Project topics and materials
Get Complete Project Material File(s) Now! »
CHAPTER 3: RESEARCH METHODOLOGY
INTRODUCTION
A research methodology is a systematic process followed to conduct a research study [Kothari, 2005]. The research methodology assists in fulfilling the purpose of the particular study.
In the previous chapter, a literature review was presented. In this chapter, the research methodology for this study is described. A theoretical discussion on the research process onion is presented in Section 3.2. This consists of: the research philosophy, the research approach, the research strategy, the time horizon and the data collection method. Each layer of the research process onion is then used to discuss the research methodology that was followed in this study (Section 3.3). The design of the questionnaire (Section 3.4) and interview (Section 3.5) are presented before qualitative, quantitative and mixed method data analysis are described (Section 3.6). The importance of data triangulation is discussed in Section 3.7. The ethical manner in which the research was conducted is presented in Section 3.8. Finally, a summary highlighting the central elements of the research methodology adopted is presented in Section 3.9.
THE RESEARCH PROCESS
Research methodology and design can be viewed from many perspectives and is often seen as controversial [Knox, 2004]. Each study can follow a unique methodology in order to fulfil the purpose of the study. The Research Process Onion by Saunders, et al. [2003], Figure 10, is made up of different layers. The onion serves as a guide of to how to methodically approach research using different research philosophies, research approaches, research strategies, point of time horizons and data collection methods. A revised research process onion is presented in Saunders, et al. [2003]. However, the research process onion as presented in Saunders, et al. [2003, p. 83] is considered suitable for adaption and adoption for purposes of this study.
The process followed to define and design the research methodology involved peeling away each layer of the research process onion. This process starts with the outermost layer of the onion (research philosophy) and proceeds inwards until the core of the onion is reached (data collection methods). Each of these layers is discussed in Sections 3.2.1 to 3.2.5.
3.2.1 Research philosophy
Research stems from an underlying philosophical paradigm [Oates, 2006, p. 13]. „A paradigm is a set of shared assumptions or ways of thinking about some aspect of the world‟ [Oates, 2006, p. 282]. There are different research philosophical paradigms, but the most prominent are positivism, realism and interpretivism. These are illustrated in the outermost layer of the research process onion.
Positivism also referred to as the scientific method [Oates, 2006, p. 283], views the world as structured and ordered. Positivist researchers aim to look at the world from an objective and „real world‟ perspective [Cornford & Smithson, 1996, p. 37].
The realistic philosophy also commonly referred to as critical realism, states that „there is a reality independent of our thinking about it that science can study‟ [Trochim & Donnelly, 2007]. Critical realist researchers, who subscribe to this philosophy, believe that what we perceive as reality now can be altered later. The goal is to seek out new understandings of reality [Trochim & Donnelly, 2007].
The interpretivist philosophy is concerned with the social context of phenomena [Klein & Myers, 1999]. The focus is on interpreting meaning. „The purpose is to understand how others construe, conceptualize, and understand events and concepts‟ [Meredith, Raturi, AmoakoGyampah, & Kaplan, 1989, p. 307].
Oates [2006] states that the choice of the philosophical paradigm and execution of appropriate research approaches, research strategies, and data collection methods results in findings that correspond to the chosen philosophical paradigm.
Certain research philosophies correspond better to specific research approaches, research strategies, and data collection methods. However, the decision to adopt a research approach, strategy or data collection method should be considered on a case by case basis [Knox, 2004].
Section 3.3.1 discusses the research philosophy adopted in this study.
3.2.2 Research approach
The research approach affects how the research is carried out, that is, from a more general standpoint or from a more specific standpoint. Two research approaches, namely deductive and inductive reasoning, are illustrated in the second layer of the research process onion.
Inductive reasoning begins its focus from a specific view and works towards a more general and conceptual understanding of theory [Wills, 2007, p. 213]. The process of inductive
reasoning is also referred to as the „bottom-up‟ approach [Trochim & Donnelly, 2007], as depicted in Figure 11.
Deductive reasoning begins its focus on a general, holistic understanding of the theory and then abstracts to a specific subject of focus [Wills, 2007, p. 213]. The process of deductive reasoning is also referred to as the „top-down‟ approach [Trochim & Donnelly, 2007], as shown in Figure 12.
3.2.3 Research strategy
A research strategy provides pre-specified procedures that should be followed to address research questions and fulfill research objectives [Oates, 2006; Yin, 2003]. Six of the popular research strategies are illustrated in the third layer of the research process onion. Each of these six research strategies is briefly discussed next.
The first research strategy considered is an experiment. An experiment „is defined as a particular kind of research strategy that aims to isolate cause and effect by manipulation of what is thought to be the causal, or independent, variable and measurement of its effect on the dependent variable(s)‟ [Oates, 2006, p. 128]. Experiments can take place in a laboratory setting or in the field (also referred to as field experiments, quasi-experiments, natural setting experiments) [Oates, 2006]. Davis, et al. [1989] study on the TAM made use of laboratory experiments. The assessment of TAM in a laboratory setting was critiqued by Legris, et al. [2003], who recommended that real world, social context assessment of technology acceptance be carried out. However, experiments in a real world setting make it challenging to keep control over. Experiments require a significant representative sample of research participants in order for the research to be of any value [Oates, 2006].
The second research strategy considered is surveys. Surveys provide a means to „obtain the same kinds of data from a large group of people (or events), in a standardized and systematic way‟ [Oates, 2006, p. 93]. Several studies [Bueno & Salmeron, 2008; Buonanno, et al., 2005; Laukkanen, et al., 2007] evaluate the acceptance, adoption, and use of ERP systems by small and medium enterprises using the survey research strategy. What was common in these surveys was the number of research participants. Surveys usually involve a significant sample size of research participants [Oates, 2006] in order to support generalisation of findings. Surveys tend to focus on quantifiable findings and not necessarily on non-quantifiable findings [Oates, 2006].
The third research strategy considered is the case study. Yin [2003, p. 13] defines a case study as „an empirical inquiry that investigates a contemporary phenomenon within its real-life context, especially when the boundaries between phenomenon and context are not clearly evident.‟ A number of researchers [Illa, Franch, & Pastor, 2000; Iskanius, et al., 2009; Liang & Xue, 2004; Light, 2005; Molla & Bhalla, 2006; Muscatello, et al., 2003; Olsen & Sætre, 2007; Verville, et al., 2005] promote the use of the case study research strategy to assess user perceptions and evidence of ERP system intervention [Huang & Palvia, 2001; Loh & Koh, 2004; Markus & Tanis, 2000; Robert Jacobs & Ted’ Weston, 2007]
The fourth research strategy is grounded theory. The grounded theory research strategy „is a particular approach to qualitative research where the intention is to do field research and then analyse the data to see what theory emerges, so that the theory is grounded in the field data‟ [Oates, 2006, p. 274]. Glaser and Strauss [1967] advocate for this inductive reasoning strategy to inform theory. Although it is recognised as a research strategy in the information systems discipline, Oates [2006, p. 276] warns about the use of a specific version of the grounded theory research strategy. There is little research that makes use of grounded theory in the ERP system acceptance research domain. One recent example is a study by Nah, et al. [2005], in which the inductive reasoning research approach is used to develop a theoretical model of enduser acceptance of ERP systems.
The fifth research strategy that is considered is ethnography. Ethnography can be regarded as „an art and science of describing a group or culture‟ [Fetterman, 2010, p. 11]. Ethnography provides a rich account of what has been investigated in a particular context. However, this account may not be generalised to other contexts [Oates, 2006, p. 182]. There is little research that makes use of ethnography in the ERP system acceptance research domain. The study done by Lee and Meyers [2004] made use of a critical ethnography research strategy to understand factors impacting on an ERP system implementation within a small-to-medium sized enterprise. The findings from the ethnographic study showed that dominant actors, political agendas and changes in strategy all influenced the implementation of the ERP system.
The final research strategy represented is action research. Action research is a research strategy that involves the participation of the researcher. The researcher diagnoses a problem in a specific context. The researcher then plans to resolve the problem. The plan is then implemented. The second last activity involves evaluation to assess the resolution of the problem. Lastly, the researcher reflects on the outcomes from the intervention to resolve the problem [Oates, 2006]. Some examples of ERP system acceptance research that makes use of action research is [Akkermans & Helden, 2002; Deep, et al., 2008; Lim, et al., 2005].
3.2.4 Time horizon
Time horizons relate to when research is conducted. This can either be at a specific point in time, or across a specific period. Two time horizons, namely cross-sectional and longitudinal, are illustrated in the fourth layer of the research process onion.
Cross-sectional research takes place at a single, specific point in time [Trochim & Donnelly, 2007]. This involves collecting information at a particular section or slice of time. This can be data about the past, the present or the future.
Longitudinal research takes place over a specific period of time [Trochim & Donnelly, 2007]. This involves collecting information from one point in time across to another point in time.
Karahanna, et al.‟s [1999] study focuses on the pre-adoption and post-adoption of information technology and makes use of cross-sectional research. Karahanna, et al. [1999] state that findings from the cross-sectional study may not be conclusive in describing the complexity and time period of information technology adoption. Karahanna, et al. [1999] suggest follow up longitudinal studies to compare findings and give more insight to the process of the adoption of information technology over time.
3.2.5 Data collection method
A data collection method, also referred to as data generation method, provides a means to gather research findings [Oates, 2006, p. 36]. There are five data collection methods, including: sampling, secondary data, observations, interviews, and questionnaires. These are illustrated in the core (fifth layer) of the research process onion.
The first data collection method is sampling, which refers to the process of selecting a sample from a whole population, which is then included in the research strategy [Oates, 2006, p. 95]. There are several aspects to consider when sampling. These sampling aspects include identifying the sampling frame (potential sample), sampling technique (probability and nonprobability sampling techniques to select the actual sample), response rate and non-responses, and sample size. It is recommended that an adequate sample size form part of the study in order to assist in generalisation of research findings [Bartlett, Kotrlik, & Higgins, 2001]. Nonprobability purposive sampling techniques were used in some ERP system acceptance and adoption studies , such as that done by Poba-Nzaou, et al. [2008]; non-probability convenience sampling techniques were used in the study done by Ramayah and Lo [2007]. Nonprobabilistic sampling means that it is not necessary to have a representative sample. The researcher can also hand pick (purposive sampling) the sample to fulfil the purpose of the study or select participants who are convenient (convenience sampling) to reach. The second data collection method is secondary data. Secondary data includes documents, data, and information from previous studies that a researcher might use in a new study [Oates, 2006, p.234]. Some ERP system studies [Finney & Corbett, 2007; Hakim & Hakim, 2010] make use of secondary data as part of the research methodology.
The third data collection method is observations. Observations are data collection methods that observe what participants actually do [Oates, 2006, p. 202]. This is not just about seeing participants act within a context; it involves a careful assessment of the environment and the behaviour of the participant under observation. Observation could involve the senses of sight, sound, touch, taste and smell, depending on the context. The researcher could act as an invisible observer or active participant in the research process. Some ERP system studies that made used of on-site observations include: [Häkkinen & Hilmola, 2008; Lim, et al., 2005].
The fourth data collection method presented was interviews. Interviews allow the researcher to constructively communicate with the research participant to obtain detailed information that cannot otherwise be obtained using other data collection methods [Oates, 2006, p. 187]. The researcher also observes the research participant during the interview interaction to assess possible changes in emotion or emotional responses to sensitive questions [Nandhakumar & Jones, 1997]. There are three forms of interviews [Oates, 2006, pp. 187-188]. Interviews can be structured (pre-defined set of standard questions only), semi-structured (research participant responses can change the standard structure of pre-defined questions), and unstructured (no pre-defined set of standard questions). Many ERP system research studies make use of interviews. The study by Deep, et al. [2008], for example, made use of semistructured interviews to assess ERP system limitations and suggestions for ERP system improvements. A study by Buonanno, et al. [2005] made use of personal, direct interviews that accounted for a good response rate in order to assess factors affecting ERP system adoption.
The final data collection method assessed was questionnaires. Questionnaires make it easier to collect large amounts of pre-defined data in a pre-determined order over a shorter period of time [Oates, 2006]. Questionnaires are often related to the survey research strategy although questionnaires can be used as part of other research strategies, such as case studies [Oates, 2006, p. 219]. The questionnaire data makes it easier for researchers to look for patterns within the research findings. These patterns can be used to generalise findings from the sample to the larger population under study [Oates, 2006]. Questionnaires can be self-administered (completed by the research participant without the assistance of the researcher) or researcheradministered (the researcher completes the questionnaire after asking for a response from the research participant). Several ERP system research studies have made use of formal questionnaires to gather responses from users of ERP systems [Reuther & Chattopadhyay, 2004], top level management [Raymond & Uwizeyemungu, 2007], as well as small, medium and large enterprises [Bernroider & Koch, 2000; Koh & Simpson, 2005] regarding perceptions about ERP system use.
APPLIED RESEARCH PROCESS
The research methodology adopted in this study is discussed in Sections 3.3.1 to 3.3.5.
3.3.1 Research philosophy
The interpretivist research philosophy was adopted, as it was the interpretation and meaning of the findings [Klein & Myers, 1999; Walsham, 1995] of small manufacturing enterprise ERP system acceptance that was of interest to this study.
Multiple and diverse facets of reality were considered, rather than one view that is evident of „the truth‟ [Oates, 2006, p. 292]. A small manufacturing enterprise is not a generic enterprise. There are a number of small manufacturing enterprises operating in different manufacturing sectors, each with a diverse understanding of sector-specific terminology. Small manufacturing enterprises in one manufacturing sector have similarities with small manufacturing enterprises in other manufacturing sectors as well as differences. In terms of objectivity of research, research reflexivity [Oates, 2006] was maintained. In interpretivism, it is believed that the interpretivist researcher, through detailed understanding and analysis infers own beliefs in the interpretation of findings. Walsham [1995] refers to this dilemma as second-order concepts in interpretive studies. All perspectives were taken into account and distinguished to differentiate interpretation of field findings from personal interpretation.
3.3.2 Research approach
While inductive reasoning is commonly associated with interpretivist views [Knox, 2004], the deductive reasoning research approach is seen as more appropriate for adoption. This is in line with the view of Knox [2004] that the research philosophy should not limit the selection of research approach, as it should be carefully considered in terms of the context and appropriateness of the study. Deductive reasoning allowed me to first concentrate on collecting literature findings on small manufacturing enterprise ERP system acceptance factors and then to focus on obtaining field findings to confirm or refute the findings and to comment on these literature findings. Figure 13 illustrates the top-down research approach followed. This research approach involved first reviewing literature that focuses on small manufacturing enterprises and ERP system acceptance. The next step was to clearly define the research questions to be addressed. The third step involved conducting „observations‟. Two forms of „observations‟ were conducted. The first „observation‟ involved gathering data from small manufacturing enterprises through the use of a survey questionnaire. The second „observation‟ involved gathering data from ERP system consultants through the use of interviews. The final step involved analysing the findings collected from the research participants to confirm or refute the findings, and comment on the literature findings, that is, theory of small manufacturing enterprise acceptance of ERP systems.
3.3.3 Research strategy
An interpretive survey research strategy was adopted. Surveys are traditionally associated with positivistic research, where the focus is on discovering the same patterns of findings across a substantial number of participants under study, and then generalising these to the larger population [Oates, 2006]. Surveys can, however, be associated with interpretive studies [Oates, 2006, p. 93].
The motivation for using this research strategy was not only to inform an understanding of acceptance of ERP systems by small manufacturing enterprises, but to find possible patterns of ERP system acceptance that could not easily be collected using alternate research strategies.
The survey was not intended to generalise the findings of the survey, since the sample size (Section 3.3.3.4) was not significantly representative of small manufacturing enterprises in South Africa. The interpretive findings provided insight into acceptance, which can be replicated in further, large scale surveys of small manufacturing enterprises in terms of their experiences and expectations for using ERP systems. Sections 3.3.3.1 to 3.3.3.4 discuss the process of sampling research participants.
3.3.3.1 Sampling frame
The small manufacturing enterprises were selected by consulting business databases and telephone directories of small manufacturing enterprises in Gauteng, South Africa. These databases and directories contained contact information for potential small manufacturing enterprises to be sampled from.
3.3.3.2 Sampling technique
As indicated in Section 3.2.5, studies make use of various sampling techniques and the use of non-probability purposive sampling techniques, e.g. Poba-Nzaou, et al. [2008], are common when researchers prefer not to generalise, but to understand the topic in-depth. This survey used a non-probability purposive sampling technique, as it enabled more freedom in terms of the number of issues that could be investigated, such as instances that may be different, „extreme, unusual or somehow atypical‟ [Oates, 2006, p. 98].
3.3.3.3 Response rate and non-response
Telephone calls and physical visits to small manufacturing enterprises assisted in obtaining research participants. It was easier to generate interest in participation with physical visits rather than via telephonic conversations with small manufacturing enterprise representatives. There were three main reasons for non-participation of small manufacturing enterprises either the enterprises were: too busy during the period August to November 2009; interested but not keen; or not interested at all in participating in the study.
3.3.3.4 Sample size
The final sample size of small manufacturing enterprises that participated in the survey was 16. It is realised that although the sample size may not yield statistically interesting results, nonetheless, it provided important insights into ERP system acceptance by small manufacturing enterprises. As mentioned, the focus of this study is not to generalise, but to explore this relatively young research domain.
3.3.4 Time horizon
A cross sectional time horizon was adopted for this study. This study did not wish to examine the longitudinal changes to ERP system acceptance.
There are several periods associated with cross sectional research [Oates, 2006]. Historical and short-term studies are examples of two types of studies based on the aspect of time. Historical studies focus on past events; short-term studies examine present events.
A short-term study approach was adopted. However, because of the dynamic nature of this study, i.e. incorporating small manufacturing enterprises that use ERP systems and those that have not; questions relating to previous and current experiences, as well as to future expectations regarding ERP systems were asked as part of the survey.
3.3.5 Data collection method
Questionnaires and interviews are widely used data collection methods in information systems research [Oates, 2006]. However, before one selects a data collection method, one needs to understand the type of data that needs to be collected. It was important to collect two types of data in the survey: data indirectly related to the study and data directly related to the study.
Indirectly related data included demographic information, such as manufacturing sector, year of establishment of enterprise, age of employees, and gender of employees. Data indirectly related to the study are presented in Chapter 4: Sample of Small Manufacturing Enterprises Surveyed.
Data that focuses on the acceptance of ERP systems, particularly strategic, business, technical, and human acceptance factors relate to data that is directly related to the study. Data that is directly related to the study are presented in Chapter 5: Experiences of Using ERP Systems and Chapter 6: Expectations for Using ERP Systems.
The design of the questionnaire used as part of the survey, is discussed in Section 3.4. The details of the interview and questions used as part of the data triangulation exercise (Section 3.7) follows in Section 3.5.
Table of Contents
CHAPTER 1: INTRODUCTION
1.1 INTRODUCTION
1.2 BACKGROUND
1.3 THESIS STATEMENT AND PURPOSE OF STUDY
1.4 RESEARCH OBJECTIVES AND QUESTIONS
1.5 THE SCOPE AND CONTEXT OF THE STUDY
1.6 RATIONALE BEHIND STUDY
1.7 SIGNIFICANCE AND POTENTIAL CONTRIBUTION OF THE STUDY
1.8 RESEARCH APPROACH
1.9 ETHICAL CONSIDERATIONS
1.10 DISSERTATION STRUCTURE
CHAPTER 2: THEORETICAL FRAMEWORK
2.1 INTRODUCTION
2.2 UNDERSTANDING TECHNOLOGY ACCEPTANCE AND TECHNOLOGY ADOPTION
2.3 ERP SYSTEMS
2.4 SMALL MANUFACTURING ENTERPRISES
2.5 TECHNOLOGY ACCEPTANCE OF ERP SYSTEMS
2.6 CURRENT STATUS OF RESEARCH ON ACCEPTANCE OF ERP SYSTEMS BY SMALL MANUFACTURING ENTERPRISES
2.7 SUMMARY
CHAPTER 3: RESEARCH METHODOLOGY
3.1 INTRODUCTION
3.2 THE RESEARCH PROCESS
3.3 APPLIED RESEARCH PROCESS
3.4 QUESTIONNAIRE DESIGN
3.5 INTERVIEW DESIGN
3.6 QUALITATIVE AND QUANTITATIVE DATA ANALYSIS
3.7 DATA TRIANGULATION
3.8 ETHICS
3.9 SUMMARY
CHAPTER 4: SAMPLE OF SMALL MANUFACTURING ENTERPRISES SURVEYED
4.1 INTRODUCTION
4.2 SAMPLE SURVEYED
4.3 MANUFACTURING SECTORS
4.4 YEARS OF ENTERPRISE OPERATION
4.5 NUMBER OF EMPLOYEES
4.6 SUMMARY
CHAPTER 5: EXPERIENCES OF USING ERP SYSTEMS
5.1 INTRODUCTION
5.2 FAMILIARITY WITH ERP SYSTEMS
5.3 FACTORS THAT INFLUENCE THE USE OF ERP SYSTEMS
5.4 FACTORS THAT INFLUENCE THE ACCEPTANCE OF ERP SYSTEMS
5.5 SUMMARY
CHAPTER 6: EXPECTATIONS FOR USING ERP SYSTEMS
6.1 INTRODUCTION
6.2 FAMILIARITY WITH ERP SYSTEMS
6.3 FACTORS THAT INFLUENCE THE ACCEPTANCE OF ERP SYSTEMS
6.4 SUMMARY
CHAPTER 7: DISCUSSION OF RESEARCH FINDINGS
7.1 INTRODUCTION
7.2 TRIANGLUATION EXERCISE
7.3 ERP SYSTEM ACCEPTANCE FACTORS
7.4 UTAUT ACCEPTANCE FACTORS
7.5 PERCEIVED FACTORS INFLUENCING THE ACCEPTANCE OF ERP SYSTEMS BY SMALL MANUFACTURING ENTERPRISES
7.6 SUMMARY
CHAPTER 8: CONCLUSION
8.1 INTRODUCTION
8.2 SUMMARY OF KEY CHAPTERS
8.3 REFLECTION OF KEY FINDINGS
8.4 REFLECTION ON EMERGING THEMES
8.5 SIGNIFICANCE AND CONTRIBUTION OF RESEARCH
8.6 LIMITATIONS OF RESEARCH
8.7 RECOMMENDATIONS FOR FURTHER WORK
8.8 CONCLUDING STATEMENTS
Bibliography
GET THE COMPLETE PROJECT
EXPLORING THE ACCEPTANCE OF ENTERPRISE RESOURCE PLANNING SYSTEMS BY SMALL MANUFACTURING ENTERPRISES
