CHAPTER 3 METHODOLOGY AND DESIGN
This chapter presents the research methodology and provides a detailed description or overview of how this research project was conducted. The application of the research objectives as described in Chapter 1 will be reviewed and the steps and processes that were undertaken will be described.
The reader will thus be led through the research process from the planning stage, through to the data-collection phase, then the data analysis and the development of the metric and what it means. Consequently, the conceptual model that the researcher developed in Chapter 1 will be clarified.
As expected in any research study such as this, the credibility of the findings, recommendations made, and framework developed are closely linked to the research design, data-collection techniques, and data analysis (Birkhofer 2011:17). Thus the research method and procedures that were used to obtain the analysed data will be described and closely examined to support the veracity of the findings (Birkhofer 2011:11).
This research study was conducted in a field where very little research has been conducted before, and consequently there is limited literature available locally (Babbie, Mouton, Payze, & Prozesky 2010:160). Research projects in the academic field or in industry could be a systematic endeavour to unearth answers to problems that may not easily be understood, defined, or demonstrated without intensive intellectual work (Simons 2009:10). That is how researchers see it and want to engage with it.
Before any scientific research project can be undertaken within the broader research community, the researcher should first identify the methodology that will be used in order to find answers to the research questions that were formulated (Babbie et al. 2010:161). The actual research can only begin once the methodology has been clearly disclosed, discussed, and agreed upon by all the stakeholders (Bwalya 2009:49).
GOALS AND PURPOSE OF THE METHODOLOGY
The purpose of the methodology chapter is to clarify the procedures before the study was conducted, and to describe the conditions under which the data were collected and analysed, and to explain the core of the research. The credibility of the results is linked to the research principles as stipulated in the methodology and data collection (Ridley 2009:19).
It is necessary to specify the population, sampling techniques, data-collection approach, research design, and any other element that may have an effect on the scientific authenticity of the manner in which the research was conducted (Ridley 2009:11).
Any research outcome that cannot be replicated is known as an isolated occurrence (Welman & Kruger 2003:41). The replication of a study is meant to ensure the scientific nature of the previous outcome; in other words, if correctly implemented, the outcome can be replicated by another researcher (Simons 2009:67).
In many instances, scientific researchers place great emphasis on the possible replication of the outcome whenever embarking on a new research venture, and this can only be irrefutable if the correct research methodology was followed and implemented (Ridley 2009:17).
Any research project can fall under a variety of categories, depending on the anticipated outcomes and expectations of the researcher. The approach taken by the researcher could help determine the classification of the research project (Ridley 2009:15).
In most academic environments, there are two distinctive types of research: basic and applied research. Although some researchers use these terms interchangeably, they should be seen and defined differently (Welman & Kruger 2003:15).
Applied research is fundamentally built on systematic inquiry involving the practical application of the science that guides it. It pays attention to the research community, or mostly the academic research sector, by accumulating theories, knowledge, methods, and techniques for specific outcomes and deals with solving practical problems based on empirical methodologies (Krueger & Casey 2014:47).
Ridley (2009:305) stated, “Applied research refers to scientific study and research that seeks to solve practical problems. Applied research is used to find solutions to everyday problems, cure illness, and develop innovative technologies.” Simply put, applied research is used to solve immediate and practical problems facing the scientific world.
Based on the above knowledge of both approaches and the type of research expected, the researcher believes that applied research was appropriate for this study since the intention of this research project was to solve an existing problem by finding ways to explore computing usability problems for adult first-time users and thus narrowing the digital gap in the most remote and impoverished rural areas in South Africa (Mphidi 2011:51).
Exploratory study research approach
Generally speaking, a research approach may be defined as the process of developing new knowledge or re-evaluating the outcomes of existing ones. It enables the researcher to have a clear view or perspective of the whole project from the onset, and enables the reader to have a preview of what the outcome may be (Bauman 1991:31).
There are a few orientations, of which the most popular are exploratory, descriptive, and explanatory studies. The following paragraphs will elucidate the important facets of exploratory studies, why it is the most appropriate research approach, and also shed some light on its application.
Exploratory studies are conducted when insufficient research has been conducted in a certain area of interest. The main purpose of conducting an exploratory study is to help understand and determine the best manner in which to approach the problem emphasised in the research design (Bruyn 1966:41).
Data collection, methods, and the selection of the subjects will be thoroughly investigated during exploratory research (Buchanan et al. 2006). When choosing an exploratory research approach, the outcome could be contradictory when compared with the earlier assumptions made by the researcher. More often than not, exploratory research enables the researcher to rapidly unearth new meaning (Cheng, Ernesto & Truong 2008:77). Exploratory research may be a better alternative to gather preliminary information that will help the researcher to develop a hypothesis (Mehl et al. 1994:28).
Particularity of the current project
After having listed and defined the existing types of research approaches, the researcher was of the view that an exploratory research approach was most appropriate as it enabled the provision of an in-depth explanation of each sub-question. There are very few locally based research projects which identify the extent of the digital gap in HCI efficiency, effectiveness, adult first-time user preferences, and the negative usability impact on adult first-time users of handheld devices.
As a result of this lack of literature on the topic, a number of questions were raised; for example: What is a first-time computing user? What is good usability and what is not? What is system adaptability time, good and poor usability, speed of operation, eye-tracking processes, etc.?
Adult first-time users of handheld computing devices are the focus of this research study, and local users, like computing users in other underdeveloped nations, have their own particular needs which should be explored before being classified. Computing systems usability causes a variety of anxieties, as reported by many international studies; however, in South Africa, these aspects regarding frustration need to be investigated through exploratory research (Mphidi 2011:17).
The research method describes the process of collecting and analysing data with the aim of making interpretations based on preliminary criteria as structured by the researcher. The goal of the research method is to define the analysis approach (Elmer-DeWitt 1996:7). In so doing, future researchers who may want to replicate the research will be able to understand what benchmarks were considered at the time (Morgan 1993:5).
There are many types of research methods that are used for scientific research. Quantitative data-collection methods involve numerical data, or any form of data that can be quantified (Fowler et al. 1999:61), while qualitative methods are mostly based on literary explanations of the phenomenon and give reasons, motivations, and opinions about a potential exploratory question (Fowler et al. 1999:111). Primarily, it helps to understand the research question as well as the answers to those questions (Gamson 1992:71).
The mixed-methods approach combines both quantitative and qualitative research methods, and aims to support each area and provide a deeper understanding of where there are shortfalls (Gill 2008:40).
The mixed-methods approach is used in this study based on the fact that where justification or substantiation is required for an answer to an open-ended question, this approach is used in order to fully comprehend the outcome of a numerical answer.
In any research project that involves more than one person, considerable resources are required (Groves 1990:41). There should be an agreed-upon blueprint that should stipulate the research process from the beginning of the study to the finalisation of the study (Grimes 1991:13).
Research design also helps to determine the type of data-collection design that will be implemented. There are three types of research design; namely experimental, quasi-experimental, and non-experimental (Parker 1992:67).
The researcher chose the quasi-experimental design for this study, as random sampling was applied when choosing the population but there was control to identify their place of origin and the characteristics for their selection.
The research was conducted in all nine provinces in South Africa. The purpose of defining the research population is to ensure that the researcher was clear about where and from whom data were collected. It also meant conducting a meaningful assessment of the potential participants, knowing who they were, where they were, and the exact number of participants in each group.
When dealing with a large population participating in a research study, it is important to group them in a cluster called a target population. A target population describes a group of individuals, objects, and aspects that may contribute to the meaningful and positive identification of the type of input they may provide (Kahane 1992:41). The target population for this research study are those who have neither been exposed to any form of meaningful ICT device nor any handheld computing device apart from cell phones. Because of the geographical location of the target population and their usability of handheld devices, selection criteria depended on those who may have been exposed to handheld computing devices despite not having owned one at the time of conducting this research.
The researcher approached all participants in their most natural environment, where questionnaires were administered. Usability demonstrations and observations before and after training were recorded on video. In order to access the most important participants in each province, age, gender, and education level were not considered as restrictions but rather as independent demographics aspect of the whole study.
The previous section described the population of the research project, but one also needs to be aware of the fact that the population as described did not at all times participate in the study, and therefore other means of selection were needed to help focus on those who are more likely to participate in the research. Sampling was used to specify the demographics of the participants who were expected to participate in the study (LeCompte & Preissle 1993:17).
The sample was selected from potential computing users from rural communities in all the South African provinces. To be eligible for selection, the participants had to be able to read and write in English and have an understanding of the importance of a computing device and its usefulness. A stratified sampling method was used to select the participants from whom data would be obtained for analysis (Lofland 1995:31).
Upon arrival at the identified venue, all participants received a brief description of the research study and what it entailed. After having completed the administrative procedures, which required the completion and signing of a waiver and ethical clearance documents, the participants were placed in a control location where they began interacting with the setup of computing devices. This was recorded on a fixed digital camera. All participants contributing to system usability were recorded on video. The testing device was installed at the venue in advance and the participants were informed that they were being recorded. All data collected in this process were analysed.
In this experiment, the first-time user participants demonstrated their usability of handheld devices. The researcher travelled around the community and explored aspects that could contribute to displaying technological limitations, user frustration, and system adaptation for that specific community. All aspects pertaining to the above were recorded on video for later analysis.
The researcher approached communities, churches, schools, and business leaders and administered questionnaires which consisted of open-ended questions to the participants. These questionnaires enabled participants to answer questions based on their own experience with a handheld computing system and the impact of system usability as a hurdle. The types of questions used in the questionnaire were meant to enable participants to air their views and opinions.
The questionnaires had blank spaces for written comments. Closed-ended questions were included and were structured in such a manner that participants could only answer them in a pre-defined section. These questions were mostly entry-level questions or questions that did not lead to long answers. The questionnaires are included in Annexure 2, 4 and 5.
At this stage, it is important to explain that not only surveys were used as a technique to collect data; other techniques were also used, which will be introduced as the process unfolds. These included experiments and interviews.
Lazar et al. (2010:68) stated,
“One data-collection effort does not lead to a definitive answer on a question of research. In scientific communities, the goal is generally for multiple teams to examine the same research question from multiple angles over time. All of these efforts come up with same general findings over time, give evidence for the scientific truth of the findings; this is often known as triangulation.”
The researcher attempted to apply this approach as closely as possible.
DATA ANALYSIS AND PRESENTATION OF FINDINGS
Once the data were collected, the researcher made use of QDA Miner to analyse the quantitative data. It features built-in capabilities that enable the analysis of both qualitative and quantitative data and is widely used in the fields of social sciences, medicine, sociology, political sciences, and psychology. This software is best used for coding, annotation, retrieving, and analysis of small and large documents.
Qualitative data analysis
The qualitative data were analysed using ATLAS.ti, which is a well-known commercial software system. The purpose was mostly to provide an explanation, understanding, or interpretation of what the respondents said during their interview. The researcher took some time to review the theories under investigation, which led to the main research question of this study.
The researcher then themed important points of interest to the responses that led to the question at hand; this was achieved by grouping lower-level data across the corresponding level. The researcher then identified the ideal characteristics, which are also known as single items or responses given by a single respondent.
Coding the data was one of the most time-consuming activities during this research. During this phase, the researcher attached important labels to lines of text for comparison purposes and later sorted them according to the lines of ideas; indexing was also conducted as part of the same process. Narrative analysis was systematically used, which is based more on transcribed data and which is an ideal method for the core activity of reformulating stories presented by participants in a manner that is better understood.
Quantitative data analysis
In this section, there were two separate sets of quantitative data, namely data obtained as the result of the experiments and computed using various mathematical formulas, and data collected before and after the experiment and analysed using the Statistical Package for the Social Sciences (SPSS), which is basically a quantitative data analysis software system. The results were all interpreted by the researcher and the report is presented in corresponding chapters for reader perusal.
This study used a variety of research methods which were independent in their own right and needed to be incorporated into a single research objective for global or broad analysis. For this reason, the research approaches were mixed to enable cross-completion of various sub questions which were all part of the same main research question. For example, if the objective, which is the main research question of the chapter, contains three or four sub-questions which are answered using multiples methods or approaches, one method would then be used to support the other one and vice versa.
The researcher then introduced the quantitative report pertaining to the same main objective, which is based on the question under investigation, followed by the respondents’ responses, then the quantitative data and analysis, as well as implementing a mixed interpretation in support of the quantitative data. The third section to complete the triangulation of the data is the result of the experiment where applicable; it is located at the end of the individual section.
TABLE OF CONTENTS
TABLE OF CONTENTS
LIST OF FIGURES
LIST OF TABLES
LIST OF PAPERS PUBLISHED
CHAPTER 1 STUDY PRELIMINARY
1.3 RATIONALE FOR THE STUDY
1.4 MAIN RESEARCH QUESTION
1.7 OBJECTIVES OF THE STUDY
1.8 PRELIMINARY METHODOLOGY
1.9 BENEFITS OF THE RESEARCH STUDY
1.10 ETHICAL CONSIDERATIONS
1.11 OUTLINE OF THE RESEARCH
CHAPTER 2 LITERATURE REVIEW
2.3 WHY IS THIS RESEARCH OF INTEREST TO ACADEMICS?
2.4 HANDHELD COMPUTING DEVICES
2.5 COMPUTING MOBILITY NEED
2.7 HANDHELD COMPUTING SYSTEMS
2.8 THE USER
2.9 INTERACTION WITH HANDHELD SYSTEMS
2.10 DIGITAL DIVIDE IN SOUTH AFRICA
2.11 ADULT FIRST-TIME USERS OF HANDHELD COMPUTING DEVICES
2.12 LINK BETWEEN ADULT FIRST-TIME USERS, DIGITAL GAP, USABILITY, AND TABLET SYSTEMS
2.13 FITTS’ LAW
2.14 CURRENT AND SIMILAR PROJECTS
2.15 THE ROLE OF METRICS
2.16 CONCEPTUAL FRAMEWORK
CHAPTER 3 METHODOLOGY AND DESIGN
3.2 GOALS AND PURPOSE OF THE METHODOLOGY
3.3 RESEARCH APPROACH
3.4 RESEARCH METHODS
3.5 DATA-COLLECTION PROCEDURE
3.6 DATA ANALYSIS AND PRESENTATION OF FINDINGS
3.9 PROJECT APPLICATION FRAMEWORK
3.10 EXPERIMENT AND LOGISTICS
3.11 FORMULA EXPLORATION
3.12 PARTICIPANT TRAINING
3.13 EXPERIMENT PHASES
3.14 AREAS WHERE DATA WERE COLLECTED
3.15 APPLICATION OF THE METHODOLOGY IN EACH OBJECTIVE
3.16 QUESTIONNAIRES, ACTIVITIES, AND TASKS FOR EXPERIMENT
3.17 ETHICAL CONSIDERATIONS
3.18 ACCURACY AND ERROR CHECKING
CHAPTER 4 QUANTITATIVE AND QUALITATIVE DATA ANALYSIS
CHAPTER 5 EXPERIMENT DATA AND ANALYSIS
5.2 EXPERIMENT SITE SETUP
5.3 PRE-TRAINING, EXPERIMENT FINDINGS, AND REPORT ANALYSIS
5.4 TRAINING AND TABLET SYSTEMS INDUCTION
5.6 RESULTS INTERPRETATION
5.7 PRELIMINARY SUMMARY
CHAPTER 6 UX METRIC (UXM), POST-EXPERIMENT, AND FRAMEWORK
6.2 THE METRIC
6.3 THE LINK BETWEEN UX AND THE DIGITAL GAP
6.4 POST-EXPERIMENT QUESTIONNAIRE AND REPORT
6.5 DIGITAL CLOSURE FRAMEWORK
CHAPTER 7 CONCLUSION
7.2 REFLECTING ON THE JOURNEY
7.3 RESEARCH AIMS
7.4 RESEARCH QUESTIONS
7.5 CONTRIBUTION MADE TOWARDS THE BODY OF KNOWLEDGE
7.6 FUTURE RESEARCH CONTRIBUTION
GET THE COMPLETE PROJECT