Students' intention to adopt Internet-based personal safety wearable devices: Extending UTAUT with trusting belief
First Monday

Students' intention to adopt Internet-based personal safety wearable devices: Extending UTAUT with trusting belief by Devendra Potnis, Dawit Demissie, and Kanchan Deosthali



Abstract
In the backdrop of growing violence and burgeoning crime rates on campus, student safety is one of the topmost priorities for North American universities. While the promises of Internet-based personal safety wearable devices (PSWDs) are highly touted by manufacturers and the academic campuses that adopt them, there is a lack of empirical data on the level of user (student) acceptance. Drawing on the literature on IT adoption, in particular the Unified Theory of Acceptance and Use of Technology (UTAUT) model and trusting beliefs, we propose a model to investigate the factors influencing the intention of 405 undergraduate students to voluntarily adopt POM, a personal safety wearable device, at a four-year college in the Northeast portion of the United States. The empirical analysis of the model using structural equation modeling (SEM) indicated that social influence, facilitating conditions in the form of resources, effort expectancy, and trusting beliefs influence the intentions of students to use POM.

Contents

Student safety on campus
Literature review
Method
Findings and discussion
Conclusions, limitations, and implications

 


 

Student safety on campus

For more than 300 years, North American colleges and universities have experienced “violence, vice, and victimization” on their campuses (Sloan and Fisher, 2010). Despite strict rules and punishment, crime incidents like vandalism of college property, public brawls among student groups, theft from students and faculty, shootings, and stabbings of students and faculty on academic campuses is on the steady rise (Sloan and Fisher, 2014). There has been a growing security concern among academic administrators, law enforcement officials, students, parents, and the community at large (Addington, et al., 2002). In the backdrop of this growing security concern in the last decade or so, mass shootings on the campuses of Virginia Tech, Northern Illinois University, and the University of Alabama at Huntsville resulted in scores of faculty members and students being killed and injured (Sloan and Fisher, 2010). Traditionally, the larger responsibility to create a “safe environment” has been left to the academic institutions and their local community (Green, 1999). Hence, student safety is one of the topmost priorities of academic institutions in the U.S.

Information technology (IT) plays a key role in helping administrators mitigate and prevent a wide range of criminal incidents on academic campuses. Closed-circuit television, remote video monitoring, and night-vision equipment and aircraft for expanding the scope and nature of campus-wide security arrangements (Lonsway, et al., 2009), geospatial technology and global positioning system for campus crime modeling and mapping (Leitner, 2013), “blue light” emergency phones connecting the caller directly to campus police or security (Burling, 2003), automatic license plate recognition system (National Law Enforcement and Corrections Technology Center, 2012), RFID computer chips embedded in student ID cards, and large-scale X-ray and mobile truck X-ray for screening and detecting radiation-related threats (Purpura, 2013) are some of the most popular physical security technologies deployed for addressing the urgent need of keeping academic campuses safe.

As of June 2017, there are numerous Internet-based, low-cost personal safety wearable devices (PSWDs), like REACT Sidekick, Wearsafe, Peace of Mind (POM), Artemis, Safelet, Stiletto, Silent Beacon, Revolar, and Nimb, and mobile apps (e.g., bSafe, Kitestring, Bugle, SafeTrek, Safety Assistance, Watch Over Me, Witness, SafeSnapp, and Emergensee) for students, which equip them to protect themselves by seeking help in case of emergency on academic campuses and beyond. A majority of PSWDs, like POM, REACT Sidekick, Tapshield, and Guardly, rely on corresponding mobile apps for offering enterprise-grade security, which enables academic institutions to enhance security responses campus-wide.

Due to the rapid proliferation of smart phones among college students, PSWDs equipped with mobile apps seem to present an enticing solution to secure campuses of academic institutions. Students can flaunt PSWDs as fashion accessories, “cool gadgets,” or jewelry, which is another reason for the rising popularity of PSWDs among students (Chuah, et al., 2016; Wearable Technologies, 2016). As a result, a growing number of academic campus security teams are beginning to utilize a myriad of PSWDs to enhance safety and emergency response on campus (Zuckerman, 2014). In 2014, the market for PSWDs and related mobile apps was over US$340 million and was expected to grow exponentially in the near future (Gallup, 2014).

This study focuses on Internet-based POM devices which cost in the range of US$30 to US$45 (see Figure 1). POM devices are compatible with iOS and Android operating systems. POM devices serve as a solution to personal safety problems and aims at providing peace of mind for parents, students, and families.

 

A POM device
 
Figure 1: A POM device. Source: https://pom-co.com.

 

This next-generation, silver dollar-sized, gender-neutral personal safety device is connected to the user’s smart phone via Bluetooth. POM is a streamlined device with a single purpose: a simple, discreet push of a button in an emergency, which instantly connects the user with campus security and records the call (POMCO, 2015). At the push of a button, POM can make a phone call to the campus police department and send a real-time GPS location to the dispatch center. The dispatch center also receives the entire profile information of the caller, including their picture, medical conditions, allergies, and emergency contact information. In addition to a direct line to authorities, the device can also activate a silent or audible alarm function.

Research question

In 2015, a four-year college in the northeast section of the United States handed out complimentary POM devices to all of its students (i.e., approximately 1,400 students). None of the students were using POM before the college introduced them to POM devices. The complimentary POM devices were part of the institution’s effort to demonstrate their commitment to keep students safe on campus, although the college does not require students to use POM devices on a daily basis. However, free access to POM devices might not necessarily translate into their adoption. Hence, this study investigates the following research question: What are the factors influencing voluntary adoption of complimentary PSWDs by students?

This paper is organized as follows: the second section reviews relevant literature to propose a theoretical model for predicting the intention of students to adopt POM. The third section outlines methodological details from the study including data collection and analysis. The fourth section discusses the findings of the study. The final section concludes with theoretical and practical implications of the findings.

 

++++++++++

Literature review

A wearable device is an electronic computing device embedded in the personal space of a user, and can be worn, carried, or attached to the body (Choi, 2014; Jiang, et al., 2015). Wearable devices can be broadly divided into three categories: “notifiers” which give users information about the world around them (e.g., smart watches), “glasses” which use eyeglasses to create augmented virtual reality (e.g., Google Glass, smart glasses, etc.), and “trackers” like sports and fitness trackers which use sensors to record data (Stein, 2014). PSWDs like POM represent a combination of “notifiers” and “trackers.”

Adoption of wearable devices

A large majority of wearable devices rely on smart phones, which are the most widely used technology across all adult demographic groups (Nielsen, 2016). Over half of the world’s population uses mobile broadband services (International Telecommunications Union, 2015). With the rising popularity of smart phones, there is an increase in the use of complementary applications and syncing devices like wearable devices (Lunney, et al., 2016). Existing research on the adoption of wearable devices is in its infancy and focuses mainly on the usability, accuracy, and reliability of these devices (Byun, et al., 2016; Diaz, et al., 2015). Most of the existing studies on wearable devices like trackers and notifiers are exploratory. Apart from technological papers introducing diverse plausible forms of wearable computing, few studies address the challenges and adoption-related research issues related to the widespread use of wearable devices (Choi and Kim, 2016; Jiang, et al., 2015).

Choi (2014) identified the following six factors affecting the attitude or behavior of wearable device adoption: fundamental needs, cognitive activity (e.g., perceived usefulness, perceived ease of use), social aspects (e.g., social inuence, privacy, culture), physical aspects (e.g., physical comfort, appearance), demographic characteristics (e.g., age, gender), and technological experiences. Perceived usefulness emerges as the strongest factor supporting the acceptance of wearable technologies (Gribel, et al., 2016). After analyzing the wearable technology adoption behavior of 375 respondents, Yang, et al. (2016) found perceived value to be a clear antecedent of using wearable technologies in Korea. Perceived benefits — including perceived usefulness, enjoyment, and social image — seem to have a greater impact on perceived value than perceived risks. Turhan (2013) discovered that subjective norms, perceived usefulness, attitude and perceived behavioral control explain users’ intention to use wearable devices in Turkey.

A study focusing on users’ complex analytic processes comprising the Unified Theory of Acceptance and Use of Technology (UTAUT) model and Analytical Network Process generated a wearable technology acceptance model (WTAM) that consisted of four main clusters of factors: performance expectancy, effort expectancy, social inuence, and facilitating conditions, which inuence the intention to use as well as actual potential use behavior (Chen and Shih, 2014). Although this model is based on UTAUT it does not test it empirically. This current study fills in the gap.

Little research has been conducted on the nature of acceptance and use of PSWDs on academic campuses. Despite the overwhelming promises these PSWDs offer, academic campuses are confronted with finding creative ways to encourage students to use these systems in a large scale (Horvath and Pisciotta, 2015). In contrast, this study seeks to better understand how, why, and with what effect students are using wearable activity devices. In particular, UTAUT (Venkatesh, et al., 2003), which this study employs, explains user motivations in using an information technology (IT) solution and subsequent behaviors.

UTAUT & IT adoption

Past research on IT adoption has employed a number of theoretical models (Chan, et al., 2010) but has culminated in the UTAUT, synthesizing previous adoption models (Venkatesh, et al., 2003). UTAUT is a combination of comprehensive review and empirical comparison of eight major competing models of technology acceptance and use by individuals. Venkatesh, et al. (2003) reviewed and empirically tested and combined the following methodologies: theory of reasoned action, technology acceptance model, motivational model, theory of planned behavior, a combined theory of planned behavior/technology acceptance model, model of PC utilization, innovation diffusion theory, and social cognitive theory. The subsequent longitudinal study indicates that this new framework accounts for 70 percent of the variance in behavioral intention to use and about 50 percent in actual use (Venkatesh, et al., 2003). Considered one of the most robust theoretical models for explaining IT adoption and use, UTAUT has been tested extensively and has achieved a wider recognition as a result. By building on UTAUT, this research study identifies factors responsible for the adoption and use, or lack thereof, of POM.

Previous studies have suggested that the four core determinants in UTAUT — performance expectancy, effort expectancy, social influence, and facilitating conditions — are key general beliefs that influence user adoption (e.g., Venkatesh, et al., 2003). Venkatesh, et al. (2003) define performance expectancy (also known as perceived usefulness) as the degree to which an individual believes that using the IT will help him or her to attain goals in job performance, and effort expectancy as the degree of ease associated with the use of IT. Social influence is defined as the degree to which an individual perceives that others believe he or she should use new IT, and facilitating conditions as the degree to which an individual believes that organizational and technical infrastructure exists to support use of IT (Venkatesh, et al., 2003).

As of February 2017, there were few studies extending the UTAUT framework for examining the adoption of PSWDs like POM on academic campuses. This study fills in the gap by extending UTAUT with trusting beliefs.

Performance expectancy (PE)

From a motivation perspective, performance expectancy is a measure of the user’s level of extrinsic motivation and outcome expectancy (Kim, et al., 2007). Therefore, it is argued to influence behavioral intention of external rewards (Chuah, et al., 2016). In a study with smart glass consumers, Rauschnabel, et al. (2016) found performance expectancy (i.e., utilitarian benefits) to be a core determinant of wearable technology adoption. POMCO (2015) claims that because of POM’s compact size, students or other users do not have to worry about fumbling for their phone and finding an app, entering a passcode, or even dialing 911 if they find themselves in an emergency. In this way, the POM device is expected to help students attain the goal of safety. Hence, this study hypothesizes:

H1: Performance expectancy will positively influence the intention of students to use POM devices on academic campuses.

Effort expectancy (EE)

Effort expectancy is the degree of ease when using any information technology or information system (Venkatesh, et al., 2003). Kim and Shin (2015) examined psychological determinants of smart watch adoption. Findings revealed that mobility and availability led to a greater perceived ease of use, which in turn affected effort expectancy. Designed with the college student in mind, the POM is compact, portable, convenient to attach to keys, lanyards or backpacks, and fits easily in a purse or pocket. POM only needs to be charged every 10–14 days. When the charge gets too low, the POM automatically puts itself to sleep to preserve battery life, ensuring the device is still able to make an emergency call lasting up to eight minutes (POMCO, 2015). In addition, the cloud-based software requires no installation or set-up, allows for easy upgrades, and utilizes existing cellular networks to connect users. Thus, students are not required to make any extraordinary changes in their lifestyle or invest any extra effort for using POM when needed. Hence, this study’s second hypothesis is:

H2: Effort expectancy, i.e., the ease of using the system, will positively influence the intention of students to use POM devices on academic campuses.

Social influence (SI)

A positive social acceptance of wearable technologies encourages students to buy and use such innovative devices (Jeong, et al., 2017). For instance, subcultural appeal of mobile devices, affective quality, and relative advantage enhance the perceived usefulness, and hence, intentions of consumers to buy and use smart watches (Kim and Shin, 2015). Consumers who perceive the potential for social conformity of smart glasses tend to adopt them easily (Rauschnabel, et al., 2016). WTAM identifies social influence as one of the significant factors affecting students’ intention to use technology such as Google Glass (Woodside, 2015). The four-year college under consideration in this study distributed POM devices for free to potential study participants and encouraged them to use it for safety. Hence, this study hypothesizes:

H3: Social influence will positively influence the intention of students to use POM devices on academic campuses.

Facilitating conditions (FC)

Facilitating conditions include the extent and type of support provided to individuals that influence their use of the technology (Triandis, 1979; Potnis and Deosthali, 2014). Woodside (2015) found that facilitating conditions in the form of organizational support was a key factor affecting the intention of adoption of Google Glass by 57 students at a small college in the U.S. Just as that device’s portability and low maintenance make it ideal for students, the POM system’s streamlined implementation process is designed to be easy for universities to adopt (POM, 2015). This facilitating support from POM creates a conducive environment for its potential users. For instance, with the POM, students are instantly connected with campus security. They know where you are, who you are, and can even talk to you directly. The POM system enhances safety and security on campuses by improving accessibility, connectivity, and the quality of information exchanged between campus community members and their security departments.

In addition to providing a supportive technical environment, POM offers flexible pricing models to colleges and universities, which can be tied into existing student fees or offered on an individual subscription basis. The low annual subscription fee of approximately US$200 a year, tailored to students and parents, includes the POM and its mobile app, which is less than the cost of most textbooks (POM, 2015). Gribel, et al. (2016) discovered political and legal environment, pricing structure, and competitors supplying wearable devices to be key macro-environmental factors (i.e., facilitating conditions) influencing the adoption of wearable devices including PSWDs like POM. All of the participants in this study received complimentary POM devices. In addition, their academic institution provides support for addressing technical difficulties associated with using POM. Hence, the study’s fourth hypothesis is:

H4: Facilitating conditions will positively influence the intention of students to use POM devices on academic campuses.

Trusting beliefs (TB)

Trusting beliefs play a key role in the adoption of innovative IT-based services (Carter and Bélanger, 2005; Teo, et al., 2008; Warkentin, et al., 2002). Trust is critical in influencing the user’s intention to continue using wearable health devices (Li, et al., 2016). Lunney, et al. (2016) recommended studying the effect of trust on wearable technology adoption after discovering a negative relationship between perceived usefulness of wearable devices and untrustworthy data generated by the devices. Gribel, et al. (2016) found trust in consequences of usage and trust in the system’s functionality to be key determinants of the adoption of wearable technologies by users. Trust in technology itself, including its functionality and predictability, directly affects risk perception of potential users. Hence, this study hypothesizes:

H5: Trusting beliefs will positively influence the intention of students to use POM devices on academic campuses.

Behavioral intention (BI)

Behavioral intention to use is the degree of the psychological state of people’s general minds to use specific services and systems (Davis, et al., 1989). Brown, et al. (2002) found that users’ attitudes toward using a technology, and their intentions to use that technology are related in voluntary use environments. They noted that intention to use a technology is related more to other beliefs, such as the associated rewards and punishments, than to beliefs about the technology itself.

Research model

Based on the existing relevant literature, we propose a theoretical model of six variables: performance expectancy, effort expectancy, social influence, facilitating conditions, and trust as independent variables, and behavioral intention as a dependent variable (see Figure 2).

 

Proposed theoretical model
 
Figure 2: Proposed theoretical model.

 

 

++++++++++

Method

A total of 1,400 undergraduate students with the possession of POM gifted by their academic institution were invited to participate in this study. Of these, 405 surveys were returned (response rate 28.9 percent). They were informed that participation in the study was voluntary. All study participants were at least 18 years of age.

Data was collected anonymously in 2016–17, almost a year after POM devices were distributed for free among students. No compensation was awarded to complete the Qualtrics-hosted online survey. It took approximately 15 minutes to complete the survey. Prior to any data analysis, the data was evaluated for duplicate participants and missing data. SEM was used to quantitatively analyze data with AMOS software version 24.

The questionnaire used for data collection is divided into two sections. The first section contains relevant demographic information to the study, such as gender, year in college, age, location or residence, training for using POM, level of comfort for using POM, duration of using POM, and experience of using POM.

The second section includes the following six subcomponents: Performance expectancy: Four items modified from the UTAUT survey were included in this study. Participants rated their expectations on the use of POM in relation to how it improves safety issues, and as a result, improves quality of life. Effort expectancy: Participants rated their interaction with POM in responding to five survey items that asked the effort it takes to learn and familiarize themselves with the technology of POM. Social influence: Five of the items in the survey asked participants to rate the influences of individuals in their social circles regarding their uses of POM. Facilitating conditions: Participants rated six facilitating condition items, such as instructions, assistances, and encouragements given to students both by their respective institutions and the POM Company. Trusting beliefs: Participants rated their trusting beliefs toward POM and also rated the dependability and reliability of POM. Behavioral intention: The last subsection of part two of the survey contains four questions related to the intentions of participants about their continued use of POM.

We also evaluated data for duplicate participants and missing data. The data was tested for multicollinearity. Tolerance was greater than 0.10 (0.63) and the variance inflation factor was less than 10 (1.63) thus demonstrating that multicollinearity is not an issue.

Measures

The questionnaire consists of six major sections that assess: (1) performance expectancy; (2) effort expectancy; (3) social influence; (4) facilitating conditions; (5) trusting beliefs; and, (6) behavioral intention (see Appendix).

Performance expectancy

Performance expectancy (Davis, et al., 1989) is measured using a four-item, five-point Likert-type scale ranging from Strongly Disagree (score = 1) to Strongly Agree (score = 5). Sample items include: “Using POM would enable me to report incidences/emergencies more quickly,” “Using POM gave me greater control over emergency reporting process in this college/university,” etc. Items measured were collapsed to form a single measure (M = 3.44, SD = 0.50). The internal consistency reliability estimate of this scale was 0.63.

Effort expectancy

Effort expectancy (Venkatesh, et al., 2003) is measured using a four-item, five-point Likert-type scale ranging from Strongly Disagree (score = 1) to Strongly Agree (score = 5). Sample items included: “My interaction with POM would be clear and understandable,” “It would be easy for me to become skillful at using POM,” etc. Items measured were collapsed to form a single measure (M = 3.06, SD = 0.42). The internal consistency reliability estimate of this scale was 0.78.

Social influence

The measure of social influence uses an established five-item, five-point Likert-type scale operationalized by Ajzen (1985) and adapted by Norman and Smith (1995), ranging from Strongly Disagree (score = 1) to Strongly Agree (score = 5). Sample items include: “People who influence my behavior (e.g., my professors and parents) think that I should use POM,” “My friends at school thinks that I should use POM,” “My siblings think that I should use POM,” etc. Items measured were collapsed to form a single measure (M = 3.12, SD = 0.56). The internal consistency reliability estimate of this scale was 0.88.

Facilitating conditions

Facilitating conditions (Venkatesh, et al., 2003) were measured using a six-item, five-point Likert-type scale ranging from Strongly Disagree (score = 1) to Strongly Agree (score = 5). Sample items included: “The school has been helpful in encouraging the use of POM for the purpose of reporting incidences/emergencies,” “The personal safety service provider (the maker of POM) has supported the use of POM,” etc. Items measured were collapsed to form a single measure (M = 3.05, SD = 0.54). The internal consistency reliability estimate of this scale was 0.81.

Trusting beliefs

Trusting beliefs (McKnight, et al., 2002) were measured using a five-item, five-point Likert-type scale ranging from Strongly Disagree (score = 1) to Strongly Agree (score = 5). Sample items included: “The POM system is dependable and reliable,” “I trust the POM system because it serves my best interests,” etc. Items measured were collapsed to form a single measure (M = 3.14, SD = .72). The internal consistency reliability estimate of this scale was 0.87.

Behavioral intention

Behavioral intention (Brown, et al., 2002) was measured using a three-item, five point Likert-type scale ranging from Strongly Disagree (score = 1) to Strongly Agree (score = 5). Sample items included: “I will primarily use POM for my safety needs.” Items measured were collapsed to form a single measure (M = 3.1, SD = .62). The internal consistency reliability estimate of this scale was 0.8.

 

++++++++++

Findings and discussion

A majority of respondents were female in the age group of 18–20 (see Table 1).

 

Table 1: Demographics of respondents.
Demographic variablesResponses
(N = 405)
GenderMale: 11.86%
Female: 88.14%
Age (years)18–20: 58.27%
21–22: 27.65%
23–25: 6.17%
25 or more: 7.91%
Academic statusFreshmen: 19.01%
Sophomore: 22.48%
Junior: 27.9%
Senior: 30.61%

 

Over 50 percent of respondents had neither used POM before receiving it for free from their institution nor received any training for using POM (see Table 2), which warrants for eliciting qualitative responses from participants for the barriers to using POM. We present these qualitative findings later in this paper.

 

Table 2: Comfort level, training, and duration of using POM.
VariablesData
Training for using POMNo training at all: 51.62%
Less than 30 minutes: 39.25%
30–45 minutes: 7.65%
More than 45 minutes:1.48%
Duration of using POMNever used: 48.17%
1 month or less: 7.65%
1–6 months: 19.01%
6 months — 1 year: 9.87%
More than 1 year: 15.30%
Comfort level with using wearable devices like POMVery uncomfortable: 15.1%
Uncomfortable: 5.9%
Neutral: 23.95%
Comfortable: 30.61%
Very Comfortable: 24.44%

 

Analysis

The current study focused on assessing the factors influencing voluntary adoption of POM by undergraduate students. The data was analyzed with path analysis structural equation modeling (SEM) using AMOS 24.

Numerous overall fit indexes for SEM have been proposed over the last 20 years, and there is no general agreement on which measures are preferred (Kline, 2015; Myers, et al., 2000; Maruyama, 1998). Myers, et al. (2000) proposed that absolute fit and relative fit indicators should be assessed. Using the guidelines by Myers, et al. (2000), the hypothesized model was evaluated by AMOS 24 using the comparative fit index (CFI) and the normed fit index (NFI).

Assessing model fit

Based on the fit indices, the model exhibited a good fit to the data. The values for Comparative Fit Index (CFI) and Normed Fit Index (NFI) were 1 and 0.99 (values closer to 1 represent a very good fit) respectively reflecting a very good fit of the model to the data.

Control variables

We did not find gender, age, and academic status of respondents to be statistically significant in our model. Importantly, their inclusion did not change the estimates (coefficients) of our independent variables. Hence, we did not control the effects of these variables in the model.

SEM results for the whole sample

The measurement model explained 53.4 percent of the variance in behavioral intentions to voluntarily adopt POM. All of the hypotheses were supported, except for H1 (see Table 3). H1 theorized that performance expectancy would be positively related to intentions to use POM. However, this hypothesis was not supported (β = - 0.061, n.s.). In support of H2, effort expectancy and intentions to use POM were positively related (β = 0.28, p ≤ 0.001). H3 theorized a positive relation between social influence and intentions to use POM. This hypothesis was supported (β = 0.21, p ≤ 0.001). In support of H4, facilitating conditions were positively related to intentions to use POM (β = 0.13, p ≤ 0.001). H5 theorized that trusting beliefs would be positively related to intentions to use POM, and this hypothesis was supported (β = 0.36, p ≤ 0.001).

 

Table 3: Results of structural equation modeling for the whole sample.
Note: Statistically significant as *. p ≤ 0.05, **. p ≤ 0.01, ***. p ≤ 0.001.
VariablesSEM results
Performance expectancyβ = -.061
Effort expectancyβ = 0.28***
Social influenceβ = 0.21***
Facilitating conditionsβ = 0.13**
Trusting beliefsβ = 0.36***

 

Female vs. male estimates

Since our sample consisted of a majority of women (i.e., 357 out of 405), a multigroup analysis was performed to check if the proposed model performs differently for male and female respondents (see Table 4).

 

Table 4: Comparing SEM results of male and female respondents.
Note: Statistically significant as *. p ≤ 0.05, **. p ≤ 0.01, ***. p ≤ 0.001.
VariablesMenWomen
Performance expectancyβ = -0.35***β = -0.04
Effort expectancyβ = 0.36***β = 0.26***
Social influenceβ = 0.04β = 0.23***
Facilitating conditionsβ = 0.30*β = 0.12*
Trusting beliefsβ = 0.57***β = 0.36***

 

Social influence does not seem to affect the intention of men to use POM. In addition, unlike for female performance, expectancy negatively influences the intention of male students to use POM. Due to the small sample size of male respondents (n = 48), it would be inappropriate to interpret these results. In any event, the statistical differences show that our proposed model performs differently for male and female respondents. Further research with a larger sample size of male population is required.

Barriers to Using POM

This study elicited qualitative responses from students in order to learn about the obstacles to using POM devices, especially since a majority of the study participants had not used them after receiving them for free. This study applied grounded theory principles to analyze the corpus of qualitative data collected from study participants, which was originally conceptualized by Glaser and Strauss (1967). The open coding, i.e., line-by-line coding, of more than 50 challenges, led to 12 axial codes that were further clustered into five select codes (see Table 5). These five codes represent factors that negatively influence the behavioral intentions of students in using POM devices on campus. These are: lack of perceived usefulness/low performance expectancy, lack of trust, insufficient facilitating conditions, high-effort expectancy to use POM, and lack of interest and time.

 

Table 5: Analyzing barriers to using POM.
Open coding →Axial coding →Selective coding
I don’t feel the need to use itLack of need to use POMLack of perceived usefulness/Low performance expectancy
Haven’t needed to yet
Don’t think I need one
I have never been in a situation where I would need to use the POM device
Haven’t had a reason to use it yet
No need — it is hardly useful
I do not think it is helpfulNot helpful in general or on campus
Not useful on campus much
Too many accidental calls — It’s annoying to maintainUnsatisfactory performance
Has difficulty connecting to Bluetooth at times, inconvenient as I am a commuter so it doesn’t work once I leave campus
Change of phones, can’t log in to my old account nor can I connect my new phone to the POM that I originally hadPOM does not work as expected
I’m afraid I’ll set it off by accident or that it will go off in classUndesired consequences of POM
The sound is loud and seems unsafe
Annoying to carry and kills phone batteryUnhappy with the way POM works/Unhappy with dependency on mobile phone
I don’t like the fact that it requires my phone. I don’t always carry my phone with me
Don’t like the app on my phone
Drains battery and button hold is too longDissatisfied with POM functionalities and operation
Kills phone battery
Drains battery long time to press for help
It uses a lot of data and battery (Bluetooth drains battery)
Always have to charge my device
Have to charge it too frequently
I mistakenly hit the button once and the way that was handled made it clear that in a real situation there would be a lack of public safety competencePOM penalizes users for their mistakesLack of trust
Uses up phone battery and goes off accidentally a lotUsers cannot rely on POM
It drains the battery from my phone, meaning I can’t call anyone, which defeats the purpose
Username and password would not work sometimes
It turns on by itself and sometimes goes off on its own
I don’t think I can rely on POM. Hence, I carry pepper spray and have my phone on public safety number when I go for a walk so I can just hit it
Uses up phone battery by using Bluetooth and turns on accidentally in pocket
Drains battery
Too short battery life
Drains phone battery too quickly because of Bluetooth
Dies too quickly. Brings down the battery life on my phone
When I tried to get it upon transferring here, no one could tell me the right place to pick it up. Therefore, I never got it.Lack of support for using POMInsufficient facilitating conditions
Setting up the deviceUnable to set upHigh effort expectancy to use POM
I couldn’t set it up because of the unclear directions on the box
I don’t know how to set it up
Lack of time to learn and use itInert userLack of interest and time
Lost the charger and once it died it stayed dead

 

The overlap between the study’s qualitative findings in Table 5 and proposed hypotheses (see Figure 2) supports the study’s proposed theoretical model, and highlights its significance in studying the adoption of PSWDs like POM. For instance, four out of five independent variables (i.e., performance expectancy, effort expectancy, facilitating conditions, and trusting beliefs) in the proposed theoretical model are confirmed by qualitative responses. Our first hypothesis states that “performance expectancy will positively influence the intention of students to use POM devices on academic campuses,” which is not supported by SEM results. Qualitative data analysis also supports SEM results by revealing lack of perceived usefulness or low performance expectancy as the most frequently reported barrier to using POM. Clearly, POMCO’s claim about high performance of POM devices is refuted by this study’s participants. Reliance on smart phones for using POM (which could lead to the inability of POM to operate if the smart phone is lost or discharged), severe unintended consequences of pressing wrong buttons on POM (i.e., POM does not give users a second chance to correct their mistakes of accidentally pressing incorrect buttons), and unreliable performance discourage students from relying on the device to be safe. These drawbacks lower students’ outcome expectancy from POM and negatively influence their motivations to use the device for safety purposes. The inability of POM to function as promised makes students dissatisfied about its performance, thereby negatively affecting their behavior intentions to continue using POM.

Gribel, et al. (2016) found “perceived IT security risks” to be the main reason for resistance to wearable technologies in Europe. However, participants in this study did not report a potential security compromise of personal information as a barrier to using POM.

 

++++++++++

Conclusions, limitations, and implications

With the exception of a few studies establishing technical usability, accuracy, and reliability, research on PSWDs on academic campuses is still in its infancy (Byun, et al., 2016; Diaz, et al., 2015). This study advances our understanding of the factors influencing students’ intention to use PSWDs by extending UTAUT with trusting beliefs. The SEM results of this study help us understand the level and drivers of acceptance of PSWDs like POM. This study identifies effort expectancy, social influence, facilitating conditions, and trusting beliefs to be key factors positively affecting students’ intention to use POM. In a complimentary relationship, the analysis of qualitative responses suggests perceived usefulness, lack of student interest and time, and POM’s ability to meet student expectations as factors that negatively affect the intention of students to use POM. This complimentary portrayal of factors that encourage and discourage students from using POM devices is a unique contribution of this paper to the existing literature on wearable technology acceptance by students.

Limitations

The findings of this study do not hold across American universities. Most post-secondary institutions do not inform students about POM devices and/or give away these devices to all students. These two factors could be major hurdles to adoption of POM devices. However, the sample population in this study did not experience these barriers to adopting POM devices. We did not collect any data on prior experience of using PSWDs, which is another limitation of this study, although none had used POM before receiving it free from their institution.

Practical implications

Based on our study findings, we have the following set of recommendations for academic institutions that plan to deploy PSWDs among students.

  1. Survey safety concerns, needs, and related existing services by involving key stakeholders, such as students and their parents, faculty, staff, and police on campus.
  2. Select an appropriate technology vendor that meets your campus needs. Make sure to verify claims of their products like PSWDs and services (i.e., corresponding mobile apps) in terms of features, performance, and usability, with the help of experts.
  3. Create awareness about PSWDs in general through print and digital media (e.g., campus newspapers, campus radio stations, etc.). Introduce pros and cons of using PSWDs among key stakeholders.
  4. Involve other academic units, like libraries and offices of IT, for creating and running (a) educational videos highlighting the benefits of using PSWDs, and (b) instructional videos on how to use PSWDs on campus. Make sure to emphasize that the introduction of PSWDs is a proactive, precautionary measure by the institution.
  5. Pilot test PSWDs with key stakeholders before introducing and championing for any specific personal safety wearable device and corresponding mobile apps.
  6. Provide more charging stations equipped with electric cords so that students can charge their smart phones across the campus.
  7. Make sure to involve campus police in the rollout of PSWDs for students.

The wearable technology market is diversifying at an extraordinary pace, which further underlines the significance of this study and its findings. Based on this study, academic institutions can make data-driven decisions to develop a better student safety strategy and training plan using PSWDs like POM. Study findings could assist colleges and universities in calculating return-on-investment in PSWDs, where cost and return can be measured in terms of students’ personal safety and institutional benefits in terms of providing a safe environment Additionally, it important to understand how technologies are adopted and managed in higher education institutions and therefore directly or indirectly affect academic business practices. The outcomes of this study serve as a feedback for investing in and deploying PSWDs in academia. Overall, there is a significant and growing market opportunity for PSWDs, especially for the safety of women and elderly populations. The findings could inform institutions on strategies to deploy PSWDs for the safety of vulnerable populations.

Theoretical implications

This study introduced and subsequently validated an additional construct (i.e., trusting beliefs) to the UTAUT model in exploring usage behavior. Thus, this study contributed to the existing literature on user acceptance of PSWDs on academic campuses, an area that has not been studied in this specific fashion.

The wearable technology adoption model proposes perceived enjoyment, perceived usefulness, organizational support, privacy, social influence, and perceived ease of use to be key determinants of students’ intentions to use wearable technology like Google Glass (Woodside 2015). It is important to note that even though PSWDs do not belong to the “glasses” sub-category of wearable devices, there is an overlap between the factors the influence user intentions to adopt PSWDs like POM and Google Glass. At the same time, due to the differences in the three sub-categories of wearable devices, perceived enjoyment is less likely to influence the intention of users to adopt “notifiers” and “trackers” like POM devices than “glasses.” End of article

 

About the authors

Devendra Potnis is Associate Professor in the School of Information Science at the University of Tennessee at Knoxville.
E-mail: dpotnis [at] utk [dot] edu

Dawit Demissie is Assistant Professor and faculty program director for the CIS and IT-Cybersecurity programs at the Sage Colleges in Albany, N.Y.
E-mail: demisd [at] sage [dot] edu

Kanchan Deosthali is Assistant Professor of Management in the School of Business at the University of Mary Washington in Fredericksburg, Va.
E-mail: kdeostha [at] umw [dot] edu

 

References

L.A. Addington, S.A. Ruddy, A.K. Miller, and J.F. DeVoe, 2002. “Are America’s schools safe? Students speak out: 1999 school crime supplement. Statistical analysis report,” at https://eric.ed.gov/?id=ED472826, accessed 19 February 2017.

I. Ajzen, 1985. “From intentions to actions: A theory of planned behavior,” In: J. Kuhl and J. Beckmann (editors). Action control: From cognition to behavior. Berlin: Springer-Verlag, pp. 11–39.
doi: https://doi.org/10.1007/978-3-642-69746-3_2, accessed 10 August 2017.

S.A. Brown, A.P. Massey, M.M. Montoya-Weiss, and J.R. Burkman, 2002. “Do I really have to? User acceptance of mandated technology,” European Journal of Information Systems, volume 11, number 4, pp. 283–295.
doi: https://doi.org/10.1057/palgrave.ejis.3000438, accessed 10 August 2017.

P. Burling, 2003. Crime on campus: Analyzing and managing the increasing risk of institutional liability. Second edition. Washington, D.C.: National Association of College and University Attorneys.

W. Byun, A. Barry, and J.-M. Lee, 2016. “Accuracy of the Fitbit for measuring preschoolers’ physical activity,” Medicine & Science in Sports & Exercise, volume 48, number 5S, p. 778.
doi: https://doi.org/10.1249/01.mss.0000487337.24755.a2, accessed 10 August 2017.

L. Carter and F. Bélanger, 2005. “The utilization of e-government services: Citizen trust, innovation and acceptance factors,” Information Systems Journal, volume 15, number 1, pp. 5–25.
doi: https://doi.org/10.1111/j.1365-2575.2005.00183.x, accessed 10 August 2017.

F.K.Y. Chan, K.Y. Thong, Y.L. James, V. Vankatesh, S.A. Brown, P.J.-H. Hu, and K.Y. Tam, 2010. “Modeling citizen satisfaction with mandatory adoption of an e-government technology,” Journal of the Association for Information Systems, volume 11, number 10, p. 519–549.

C.-C. Chen and H.S. Shih, 2014. “A study of the acceptance of wearable technology for consumers — An analytical network process perspective,” at http://www.isahp.org/uploads/p729838.pdf, accessed 19 February 2017.

J. Choi and S. Kim, 2016. “Is the smartwatch an IT product or a fashion product? A study on factors affecting the intention to use smartwatches,” Computers in Human Behavior, volume 63, pp. 777–786.
doi: https://doi.org/10.1016/j.chb.2016.06.007, accessed 10 August 2017.

M.-S. Choi, 2014. “Human movement searching in video streams using shape sequence,” International Journal of Multimedia and Ubiquitous Engineering, volume 9, number 4, pp. 103–108.
doi: https://doi.org/10.14257/ijmue.2014.9.4.11, accessed 10 August 2017.

S. H.-W. Chuah, P.A. Rauschnabel, N. Krey, B. Nguyen, T. Ramayah, and S. Lade, 2016. “Wearable technologies: The role of usefulness and visibility in smartwatch adoption,” Computers in Human Behavior, volume 65, pp. 276–284.
doi: https://doi.org/10.1016/j.chb.2016.07.047, accessed 10 August 2017.

F.D. Davis, R.P. Bagozzi, and P.R. Warshaw, 1989. “User acceptance of computer technology: A comparison of Two theoretical models,” Management Science, volume 35, number 8, pp. 982–1,003.
doi: https://doi.org/10.1287/mnsc.35.8.982, accessed 10 August 2017.

K.M. Diaz, D.J. Krupka, M.J. Chang, J. Peacock, Y. Ma, J. Goldsmith, J.E. Schwartz, and K.W. Davidson, 2015. “Fitbit®: An accurate and reliable device for wireless physical activity tracking,” International Journal of Cardiology, volume 185, pp. 138–140.
doi: https://doi.org/10.1016/j.ijcard.2015.03.038, accessed 10 August 2017.

Gallup, 2014. “In U.S., 37% do not feel safe walking at night near home” (24 November), at http://www.gallup.com/poll/179558/not-feel-safe-walking-night-near-home.aspx, accessed 19 February 2017.

B.G. Glaser and A.L. Strauss, 1967. The discovery of grounded theory: Strategies for qualitative research. Chicago: Aldine.

L. Gribel, S., Regier, and I. Stengel, 2016. “Acceptance factors of wearable computing: An empirical investigation,” Proceedings of the Eleventh International Network Conference (INC 2016), pp. 67–72, and at https://www.cscan.org/default.asp?page=openaccess&eid=18&id=321, accessed 19 February 2017.

J. Horvath and F. Pisciotta, 2015. “10 common dorm security challenges and their solutions” (25 March), at http://www.campussafetymagazine.com/article/solutions_to_10_common_dorm_security_challenges, accessed 19 February 2017.

International Telecommunications Union (ITU), 2015. “ICT facts & figures: The world in 2015,” at https://www.itu.int/en/ITU-D/Statistics/Documents/facts/ICTFactsFigures2015.pdf, accessed 19 February 2017.

S.C. Jeong, S.-H. Kim, J.Y Park, and B. Choi, 2017. “Domain-specific innovativeness and new product adoption: A case of wearable devices,” Telematics and Informatics, volume 34, number 5, pp. 399–412.
doi: https://doi.org/10.1016/j.tele.2016.09.001, accessed 10 August 2017.

H. Jiang, X. Chen, S. Zhang, X. Zhang, W. Kong, and T. Zhang, 2015. “Software for wearable devices: Challenges and opportunities,” arXiv (3 April), at http://arxiv.org/abs/1504.00747, accessed 19 February 2017.

H.-W. Kim, H.C. Chan, and S. Gupta, 2007. “Value-based adoption of mobile Internet: An empirical investigation,” Decision Support Systems, volume 43, number 1, pp. 111–126.
doi: https://doi.org/10.1016/j.dss.2005.05.009, accessed 10 August 2017.

K.J. Kim and D.-H. Shin, 2015. “An acceptance model for smart watches: Implications for the adoption of future wearable technology,” Internet Research, volume 25, number 4, pp. 527–541.
doi: https://doi.org/10.1108/intr-05-2014-0126, accessed 10 August 2017.

R.B. Kline, 2015. Principles and practice of structural equation modeling. Fourth edition. New York: Guilford Press.

M. Leitner (editor), 2013. Crime modeling and mapping using geospatial technologies. Dordrecht: Springer.
doi: https://doi.org/10.1007/978-94-007-4997-9, accessed 10 August 2017.

H. Li, J. Wu, Y. Gao, and Y. Shi, 2016. “Examining individuals’ adoption of healthcare wearable devices: An empirical study from privacy calculus perspective,” International Journal of Medical Informatics, volume 88, pp. 8–17.
doi: https://doi.org/10.1016/j.ijmedinf.2015.12.010, accessed 10 August 2017.

K.A. Lonsway, V.L. Banyard, A.D. Berkowitz, C.A. Gidycz, J.T. Katz, M.P. Koss, P.A. Schewe, and S.E. Ullman, 2009. “Rape prevention and risk reduction: Review of the research literature for practitioners,” at http://www.councilofpresidents.org/docs/SA_Conf/Berkowitz_AR_RapePrevention_Lonsway_et_al.pdf, accessed 19 February 2017.

A. Lunney, N.R. Cunningham, and M.S. Eastin, 2016. “Wearable fitness technology: A structural investigation into acceptance and perceived fitness outcomes,” Computers in Human Behavior, volume 65, pp. 114–120.
doi: https://doi.org/10.1016/j.chb.2016.08.007, accessed 10 August 2017.

G.M. Maruyama, 1998. Basics of structural equation modeling. Thousand Oaks, Calif.: Sage.

D.H. McKnight, V. Choudhury, and C. Kacmar, 2002. “Developing and validating trust measures for e-commerce: An integrative typology,” Information Systems Research, volume 13, number 3, pp. 334–359.
doi: https://doi.org/10.1287/isre.13.3.334.81, accessed 10 August 2017.

M.B. Myers, R.J. Calantone, T.J. Page, and C.R. Taylor, 2000. “An application of multiple-group causal models in assessing cross-cultural measurement equivalence,” Journal of International Marketing, volume 8, number 4, pp. 108–121.
doi: https://doi.org/10.1509/jimk.8.4.108.19790, accessed 10 August 2017.

National Law Enforcement and Corrections Technology Center, 2012. “License plate recognition systems (LPRs): Function, performance, and considerations for law enforcement agencies,” at https://www.justnet.org/pdf/LPR-Report-Lowres.pdf, accessed 10 August 2017.

Nielsen, 2016. “Daily dose: Smartphones have become a staple of the U.S. media diet” (21 April), at http://www.nielsen.com/us/en/insights/news/2016/daily-dose-smartphones-have-become-a-staple-of-the-us-media-diet.html, accessed 19 February 2017.

P. Norman and L. Smith, 1995. “The theory of planned behaviour and exercise: An investigation into the role of prior behaviour, behavioural intentions and attitude variability,” European Journal of Social Psychology, volume 25, number 4, pp. 403–415.
doi: https://doi.org/10.1002/ejsp.2420250405, accessed 10 August 2017.

POMCO, 2015. “POMCO introduces the next generation of personal safety for college students” (30 September), at http://www.businesswire.com/news/home/20150930005146/en/POMCO-Introduces-Generation-Personal-Safety-College-Students, accessed 19 February 2017.

D. Potnis and K. Deosthali, 2014. “Factors influencing adoption of Web 1.0, Web 2.0, and mobile technologies by the growth engine of the U.S. economy,” First Monday, volume 19, number 9, at http://firstmonday.org/article/view/5419/4112, accessed 31 May 2017.
doi: http://dx.doi.org/10.5210/fm.v19i9.5419, accessed 10 August 2017.

P.P. Purpura, 2013. Security and loss prevention: An introduction. Sixth edition. Waltham, Mass.: Butterworth-Heinemann.

Rauschnabel, P.A., D.W.E. Hein, J. He, Y.K. Ro, S. Rawashdeh, and B. Krulikowski, , 2016. Fashion or technology? A Fashnology perspective on the perception and adoption of augmented reality smart glasses, i-com, volume 15, number 2, pp. 179194. doi: 10.1515/icom-2016-0021.

J.J. Sloan and B.S. Fisher, 2014. “Campus crime,” In: G. Bruinsma and D. Weisburd (editors). Encyclopedia of criminology and criminal justice. Hoboken, N.J.: Wliey.
doi: http://dx.doi.org/10.1002/9781118517383.wbeccj236, accessed 10 August 2017.

J.J. Sloan and B.S. Fisher, 2010. The dark side of the ivory tower: Campus crime as a social problem. New York: Cambridge University Press.

S. Stein, 2014. “Wearable tech at CES 2014: Many, many small steps” (9 January), at https://www.cnet.com/news/wearable-tech-at-ces-2014-many-many-small-steps/, accessed 19 February 2017.

T.S.H. Teo, S.C. Srivastava, and L. Jiang, 2008. “Trust and electronic government success: An empirical study,” Journal of Management Information Systems, volume 25, number 3, pp. 99–132.
doi: http://dx.doi.org/10.2753/mis0742-1222250303, accessed 10 August 2017.

G. Turhan, 2013. “An assessment towards the acceptance of wearable technology to consumers in Turkey: The application to smart bra and t-shirt products,” Journal of the Textile Institute, volume 104, number 4, pp. 375–395.
doi: http://dx.doi.org/10.1080/00405000.2012.736191, accessed 10 August 2017.

V. Venkatesh, M.G. Morris, G.B. Davis, and F.D. Davis, 2003. “User acceptance of information technology: Toward a unified view,” MIS Quarterly, volume 27, number 3, pp. 425–478, and at http://aisel.aisnet.org/misq/vol27/iss3/5/, accessed 10 August 2017.

M. Warkentin, D. Gefen, P.A. Pavlou, and G.M. Rose, 2002. “Encouraging citizen adoption of e-government by building trust,” Electronic Markets, volume 12, number 3, pp. 157–162.
doi: http://dx.doi.org/10.1080/101967802320245929, accessed 10 August 2017.

J.M. Woodside, 2015. “Wearable technology acceptance model: Google Glass,” Proceedings of SITE 2015 — Society for Information Technology & Teacher Education International Conference, pp. 1,800–1,802, and at https://www.learntechlib.org/p/150242/, accessed 10 August 2017.

H. Yang, J. Yu, H. Zo, and M. Choi, 2016. “User acceptance of wearable devices: An extended perspective of perceived value,” Telematics and Informatics, volume 22, number 2, pp. 256–269.
doi: http://dx.doi.org/10.1016/j.tele.2015.08.007, accessed 10 August 2017.

J. Zuckerman, 2014. “5 tips to help your campus select a mobile security solution” (16 May), at http://www.campussafetymagazine.com/article/5_tips_to_help_your_campus_select_a_mobile_security_solution/blog, accessed 19 February 2017.

 

Appendix: Measurement items

Performance expectancy (Davis, et al., 1989)

PE1: Using POM would enable me to report incidences/emergencies more quickly.
PE2: Using POM would increase my overall safety at my College/University.
PE3: Using POM gave me greater control over emergency reporting process in this college/university.
PE4: Using POM would increase my quality of life as a student.

Effort expectancy (Venkatesh, et al., 2003)

EE1: My interaction with POM would be clear and understandable.
EE2: It would be easy for me to become skillful at using POM.
EE3: POM made it easier for me to report incidents/emergency situations
EE4: Learning to use POM will be easy for me.

Social influence (Ajzen, 1985; Lu, et al., 2005; Norman and Smith, 1995)

SI1: People who influence my behavior (e.g., my professors and parents) think that I should use POM.
SI2: My friends at school think that I should use POM.
SI3: My parents think that I should use POM.
SI4: My siblings think that I should use POM.
SI5: Members of my extended family think I should use POM.

Facilitating conditions (Venkatesh, et al., 2003)

FC1: The school has been helpful in encouraging the use of POM for the purpose of reporting incidences/emergencies.FC2: The personal safety service provider (the maker of POM) has supported the use of POM.
FC3: I have the resources necessary to use POM for the purpose of reporting incidences/ emergencies.
FC4: I have the knowledge necessary to use POM.
FC5: At my school, a specific person (or group) is available for assistance with difficulties associated with the use of POM.
FC6: Specialized instruction concerning the use of POM was available to me.

Trusting beliefs (McKnight, et al., 2002)

TB1: The POM system is dependable and reliable.
TB2: I trust the POM system because it serves my best interests.
TB3: I tend to trust the POM even though I have little knowledge of it.
TB4: Trusting the POM system is not difficult.

Behavioral intention (Armitage and Conner, 2001)

BI1: I will primarily use POM for my safety needs.
BI2: I predict many colleges will use POM in the future.
BI3: I recommend POM services to others.

 


Editorial history

Received 6 April 2017; revised 16 June 2017; revised 21 July 2017; accepted 22 July 2017.


Creative Commons License
“Students’ intention to adopt Internet-based personal safety wearable devices: Extending UTAUT with trusting belief” by Devendra Potnis, Dawit Demissie, and Kanchan Deosthali is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Students' intention to adopt Internet-based personal safety wearable devices: Extending UTAUT with trusting belief
by Devendra Potnis, Dawit Demissie, and Kanchan Deosthali.
First Monday, Volume 22, Number 9 - 4 September 2017
http://www.firstmonday.org/ojs/index.php/fm/article/view/7808/6524
doi: http://dx.doi.org/10.5210/fm.v22i19.7808





A Great Cities Initiative of the University of Illinois at Chicago University Library.

© First Monday, 1995-2017. ISSN 1396-0466.