zero (see Appendix B), it was concluded for the data to be linear.
Altogether, these plots did not show clear violations of the assumptions of a regression analysis. Therefore, the hypotheses were tested statistically thereafter. The multiple linear
regression model including the two independent variables perceived accuracy of smartwatches and privacy concerns and the outcome variable usage intention of
smartwatches, proved to be statistically significant, F (2,183) = 3.19, p = .044. This indicates that the regression model was a good fit for the data. However, the model only explained 3% (R2 = .03) of the variance in the dependent variable usage intention, which is rather low. The predictor perceived accuracy was found to make no significant contribution to change in the dependent variable, t = -1.71, p = .089, 95% CI [-0.33, 0.02]. Additionally, also the
independent variable privacy concerns showed to have no significant effect, t = -1.58, p = .115, 95% CI [-0.32, 0.04].
To check for the moderation effect of health consciousness on the relationship between privacy concerns and usage intentions, a new interaction term was created by multiplying the variable privacy concerns with health consciousness. A multiple linear
regression was run with the interaction term to obtain the amount of variance accounted for by the predictors with the interaction. The model showed to be of high statistical significance, F (2,181) = 6.96, p = .001. This means that there is evidence for an interaction effect between perceived accuracy and health consciousness. Furthermore, results showed that the model indeed did account for significantly more variation than did the model with just the two independent variables by themselves (ΔR2= .07, p = .001). The model accounts for 10% of the change in variance of usage intentions (R2 = .10, p = .001). Additionally, the interaction coefficient is statistically significant, b* = -1.36, t = -2,65. p = .009, 95% CI [-0.49, -0.07].
In a further step, moderation analysis was conducted by using PROCESS by Andrew Hayes in SPSS, in order to run subgroup analyses and visualize the regression lines for different values of the moderator. The continuous variable health consciousness was automatically divided by PROCESS into three categories: firstly, low health consciousness with a value of 4.63 was one Standard Deviation below the mean of health consciousness. Secondly, the average value of health consciousness among participants (n = 186) was 5.35
and lastly the group of high health consciousness - one Standard Deviation above the mean - was denoted with a value of 6.08.
The Johnson-Neyman's region of significance indicates that at a value of 5.18 of the moderator health consciousness its interaction effect on the link between privacy concerns and usage intentions becomes significant. Consequently, the effect of privacy concerns on usage intentions was statistically not significant for those low in health consciousness (one SD below the mean), p = .847. At the mean level of health consciousness, one unit increase in privacy concerns accounted for a statistically significant decrease of 0.22 in usage intentions, b = -0.22, t = -2.47, p = .015. Lastly, among those individuals scoring one standard deviation above the mean on health consciousness, there was a 0.41 decrease in smartwatch usage intentions for every unit increase in privacy concerns, b = -0.41, t = -3.26, p = .001. (Appendix C).
Discussion
The results of the multiple linear regression analysis in order to test H1 and H2 revealed that none of the two predictors perceived accuracy of the wearable device and privacy concerns made a contribution to change in the outcome variable usage intentions. Consequently, neither H1, suggesting a positive relationship between perceived accuracy and usage intentions, nor H2, anticipating a negative relationship between privacy concerns and usage intentions, was confirmed.
Regarding H1, an explanation for the relationship of perceived accuracy on usage intentions not being significant, might be found in the rather low number of respondents actually owning a smartwatch and hence having the possibility to actually detect possible inaccuracies. Those participants not owning a smartwatch were left only with guessing and estimations about the device’s accuracy. Therefore, people never having interacted with a smartwatch might have been uncertain as to what to exactly believe about the technology’s functioning, leading to inconclusive findings.
Coming to the second established hypothesis of this paper it needs to be highlighted, that even though the assumed effect in H2 was statistically not significant, the negative effect of privacy concerns on usage intentions pointed in the anticipated direction. An explanation for this effect not being significant might be found in the notion that currently privacy concerns seem to be pervasive and not restricted to one device only, which is especially apparent through the Internet of Things (IoT) (Yang et al., 2017; Xiang, Jianlin & Jin, 2012). The IoT describes a collection of devices, sensors and things interconnected by means of the internet, serving the underlying purpose of communicating directly with each other and exchanging data (Alaba et al., 2017; Atzori, Iera & Morabito, 2010). Whether people check their social media profile, surf an online shop, look up things online or take their smartphone with them everywhere they go, they are facing the risk of privacy loss (Felt et al., 2011; Malhotra, Kim, Agarwal, 2004). Another insight from a qualitative study gained, showed that people would not use their smartwatch any different, as they state being concerned about privacy issues whenever using a computer or a phone anyways (Udoh & Alkharashi, 2016). This leads to the assumption that privacy might be a pervasive issue across modern
technology and is not to be restricted to one device only, rendering device-specific concerns less strong. Therefore, even though participants might have had concerns, they could have still planned on using a smartwatch, as the threat of a smartwatch leaking private data might have been perceived as being only the proverbial drop in the ocean. Future research should take the pervasiveness of privacy threats as a moderating factor into consideration. In order to evaluate whether these assumptions indeed hold to be true, an ethnographic study in the form of interviews about perceived pervasiveness of privacy threats could be conducted in a first step. Building up on the respective insights gained, an experiment with three groups seems to be fitting. Within the two treatment groups one could be primed with a stimulus
communicating pervasiveness of privacy threats, whereas the other would be presented with a stimulus diminishing and downplaying pervasiveness (e.g. maybe by showing fake studies to
the respondents in which only mobile phones are found to be the main culprits of privacy loss). The third group, the control group, should be presented with no stimulus at all.
Addressing H3, the suggested moderation effect of health consciousness on the relationship between privacy concerns and usage intentions indeed proved to be significant, nevertheless pointed to the wrong direction. H3 predicted the moderation effect to weaken the negative relationship between privacy concerns and usage intentions. Results, however, showed that at high levels of health consciousness the effect became stronger. In other words: people high in health consciousness showed a steeper decline of usage intentions with
increasing privacy concerns than did those scoring low on the scale. The effect hence was strengthened by higher levels of the moderator, instead of weakened. H3, therefore, only could be confirmed partly.
Visualization showed that at the highest levels of health consciousness, the effect indeed was most strong and negative (see Appendix B). When privacy concerns among this group became higher, usage intentions of smartwatches decreased significantly. In contrast to this, however, privacy concerns seemed to have no significant influence of usage intentions among participants with low levels of health consciousness, confirming the statistical results of the Johnson-Neyman's test.
Limitations & Future Research
The Health Consciousness Scale was adopted from Hong (2009). However, upon finding three subscales which the 11 items loaded on, the author did not conduct reliability analysis for each of these dimensions, but instead reported Cronbach’s Alpha for the total scale. Within this analysis Hong’s example was being followed. Even though the items also were found to load on three factors and the overall Cronbach’s alpha resembled Hong’s, the results were ambiguous and showed different items to load on different factors than within the original work. Moreover, the eigenvalue for factor 3 was only marginally above 1 (1.006) with only one item loading on it and the scree plot additionally suggesting retaining only two
factors. Scale development, however, was not the main concern of this study, which is why future research is suggested to analyze and refine the scale, focusing on item development to support the factor loadings and the overall reliability of each factor, such as to further validate the scale and render it more applicable. Moreover, focusing on the subscales adds more information into the consumer’s health consciousness, as the three factors found (i.e. self-health-awareness, personal responsibility for one’s health and health motivation) seem to be distinct enough to gain deeper insights if they are regarded separately. This scale could be developed through literature review, identifying possible items in a first step and a subsequent reduction thereof through a factor analysis.
Another interesting starting point for future research concerns differences in
perception of various wearable producers. The current research only focused on the generic of a smartwatch and therefore was not occupied with one specific brand. However, especially with big tech giants like Apple having entered the smartwatch market, it could be plausible for privacy concerns to vary by smartwatch brand, as indicated by a small-scale in-depth interview study of Udoh and Alkharashi (2016). Nevertheless, in order to gain a clearer picture of this assumption, a larger sample size should be used.
My moderator of health consciousness focused on participant’s holistic mental state towards their health, whereas smartwatches tend to be used mainly as a means to track sport results (Chun, Dey, Lee & Kim, 2018). Future research could try and find a moderator specifically leaning on the fitness aspect of a person’s health behavior. As there was no respective validated scale to be found for this, a scale construction for a term the author calls “sports-enthusiasm”, focusing not only on frequency or intensity of but also dedication to and passion about actively pursuing sports, is suggested. Applying a moderator that rather focuses on the physical activity aspect of one’s health might relate more closely to the topic under investigation (i.e. smartwatch usage), hence leading to more concise and meaningful
one´s general health including various external factors, that cannot be counteracted by mere smartwatch usage, such as disadvantageous genetics or chronic and mental illness. Thus, by applying a more context-specific approach with sports enthusiasm as a moderator, stronger interaction effects could be scored, as people high in sports enthusiasm might be even more readily exchanging benefits derived out of smartwatch usage for their private data, than those scoring high in health consciousness. This in turn might lead to a more meaningful
moderation effect that might even point into a different direction, than the one found in this study. Moreover, by differentiating health and fitness, but testing both scales on participants, a clearer picture of how health consciousness and fitness enthusiasm interact can be drawn, leading to easier segmentation of the target population for health interventions and marketing efforts, from which future researchers can draw.
Conclusion & practical implications
Having functionalities like tracking fitness-related and monitoring health-related data,
smartwatches offer lay people the opportunity to take control over their own health by means of easily interpretable and visualized information (Lupton, 2014), rendering these devices a valuable asset with regard to a possible health improvement of society at large. Research into this uprising field of wearable fitness trackers, providing new insights and shedding light onto new aspects, is therefore a viable part of its development, acceptance and adoption within society.
Overall, this study contributed to gaining deeper insights into the rocketing
smartwatch market and specifically highlighted a moderation of health consciousness to be apparent among privacy concerns and usage intentions. These findings indicate that by simultaneously targeting health conscious people, as well as trying to dissolve privacy concerns, marketeers might be able to generate high usage intentions for their products
through their marketing strategies. Moreover, based on the results and insights from this study deals with physicians could be negotiated for them to actively enhance their patient´s health
consciousness while simultaneously promoting WFT’s as they see fit to get individuals to more consciously control and take part in their fitness behavior. This will likely strengthen the image of smartwatches, increase their sales and the industry overall. Thus, the overarching goal of leading society towards a healthier and more conscious lifestyle could be facilitated with this approach.
References
Adapa, A., Nah, F., Hall, R., Siau, K., & Smith, S. (2018). Factors Influencing the Adoption of Smart Wearable Devices. International Journal of Human-Computer Interaction, 34(5), 399–409. https://doi.org/10.1080/10447318.2017.1357902
Ajzen, I. (1987) ‘Attitudes, traits, and actions: Dispositional predictions of behavior in personality and social psychology’, Advances in Experimental Social Psychology 20, 1-63. https://doi.org/10.1016/S0065-2601(08)60411-6
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T Alaba, F., Othman, M., Hashem, I., & Alotaibi, F. (2017). Internet of Things security: A
survey. Journal of Network and Computer Applications, 88, 10–28. https://doi.org/10.1016/j.jnca.2017.04.002
Atwal, S., Porter, J., & Macdonald, P. (2002). Cardiovascular effects of strenuous exercise in adult recreational hockey: the Hockey Heart Study. CMAJ : Canadian Medical
Association Journal = Journal de l’Association Medicale Canadienne, 166(3), 303– 307.
Atzori, L., Iera, A., & Morabito, G. (2010). The Internet of Things: A survey. Computer Networks, 54(15), 2787–2805. https://doi.org/10.1016/j.comnet.2010.05.010. Bai, Y., Hibbing, P., Mantis, C., & Welk, G. (2018). Comparative evaluation of heart rate-
based monitors: Apple Watch vs Fitbit Charge HR. Journal of Sports Sciences, 36(15), 1734–1741. https://doi.org/10.1080/02640414.2017.1412235
Barnes, S. B. (2006). A privacy paradox: Social networking in the United States. First Monday, 11(9). https://doi.org/10.5210/fm.v11i9.1394
Boerman, S., Kruikemeier, S., & Zuiderveen Borgesius, F. (2017). Online Behavioral Advertising: A Literature Review and Research Agenda. Journal of Advertising, 46(3), 363–376. https://doi.org/10.1080/00913367.2017.1339368
Web-based Learning Technologies in a Developing Country. EJISDC: The Electronic Journal on Information Systems in Developing Countries, 9.
http://dialnet.unirioja.es/servlet/oaiart?codigo=3665169
14, 2020, from https://dictionary.cambridge.org/de/worterbuch/englisch/fitness-tracker Canhoto, A., & Arp, S. (2017). Exploring the factors that support adoption and sustained use
of health and fitness wearables. Journal of Marketing Management: The Internet of Things (IoT) and Marketing: The State of Play, Future Trends and the Implications for Marketing, 33(1-2), 32–60. https://doi.org/10.1080/0267257X.2016.1234505
Chau, K., Lam, M., Cheung, M., Tso, E., Flint, S., Broom, D., … Lee, K. (2019). Smart technology for healthcare: Exploring the antecedents of adoption intention of healthcare wearable technology. Health Psychology Research, 7(1), 8099. https://doi.org/10.4081/hpr.2019.8099
Chu, E., Dunn, J., Roy, D., Sands, G., & Stevens, R. (2017, December). AI in storytelling: Machines as cocreators. McKinsey. Retrieved March 25, 2020 from
https://www.mckinsey.com/industries/technology-media-and-telecommunications/our-insights/ai-in-storytelling
Chuah, S., Rauschnabel, P., Krey, N., Nguyen, B., Ramayah, T., & Lade, S. (2016). Wearable technologies: The role of usefulness and visibility in smartwatch adoption. Computers in Human Behavior, 65(C), 276–284. https://doi.org/10.1016/j.chb.2016.07.047 Chun, J., Dey, A., Lee, K., & Kim, S. (2018). A qualitative study of smartwatch usage and its
usability. Human Factors and Ergonomics in Manufacturing & Service Industries, 28(4), 186–199. https://doi.org/10.1002/hfm.20733
Claussen, J., Essling, C., & Kretschmer, T. (2015). When less can be more – Setting
technology levels in complementary goods markets. Research Policy, 44(2), 328–339. https://doi.org/10.1016/j.respol.2014.10.005
Clement, J., (2019, February 11). Global opinion on concern about online privacy 2019, by region. Statista. Retrieved March 21, 2020, from
https://www.statista.com/statistics/373338/global-opinion-concern-online-privacy/ Clement, J., (2020, January 07). Global number of internet users 2005-2019. Statista.
Retrieved March 21, 2020, from https://www.statista.com/statistics/273018/number-of-internet-users-worldwide/
Datta, P., Namin, A., & Chatterjee, M. (2018). A Survey of Privacy Concerns in Wearable Devices. 2018 IEEE International Conference on Big Data (Big Data), 4549–4553. https://doi.org/10.1109/BigData.2018.8622110
Davis, F. (1989). Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology. MIS Quarterly, 13(3), 319–340.
https://doi.org/10.2307/249008
Dinev, T., Hart P. (2006). An Extended Privacy Calculus Model for E-Commerce Transactions. Information Systems Research, 17(1), 61–80.
https://doi.org/10.1287/isre.1060.0080
Dutta-Bergman, M. (2006). A Formative Approach to Strategic Message Targeting Through Soap Operas: Using Selective Processing Theories. Health Communication, 19(1), 11– 18. https://doi.org/10.1207/s15327027hc1901_2
EU General Data Protection Regulation (2016). Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the Protection of Natural Persons with Regard to the Processing of Personal Data and on the Free Movement of such Data, and Repealing Directive 95/46/EC (General Data Protection Regulation). Official Journal of the European Union, 119 (1). https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:32016R0679&from=EN
Euromonitor. (2017, September 18). Megatrend: Healthy Living. Euromonitor International. https://blog.euromonitor.com/megatrend-healthy-living/
Eshet-Alkalai, Y., & Chajut, E. (2009). Changes Over Time in Digital Literacy.
CyberPsychology & Behavior, 12(6), 713–715. https://doi.org/10.1089/cpb.2008.0264 Eysenbach, G., Sanders, J., Wark, P., Winfree, K., Fallahzadeh, R., Fernández, C., Henriksen,
A., Haugen Mikalsen, M., Woldaregay, A.Z., Muzny, M., Hartvigsen, G., Hopstock, L.A., Grimsgaard, S. (2018). Using Fitness Trackers and Smartwatches to Measure Physical Activity in Research: Analysis of Consumer Wrist-Worn Wearables. Journal of Medical Internet Research, 20(3), e110. https://doi.org/10.2196/jmir.9157
Felt, A., Chin, E., Hanna, S., Song, D., Wagner, D.: Android permissions demystified. Proceedings of the 18th ACM Conference on Computer and Communications Security, pp. 627–638. ACM (2011). https://doi.org/10.1145/2046707.2046779 Goldfarb, A., & Tucker, C. (2012). Shifts in Privacy Concerns. American Economic Review,
102(3), 349–353. https://doi.org/10.1257/aer.102.3.349
Helbostad, J., Vereijken, B., Becker, C., Todd, C., Taraldsen, K., Pijnappels, M., Aminian, K., & Mellone, S., (2017). Mobile health applications to promote active and healthy ageing. Sensors, 17(3). https://doi.org/10.3390/s17030622
conceptualization. Paper presented at the International Public Relations Research Conference. Miami, FL.
Huang, D., Tory, M., Aseniero, B., Bartram, L., Bateman, S., Carpendale, S., Tang, A., Woodbury, R. (2015). Personal Visualization and Personal Visual Analytics. IEEE Transactions on Visualization and Computer Graphics, 21(3), 420–433.
https://doi.org/10.1109/TVCG.2014.2359887
Jiang, F., Jiang, Y., Zhi, H., Dong, Y., Li, H., Ma, S., Wang, Y., Dong, Q., Shen, H., Wang, Y. (2017). Artificial intelligence in healthcare: past, present and future. Stroke and Vascular Neurology, 2(4), 230–243. https://doi.org/10.1136/svn-2017-000101 Johnson, K., Torres Soto, J., Glicksberg, B., Shameer, K., Miotto, R., Ali, M., Euan, A.,
Dudley, J. (2018). Artificial Intelligence in Cardiology. Journal of the American College of Cardiology, 71(23), 2668–2679. https://doi.org/10.1016/j.jacc.2018.03.521 Jovanov, E. (2015). Preliminary analysis of the use of smartwatches for longitudinal health
monitoring. Conference Proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2015, 865–868.
https://doi.org/10.1109/EMBC.2015.7318499
Khushhal, A., Nichols, S., Evans, W., Gleadall-Siddall, D., Page, R., O’Doherty, A., Carrol, S., Ingle, L., & Abt, G. (2017). Validity and Reliability of the Apple Watch for
Measuring Heart Rate During Exercise. 1(6), E206–E211. DOI: 10.1055/s-0043-120195
Kuo, Y., & Yen, S. (2009). Towards an understanding of the behavioral intention to use 3G mobile value-added services. Computers in Human Behavior, 25(1), 103–110. https://doi.org/10.1016/j.chb.2008.07.007
Kraft, F. B., & Goodell, P. W. (1993). Identifying the Health Conscious Consumer. Journal of Health Care Marketing, 13(3), 18–25.
Levy, D. (2014). Emerging mHealth: Paths for growth. mHealth Team for PwC, PwC: 40 Liao, Z., & Cheung, M. (2002). Internet-based e-banking and consumer attitudes: an
empirical study. Information & Management, 39(4), 283–295. https://doi.org/10.1016/S0378-7206(01)00097-0
Malhotra, N., Kim, S., & Agarwal, J. (2004). Internet Users’ Information Privacy Concerns (IUIPC): The Construct, the Scale, and a Causal Model. Information Systems
Research, 15(4), 336–355. https://doi.org/10.1287/isre.1040.0032
University. http://jmc.stanford.edu/articles/whatisai/whatisai.pdf
McQuail, D. (2010). McQuail’s mass communication theory (6th edition.). London: SAGE. Milanović, Z., Pantelić, S., Trajković, N., Sporiš, G., Kostić, R., & James, N. (2013). Age-
related decrease in physical activity and functional fitness among elderly men and women. Clinical interventions in aging, 8, 549. DOI: 10.2147/CIA.S44112 Mordor Intelligence, (2019). SMARTWATCH MARKET - GROWTH, TRENDS, AND
FORECAST (2020 - 2025). Mordor Intelligence.
https://www.mordorintelligence.com/industry-reports/global-smart-watches-market-industry
Murrell, S., & Norris, F. (1983). Resources, Life Events, and Changes in Psychological States: A Prospective Framework. American Journal of Community Psychology, 11(5), 473–491. https://doi.org/10.1007/BF00896800
O’Dea, S., (2020, February 08). Smartphone penetration worldwide as share of global population 2016-2020. Statista. Retrieved March 22, 2020 from
https://www.statista.com/statistics/203734/global-smartphone-penetration-per-capita-since-2005/
Pal, D., Funilkul, S., & Vanijja, V. (2018). The future of smartwatches: assessing the end- users’ continuous usage using an extended expectation-confirmation model. Universal Access in the Information Society, 1–21. https://doi.org/10.1007/s10209-018-0639-z Rock Health, (2019). Digital Health Consumer Adoption Report 2019. Rock Health and
Stanford Center for Digital Health. https://rockhealth.docsend.com/view/i6j4ieu Smith, H., Milberg, S., & Burke, S. (1996). Information Privacy: Measuring Individuals’
Concerns about Organizational Practices. MIS Quarterly, 20(2), 167–196. https://doi.org/10.2307/249477
Spreer, P., & Rauschnabel, P. (2016). Selling with technology: understanding the resistance to mobile sales assistant use in retailing. Journal of Personal Selling & Sales
Management, 36(3), 240–263. https://doi.org/10.1080/08853134.2016.1208100 Statista Research Department (2016, December 19). U.S. wearable user penetration 2017, by
age. Statista. Retrieved March 22, 2020 from
https://www.statista.com/statistics/739398/us-wearable-penetration-by-age/
Statista Research Department, (2020, March 04). Apple Watch shipments worldwide 2017- 2019. Statista. Retrieved March 20, 2020 from
https://www.statista.com/statistics/526005/global-apple-watch-shipments-forecast/ Statista Research Department, (2020, March 05). Fitbit unit sales worldwide 2010-2019.
Statista. Retrieved March 2020 from https://www.statista.com/statistics/472591/fitbit-devices-sold/
Taylor, S., & Todd, P. (1995). Assessing IT Usage: The Role of Prior
Experience. MIS Quarterly, 19(4), 561–570. https://doi.org/10.2307/249633
Trommler, D., Attig, C., & Franke, T. (2018). Trust in activity tracker measurement and its link to user acceptance. Mensch und Computer 2018-Tagungsband. DOI:
10.18420/muc2018-mci-0361
Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of choice. Science, 211 (4481), 453-458. Retrieved May 12, 2020, from
www.jstor.org/stable/1685855
Udoh, E., & Alkharashi, A. (2016). Privacy risk awareness and the behavior of smartwatch users: A case study of Indiana University students. 2016 Future Technologies Conference (FTC), 926–931. https://doi.org/10.1109/FTC.2016.7821714 Ursachi, G., Horodnic, I. A., & Zait, A. (2015). How reliable are measurement scales?
External factors with indirect influence on reliability estimators. Procedia Economics and Finance, 20(15), 679-686.
Wu, Q., Sum, K., & Nathan-Roberts, D. (2016). How Fitness Trackers Facilitate Health Behavior Change. Proceedings of the Human Factors and Ergonomics Society Annual Meeting, 60(1), 1068–1072. https://doi.org/10.1177/1541931213601247
Xiang, G., Jianlin, Q., & Jin, W. (2012). Research on trust model of sensor nodes in WSNs. Procedia Engineering, 29, 909-913. https://doi.org/10.1016/j.proeng.2012.01.063 Yang, Y., Wu, L., Yin, G., Li, L., & Zhao. H. (2017). A Survey on Security and Privacy
Issues in Internet-of-Things. IEEE Internet of Things Journal, 4(5), 1250–1258. https://doi.org/10.1109/JIOT.2017.2694844
Yousafzai, S., Foxall, G. & Pallister, J. (2007), Technology acceptance: a meta‐analysis of the TAM: Part 1. Journal of Modelling in Management, 2(3), 251-280.
https://doi.org/10.1108/17465660710834453
Zarouali, B., Ponnet, K., Walrave, M., & Poels, K. (2017). “Do you like cookies?”
Adolescents’ skeptical processing of retargeted Facebook-ads and the moderating role of privacy concern and a textual debriefing. Computers in Human Behavior, 69, 157– 165. https://doi.org/10.1016/j.chb.2016.11.050
Usability Problems of Smartwatches. In: K., Miesenberger, G., Kouroupetroglou (Eds.), Computers Helping People with Special Needs. ICCHP 2018. Lecture Notes in Computer Science, vol 10897. Springer, Cham.
Appendix A Scales
Perceived Accuracy Pal, Funilkul, & Vanijja (2018):
Perceived Accuracy (1) I doubt whether the fitness data collected by a smartwatch is accurate
(2) A smartwatch fails to detect and track all of my fitness activities
(3) I would rely little on the various types of health data recorded by my smartwatch
Privacy Concerns Pal, Funilkul, & Vanijja (2018):
Privacy Concerns (1) Smartwatches can use my personal information without my knowledge violating my privacy
(2) Using smartwatches can result in a loss of my personal health and financial data
(3) The information gathered by smartwatches can be tracked, analyzed and misused.
Health Consciousness Hong (2009)
Health Consciousness (1) I am very self-conscious about my health.
(2) I´m generally attentive to my inner feelings about my health.
(3) I reflect about my health a lot.
(4) I´m concerned about my health all the time.
(5) I notice how I feel physically as I go through the day. (6) I take responsibility for the state of my health.
(7) Good health takes active participation on my part. (8) I only worry about my health when I get sick.
(9) Living life without disease and illness is very important to me.
(11) Living life in the best possible health is very important to me.
Behavioral intention to use the smartwatch Ajzen (1987); Davis (1989); Kuo & Yen (2009)
Intention to use I intend to use my smartwatch.
I predict that I will use my smartwatch on a regular basis in the future.
Appendix B Figure 1
Figure 2 Scatterplot -4 -3 -2 -1 0 1 2 -3 -2 -1 0 1 2 3 Re gre ss ion Sta n d ar d ized Re sid u al
Regression Standardized Predicted Value
Scatterplot
Appendix C
Figure 1
