**6.4.1 Open coding**

An open code refers to "the analytic process through which concepts are identified and their properties and dimensions are discovered in data" (Strauss and Corbin 1998, p. 101). In this study, open coding was considered in the initial phase of the analysis process. A total of 115 open codes were created based on 15 interviews and 311 questionnaires. During this stage, the analysis was done by using phrase-by-phrase coding. In order to capture what has been said in the interview, conceptual labels were appended to almost every phrase. These labels were mostly too close to the exact words and context of the interview. Phrases with the same idea were attached with the same open code; otherwise, a new open code was created if the existing one did not fit. As the coding process continued and the researchers became well focused and confident in the process, the codings were revisited and refined so that they were reasonably understandable and provided more meaningful concepts, taking into account that they truly represented the empirical data.

However, it is important to note that some sentences represent only one concept, while others represent more the one concept. For example, the following sentence represents the code, financial benefits: "Also, it may have financial benefits later". Whereas the next sentence represents four concepts: lack of m-government services, lack of e-government services, enquiry services, and target users: "The most current m-services as well as eservices are enquiry services for both citizens and residents".

Developing a Theoretical Framework

**6.4.3 Selective coding** 

Organisational factors

Organisation's strategies

Organisation's services

Organisation's needs

Organisation's perceptions

Fig. 2. Core category and relationships

as justified in Charmaz (2006).

developed based on the questionnaire results.

core category as illustrated in Figure 2.

for the Adoption of Biometrics in M-Government Applications Using Grounded Theory 197

users' needs, users' perceptions, users' concerns, organisation's strategies, organisation's services, organisation's needs, and organisation's perceptions. However, it is important to note that the categories of users' needs, users' perceptions, and users' concerns were

Selective coding can be identified as "the process of integrating and refining the theory" (Strauss and Corbin 1998, p.143). According to Strauss and Corbin (1990), the procedure of selective coding requires a selection of a focal core category which is the central phenomenon that has emerged from the axial codes and relating it to other categories in addition to validating those relationships. In this study, the acceptance factors code is the

> Acceptance factors

Challenges

Contributory factors

The identification of the core category informs substantive theory that identifies the factors that influence the adoption of biometric authentication in m-government security. However, after becoming more familiar with the area and being well focused according to the suggestive categories, some of the categories were combined and incorporated with others,

More specifically, system requirements and procedural issues were combined to be in the category of 'system factors', because they were all mentioned by the interviewees as part of

System factors

System requirements

Users' needs

Users' concerns

Users' perceptions

Procedural issues

Users' factors

The aim of this procedure is to begin the unrestricted labelling of all data and to allocate representational and conceptual codes for all incidents highlighted within the data.

## **6.4.2 Axial coding**

Axial coding refers to the process of making connections and links between the categories (Strauss and Corbin 1998). In this stage, the open codes were put together in new ways by making connections and relationships between and among codes and concepts in order to develop core codes. Figure 1 below shows the emergent categories and the relationships between them; however, further developing was made as will be discussed in the following section.

Fig. 1. Categories and relationships

The most closely interrelated open codes were aggregated under core categories. This is was done taking into account constant comparison of the open codes listed under that axial code with each other by asking questions such as: "How do these axial codes show connections and explain factors that influence the adoption of biometric authentication in m-government security?"

Through the process of axial coding, the following twelve categories were created: acceptance factors, contributory factors, challenges, system requirements, procedural issues, users' needs, users' perceptions, users' concerns, organisation's strategies, organisation's services, organisation's needs, and organisation's perceptions. However, it is important to note that the categories of users' needs, users' perceptions, and users' concerns were developed based on the questionnaire results.

## **6.4.3 Selective coding**

196 Security Enhanced Applications for Information Systems

The aim of this procedure is to begin the unrestricted labelling of all data and to allocate

Axial coding refers to the process of making connections and links between the categories (Strauss and Corbin 1998). In this stage, the open codes were put together in new ways by making connections and relationships between and among codes and concepts in order to develop core codes. Figure 1 below shows the emergent categories and the relationships between them; however, further developing was made as will be discussed in the following

> Acceptance factors

Relative advantages

Compatibility

Users' needs

M-government services including payment services

Users' perceptions

And others..

Applying biometrics in both to log onto mobile device and to gain the service access

And others..

And others..

System requirements

with ID number Balance between

Users' concerns

Applying biometrics along

Use of mobile by someone else

And others..

And others..

Lack of current authentication

Disclosure of sensitive information

security and usability

Challenges

Theft possibility for biometric capture

Authentication responsibility

> Procedural issues

> > Infrastructure

And others..

And others..

Contributory factors

The most closely interrelated open codes were aggregated under core categories. This is was done taking into account constant comparison of the open codes listed under that axial code with each other by asking questions such as: "How do these axial codes show connections and explain factors that influence the adoption of biometric authentication in m-government

Through the process of axial coding, the following twelve categories were created: acceptance factors, contributory factors, challenges, system requirements, procedural issues,

Availability And others.. Economical aspects

representational and conceptual codes for all incidents highlighted within the data.

Importance of information in mobile devices

Protection of information in mobile devices

**6.4.2 Axial coding** 

Planning to provide a high level of security

> Organisation's strategies

Providing secure services

And others..

And others..

And others..

And others..

Organisation's services

Authentication services M- Services

Advanced in SIM card

authentication system

Positive perceptions towards applying biometrics in mgovernment

security?"

Organisation's needs

Applying biometrics

Organisation's perceptions

Fig. 1. Categories and relationships

Positive perception towards applying biometric in some applications

section.

Selective coding can be identified as "the process of integrating and refining the theory" (Strauss and Corbin 1998, p.143). According to Strauss and Corbin (1990), the procedure of selective coding requires a selection of a focal core category which is the central phenomenon that has emerged from the axial codes and relating it to other categories in addition to validating those relationships. In this study, the acceptance factors code is the core category as illustrated in Figure 2.

Fig. 2. Core category and relationships

The identification of the core category informs substantive theory that identifies the factors that influence the adoption of biometric authentication in m-government security. However, after becoming more familiar with the area and being well focused according to the suggestive categories, some of the categories were combined and incorporated with others, as justified in Charmaz (2006).

More specifically, system requirements and procedural issues were combined to be in the category of 'system factors', because they were all mentioned by the interviewees as part of

Developing a Theoretical Framework

Ease of use

Trialability

Observability

Trust

Privacy

Awareness

Legislation

Economical aspects

Social and cultural aspects

the KSA

Availability

Compatibility

Relative advantage

**7. The development of the theoretical framework** 

factors influencing the adoption of biometric authentication.

Enabling factors

Contributory factors

Acceptance factors

for the Adoption of Biometrics in M-Government Applications Using Grounded Theory 199

the biometric registration and enrolment process and the current lack of research into security and m-government in the Kingdom of Saudi Arabia. These challenges, which directly influence the acceptance factors, need to be considered before starting the implementation of biometrics in m-government in order to enhance the acceptance factors.

Figure 3, below, illustrates the findings, focusing on the categories that have emerged from the open, axial, and selective coding phases. Based on the application of Grounded Theory, this theoretical framework, pictured, encompasses and organizes the concepts that form the

Challenges

**Factors influencing the adoption of biometric authentication in mgovernment security** 

Fig. 3. Theoretical framework for the adoption of biometrics in m-government security in

also be influenced by responses to technical and non-technical challenges.

Figure 3 depicts a new theoretical framework for the factors influencing adoption of biometric authentication in m-government security in Saudi Arabia, which was derived by the use of Grounded Theory as described above. Analysis and discussion of the results indicated that "entities factors", which include users', organisational, and system factors, as well as enabling factors involving acceptance and contributory factors, influence the adoption of biometric authentication in m-government security. As reported earlier in the previous section, this will

Related challenges

Technical challenges

> Entities' factors

> > System factors

Users' factors

Organisational factors

> Procedural issues

Requirements

Perceptions

Concerns

Needs

Needs

Perceptions

Services

Strategies

the system factors. Similarly, the categories of organisation's strategies, organisation's services, organisation's needs, and organisation's perceptions were combined to be 'organisational factors' as all of them related to the organisation and were mentioned as the organisation's viewpoints. Acceptance factors were also combined with contributory factors to be called 'enabling factors' as all the factors in this category were mentioned as factors that would enable the achievement of successful adoption of biometric authentication in mgovernment security.

It is noteworthy that the presented codes and categories in the above sections are the final set of the re-coding process. As suggested by (Miles and Huberman 1994), in order to ensure that codes and categories are applied consistently, it is significant for the researchers to verify all codes and categories that are assigned to the data. The benefit of re-coding process consisted of reconfirming and refining the codes and categories. Moreover, a comparison between newly and previously assigned codes and categories assisted us as well to check whether the codes and categories were reliable and truly represented the empirical data. The initial codes were mostly too close to the exact words of the data while the final codes provide more meaningful concepts.

As evidenced from the data analysis, one concept that was frequently stressed by the participants was the idea of acceptance. It had been introduced as a category involving: relative advantages, compatibility, ease of use, trialability, observability, trust, and privacy; and had been promoted to the "core" category. This, as known in Grounded Theory methodology, was because it linked up all of the other categories.

Another important category is the category of contributory factors which include availability, awareness, legislation, economical aspects, as well as social and cultural aspects. During the data analysis, it is found that both acceptance factors and contributory factors are involved in promoting the adoption of biometric authentication in m-government security.

Furthermore, the emerged categories show that organisational factors including organisation's strategies, services, needs, and perceptions influenced by users' factors which consist of users' needs, concerns, and perceptions. At the same time, both organisational and users' factors influence acceptance factors in different manners. For instance, applying biometric authentication along with a public key infrastructure (PKI) in m-government services is an important need of service providers due to the security relative advantage of this combination between biometrics and the PKI. This application also meets the users' need for protection of their personal and sensitive information through the use of mgovernment services. This combination of factors can consequently be seen to positively influence the acceptance of biometrics in m-government security among both users and service providers.

Similarly, system factors including system requirements and other related system issues such as the authentication responsibility and user registration at a website influence the acceptance factors and are at the same time influenced by users and organisational factors. This is because the majority of the system requirements, for example, have basically emerged due to organisations' and users' needs and perceptions.

It was also found that several challenges influence the adoption of biometric authentication in m-government security. These challenges include technical and related challenges such as the biometric registration and enrolment process and the current lack of research into security and m-government in the Kingdom of Saudi Arabia. These challenges, which directly influence the acceptance factors, need to be considered before starting the implementation of biometrics in m-government in order to enhance the acceptance factors.
