The Use of Self Service Technologies in Dutch
Municipalities
An investigation into the antecedents of self service technology use, and the design of the
service delivery system.
Master Thesis Operations and Supply Chains
Author Erik Douma Studentnumber: 1341677 E-mail: erikdouma84@hotmail.com Joan Maetsuyckerstraat 85 2593 ZE Den Haag The Netherlands
University of Groningen – Faculty of Economics and Business Administration First supervisor: Dr. H. Broekhuis
Second supervisor: M. R. Van der Laan Nettelbosje 2
Preface
This master thesis is about the use of self service technology in Dutch municipalities. A subject which combines my backgound as a student in operations management, and my ambition to have a career in the public environment. Writing this thesis has been the concluding part in achieving the title of Master of Science in Business Administration at the university of Groningen.
I would like to thank my first supervisor, Manda Broekhuis, particularly emphasizing her scientific contribution to this research. She always supported me with valuable advices, which was a crucial part of this research from the proposal development untill the end of report writing. I would also like to thank my second supervisor Justin Drupsteen for being the second supervisor to this research and for his valuable comments.
I hope you will enjoy reading this paper.
Abstract
In recent years, municipalities have invested in the provision of their services via self service
technologies (SST’s). The use of SST’s can offer large advantages to municipalities in terms of cost savings, due to the fact that work is transferred to citizens and automated systems. It is evident that these advantages will only materialize when sufficient use is made of the SST’s. However, research finds that SST use is below expectations. Therefore it is important to understand the antecedents of SST use, and to know what managerial options there are to affect SST use. This paper is the report of a research that aims to provide insights into the antecedent factors that affect SST use by citizens, and into the operational design elements that can be managed by municipalities to affect SST use by citizens.
Two research questions are investigated:
• Research question 1: What are the antecedent factors affecting SST use by citizens in Dutch municipalities?
• Research question 2: What operational design elements should be managed by municipalities to maximize the chance of citizen acceptance of SST’s?
Theoretical exploration
Traditional service literature suggests that the choice of a particular service channel in a multi channel environment will depend on the advantages it offers compared to other channels. In addition, the ease with which the channel can be used is of importance. It is also suggested that people may differ in how they perceive the advantages of a specific service channel, or the ease with which it can be operated, based on differences in personal characteristics.
Research methodology
In order to gain a deeper understanding of the antecedent factors and the operational design elements, two literature reviews were conducted.
In the first review, a broad search method was applied to find articles that describe specific antecedents of SST use. These articles were then analyzed to find out which antecedent factors are relevant for this research, and how they relate to SST use. The second review was performed to find operational design elements that can be managed to affect SST use. The search approach was to use both broad search terms, such as “service
design”combined with “SST use”, and more specific search terms that aimed to find articles that described the application of traditional service design elements in an SST setting. Included in the research were articles that described specific design elements and related these to SST use, or antecedent factors of SST use. These design elements were then listed in order to create an overview of managerial ways to affect SST use.
In addition to the literature reviews, interviews with 6 municipality managers were held. These were meant to compare theoretical insights with knowlegde ‘from the field’. Further, they were meant to uncover possible additional managerial methods to affect SST use.
The research results and theoretical and practical implications will now be discussed for each research question in turn.
Findings regarding antecedent factors of SST use
Regarding the first research question, the literature search revealed that quite a lot of research has been conducted regarding antecedent factors of SST use. Three groups of antecedent factors have been identified in the literature search; utility factors that resemble the avantages that the use of the SST channel has for the citizen, system factors that resemble the usability of the SST channel , and personal factors that resemble differing character traits and preferences of the population of potential users. The antecedent factors and their interrelations are shown in the figure on the next page.
However, it must be noted that further empirical research is nesscary. The articles included in the research are mostly based on survey studies, and can not make definite conclusions on causality between antecedent factors and SST use, but can only indicate that there is a correlation. Further, it is important that the utility factor is decomposed into different aspects of utility, because in theory there may be differences between the different aspects of utility. For instance some research stress the use of incentives as a way to increase utility while others claim that 24/7 accessibility is important. Managers need more specific information regarding the relative imporatance of these aspects of utility in order to make informed decisions about the design of the service delivery system.
Findings regarding operational design elements to affect SST use
The second research question is focussed towards managerial ways to affect SST use. The literature review confirmed that relatively little research has been performed to investigate what operational design elements can be managed to affect SST use. However, 15 articles were identified that discussed specific relationships between design elements of the service delivery system and SST use. The methods identified can be grouped into four categories. The first category is that of SST interface elements. The design of the
Antecedent factors affecting SST use
SST use
Positive relation Negative relation Moderating effect
Utility factors
• Perceived utility of SST’s compared to other channels
• Expected control • Expected enjoyment
System factors
• Provision and knowledge of services • Ease of use
• Task complexity
Personal charateristics
• Social acceptance • Social anxiety
interface, whether this is a web site or a service kiosk, can do much to increase utility, trust, and quality perceptions which have all been identified as relevant antecedent factors of SST use. A specific managerial method that can be applied to increase SST use is the provision of a personalized web page. This will increase information quality in terms of relevance to the user. Another method is to provide a privacy policy statement, which will diminish perceptions of risk and increase perceptions of trust. Online order tracking is also an important tool. It increases utility by allowing citizens to retrieve information about their service requests anytime they want. It also enhances trust. The next category of design considerations is that of physical delivery. Although services delivered through SST are mostly informational in nature, there is often a physical component involved. This is the case when physical documents have to be sent to the citizen. The relevant design element found in this category is fulfillment. Reliability of delivery will enhance trust and quality perceptions for the SST channel. The third category is integration. In a multi channel environment it is
important that the various channels are integrated with each other. Citizens should receive the same response to their request regardless of the channel that is accessed. This response should be of the same quality standard in terms of reliability and accurateness and speed. Related to this is that citizens should not have to provide personal information that they have already provided to the municipality using other channels. Managerial methods to achieve these goals are to have a single backoffice for multiple front offices, and to have the same design for multiple front offices in terms of look and feel. A last category is service recovery and assistance. People that experience some kind of failure in the provision of their services, or are not sure how to operate the SST channel should receive adequate support. Managerial methods are to apply failsafe service design and to offer sufficient technological feedback mechnisms. Further assistance can be provided in the form of user training. The table summarizes the above, and presents the differing managerial methods that exist to affect SST use.
Operational design elements that affect SST use Service delivery system design
elements
Affects Managerial tools Support
SST interface design elements Information availability and content
Information quality. The extent to which the information is relevant to the user, easy to understand and up-to-date
Personalized web page Sousa et al, 2008; Tarafdar and Zhang, 2008; Santos, 2003
Usability Ease of use. The ease with which a customer can employ a site to achieve a specific goal
Easy to understand navigation structure
Zeithaml et al, 2002; Casalo et al 2010; Massey et al, 2007; Tarafdar and Zhang, 2008; Pearson et al, 2007; Santos, 2003
Appearance Quality. The overall appeal of the website
Appealing interface design Zeithaml et al, 2002; Santos, 2003
Privacy and security Trust and perceived risk. The risk that personal information is provided to third parties, and
Privacy policy published on web site
the risk that personal information is used for fraud
Line of visibility Customer control and utility Online order tracking Tate and Johnstone, 2011 Physical components of service delivery design elements
Fulfillment Quality (dependability and reliability)
Online order tracking Semeijn et al, 2005; Zeithaml et al, 2002; Heim and Field, 2007 Integration of service channels design elements
Content consistency Quality (dependability) Single backoffice for multiple frontoffices
Sousa and Voss, 2006; Tate and Johnstone, 2011; Oh and Teo, 2010
Process consistency Quality Consistency in the design of multiple frontoffices
Sousa and Voss, 2006; Tate and Johnstone, 2011; Oh and Teo, 2010
Failure recovery and assistance design elements
Service failure recovery Quality Fail-safe process design and technology based feedback mechanisms
Sousa and Voss,2009; Meuter et al, 2000; Forbes, 2008
User training Self efficacy, ease of use Demonstrations and written instructions
Zhao et al, 2007
Interviews with municipality managers revealed that they already apply a lot of the methods in practice, or are planning to implement them. Two municipalities already offer order tracking options and three more are planning to do so. Further, two municipalities already offer a personalized web page, and one more is planning on doing so. A method that municipalities have an obligation to apply is to post a privacy statement on the web site. So they all apply this method. In short, all municipalities apply one or more of the methods found in scholarly literature. In order to integrate their service channels, municipalities are investigating the advantages of a midoffice. This is an application that connects various frontoffice systems with various backoffice systems. It resembles the integration methods described in the literature review in the sence that it enables multiple front offices to access the same back office sources. When municipalities have made only limited efforts in applying these methods, this is mainly due to the fact that they are still in the early stages of SST
implementation and the expansion of service options is not a priority yet.
Contents
PREFACE ... 3
ABSTRACT ... 4
CONTENTS ... 10
1
INTRODUCTION ... 12
1.1 Search for low cost and high quality ... 12
1.2 The application of technology ... 13
1.3 The use of self service technology ... 14
1.4 Research goal ... 14
1.5 Research model ... 15
1.6 General outline of this paper ... 16
2
THEORETICAL EXPLORATION ... 17
2.1 Definition of SST use ... 17
2.2 Exploration of antecedent factors affecting SST use ... 18
2.3 Exploration of relevant operational design elements ... 21
3
METHODOLOGY ... 27
3.1 Search strategy literature search research question 1 ... 27
3.2 Search strategy literature search research question 2 ... 28
3.3 Interviews with municipality managers ... 30
4
FINDINGS RESEARCH QUESTION 1 ... 32
4.1 Discussion of research background and methodology of articles included in the review ... 32
4.2 Antecedent factors ... 35
4.3 Model of antecedent factors ... 45
5
FINDINGS RESEARCH QUESTION 2 ... 46
5.1 Search results ... 46
5.2 Design elements related to the SST interface ... 48
5.3 Design elements related to the physical components of service delivery ... 51
5.3 Design elements related to the integration of service channels ... 51
5.4 Design elements related to service recovery and assistance... 52
5.5 Overview of design elements ... 53
6
FINDINGS CASE RESEARCH ... 55
6.1 Characteristics of SST implementation ... 55
6.2 Judgment of antecedent factors ... 57
6.3 Managerial ways to increase SST use: experiences from practice... 61
7
DISCUSSION ... 66
7.1 Discussion findings antecedent factors ... 66
8
CONCLUSION ... 71
8.1 Summary of the findings ... 71
8.2 Implications for management ... 71
8.3 Implications for theory ... 72
8.4 Limitations of this research and opportunities for further investigation ... 72
1
Introduction
1.1 Search for low cost and high quality
Dutch municipalities face some big challenges in the years to come. On the one hand, the economic tide has urged municipalities to economize. Funding from the central government for the municipalities is decreasing (Allers, 2009). Other sources of income are also under pressure. For instance the income derived from land and building projects is decreasing as a result of a decline in building activity. On the other hand, the central government has a policy towards decentralization; more and more public tasks are assigned to the municipalities. Meanwhile, programs are in place to assure an ever higher level of service quality. Dutch municipalities face the problem many organizations face: how to increase output of high quality and reduce costs at the same time.
In various ways the Dutch municipalities try to cope with these challenges. To improve the quality of services municipalities increasingly seek to develop citizen-centered processes in which the needs of the citizen are central. The ambitions of the Dutch municipalities regarding service standards are reflected in the report ‘Dienstverlening draait om mensen’ (Service is about people)(Vereniging Nederlandse Gemeenten, 2010). This report describes a roadmap towards more service oriented municipalities. Some of the ambitions expressed in the report are: people will not be required to provide information that the municipality already posseses, municipalities will be transparant about their services and their quality standards, the municipality will be oriented on demand instead of supply. In order to meet the demand for more efficiency many municipalities are setting up a customer service centres. These provide a scale of operations that makes it worthwhile to hire people with expertise in a range of different areas.
In the efforts to meet both quality performance and efficiency much is expected of the opportunities that innovations in information and communication technology have to offer. Specifically relevant to the issues at hand is a nationwide program called the ‘Nationaal Uitvoerings Programma dienstverlening en e-overheid’ (NUP / National Executing Program service and e-government). This program is geared towards the application of innovations in information and communication technology to assure high service quality at low costs. The program aims to build a nationwide ICT infrastructure with which developments in e-governance can be supported throughout all layers of government. This enables government organizations to quickly and
1.2 The application of technology
The NUP reflects a trend whereby more and more technology is integrated in the provision of services (Meuter, 2000). This has increasingly changed the way citizens interact with municipalities to create service outcomes (Meuter, 2000). The use of technology has expanded the number of contact channels through which a citizen can interact with a municipality (Ebbers et al, 2008). Besides traditional channels such as desk, phone and conventional mail, e-mail and web sites have been added to the range of channels. This has urged municipality managers to consider their multi channel policies. The multichannel policy of an organization includes all means of contacting with customers, like face-to-face contact, contact by phone and online contact (Sousa et al, 2006). From the point of view of the customer a good multi channel policy has the advantage that he or she can contact the organization in any way he or she likes, which gives the customer control and satisfaction. From the point of view of the organization, a multi channel policy makes it possible to direct customers to more cost effective channels, increasing the efficiency of providing the service. However, integrating all the channels can be a real challenge (Sousa et al, 2006).
A particular kind of technological advancement is the use of self service technology. Self-service technologies (SST’s) are increasingly replacing face-to-face contacts between buyers and suppliers (Eriksson and Nilsson, 2007). Self service is defined as “service in which there is no direct assistance from or interaction with a human service agent” (Rowley, 2006). Examples of self service are the use of service kiosks (like ATM-machines) or online purchasing. The implementation of self service technologies offers a number of advantages to service organizations. Potential cost savings are probably the most prominent reason (Bitner et al 2002; Dabholkar, 1996). In self service the customer performs the role of co-producer and is a productive resource for the service provider (Bitner et al, 2007). Putting the customer in the role of co-producer has the obvious advantage that the customer performs work that would otherwise be performed by the service employee. The
1.3 The use of self service technology
To the service provider, the use of technology is attractive because it can reduce costs, and improve service quality. However, these results will fail to materialize when people will not accept the SST channel. In that case, it can be a significant drain on resources (Curran and Meuterl, 2005). Unfortunately, the public still has a preference for more traditional channels such as desk, phone and mail, which can all be classified as employee service channels. There is a gap between the service channels governments prefer and those prefered by citizens (Ebbers et al, 2008). A number of researches in various countries demonstrate that citizens are reluctant to use self service channels. In the Netherlands research shows that citizens do not choose self service channels over other channels. Citizens use the phone most, followed by front desk and written letters (Pieterson and Ebbers, 2008). This corresponds with experiences in other countries. Research in Australia and New Zealand shows that most citizens are reluctant to use some of the more sophisticated, transactional services offered via SST’s, and less than half of the population has even visited a government website (Goald et al 2010). Further, the great majority of the population still prefers to deal with the government through non-digital means. Studies in Canada, Spain and Switzerland also confirm that citizens still prefer the traditional channels to contact the government (Ebbers et al, 2008; Belange et al, 2010).
1.4 Research goal
The low use of SST’s is a concern for municipal managers. The benefits of offering services over the SST’s are achieved only when sufficient use is made of the SST’s. The more citizens use SST’s to fulfill their service needs themselves, the more scalable and cost-effective the business model will be (Rowley, 2006). Therefore it is important to understand what makes citizens choose the SST channel over alternative service delivery options (Bitner et al, 2002). Further, it is imperative to understand how element of the service delivery system can be managed in order to increase the change of SST adoption by citizens (Curran and Meuter, 2005). The goal of this research is to provide guidance to municipality officials that want to increase the use of their self service channels. Therefore the research objective is to provide insights into the factors that affect SST use by citizens, and into the operational design elements that can be managed by municipalities to affect SST use by citizens.
Research objective: To provide insights into the antecedent factors that affect SST use by citizens, and into the operational design elements that can be managed by municipalities to affect SST use by citizens.
Research question 1: What are the antecedent factors affecting SST use by citizens in Dutch municipalities?
Research question 2: What operational design elements should be managed by municipalities to maximize the chance of citizen acceptance of SST’s?
The focus of this research is on SST use for all services that the municipalities offer via SST. These include complicated services, such as building permits, and simpler services such as a request for a birth certificate. This research takes a broad view of services, whereby answering questions from citizens, and making appointments are also regarded as services. Further, the search for service design elements which can be managed to affect SST use is not restricted to the management of the SST itself. Rather, the aim is to identify and discuss all elements of the service delivery system that may affect SST use, from service ordering to fulfilment. Finally, the research includes both online SST’s, such as websites, and on-site SST’s, such as self service kiosks. Paragraph 2.1 will further discuss the definition of SST. A research model is constructed that indicates how the two research questions are related to each other, and how they are related to the goal of increased SST use. This research model is presented in the next section.
1.5 Research model
The research model depicted in figure 1 summarizes much of what has been discussed so far. This research does not investigate the relationship between SST use and the advantages as depicted in the research model. However, it is assumed that SST use is positively related to the quality of service delivery and the efficiency of service delivery. This is depicted by the right part of the model. The first research question aims to provide insights into the relationship between antecedent factors affecting SST use and SST use in Dutch municipalities. This relationship is depicted by the arrow that is labelled RQ 1 (research question 1). Investigating this
Figure 1: Research model
1.6 General outline of this paper
2
Theoretical exploration
The central aim of this research is to to provide insights into the factors that affect SST use by citizens, and into the methods that can be applied by municipalities to affect SST use by citizens. The central concept in this objective is SST use. Therefore, the first paragraph of this theoretical exploration will explore the concept of self service technology, and of self service technology use. The next two paragraphs will provide a first exploration on the literature concerning the two research questions.
2.1 Definition of SST use
Self service technology
The term self service technology, or SST has been used by a number of authors (Curran and Meuter, 2007; Meuter et al, 2000; Dabholkar, 1996). This research will adopt the definition of self service technology provided by Meuter et al (2000) which defines self service technologies as “a technological interface that allows
customers to produce and consume services without direct assistance from employees” (Meuter et al, 2000). Examples of SST’s include ATM’s, pay at the pump gas station terminals, online banking transactions, touch screens in department stores, information kiosks at hotels, and self-scanning in grocery stores. (Curran and Meuter, 2005; Dabholkar and Bagozzi, 2002).The two defining parts in this definition are the absence of employee assistance and the technological interface. These defining characteristics have implications that make service through SST channels different from traditional service.
The first defining characteristic is the absence of employee assistance. Service channels without employee assistance are called self service channels, as opposed to service channels with employee assistance, which are called employee service channels. Traditionally the service literature has placed much emphasis on the interpersonal nature of service provision, examining the interaction between service employee and the recipient of the service (Meuter et al, 2000). But in technology mediated services the emphasis shifts from citizen-employee to citizen-organization interaction (Meuter et al, 2000). The absence of an employee in the service encounter often requires increased work or involvement on the part of the customer (Curran and Meuter, 2005). It is not necessary that the service process is completely automated, the term self service is related to the interaction between citizen and organization. It is possible that this interaction is through SST, while afterwards the service order is processed manually.
Self service technology use
The use of the self service technology is the target variable in this research. Some scholars have investigated the use of technology in a direct manner (Montoya-Weis et al, 2003). Other scholars have used behavioural intention to use as a proxy for actual use (Lee et al, 2009). Due to the difficulties to measure real behaviors, intentions have been widely used to measure citizen behaviour (Belange et al, 2010). This is because behavioural intention to use is significantly correlated with use (Davis et al, 1989). Yet others have called the dependent variable Intention to change behaviour (Curran and Meuter, 2007), or online channel preference (Frambach et al, 2007; Looney et al, 2008). The usage intention measure is particularly useful in the public service context. Government services include a huge collection of varying services, many of which are irrelevant to many citizens for most of their lives (Lee et al, 2009). An intention measure is a useful proxy measure of actual behavior because it can be easily designed to extract the possibility of potential use when the needs arise (Lee et al, 2009).
2.2 Exploration of antecedent factors affecting SST use
It is important to understand the mechanisms behind the acceptance and use of SST’s. This paragraph will explore research streams that are relevant to the first research question; “What are the antecedent factors affecting SST use?”.
Two important research streams are explored for answering this question. The first is the research stream that deals with acceptance and use of technology. It is assumed that this will be an important stream of research because SST’s are a relatively new way of interacting with government agencies, with which people are still unfamiliar. Technology acceptance models that will be discussed are the Technolology Acceptance Model (TAM) created by Davis, Bagozzi and Warshaw (1989), and the Unified Theory of Acceptance and Use of Technology (UTAUT) created by Venkatesh, Morris and Davis (2003). These models are well-known and they model the antecedents of the acceptance of technology that is introduced in a working environment. However, encouraging citizens to use new technologies in service encounters is considered more challenging than encouraging employees to use new technologies (Curran and Meuter, 2005). This is especially true in a multi-channel environment where citizens can choose from different multi-channels (Montoya-Weis et al, 2003; Xue et al, 2007). In such an environment, the comparative quality of a service channel becomes more important. That is why service quality research is the second research stream that will be explored.
Technology Acceptance Model
defined as the degree to which an individual thinks that using a particular system would enhance his or her job performance. This construct is positively related to the behavioural intention to use a technology both directly and indirectly, through the effect on attitude towards using a technology. Perceived ease of use is defined as the degree to which an individual thinks that using a particular system would be free of effort. It is positively related to the attitude towards using a technology, and also to the perceived usefulness. Because the easier it is to use a system, the more useful it can be (Davis et al, 1989), although this effect will become less important when the user becomes more experienced with a particular technology (Venkatesh and Goyal, 2010). Attitude is defined as an individual’s thoughts about performing the intended behavior (Davis et al. 1989). According to the TAM a users’ attitude towards using a technology is determined by the perceptions of usefulness and ease of use. TAM also suggests that the effects of external variables, such as training and system design
characteristics, on behavioural intention and use are mediated by the two key beliefs: perceived usefulness and perceived ease of use (Davis et al,1989; Venkatesh and Goyal, 2003).
Figure 2: TAM model by Davis et al (1989)
In research on technology adoption the TAM model is the most widely employed theoretical model (Brown et al, 2010). Many authors have used the TAM model in explaining antecedents of technology acceptance and usage and incorporated variables from the model in their own models. Others have made efforts to expand the model. Although attitudes had been shown to play a key role in SST adoption, research demonstrated the need to explore additional factors (Meuter et al. 2005; Parasuraman 2000; Walker et al. 2002). They found that attitudes alone cannot explain behaviour (Curran and Meuter, 2007).
Unified Theory of Acceptance and Use of Technology
Figure 3: UTAUT model by Venkatesh et al (2003)
The UTAUT model took from the TAM model the constructs of performance expectance and effort expectancy. The construct of attitude as an influencing factor was dropped because the authors argued that it did not have a direct influence on technology acceptance. Also the constructs of self efficacy and anxiety were not
incorporated in the UTAUT model. These constructs were mentioned in some of the other models that were used as the basis of the UTAUT model. The authors also added some constructs; the constructs of social influence and facilitating conditions were added. Social influence is defined as the degree to which an Individual perceives that important others believe he or she should use the new system. This may mean that certain behaviour is mandatory and that rewards are given to those who use a new kind of technology. It can also work trough status enhancement. If someone thinks that his or her status will increase if the new
technology is used he or she will be more inclined to use it. Facilitating conditions are defined as the degree to which an individual believes that an organizational and technical infrastructure exists to support use of the system. This construct was not found to be of influence on intention to use, because it worked in a similar way as effort expectancy. Good facilitating conditions make that less effort is required in using the technology. No additional explaining power was attributed to facilitating conditions alone when it comes to the effect on intention to use. However, the authors suggest that there is a direct effect on technology use. A possible explanation provided by the authors is that more experienced employees know more ways to get assistance on the job and can more easily overcome impediments to the use of new technology.
Service quality models
In a multi channel setting a service recipients’ choice for a particular service channel is not only based on the quality of the service provided through that channel, but rather it is based on a comparison of quality provided through alternative service channels (Montoya-Weis et al, 2003). To compare the quality of service delivery among channels, a method for assessing service quality is required. To measure service quality, research has developed sets of service quality dimensions. These dimensions are viewed both as antecedents to the adoption of e-service, and as the parameters used in the judgement of e-service quality (Rowley, 2006).
A well known measurement model for measuring service quality is the servqual model (Parasuraman et al, 1991). This measurement instrument measures service quality along five dimensions: reliability,
responsiveness, assurance, empathy, and tangibles. The servqual model is widely accepted as a quality measure in traditional service delivery. However, SST’s represent a unique form of service delivery, and the dimensions of service quality suggested in traditional models may not apply (Dabholkar, 1996). Therefore another, similar model was developed by the same authors to measure the quality of services provided through websites. This is the E-squal model (Parasuraman et al, 2005). This quality measurement model includes items such as efficiency, fulfillment, system availability and privacy.
Further investigation
The exploration of technology acceptance models and service quality models provides a first impression of the kind of factors that are relevant to SST use. So far, it has been established that usefulness is an important antecedent variable, and that perceived service quality is an important part of usefulness. It is likely that citizens that can choose between different channel options will weigh the comparative advantages of the different options available, and select the channel that offers the highest quality. Due to the newness of the technology it is also likely that ease of use will be an important antecedent factor. However, further investigation is needed. The technology acceptance models are familiar but since their publication more research has been done. More recent research can provide new insights in the form of additional antecedent factors. Furthermore, research in other research settings may have resulted in conclusions on the external validity of the antecedent factors that are known from research in the working environment. A review of relevant theory will be performed to find additional factors and insights. The methodology to this literature search is described in chapter 3. The results are described in chapter 4. Next, the exploration of the second research question is described.
2.3 Exploration of relevant operational design elements
research question “What operational design elements should be managed by municipalities to maximize the chance of citizen acceptance of SST’s?”
While the antecedents of SST use have been quite well documented, agreement as to how services delivered through SST’s could be designed in order to maximize utility and ease of use is largely missing from the scholarly literature (Cho and Menor, 2010; Ba and Johanson, 2008). Therefore, this exploration of operational design elements will draw upon knowledge grounded in general service literature. The service triangle and the service delivery system will first be discussed and related to the research questions in this research. After that, elements of the service delivery system will be discussed. This will form the theoretical background for a more thorough literature search into the link between operational design elements and SST use. Finally, it will be specified what further knowledge is required in order to answer the research question.
The service triangle and the service delivery system
An important framework in service literature is the service triangle (figure 4). The service triangle is a commonly cited framework for the conceptualisation of the main concepts in service strategy, and is a useful starting point for the exploration of service delivery system design elements (Ponsignon et al, 2010). It emphasises the importance of the alignment between the target market, the service concept and the service delivery system (Ponsignon et al, 2010). The first part of the service triangle is the target market. The target market in this research would consist of people that need services from the municipality. The second part of the service triangle is the service concept. The service concept includes all tangible and intangible aspects of the service offering. The third part of the service triangle is the service delivery system. The service delivery system is concerned with “how” the service concept is provided to the customer. It encompasses both structural and infrastructural aspects of service delivery (Ponsignon et al, 2010).
Figure 4: The service triangle (Ponsignon et al, 2010)
The service delivery system is the part of the service triangle that is the focus of this exploration. The main message of the service triangle regarding the service delivery system is that it needs to be aligned with the requirements of the target market and with the service concept that is offered. Although Dutch municipalities have a monopoly in their service domain, it is still important to align the service delivery system with the requirements of the target polulation. The antecedent factors that are investigated in this research can be regarded as requirements in the sence that these factors should be positively affected in order to increase the
Target market Who are the customers?
Service concept What is the service package offered?
chance that citizens will choose the use of the SST channel over the use of other channels. While the first research question is concerned with the requirements, the second research question is concerned with how the service delivery system can be designed such that these requirements are met. The next section will explore specific elements of the service delivery system.
Elements of the service delivery system
The previous section has discussed the definition of the service delivery system, and the relation of the service delivery system with other key aspects of service strategy using the service triangle. The next step is to describe the elements of the service delivery system that can be managed in order to affect SST use.
Mayer et al (2003) have listed service delivery system design elements and have separated them into two primary groups. The first group consist of elements that deal with structural design choices. Included in this group are technology, process visibility, customization, appearance, accessibility, employee costumes, the amount of interaction and the delivery method. The other group consists of elements that are more situational in nature. These are highly dependent on the employee that provides the service rather than on managerial design choices. Examples of such design elements are employee appearance and assurance. Since this second group of design elements is very much related to a service setting with employee interaction, it is of little relevance in a setting where no employees are involved. Therefore this discussion of design elements is limited to the first group of elements. Each of these design elements is described, and it is discussed whether it is worthwhile to further examine the management of the design element in an SST context in order to increase SST use.
Technology
Technology refers to the use of mechanical devices and systems in the service process (Mayer et al, 2003). Technology within the service process may play different roles (Fitzsimmons and Fitzsimmons, 2008). In this research the role that SST’s perform in the service delivery process is clear; it replaces the service employee and serves as the interface between the citizen and the municipality. So the role of technology is clear, but the question what devices and systems to use is also highly influential. Therefore the design element of technology will be further explored in the literature review.
Process visibility
Process visibility is the extent to which a customer is allowed to observe all, or a portion of, the delivery system (Mayer et al, 2003). In traditional services a well known example is found in the restaurant sector, where you can often see the chef at work while he is preparing your pizza in a pizza restaurant. When services are ordered and delivered through the use of technology, the managerial options regarding visibility are not so
is stretched a bit so that the degree to which the citizen is informed about the service process, there might be managerial choices to make that do affect the citizens’choice of a service channel. Visibility in this sence means that the citizen might be informed regarding which phase of the process the delivery of his or her service is in, or who is currently working on his or her case. This kind of visibility is found in transportation companies, where people can track the packages they have sent. Using a code that corresponds with their package they can use the website of the transportation company to see where the package is, or whether the package has been delivered. Because the visibility of the process might offer possibilities to affect a citizens’ channel preference this design element will be further explored in the literature review.
Customization
Customization is the ability of the service delivery system to attend flexibly to customer needs (Mayer et al, 2003). The degree of customization can be regarded as a continuum whith highly standardized services on the one side, and highly customized services on the other. The technological nature of service delivery through SST’s fits best with a highly standardized service delivery. Besides that, the nature of the services provided by municipalities generally do not allow for customization. Laws and regulations specify what the service offering should be. For example, municipalities cannot decide to change the specifications of a building permit because of the preferences of an individual citizen.
However, customization may not only refer to the product of the service delivery, but may also relate to the process by which the service is delivered. For instance, when the goal is to have citizens use the SST channel, a choice can be offered to the citizens between using the web site or a service kiosk at the municipality office. Further, when a citizen wants to discuss something at the municipality office, he or she can choose to make an appointment, or drop by unannounced. So there is room to customize the process of service delivery, and this design element will be further investigated in the literature review.
Physical appearance
Physical appearance is defined as the physical facility design desisions made by managers that are essentially fixed (Mayer et al, 2003). This commonly refers to such things as furnishings, lighting and noise levels in traditional service encounters. In SST delivered services, the appearance of the service encounter also plays an important role. This can be the appearance of the web site, or of a service kiosk. It resembles the physical appearance design element of Mayer et al (2003) in the sence that the appearance of the web site or kiosk is also fixed. This design element will also be further investigated in the literature review.
Accessibility
conditions must be met. Potential users are required to have internet access and the right software to run the applications that enable them to make use of the SST channel. In the case of an onsite SST channel, the physical accessibility is still as important as in traditional service provision. Therefore this design element will be further investigated in the literature search.
Employee costumes
Mayer et al (2003) also mention the decision regarding the costumes of the employees as a structural decision. They define this as the managerial choices for wardrobe an other accessories attendant to the look of customer contact personnel (Mayer et al, 2003). While this may be important in traditional service settings it does not apply when services are delivered through SST. Therefore this design element will not be explored any further.
Amount of interaction
Another important design element is the amount of interaction. This is the managerial design choice involved in how customers and employees interface along a continuum of options during service delivery, which encompasses both employee involvement and customer participation in a service (Mayer et al, 2003). The line of interaction distinguishes actions that are performed by the customer from actions performed by the service employee. Mayer et al,( 2003) clearly focus on the interpersonal aspect of interaction, and this leads them to the conclusions that there is no interaction in self-service. But there are other aspects of interaction that are relevant even when a service is delivered through SST. When a service employee is replaced by SST, there are still decisions to be made regarding which tasks should be carried out by the customer and which tasks should be carried out by the SST. Therefore, the interaction design element will also be further explored in the literature search.
Delivery method
Management of the delivery method consists of design decisions that managers must make in setting up their service delivery system (Mayer et al, 2003). In their description this encompasses both the channels that are used for contacting the potential customers as the channels that are used for delivering the service. Examples of the former are hotels that use travel agencies and the internet for reaching potential customers. Examples of the latter are restaurant that may opt for table service or home delivery. These desisions are also relevant when services are delivered through SST’s. Although the service itself needs no advertising, citizens need to be made aware of the fact that they can obtain the services they require through the SST channel. Further, when they have requested a service through the SST channel, in many cases they still need to receive a physical product, mostly in the form of a document. Managerial decisions need to be made regarding the way these are delivered to the citizen. Therefore, the delivery method will also be further investigated in the literature review.
Further investigation
This exploration of operational design elements conludes that most of the design elements that are relevant in traditional service literature might also have an application in the provision of service through SST’s. From the design elements listed by Mayer et al (2003), only the design element of employee costumes has been excluded because it is not relevant in an SST environment. Technology, process visibility, customization, appearance, accessibility, the amount of interaction and the delivery method are all design elements that may deserve management attention, because they might affect SST use.
3
Methodology
In the previous sections the theoretical backgrounds of the research questions have been explored. The exploration has provided a first impression of antecedent factors of SST use and of operational design elements that are related to service quality. However, for each of the two research questions a number of issues still remain. For the antecedent factors, more recent insights may expand the factors already known. It is also possible that more recent empirical research will shed new light on existing insights. Further, the development of SST service may have inspired researchers to define quality dimensions especially for SST service. A review of relevant theory must uncover additional factors and insights. Further, more information is needed to inform managers that want to affect SST use. The exploration of operational design elements in chapter 2.3 indicates what design elements are relevant in traditional services. However, further insights are necessary regarding the question if and how these design elements apply in the SST context. The literature search intends to uncover design elements that are relevant in an SST service context, and which can be managed in order to affect SST use.
This chapter describes the methodology used in finding the answers to the research questions. Paragraph 3.1 specifies the search methodology for research question one, and paragraph 3.2 specifies the search
methodology for research question two. In addition to the literature review, interviews were held with municipality officials. This is described in paragraph 3.3.
3.1 Search strategy literature search research question 1
Search strategy
The Business Source Premier database was used to search for relevant articles. As explained, the goal of the literature search was to find additional insights into antecedent factors of SST use. Following this goal, a general search for articles related to the acceptance and use of SST’s was performed using the following search phrases:
SST AND use E-service AND use SST AND acceptance E-service AND acceptance
The search term ‘e-service’ was added because it is a more commonly used term, and adding this term to the search phrase could decrease the chance of articles being overlooked. Because quality dimensions are important in determining usefulness, an additional search for quality related papers was performed using the search phrases:
Selection criteria
The search results were examined and a selection of the articles were included in the study. The articles were selected using a number of search criteria set up to include only the articles relevant to this study. The selection criteria were the following:
Only articles specifying antecedents of SST use, or dimensions of quality are included.
Articles specifying dimensions of quality need to be tailored to service settings with a technological interface.
For each search phrase entered the result list was scanned. When the title of an article did not make clear whether the article met the criteria, the abstract was examined. When the abstract also did not make clear whether the criteria were met, the contents of the article itself were examined.
Analysis
Each of the articles included in the study was searched for relevant antecedent factors and quality dimensions. Each of the antecedent factors was listed in a table. The table list the antecedent factors found per article. For each antecedent factor, a description or definition was taken from the source article, and it was specified how, according to the article, the antecedent factor is related to SST use. The result is an overview of all the
antecedent factors found in literature, their descriptions, and their relations to SST use. The next step was to find out which of the factors in the overview are unique antecedent factors. Some factors may have been discussed in multiple articles. Some under the same label, but some may also have been discussed under different labels. An examination of the definitions and descriptions of the antecedent factors had to determine which are unique antecedent factors. A third step was to examine those antecedent factors for which there are contradictory results in literature. In such a case, arguments supporting or challenging a specific finding had to be examined, and a choice had to be made whether to include the antecedent factor in the final list of antecedent factors. The end result is a list of antecedent factors affecting SST use, with a description of each antecedent factor, and a specification of its’ relationship with SST use.
3.2 Search strategy literature search research question 2
Search strategy
As with the literature search for research question 1, the Business Source Premier database was used to search for relevant articles regarding operational design elements which may affect SST use. The literature search for research question 2 aimed to find articles that describe specific design elements and their influence on SST use.
elements. These search terms were “visibility”, “customization”, “appearance”, “accessibility”, “interaction” and “delivery method”.
For the purposes of this research articles were required that describe how a certain design element can be managed in order to affect SST use. The research model from paragraph 1.5 shows how design elements affect SST use through the effect they have on antecedent factors. Therefore a link must be found between design elements and antecedent factors. The exploration of antecedent factors in paragraph 2.2 indicated that utility and ease of use might be important antecedent factors, and that quality is an imporatant part of utility in the sence that people will compare the quality levels of multiple channels in order to inform their channel selection choice. Therefore, “utility”, “ease of use”, and “quality” were used as search terms. In addition “SST” and “self service” were added as search terms.
Together these search terms were be combined into a large number of search phrases to find articles that describe design elements on the one hand, but also describe the link with the goal of SST use on the other hand. Each design element search term was combined with each outcome search term. For instance, the combination of the design element search term “visibility” was combined with all 5 outcome search terms. The resulting search phrases were:
• Visibility AND self service • Visibility AND SST • Visibility AND utility • Visibility AND ease of use • Visibility AND quality
This was done for all design elements. This search approach was aimed to find articles on specific design elements. In addition a broad search approach was applied by using the search phrases:
• Service delivery system • Self service technology
The fact that the term “technology” was present in the broad search terms was the reason that no specific search phrases were used regarding the design element of technology.
Selection criteria
Articles were included in the study if they met the following criteria: Only articles that discuss specific design elements are included.
Only articles that describe specific relations between these design elements and performance outcomes are included.
For each search phrase entered the result list was scanned. When the title of an article did not make clear whether the article met the criteria, the abstract was examined. When the abstract also did not make clear whether the criteria were met, the contents of the article itself were examined.
Analysis
The articles included in the study were searched for manageable design elements of the service delivery system. Distinct elements were listed and described, specifying their relation with performance outcomes. Specific methods or tools for managing these elements were listed together with the description of the elements. The review resulted in a list of service delivery system design elements which can be managed to affect performance outcomes, and hence, SST use.
3.3 Interviews with municipality managers
The previous two paragraphs have described how a literature study will be performed to find scholarly insights to address the research questions. These insights will inform municipality officials about the mechanisms underlying SST use, and managerial ways that can be applied to influence SST use. However, most
municipalities already have some experience in offering services through SST, and municipality managers might have their own opinions on which factors are important antecedent factors of SST use. Further, practical experience might have resulted in the identification of useful methods for affecting SST use. Therefore interviews were held with officials of Dutch municipalities in addition to the literature study. These interviews were meant for making comparisons between theoretical findings and experiences from practice. In this way, the interviews will provide an illustration of SST management in practice. They will also serve as an exploration of differences between theory and practice, and they may provide additional insight from experiences in practice. It might be that municipalities use different strategies for affecting SST use than those found in scholarly literature. Or that methods described in scholarly literature are not familiar to municipality managers. Interviews might uncover areas where theory can be expanded and where practice can be improved.
between municipalities regarding the judgement of antecedent factors, or managerial method applied, could be related to differences in SST strategy and implementation.
The second part of the interview was used to examine whether municipality managers recognized the findings regarding antecedent factors found in scholarly literature in practice. Each manager was given a list of
antecedent factors found in scholarly literature, and was asked to score each antecedent factor on a 1-5 scale, whereby score 1 meant that the antecedent factor was regarded unimportant as an antecedent factor, and score 5 meant that the antecedent factor was regarded very important as an antecedent factor. The managers were specifically asked to imagine themselves in the position of a citizen, and indicate how relevant the thought a particular antecedent factor was in deciding whether to use the SST channel or not. Scores were averaged for each antecedent, providing a first impression of the relative importance of antecedent factors compared to other antecedent factors. Managers were also asked whether they thought antecedent factors were missing from the list.
4
Findings research question 1
The first research question is: “What are the antecedent factors affecting SST use?” This paragraph will describe the results of the literature search that was performed in order to find the answer to this question. The goal of the literature search was to find antecedent factors that affect SST use. These antecedent factors are found in literature from various research streams. This chapter will describe the antecedent factors found, and will discuss differences between the findings of the various authors. Based on this discussion some antecedent factors are found to be relevant in the public service context, and some are exluded.
The literature search as described in chapter three has resulted in the inclusion of 18 articles in this research. Table one shows the number of search result per search phrase, and the additional number of articles in the results list that met the criteria, and had not been included from the results of an earlier search phrase.
Table 1: Search results research question 1
Search phrase Number of search results* Additional articles included in the research
SST AND use 47 12
E-service AND use 112 1
SST AND acceptance 6 0
E-service AND acceptance 18 0
SST AND Quality 17 2
E-service AND Quality 132 3
Total number of articles included in the research 18 * Articles found in BSP on 15-06-2011. Search results limited to peer reviewed articles only.
The articles included in the study have been searched for antecedent factors of SST use. The articles and the antecedent factors are listed in the table in appendix A, sorted per article. The table in the appendix describes per article what antecedents were identified and what the relationship with SST use is. Next, the definitions of the identified antecedent factors were examined to see whether they were distinct concepts, or referring to the same underlying concept. This resulted in an overview of unique antecedent factors. Finally these
antecedent factors were examined to see if there were any contradictions between articles with respect to the relationship with SST use. If any contradictions were found, an examination of the arguments supporting or challenging a specific finding determined whether a specific antecedent factor was included in the final list of antecedent variables.This has resulted in a final list of 14 antecedent factors. These are discussed in paragraph 4.2. First a discussion of the research background and the research methodologies of the included articles will indicate to what extent the research findings can be externalized to the setting of the provision of services through SST by a municipality in paragraph 4.1.
4.1 Discussion of research background and methodology of articles included in the review
This research aims to identify antecedent factors of SST use in the public service sector, while the literature search has resulted in the identification of research that was performed in various sectors. However, the public service context differs from other service contexts. Firstly because providers of public services usually have monopolistic authority in their service domain. Secondly because municipalities are part of a familiar and reputable government. Thirdly because the use of SST’s has to remain voluntary. Municipal service has to be accessible to all citizens (Lee and Rao, 2009). Because of these specificalities of service provision in the public environment it needs to be examined whether there are reasons to assume that the antecedent factors found in the articles can not be genereralized to the public sector setting.
Of the 19 articles included, only four were specifically aimed at investigating SST use in the public service sector (Belange et al, 2010, Ebbers et al, 2008; Van Dijk et al, 2008; Lee and Rao, 2009). Other research has
investigated antecedents of information technology acceptance in a working environment. (Davis et al, 1989; Venkatesh et al, 2003), antecedents of SST use in the banking sector (Curran and Meuter, 2007; Curran and Meuter, 2005, Xue et al, 2011; Falk et al, 2007), banking and university services (Montoya-Weiss et al, 2008),online brokerage services (Looney et al, 2008), and online entertainment (Collier et al, 2010), and SST services in general (Meuter et al, 2003). Further, some articles were included that investigated the
determinants of service quality (Parasuraman et al, 1991; Parasuraman et al, 2005). Finally, a number of articles specifically investigated the use of onsite SST’s (Dabholkar, 1996; Dabholkar and Bagozzi, 2002; Gelderman et al, 2010).
The general conclusion regarding the research background is that the differing backgrounds of the studies included in the review is not problematic. Antecedent factors originally found in working environments have been examined in multiple customer service, and public environments. Further, the service environments in which other research took plce show many similarities with the public service environment.
Research methodology
The validity and reliability of the articles included in the review also depends on the research methodologies applied. The table in appendix A specifies for each article what research methodology is applied. Almost all studies show empirical evidence to support their conclusions. The survey method is by far the most widely applied research method among researchers investigating antecedents of SST use.
Researchers applying the survey method generally follow a process where they formulate hypotheses about the relation between independent variables and a dependent variables based on theory. Consequently a questionaire is developed with which the variables under study can be measured. This questionnaire is distributed to a relevant population. The results of the questionnaire are analyzed in order to validate the questionaire. Statistical methods are applied to ensure that the questionnaire measures what it is supposed to measure. Finaly, the data are analyzed to discover what correlations exist between the variables under study. The result in a model that best ‘fits’ the collected data. A potential weakness of the survey method is that the survey population may be different from the population that is investigated. For instance, many surveys are distributed to students, because the student population is easily accessible to researchers. However, as a group, the student population typically has characteristics such as a high education level, and above average experience with computers. Before externalizing results of a survey to other populations, potential differences between the survey population and the target population need to be taken into account.
This research is about SST use. The target population consists of users and potential users of SST’s of Dutch municipalities. This is a very broad category of people, since all people will require services from their
surveys among student populations and surveys among general populations it is likely that the latter provide more accurate results.
The use of the survey method also makes it difficult to determine whether there is a causal relation between variables (Baarda and De Goede, 2001). Using a survey, it can be determined whether the variables are
correlated, but correlation does not prove causation. Correlation does not indicate which variable is dependent and which variable is independent (Shadish et al, 2002). Therefore, sound arguments are required that make it plausible that one variable is affected by the other. Furthermore, correlations do little to rule out alternative explanations. When two variable correlate, there may be a third variable that affects both of them. In using the survey method, there is a danger that such variables are overlooked (Shadish et al, 2002).
Next to the survey method, a few articles used different methodologies. Xue et al, 2011 invetigated acceptance and use of SST’s in the banking sector by examining customer data provided by a large retail bank. They related actual use of internet banking with other variables such as the total demand for banking services per customer and the number of internet banking adopters in a specific area. In the analysis of these data they search for correlations between internet banking adoption and other variables. The difference with the survey method is that only factual information can be investigated. Opinions and preferences of banking customers are not known. Finally, Ebbers et al (2008) performed a literature study to explain a perceived gap between the service channels governments prefer and the service channels citizens prefer.
The analysis of the research methodologies of the articles included in this literature study has resulted in a few main concerns. Firstly, in some studies students are the survey population. This may lead to conclusions which are accurate for student populations, but may be less accurate for a population of citizens that require municipal services, since the latter category may differ from the former in term of education level and computer experience. Secondly, statements about the nature of the relationship between two variables may differ. In such cases, the survey approach can not give a definitive answer as two which of two explanations is more accurate. Careful consideration of the arguments provided by the researchers is therefore required when such a difference is found.
4.2 Antecedent factors
quo bias (Falk et al, 2007). In the opposite situation, where people have already tried the new channel, confirmation of positive expectations is positively related to the intention to keep using the SST (Belange et al, 2010). Perceived utility can be affected by the demand for services of an individual client. The higher the total demand for services, the higher the potential benefits of SST adoption (Xue et al, 2011).
The second group is in this research referred to as system factors. Potential user’s might consider using an SST because of the perceived utility, but obtaining the service trough the SST might become an obstacle due to the difficulties in using the SST.
The third goup is that of personal characteristics. Differences between the potential users might impact the chance that someone will actually use an SST. Included in this group of antecedent factors are factors that have to do with a person’s social needs, competences and experiences in using technology, and risk perceived in using SST’s. Not included in this group of antecedent factors are demographics (Dabholkar and Bagozzi, 2002) The reason is that differences in SST use between groups with a different demographic background are better explained using the underlying personal characteristics (Dabholkar and Bagozzi, 2002).
In the remainder of this paragraph each of the antecedent factors included in the final list will be discussed. The antecedent factor and the supporting literature will briefly be described. When contradictory results regarding an antecedent factor were found, it will be descibed how the choice was made to include the factor in the final list. Two factors that were mentioned in scholarly research, but were not included in the list will also be discussed explaining why they are not included. Finally the findings are represented in a model showing all the antecedent factors included in the list.
Table 2: Antecedent factors affecting SST use
Antecedent factor Definition / description Relation with SST use Supporting literature Group 1; Utility factors
Perceived utility of SST’s compared to other channels
The advantages of using an SST compared to using other channels as perceived by citizens.
This antecedent is positively related to SST use.
Davis et al, 1989; Belange et al, 2010; Curran and Meuter, 2005; Van Dijk et al, 2008; Falk et al, 2007; Lee and Rao, 2009; Looney et al, 2008; Collier et al, 2010
Expected control The amount of control a customer expects to have over the process or outcome of a service encounter.
This antecedent is positively related to SST use.
Dabholkar, 1996; Collier et al, 2010
Expected enjoyment No definition provided This antecedent is positively related to SST use
Curran and Meuter, 2007; Dabholkar, 1996; Dabholkar and Bagozzi, 2002
Group 2; system factors
