Public Self Service Technology (SST): Designing for Trust: Factors enhancing user’s trust towards a public SST

(1)

IN

DEGREE PROJECT INFORMATION AND COMMUNICATION TECHNOLOGY,

SECOND CYCLE, 30 CREDITS STOCKHOLM SWEDEN 2018,

Public Self Service Technology (SST):

Designing for Trust

Factors enhancing user’s trust towards a public SST

DEEPIKA DIPESH DUGAR

KTH ROYAL INSTITUTE OF TECHNOLOGY SCHOOL OF ENGINEERING SCIENCES

(2)

(3)

Public Self Service Technology (SST): Designing for Trust

Factors enhancing user’s trust towards a public SST

Deepika Dipesh Dugar

Master of Science Thesis

School of Information and Communication Technology KTH Royal Institute of Technology

Stockholm, Sweden

25 September 2018

Examiner: Associate Professor Fredrik Kilander

Industrial Supervisor: Theo Franz ´en (Fidesmo AB, Stockholm, Sweden)

(4)

(5)

Foreword

This work is presented in partial fulfillment of the requirement for the degree of Master of Science at the Department of Informatiom and Communication Tech- nology Innovation (ICT Innovation), School of Information and Communication Technology (ICT), at the Royal Institute of Technology (KTH), Stockholm, Swe- den. This work was carried out in collaboration with Fidesmo AB, Stockholm, Sweden.

i

(6)

(7)

Abstract

Public Self Service Technology (SST) has become an important part of our daily life. Advancement in technology and reduced hardware costs have motivated service providers to deploy public a SST for various important and complex tasks.

Examples of such tasks include editing and printing confidential documents, performing monetary transactions, etc. These tasks requires a user to reveal his/her personal information to a public SST. The major problem while performing these tasks using a public SST is that the user has to deal with many surrounding factors like social density, privacy and security, which might affect his/her trust towards the SST and in turn the user might refrain from using it. This study aims to find different factors that can enhance user’s trust towards a public SST, encourage to use it and complete the task even if it requires user’s personal information.

The in-depth interview method was adopted for the study to learn twelve interviewees’ experience with varied public SSTs that specifically handles personal information, in an urge to understand interviewees’ behaviour, underlying moti- vations and desires to use those public SSTs. Analysing the data collected from interviews, ten trust factors were found that emerge at various stages of interaction with a public SST. They were categorized into pre-interaction, on-interaction and post-interaction factors based on their time of interaction. Beyond the trust factors, three additional important factors have emerged from interview data, which motivates users to adopt public SSTs. They are usefulness, convenience and personality-based trust. The results may be valuable for researchers who are fo- cusing on different aspects of trust and any public artefact as well as for service providers and designers to design a trustworthy public SST. Based on the results, some practical implications for designing public SSTs are also presented.

Keyword: Public Self Service Technology (SST), trust, personal information, risk, situational context, Interaction design

iii

(8)

(9)

Sammanfattning

Offentlig självbetjäningsteknik (eng. Public Self Service Technology) har blivit en viktig del av v˚art dagliga liv. Förbättring i teknik och minskade h˚ardvarukostnader har motiverat tjänsteleverantörer att distribuera offentliga SST för olika viktiga och komplexa uppgifter. Exempel p˚a s˚adana uppgifter är att redigera och skriva ut konfidentiella dokument, utföra monetära transaktioner etc. Dessa uppgifter kräver att en användare delar med sig av personlig information till en offentlig SST. Det stora problemet med att utföra dessa uppgifter med hjälp av SST är att användaren m˚aste hantera m˚anga omgivande faktorer som social täthet, integritet och säkerhet, vilket kan p˚averka personens förtroende mot SST och i sin tur kan användaren avst˚a fr˚an att använda den. Följaktligen syftar denna studie till att hitta olika faktorer som kan öka användarnas förtroende mot offentliga SST, upp- muntra att använda dem och slutföra uppgifter även om det kräver användarens personuppgifter.

En djupintervjumetod användes i studien för att lära sig om tolv testdeltagares erfarenhet av olika offentliga SST, specifikt de som hanterar privat information, i en strävan att först˚a deltagarnas beteende, underliggande motivation och önskem˚al att använda dessa offentliga SST. Analys av de uppgifter som samlats in fr˚an inter- vjuer, upptäckte tolv förtroendefaktorer som uppst˚ar vid olika stadier av interaktion med en offentlig SST. De var därför uppdelade i pre-interaktion, interaktions- och post-interaktionsfaktorer. Fr˚an intervjuerna har ytterligare tre viktiga faktorer upptäckts som motiverar användarna att använda offentliga SST. De är använd- barhet, bekvämlighet och personlighetsbaserat förtroende. Resultaten är värde- fulla för forskare som fokuserar p˚a olika aspekter av förtroende och offentliga artefakter samt för tjänsteleverantörer och designers för att utforma en p˚alitlig offentlig SST. Baserat p˚a resultaten presenteras ocks˚a n˚agra praktiska konsekvenser för utformningen av offentliga SST.

v

(10)

(11)

Acknowledgments

My sincere thanks to my thesis academic supervisor Patric Dahlqvist and examiner Assoc. Prof. Fredrik Kilander. (School of Information and Communication Technologies, KTH Royal Institute of Technology, Stockholm, Sweden) for their valuable suggestions and recommendations. I am very grateful for their timely feedbacks. My heartfelt gratitude to my industrial supervisor Mr. Theo Franz´en (UX Designer at Fidesmo AB, Stockholm, Sweden) for sharing his valuable ideas and continuous feedback throughout the study and also helping me in conducting interviews. Much appreciation to the entire Fidesmo team for thoughtful advices and encouragement. Special thanks to Angel Anton, IOS developer at Fidesmo, for helping me in designing the Fidesmo App’s UI.

A very special thanks to Dipesh Mitthalal for thoughtful advices, language proofreading and most importantly his encouragement during the study.

Finally, I thank my family and friends for supporting me throughout this en- deavor.

-Deepika Dipesh Dugar

vii

(12)

(13)

List of Figures

1.1 Methodology Diagram . . . . 4

2.1 Relationship between trustor and trustee . . . . 10

2.2 Theory of Reasoned Action . . . . 11

2.3 Technology Acceptance Model . . . . 12

3.1 Userflow Diagram of Fidesmo’s public SST . . . . 18

3.2 Physical appearnece of Fidesmo’s public SST . . . . 19

3.3 Screen 1 - Welcome . . . . 20

3.4 Screen 2 - Place your device . . . . 21

3.5 Screen 3 - Enter your card details . . . . 22

3.6 Screen 4a - Verify Identity . . . . 22

3.7 Screen 4b - Verify Identity (entering OTP code on SMS verification) 23 3.8 Screen 5 - Installing card . . . . 24

3.9 Screen 6 - Finish . . . . 24

5.1 Thematic Map: Overall trust factors that enhance user trust towards a public SST . . . . 36

5.2 Enter your card details UI - Second Iteration . . . . 40

5.3 Fidesmo’s public SST UI with language icon - Second Iteration . 42 5.4 Physical appearnece of Fidesmo’s public SST - Second Iteration . 44 5.5 Verify identity UI - First Iteration . . . . 45

5.6 Verify identity UI - Second Iteration . . . . 46

5.7 Installing card UI - First Iteration . . . . 47

5.8 Installing card UI - Second Iteration . . . . 47

5.9 Finish UI - Second Iteration . . . . 48

5.10 Thank you UI - Second Iteration . . . . 48

xiii

(18)

(19)

List of Acronyms and Abbreviations

SST Self Service Technology

FPSST Fidesmo’s Public Self Service Technology

UI User Interface

TRA Theory of Reasoned Action

TAM Technology Acceptance Model

SMS Short Message Service

OTP One time password

ID card Identification card

xv

(20)

(21)

Chapter 1 Introduction

Self Service Technology (SST) is any technological interface, which enable users to perform tasks without the presence of the service provider and generate benefits for themselves [1, 2]. Collier et al. [3] classified SST into private and public SST based on the location and potential for social interactions. A private SST is located at private place like home or private office cabins where a user can inter- act with the SST without any interaction with others. Examples of private SST include the telephone banking access, e-commerce, internet banking. Conversely, a public SST is located in public space, where social interaction can also take place between the user and other people around. Examples of public SST include ATMs, SST to order food at restaurants, self-checkin and self-checkout kiosks.

Advances in technology, reduced hardware costs and increased labour costs have motivated service providers to deploy different types of SST based on its purpose.

Public SSTs are becoming ubiquitous in modern society offering fast, consis- tent, and accurate service to the user. At the time of writing this report, public SSTs are used for more complex and secured tasks during which the user is in- tended to enter his/her personal information like name, address, credit card details, bank details, medical information, etc. Taking food orders and collecting payments, editing and printing certified documents, purchasing a travel ticket, are some of the applications of public SSTs that require the user to enter his/her personal information. Revealing personal information in the presence of a stranger on a public display is stressful [4]. It is a matter of great sensitivity, and forms a possible threat of this information being observed and misused. This threat can lead to a psychological barrier for an individual to carry out any monetary or information transaction, if there is a lack of trustworthiness and security. The user may refrain from using a public SST, which requires him/her to reveal personal information. As a result, SST looses its purpose of deployment.

1

(22)

Fidesmo’s Public SST (FPSST)

Fidesmo AB, a technology company based in Stockholm (Sweden), wants to launch a public SST at stores. This SST will help a user (a customer) to perform tokenization. The tokenization is a process to transfer a reference of a original credit/debit payment card to a device [5]. Hence, by using the Fidesmo’s public SST (FPSST), a user (a customer) can transfer a reference of his/her original credit/debit card to a Fidesmo enabled device like a watch, and this will activate the contactless payment feature in the watch. This tokenization process entails the user (the customer) to enter his/her personal information like credit card number and its security code into the FPSST. This situation of revealing personal information at stores might put the user in a stressful moment due to the fear of losing the information to any unknown person in the surrounding. The user may not like to be observed by other people while entering personal information. The user may feel stressed, which may influence his/her attitude to further use the FPSST.

According to Artz and Gil, trust is an integral part in any kind of interaction, allowing users to act under uncertainty and with the risk of negative consequences [6]. Trust help users to overcome the perception of uncertainty and risk in the acceptance and use of any technology [7]. A Public SST is deployed to enhance customer service but it is also associated with risk and negative consequence.

Hence, it is necessary to design a public SST such that it builds the user’s trust under a uncertain surrounding situational context. This will help the user to accept and use the public SST. Therefore, the biggest challenge for Fidesmo is to build a trustworthy FPSST such that it motivates the user (the customer) to accept it. Consequently, Fidesmo can better serve their customers. To build trust, it is important for Fidesmo to identify different factors that can increase a user’s trust towards their SST.

1.1 Goal

The main goal of this study is to identity different trust factors and provide guide- lines based on the identified trust factors to designers and service providers which will help them to design a trustworthy public SST.

1.2 Problem Statement

There has been much research conducted on public SSTs to identify different factors that influence user’s attitude towards the adoption of a public SST [1, 3, 8].

(23)

1.3. RESEARCHQUESTION 3

Research has also been conducted to study different factors and their relationships that increases user’s satisfaction with a public SST experience [9,10].

However, how to evolve the feeling of trust, and how to enhance it? This is an area that has still remained largely untouched by scientific research, until now.

There has been no major study committed to the issue of trust as regards the use of the public SST that handles personal information. Collier et al. [3] has suggested for a future research considering trust as one of the constructs that influence adoption of a public SST. Excluding Dabholkar and Spaid’s work on public SST [11], other workers have not considered the issue of revealing personal information to a public SST in public space. Moreover, all the above study discussed used web-based survey as their primary source of data collection except Meuter et al.[9]. On the other hand, Little and Briggs [4] suggest to consider psycological constraints of users while designing a trustworthy public SST. Therefore, due to lack of proper research, service providers like Fidesmo, find it difficult to design a trustworthy public SST. Having a design without considering the trust factors, the public SST may be left unused, thereby loosing the purpose of having a public SST.

1.3 Research Question

What are the key factors that increases user’s trust towards a public SST usage, especially when a user is required to disclose his/her personal information in the public SST?

1.4 Purpose

The purpose of this study was to learn about trust in the form of enquiring users about their experience while providing personal information to anonymous third party systems in public spaces and through online. Also, understand users’ behaviour patterns concerning their use of different public SSTs specifically thst handles personal information. This includes ATMs, gas station SST, parking ma- chines, self-checkin, and self-checkout SSTs, FPSST.

1.5 Methods

This section introduces the research methods used in this study. The study consists of theoritical background review, iterative design process through evaluation, data

(24)

collection and result analysis. Figure 1.1 gives an overview of various methods used at different stages.

Figure 1.1: Methodology Diagram

Given the fact that there are wide variety of types and purposes of public SSTs, it is critical to understand how users feel about them, how and why they use some, and why they don’t use others? The primary focus of this study was to find sources of trust underlying various public SSTs experiences. This directs to explore user’s behaviour, beliefs, opinions, emotions, and relationships of individuals while using different types of public SSTs. Therefore, based on the analytical objective and type of data to be collected, a qualitative research method was more appropriate for this study. A semi-structured in-depth interview method was adopted for interviewing interviewees face-to-face which helped to collect a wealth of detailed data from interviewees.

(25)

1.6. CONTRIBUTIONS 5

An initial mockup of Fidesmo public SST was designed based on Fidesmo’s business case. This mock-up was used during the interviews. Each interview was divided into two phases. In the first phase of an interview, through a sce- nario based role play, an interviewee was encourgaed to use the inital mock-up of FPSST. During the use of the mock-up, personal information like credit card number and its security code was inquired from the interviewee. The idea of going through the initial mock-up in the first phase of the interview was to evoke questions about the interviewee’s willingness to provide personal information in the public space when the service required it. In the second phase of the interview, an interview guide was used. The focus of the second phase of the interview was to learn about similar experiences with other public SSTs of the interviewee. It was thought that using this inital mockup of FPSST as a tool of reference, the interviewees would be reminded of some related real-life situations with other public SSTs. This would evoke discussion about those situations during the second phase of the interview. Hence, by comparing different experiences with differnt public SSTs, the main aim was to find various factors that enhances user’s trust towards a public SST.

Two iterations of interviews were conducted. The first iteration of interviews was conducted with six interviewees in an office lobby. During this iteration, some of the usability and trust issues of FPSST were identified. But interestingly, none of the interviewees discussed much regarding the surrounding environment and its impact on using the Fidemso’s public SST. This type of results demanded more real situation to test the SST. Therefore, the mockup was redesigned based on the design flaws and trust triggers found in the first iteration. Using the redesigned mockup, the second iteration of interviews was conducted in a shopping mall with another six interviewees. As the core intention of interviews was same for both the iterations, data of both the iterations was analysed together. Using a thematic analysis technique, data were sorted into categories and subcategories that reflected the factors influencing trust while using a public SST.

1.6 Contributions

This thesis project makes the following contributions:

• Identified various trust factors contributing towards designing a trustworthy public SST handling personal information

• The mockup of the Fidesmo’s public SST was designed and used as a tool of reference for this study

(26)

1.7 Delimitation

Due to the time constraint, this study dealt only with public SSTs that handles financial information. The public SSTs that handle other types of personal information like health information were not dealt with.

1.7.1 Ethics and Sustainability

Since the research involves human subjects and private information in public space, the interviewees’ identity were kept anonymous and also interview questions were designed in such a way that interviewees were not expected to reveal any of their personal information.

Furthermore, the environment in which a public SST operates is exposed to privacy breaches or information thefts. However, deploying the SST reduces required manpower and increase companies’ and users’ efficiency. Additionally, environmental benefits can be achieved by manufacturing one device for thousands of users rather than thousand devices. Before all these effects can be achieved though, research like this needs to be conducted to ensure safety while using the public SST.

1.8 Outline

The rest of the thesis is organinzed as follows. Chapter 2, begins with an detailed description of the term ”Trust”, followed with a brief explanation of two behaviourial theories, Theory of Reasoned Action (TRA) and Technology Ac- ceptance Model (TAM), and then followed with a theoritical jusification of relationship between trust, acceptability and self-disclosure, therefore, the chapter provides the entire theoretical background for this study.

Chapter 3 explains the design process of the Fidesmo’s public SST. The chapter contains the reasoning behind the design decisions of each user interface and the physical interaction design of the SST.

Chapter 4 gives a explanation for choosing the in-depth interview method as an evaluation method by comparing with other qualitative research methods followed by complete explanation of the in-depth interview method. Later, the chapter outlines the entire evaluation plan from selecting interview location, recruiting interviewees and conducting interviews. Finally, the short description is given

(27)

1.8. OUTLINE 7

about the data analysis of collected data.

In Chapter 5, the results are analysed in detail.

Chapter 6 includes discussion of results and pratical implications to guide service providers and desigers to design a trustworthy public SST.

Chapter 7 consists of the remaining problems and expected follow-up studies.

(28)

(29)

Chapter 2 Theoretical Background

This chapter introduces trust, which is a core concept of this study. This is followed by an explanation of two behavioural theories, TRA and TAM. These theories will aid in understanding how different factors associated with a public SST can influence user’s trust and so changing user’s behaviour to use the SST. Us- ing e-commerce as an example, many earlier works are reviewed to learn how different elements of e-commerce develops user’s trust towards it and encourage the user to provide personal information when required. This built a motivation to find how to improve trust for a public SST. This chapter describes important concepts and strategies which forms a background of the study.

2.1 Trust

For years, trust has been studied in a variety of disciplines (including psychology, sociology, economy and technology) for understanding relationships between humans or between humans and organizations or between human and machine. Trust proved to be a critical factor for any relationship. The variety of contexts within which the trust has been studied led to various definitions and theories of trust.

However, there is a lack of agreement on a universal definition [12, 13]. But, the fundamental elements of trust’s definition has always been the “willingness to be vulnerable” [12, 14, 15, 16] under any uncertain situation. Hence, trust is best conceptualized as a psychological attitude. It involves a trustor’s beliefs and expectations about a trustee’s trustworthiness derived from the trustor’s experience and interactions with the trustee in situations involving uncertainty and risk [12,17,18]. The Trustor is the one who places trust and the trustee is the one who maintains the trust as shown in figure2.1. In the context of the public SST, the trustor is a public SST’s user while the trustee is public SST provider.

9

(30)

Figure 2.1: Relationship between trustor and trustee

Trust has also been interpreted as the trustor’s cooperation, confidence, and positive expectations from the trustee [19]. Trust has proved to be one of the predictors of overall satisfaction [20]. Hence, trust creates a positive attitude towards the trustee that are likely to reduce the fear of trustee’s opportunism and other related concerns [12,21]. Researchers claimed that trust would not be required if actions were performed with complete certainty and without any risk, hence the trustor need not be dependent on the trustee [22, 23]. Therefore, risk and interdependence are main conditions at which trust occurs, where risk is a probability of perceived loss, and interdependence is where the goal of the trustor cannot be achieved without relying on the trustee [12, 16, 21]. Moreover, Rousseau et al [12] expressed trust and risk as a path dependent connection where risk creates an opportunity for trust, which inturn leads to risk taking. They also proclaimed, trust is dynamic that changes based on past experience, present expectations, and the surrounding environment. Carsten D Schultz [18] developed a trust equation to calculate trust level of the trustor at any given time and situation, which incor- porates the trustor’s past experience with the trustee, current trust situation and the trust object around which the trust of the trustor revolve. This trust framework model showed that the basic concept of trust is the same for all the disciplines and by including different situation context, trust can be formulated for all disciplines.

Reduced levels of perceived risk and enhanced trust of the trustor will contribute greatly to the acceptance and continued relationship with the trustee [24].

2.2 Theory of Reasoned Action

The Theory of Reasoned Action (TRA) is a model developed by Fishbein and Ajzen [25]. Using this theory, an individual’s behaviour can be determined by his/her intention to perform that behaviour. This intention is further determined by the individual’s attitude towards the behaviour and his/her subjective norms associated with that behaviour. The attitude is defined as the individual’s belief that performing a particular behaviour will produce specific results and subjective norms is defined as the person’s perception that most people who are important to him/her think he/she should or should not perform the behaviour. The TRA also claimed that stronger intentions increase the effort to perform the behaviour,

(31)

2.3. TECHNOLOGY ACCEPTANCE MODEL 11

which in turn increases the likelihood of the behaviour to be performed. TRA proved to be a useful model in the social psychology discipline as it can explain and predict the actual behaviour of an individual [26]. Below, figure2.2 demon- strates the same.

Figure 2.2: Theory of Reasoned Action

2.3 Technology Acceptance Model

Fred D.Davis [27] adapted the TRA model to the context of user acceptance of an information system and developed a model called Technology Acceptance model(TAM). According to this model, actual use of the system is considered to be the actual behaviour. Hence, the user’s behaviour of using the system is determined by his/her intention to use the system and, his/her attitude towards the system. However, Davis made two main changes to the TRA model. First, he dropped subjective norms as they are the least understood aspects of TRA. Sec- ondly, he considered only two distinct beliefs; perceived usefulness and perceived ease of use, which were sufficient enough to predict the user’s attitude to use the system. He argued that people tend to use a system only on the belief that using the system will increase their efficiency (perceived usefulness) and also that the belief of the efforts required to use the system (perceived ease of use) might affect the behavioural intention to actually use the system. Hence, he defined perceived usefulness and perceived ease of use as follows:

• Perceived usefulness - the degree to which a person believes that using a particular system would enhance his/her job performance [27].

• Perceived ease of use - the degree to which a person believes that using a particular system would be free of effort [27].

With the further research on TAM model, it was shown that perceived usefulness and perceived ease of use were found to build a stronger behavioural intention

(32)

directly thus eliminating the need for the attitude construct. Beside eliminating attitude, external factors like system characteristics, user training, user participation in design, and the nature of the implementation process were included to the TAM model. It was claimed that these external factors might influence the person’s belief towards the system. Below figure2.3demonstrate the extended model.

Figure 2.3: Technology Acceptance Model

Due to its simplicity and ease of understanding, TAM was further adapted and extended in many ways to learn user’s behavioural intention to accept any technology. In its earlier stage, TAM was applied to office applications like Email, word processor, spreadsheet, fax, etc [26]. Later, with the development of internet, TAM was more widely applied to different implementation of SST (public and private) for the use of information kiosk, e-commerce, internet banking, ATM, etc. [26]. Hence, it found its roots in the self-service technology (SST).

2.4 Trust, acceptability and self-disclosure in SST

Applying TAM to SST that handles personal information (eg. ATM, internet banking, e-commerce, printing Identification cards (ID cards), etc.) has its own challenges. These SSTs handles personal information, such as contact and address information, social security number, credit card information, or bank account details over internet in order to complete the purchase of a product or service. Little et al. [4] conducted a short survey to determine the sensitivity of different types of personal information. Financial information like credit card details was rated as second most sensitive personal information among four categories. Due to the virtuality of service providers, online exchange of information using SST involves great risk and uncertainty. This being the case, users are concerned about the privacy of their personal information and also has the security concern about the third party having access to their information. In such online business transactions, trust is a vital factor to retain the customers (users) relationship [21,28,29]. However,

(33)

2.4. TRUST, ACCEPTABILITY AND SELF-DISCLOSURE INSST 13

unlike traditional business setting, SST is the primary source of interaction for users. Therefore, trust is even more crucial with the SST because the limited user interface does not allow users to judge its service provider’s trustworthiness.

Considering the business-to-consumer relationship in e-commerce, the fundamental lack of trust has been claimed as a prominent factor for the growth of e-commerce and why consumers were reluctant to perform online shopping [30].

On the other hand, there has been paradigm shift in e-commerce from security to trust and with trust identified as the ultimate and decisive factor for the consumer’s willingness to actually carry out online transactions [31]. With the grow- ing e-commerce technology and consumers reluctance to use e-commerce, it was important for e-vendors to build an online trust in order to alleviate consumers perceived risk and accept the new technology. Therefore in the last two decades, trust received considerable attention in the e-commerce research.

Researchers recognized the influence of trust on e-commerce adoption because the trust was believed to be a fundamental factor to diminish the uncertainty and risks of the e-commerce adoption [21, 28]. Hence, researchers inte- grated trust into TAM and studied its relationship with different components of e-commerce that influence e-commerce users’ attitude [13,21,28,32]. Cognition trust, personality trust, institutional trust, experience trust, and economic trust are five types of trust bases which were identified as sources of trust in e-commerce.

Their direct impact on the users’ behavioural intention to use e-commerce has also been studied [28,33,7]. Classifying uncertainties in online environment into behavioural and environmental uncertainty, Pavlou et al. [21] examined users’

behavioural intention on the basis of perceived risk and trust. He found that trust in the e-vendor reduce the perceived level of risk associated with the transaction process and the associated infrastructure. Therefore, this motivates users’ intention to complete the transaction. Further, Kim et al. [33], uncovered the role of trust, risk and their antecedents affecting e-commerce users’ purchasing deci- sion. Trust being dynamic and a subjective assessment of the users’ perception of the e-vendor, researchers divided the trust building process into 3 stages: 1) Building trust before interacting with e-commerce site, 2) Building trust while interacting with e-commerce site and 3) Building trust after interaction with the site [18, 34]. Hence researchers studied to foster consumer’s trust at different stages of interaction with e-vendors. This would help e-vendor not only to build initial trust but to overcome trust at any stages. Some studies try to build initial trust in the case of first-time buyers as they will perceive higher risk due to the lack of any earlier interactions with the e-vendor [35, 36]. Many studies highlighted, various website components that act as a circumstantial cues for consumers to assess their e-vendor’s trustworthiness. Some studies also claimed self-efficacy

(34)

as the trust trigger. Researchers also demonstrated that once users develop trust for online sites with perceived privacy that in-turn increases users’ willingness to disclose their personal information to complete the transactions or process on internet [37,38]. Similar studies have been done which proposed trust as one of the determinants in individual’s acceptance of other private SSTs like internet banking, e-recruitment [38,39,40].

Hence, different trust factors and their effects have been studied and consequently enhanced user’s trust to provide his/her personal information while using a private SST. However, little attention has been given on how to enhance trust towards a public SST. A public SST requires a user to enter his/her private and monetary information in a public space and in a public device. There are various problems a user has to face while using a public SST which in turn effects the user’s intention to use the SST. Unlike a private SST, where user use his/her own system at home/office and is not distracted by noise or shoulder surfing; these are some major challenges while using a public SST. The presence of other people in the surroundings or in the queue behind, effects the user’s intention to enter personal information in the public space. Using a public device to buy a product or attain a service by entering personal information may affect the user’s trust on the same. While using a private SST, user has the ability to stop the process any time whereas while using a public SST, the presence of others waiting in the queue may limit such decisions. Dabholkar and Bagozzi has shown that in-store anxiety caused by the presence of other users, lowers a public SST’s evaluations by customers [8].

Meuter et al. [9] have learned various users’ satisfaction and dissatisfaction experience while using different public SSTs. The study help them to identify numerous factors that leads to satisfying and dissatisfying experience. The result from the study was further used by Joel E.Collier and other workers to examine the effect of control and convenience in the user’s satisfying experience with a public SST [3,10,41]. Trust in the service provider was recognized as one of the predictors of the satisfying experience and further intent to use a public SST [41].

There has also been much research conducted on a public SST to identify different factors that influence user’s attitude towards and adoption of a SST [8,1, 3].

Rensel et al. studied various physical and virtual facilitating conditions that has impact on usage of public computer for online transactions [42]. Various factors have been identified that impacts the user’s experience during a public SST failure and it is found that a clerk’s assistance and a low anxiety environment reduce the user’s negative attribution towards a public SST [8]. Therefore, by summarizing all the above studies it is evident that there is a lack of studies on how to evolve trust and enhance it towards a public SST. In addition, not many of these studies

(35)

2.4. TRUST, ACCEPTABILITY AND SELF-DISCLOSURE INSST 15

gave attention to the context of user’s providing personal information while using a public SST. Dabholkar and Spaid [11] suggested in their future research to consider a public SST that handle important personal information due to their grow- ing importance. Collier et al. [3] proposed examining the trust that user place in a public SST. He claimed that it would be a fruitful pursuit in understanding user behaviours. Previous research has also recommended for exploring variables like peer group influence, knowledge efficacy, privacy, and security that influence trust in the use of public SSTs, e.g. ATM [43]. Considering all the above suggestions, this study explores the relationship between personal information like financial information in public space and trust in digital devices handling personal information. This study also aims to learn how to generate and increase trust towards a public SST. How does the interface design affects user trust in public place? How do these experiences affect future behaviours? Do a user plan to use the specific company’s public SST, Why or why not? Because so little has been published about trust and public SST, I explored these questions to further understand the complexities surrounding trust and a user’s interactions with a public SST while disclosing personal information.

All the above specified researches used the online survey as their evaluation method for data collection. These researches lack a direct interpretation from users about their behaviour and thoughts regarding trust and a public SST usage.

Florian N. Egger [34] suggest involving users during the design process to yield the high-level user requirement as trust components. Hence, a qualitative method like in-depth interviews was used to collect deeper insights from users. Accord- ing to Dabholkar and Spaid [11], it is of greater importance to collect direct user input in the development and testing of a public SST due to the sensitivity of information shared. Moreover, businesses can better understand the impact of this technology directly from their customers. This will have a important implication on customer-firm relationship.

In summary, trust is a decisive factor in many forms of human interaction where it can help user to overcome the perceived risk and its negative consequences. TRA and TAM helped to understand how related surrounding factors of a task can change the human attitude and intention to perform the task. Using e-commerce as an example, it was shown that various factors of trust increases a user’s trust to use e-commerce and encouage to provide personal information when required. Hence, it was realized that enhancing trust will increase a public SST adoption especially that handles important personal information.

(36)

(37)

Chapter 3 Design Process

Since FPSST was used as a tool of reference in interviews, a mock-up of FPSST was designed and built during the study. In this chapter, the design process of FPSST’s initial mock-up is described in detail. This chapter is divided into two sections. First, the physical interaction design section explains the design process of the FPSST’s physical appearence. Followed by, the User Interface design section explaining the design process of the FPSST’s User Interface (UI). Design decisions are documented along with the design process. This will help to understand the reasoning behind the specific design.

3.1 Initial mock-up of FPSST

FPSST will be used for the tokenization process that helps to activate the contactless payment feature in Fidemso enabled watch. To do this, a user has to place the watch on a NFC reader and enter his credit card details on the SST’s screen. Then, through the tokenization process, a reference of the original credit card is developed and transfered to the watch through the NFC reader. This will activate the contactless payment feature in the watch. Based on the above process description, I designed a userflow diagram as shown in Figure3.1, which gives the overview of FPSST. Later, each step of the userflow was sketched as a seperate screen. Once the sketches were confirmed with the industrial supervisor, I designed and developed the mock-up based on the sketches.

The Jakob Nielsen’s usability heuristic principles [44] were considered while designing the initial mock-up of FPSST. Assuming a well designed mock-up would lead participants to speak more about trust related issues rather than usability issues. FPSST will be placed in the public place and users will be in the

17

(38)

Figure 3.1: Userflow Diagram of Fidesmo’s public SST

standing posture while using it. Therefore, FPSST was designed for the purpose of being understandable and easy to use [45,46]. The following sections describes FPSST’s physical interaction and UI design.

3.1.1 Physical Interaction Design

FPSST requires two devices to complete its specified process as shown in Figure 3.1, a display and a NFC reader. The display is used to present all the screens’

content on it and thus the user can take appropriate action to complete the pro-

(39)

3.1. INITIAL MOCK-UP OFFPSST 19

Figure 3.2: Physical appearnece of Fidesmo’s public SST

cess. The NFC reader is used where a watch is placed and through which the process of tokenization is carried out. Since this SST will be used in public space and require user to enter personal information, an 12.9 inch touchscreen display was used. The privacy partitions were attached to its sides of the display, such that the user can cover the display with his/her body. Privacy partitions improves user’s perceived privacy [47]. In addition, the display height was also maintained between 0.9-1.1 meters using a high table of the same height [46]. The display and the NFC reader were made using paper and cardboard, as shown in Figure3.2.

The NFC reader was designed as a flat plane in a way that the user can easily place his watch in a sleeping position thus simplifying the interaction. The NFC reader is attached to the front of the display, imitating as if they are one connected device. A blue colored NFC logo was printed on the cardboard resembling it as an NFC reader.

3.1.2 User Interface Design

Since users spend very limited amount of time to explore a public SST, a simplest practical interface for FPSST was designed with an immediately understandable metaphor and ensuring that all the results are immediately apparent. Studies eval- uating the usability of websites using eye-tracking techniques have shown that users look for significantly on the left side of web pages for longer time [48].

Keeping this in mind, all UI screens were divided into two panes. The left pane displaying the flow of the process with the current step being highlighted in bold

(40)

letters. This will help users to easily grasp the overview of the entire process [49], also user will be well informed about the next step to be performed. The right pane displays the content of current step. Both the panes were divided with a color difference.

3.1.2.1 Screen 1 - Welcome

Figure 3.3: Screen 1 - Welcome

Being a first step, the welcome screen as shown in Figure 3.3 was designed indicating start of the process. A clear text message with a pointer image was used to indicate that by clicking anywhere on the screen, the user can start the process [43]. After clicking, the user is taken to the next screen. At the bottom of the welcome screen, all associated third party logos are displayed to inform the user about the key partners of the FPSST.

3.1.2.2 Screen 2 - Place your device

In this step, the user places his/her watch on the NFC reader. To avoid any usability confusion, a graphical approach was taken to design this screen as shown in Figure3.4. This screen was designed resembling the physical design (Figure3.2)

(41)

of the FPSST’s reader representing the position of the watch on the reader, where the user has to place his/her watch to activate it. The down arrow indicates that the reader is below the display and the watch has to be placed on it. The blue color ring was used as a signifier matching it with the NFC logo on the reader. Once the NFC reader detects the watch, the user is taken to the next screen.

Figure 3.4: Screen 2 - Place your device

3.1.2.3 Screen 3 - Enter your card details

At this step, user provides his/her credit/debit card details to perform the process of tokenization. Hence this is a crucial step as the user reveals his/her personal information in public space. Keeping this in mind, the screen had to be designed using UI elements, such that user is familiar with the UI thereby avoiding any in- convenience to complete the step. According to Gefen et al. [28], if a task appears familiar to the user, then the user map things easily into his existing cognitive memory, making the task easier to perform. Adapting this philosphy, Screen 3 (Figure3.5) was designed. This step is quite similar with the e-commerce checkout process.

(42)

Figure 3.5: Screen 3 - Enter your card details

Figure 3.6: Screen 4a - Verify Identity

(43)

3.1.2.4 Screen 4 - Verify Identity

This step is usually a part of an e-commerce check-out process, so as to verify a user’s identity after providing his/her credit card details. Therefore as shown in Figure3.6and3.7, this step is also well known and hence designed in a very similar fashion as e-commerce sites to avoid any confusion in interaction. A detailed text message is also displayed for all the users.

Figure 3.7: Screen 4b - Verify Identity (entering OTP code on SMS verification)

3.1.2.5 Screen 5 - Installing card

In this step, a reference of the credit/debit card is transferred to watch to activate the contactless payment feature thereby completing the process of tokenization.

As shown in figure3.8, the right arrows are meant as the indication for data transfer from the display screen to watch and hence activating the contactless payment feature in the watch. During this step, the user is cautioned not to remove his/her watch from the reader using a message, ”Please do not move your watch from the reader”.

(44)

Figure 3.8: Screen 5 - Installing card

3.1.2.6 Screen 6 - Finish

Figure 3.9: Screen 6 - Finish

(45)

Finish screen (Figure3.9) was designed to depict the completion of the tokenization process. The title “Finish” and the tick icon indicates that the process is completed. A text message is also displayed for the confirmation. An image of card reader and watch with logo similar to NFC represents that watch is enabled for contactless payments. A text description has also been added to avoid any misunderstanding.

To summarize, based on different requirement and design principles, the initial mock-up of FPSST was explained in this chapter. The design was evaluated as per the evaluation method explained in chapter4. Based on the results of first iteration of interviews, the initial mock-up of FPSST was re-designed. The second iteration of interviews were conducted to evaluate the final mock-up as explained in chapter 5.

(46)

(47)

Chapter 4 Evaluation Method and Evaluation Plan

This chapter starts with details of evaluation methods used for this study. Various research methods were compared and the in-depth interview method was chosen to find trust factors influencing user’s attitude towards a public SST. Followed by the evaluation method, an evaluation plan is also explained. The Evaluation plan consisted of recruitment of interviewees, selection of the interview location, and how interviews were conducted. Using this evaluation plan, two iteration of interviews were conducted. After conducting both iterations of interviews, data was analysed using a thematic analysis technique.

4.1 Selection of the Evaluation Method

Trust is indeed a multidimensional and complex concept [50] and this study being one of the first to find trust factors with regards to a public SST, it was important to explore users’ viewpoints and experiences with various public SST rather than predefined hypotheses. Therefore, with an intention to learn interviewees’

experiences with various public SST and assuming that hypotheses would have reduced interviewees’ freedom to express their thoughts, a qualitative research method was preferred. The ability of the qualitative research to provide complex and rich textual descriptions of how people experience a given research issue [51]

was relevant for the study.

After comparing various qualitative research methods like focus groups, in- depth interview, contextual inquiry, and diary method, the in-depth interview method was best suited for this study. Due to the short time span of research period, the diary method was not feasible to use. The remaining three methods

27

(48)

are helpful to collect valuable insights in short period of time but the focus group and contextual inquiry had their own drawbacks for this particular study. One of the main principles of the contextual inquiry is to conduct the inquiry while an interviewee is naturally using an artefact at their own work environment [52].

This didn’t fit the current evaluation situation, as the FPSST is a new artefact and the interviewee will be using it for the first time. This may influence his/her attitude towards the FPSST. Consequently, may affect the result of the study. On the other hand, due to the fact that during a focus group interview, a group of people are interviewed together [51], this was not best suited for acquiring information on socially sensitive topics like trust. Therefore, the in-depth interview being one-on-one interview technique was the most appropriate method. The method enabled participants to feel free to share all their personal views and experiences on the sensitive topic of trust. The method was helpful to get deeper insights from participants. The following subsection4.1.1 gives a brief theoretical description of in-depth interview method.

4.1.1 In-depth Interview

The in-depth interview is a face-to-face interview technique that involves an interviewee (a person who is interviewed) and an interviewer (a person who performs the interview). This is an open-ended, discovery-oriented method to elicit a vivid picture of the interviewee’s perspective on the research topic [53]. This method aims to explore the connections and relationships the interviewee sees between particular events, phenomena, and beliefs related to the research topic. There- fore, the interviewer has to try his/her best to learn everything the interviewee can share. This kind of information can be gathered by engaging the interviewee in a casual one-to-one conversation and allowing them to answer in their own words.

Interviewer listens attentively to the interviewee’s responses, and ask follow-up questions based on those responses. They do not lead the interviewee according to any preconceived belief, nor do they influence the interviewee’s reply by ex- pressing approval or disapproval of what they say. It is an effective method for getting the interviewee to share his/her opinions and beliefs. This leads to an opportunity to get clear insights of the interviewee’s perception about the world related to the research topic [51].

To conduct the in-depth interview, the interviewer priorly prepares an interview guide. This helps the interviewer to structure the interview and ensure that he/she doesn’t forget any important questions during the interview [53]. But, the interviewer can reorder the questions if it better suits the flow of the discussion.

The interview guide emphasizes on exploration and the depth of information that

Public Self Service Technology (SST): Designing for Trust: Factors enhancing user’s trust towards a public SST

Public Self Service Technology (SST):

Designing for Trust

Factors enhancing user’s trust towards a public SST

DEEPIKA DIPESH DUGAR

Public Self Service Technology (SST): Designing for Trust

Foreword

Abstract

Sammanfattning

Acknowledgments

Contents

List of Figures

List of Acronyms and Abbreviations

Chapter 1 Introduction

Fidesmo’s Public SST (FPSST)

1.1 Goal

1.2 Problem Statement

1.3 Research Question

1.4 Purpose

1.5 Methods

1.6 Contributions

1.7 Delimitation

1.8 Outline

Chapter 2

Theoretical Background

2.1 Trust

2.2 Theory of Reasoned Action

2.3 Technology Acceptance Model

2.4 Trust, acceptability and self-disclosure in SST

Chapter 3

Design Process

3.1 Initial mock-up of FPSST

Chapter 4

Evaluation Method and Evaluation Plan

4.1 Selection of the Evaluation Method