Memoire Online - Influence of an ERP system on the value chain process of multinational enterprises (mnes)

CHAPTER 5: RESEARCH METHODOLOGY

5.1 Introduction

In this chapter the methodology and research design utilised in this study are described. The type of research, the population, the data collection, as well as the sampling methodology and procedures, and data analysis used in this study is also outlined. Included in this chapter are the description of the methods used to collect the qualitative and quantitative data relating to the objectives and hypotheses formulated in this study.

5.2 Qualitative and quantitative study

This study involved both quantitative and qualitative research. Kruger and Welman (2002:191) stipulate that qualitative research is not concerned with the methods and techniques to obtain appropriate data for investigating the research hypothesis, as in the case of quantitative research. Qualitative data is based on meanings expressed through words and other symbols or metaphors. Qualitative studies can be used successfully in the description of groups (small) communities and organisations by studying cases that do not fit into particular theories. Thus, in this study qualitative research was applied to compare the different ways in the literature of linking strategy with IT through the value chain approach, and the general theory of ERP systems with the Axapta Microsoft software information. The quantitative research dealt with the factors associated with IT strategy, the ERP concept, usage and selection. This data was obtained through mail survey questionnaires in order to investigate some of the research objectives and hypotheses of this study (see table 5.2).

For the qualitative part of this research, the literature review dealt with the primary objective, as well as secondary objectives 1, 2, 3 and 4. For the empirical part of this research, the study involved a self-administered survey, which attempted to highlight the primary objective, as well as secondary objective 3 and hypotheses 1, 2, 4, 5 and 6 (H1, H2, H4, H5 and H6) to answer the overall hypothesis 3 (H3) of this study. Objective 5 is linked to chapter 7.

5.3 Research design

According to Kruger and Welman (2002:94), the design of a study concerns the plan to obtain
appropriate data for investigating the research hypothesis or question. Data collection tools for
surveys include interviews and questionnaires. In this study interviews allowed the researcher to

clarify answers and follow up on interesting answers, while questionnaires were designed to be self-administered, since they can be mailed to a larger number of respondents. This short-dissertation was conducted using primarily a case study, which laid the foundation for the exploratory and empirical study. Interviews were used to explore the organisations using ERP systems, i.e. SAP software. Secondary to this, the researcher conducted an exhaustive literature review of the research topic. This involved the study of appropriate textbooks and conference papers, and the maximal use of Web-based documents that were used in the discussion of literature assessed in the case study. The study falls into the category of descriptive research, which aids in accommodating larger sample sizes, thus giving the research findings more generalisability than those of exploratory or qualitative designs (Hair, Buch & Ortinau, 2003:256).

Melville and Goddard (1996:44-5) point out that non-returns are a particular problem with questionnaires. Repeated follow-ups are most effective for reducing the non-response rate. Most research in the world is hampered by constraints of resources, subjects and time. Furthermore the researcher's work is complicated by many sources of bias and error that must each be dealt with as effectively as possible to ensure the high quality of the research (Bless & Higson-Smith, 1995:79). Thus, in this study, the researcher dealt with the non-response rate through the follows-ups. Non-respondents were contacted by telephone to remind them about the questionnaires that they had been asked to complete.

5.4 Methods of collecting quantitative data

The researcher chose self-administered surveys and structured telephone interviews to achieve the research objectives of this study. Hair et al., (2003:265) note that a self-administered survey is a data collection method where respondents read the survey questions and record their responses in the absence of a trained interviewer.

In this study, the researcher first conducted an exploratory telephone interview survey to identify the participants, especially the MNEs operating with any ERP system software. From that, the researcher contacted the key informants by telephone, most of them IT managers, to explain the purpose of this study. Respondents were also asked for their participation in the survey, which therefore allowed the researcher to compile the list of companies that use ERP software. These companies were used in the self-administered survey (see appendix E).

The participants in the survey of this study were CEOs/CIOs, managers (general, senior, middle and junior) and the end-users in the MNEs that use ERP, i.e. SAP software. The questionnaires were distributed to the IT human resource department in the nominated companies. For some of the MNEs, questionnaires were distributed through the general human resource department.

In total, 11 MNEs using SAP software were identified during the preliminary structured telephone interview survey. Only five MNEs using SAP software in Gauteng were involved in the self-administered surveys. The limited number was owing to budget constraints (see appendices D and E).

The researcher chose Axapta Microsoft solution from the available ERP software to be included in the qualitative study for several reasons:

· Firstly, it is designed for MNEs and has the ability to integrate the value chain so that the organisation can act globally and increase competition.

· Secondly, it is one of the leading ERP software systems in the world for manufacturing and service organisations.

For the quantitative study, SAP software was selected owing to its popularity and utilisation in the MNEs randomly selected through the preliminary structured telephone interview survey conducted by the researcher, as indicated in table 5.1 below.

Table 5.1: The usage of ERP software in randomly selected MNEs

Number of companies using ERP	Type of ERP software used	Size of company	4
Oracle	Medium	11
SAP	Medium	4
JD Edward	Medium	1
Axapta Microsoft solution	Medium	1
ARIBA	Medium

Source: Obtained from the preliminary structured telephone interview survey conducted by the researcher.

5.5 Sampling method and sample size

5.5.1 Sampling method

Sampling is defined as the selection of a small number of respondents from a larger defined target population, assuming that the information gathered from the small group will allow the researcher to make generalisations or judgements concerning the larger group (Hair et al., 2003:333). According to Kruger and Welman (2002:46), there are two sampling designs:

· The probability sample involving simple random samples, stratified random samples, systematic samples and cluster samples; and

· The non-probability sample involving accidental or incidental samples, purposive samples, quota samples and snowball samples.

In probability sampling, the probability that any element or member of the population will have to be included in the sample can be determined. This is not the case in non-probability sampling. Considering the advantages and disadvantages of the sampling methods discussed above, it was concluded that a probability sampling design involving simple random samples had to be implemented in this study to compile the representative samples of employees in the MNEs. The researcher selected five out of eleven MNEs from a random list of the organisations using SAP in Gauteng (appendix E). By focusing only on these organisations in one province, the sample constituted a good representation of the organisations operating in Gauteng with a total turnover of nearly R1 billion per annum. Most of these organisations have been classified in the sector of financial services and manufacturing.

5.5.2 The sample size

The sample size of this study consists of five MNEs. In total 75 self-administered surveys were issued to each MNE, with the total sample size (N) equal to 375 participants. (One self-administered survey targeted one CEO/CIO with N = 1; one self-administered survey targeted managers (general, senior, middle and juniors) in each MNE with N = 30, and 44 end-users of the IT department of each MNE, N= 44). See table 5.3 (Response rate to the self administered survey).

5.6 Questionnaire design for the quantitative study

A questionnaire is a tool for collecting information to describe, compare or explain knowledge, attitudes, behaviours and/or socio-demographic characteristics of a particular target group (Rojas & Serpa, n.d.). The questionnaire used in this study to collect primary data was designed in accordance with the primary and some of the secondary objectives and research hypotheses proving the overall hypothesis 3 (H3) of this study, as will be discussed in more detail in chapter 6. The researcher compiled the questionnaire in accordance with seven basic principles of questionnaire design and layout (Dillon, Madden & Firtle, 1993:304).

The principles are as follows:

· Principle 1: Be clear and precise.

· Principle 2: Response choices should not overlap and should be exhaustive.

· Principle 3: Use natural and familiar language.

· Principle 4: Do not use words or phrases that show bias.

· Principle 5: Avoid double-barrelled questions.

· Principle 6: State explicit alternatives.

· Principle 7: Questions should meet criteria of validity and reliability.

The empirical data collection was structured and subdivided into two separate questionnaires (see appendices A and B).

· Firstly, a questionnaire was designed to target the CEOs or CIOs to get their views on the strategic part of the system within their organisation.

· Secondly, a questionnaire was designed to target the managers and the end-users in order to get
their views on the management and operational aspects of the system, its usage and awareness.

Table 5.2 contains a summary of the objectives and hypotheses as linked to the questions used in the questionnaire.

Table 5.2: Relationships between questions in self-administered questionnaire survey and the primary and secondary objectives and research hypotheses.

· Questions linked to the primary and secondary objectives

Primary objective	Questions	To determine the extent to which ERP software could facilitate integration in the MNE value chain process.
A9 & A10 B9	Secondary objectives
Questions	Objective 1: To establish the value of ERP system software alignment with the organisation's strategy and to determine the extent to which it can be seen as a strategic IT tool in order to strengthen the organisation's value chain system.
A7, A9 & A10 B9	Objective 3: To obtain the view of SAP software users in MNEs surveyed of strategic management through the value chain approach, ERP awareness and usage, and training.
A6, A7, A8, A9, A10 & A11, B6, B7, B8, B9 & B10

· Questions linked to hypothesis

Hypotheses	Questions	Hypothesis 1: The value chain integration through ERP software helps MNEs to operate efficiently due to the software's multifunctional, integrated and modular characteristics, which meet the general model of an ERP system as a value chain system.
A9 & A10 B9	Hypothesis 2: The use of strategic IT plan theory in ERP evaluation determines MNE competitiveness impact and IS usage.
A10 & B9

5.7 Response rate

The three-page questionnaire was mailed to every participant (CEOs/CIOs, managers and end-users), resulting in a total of 375 questionnaires. Therefore each of the five MNEs received 75 questionnaires. In total only three MNEs returned the completed questionnaires, with a total response of 137 constituted of 3 CIOs, 61 managers and 73 end-users (In some of the statistical analysis in the chapter six, the missing value were found therefore N value vary between 134 and

137). The total response rate therefore was 36.53 % of the sample (137 of the questionnaires returned out of 375 questionnaires issued).

Table 5.3: Response rate to the self-administered survey.

· Sample population

Questionnaire issued								CIO
Managers		End-users					N
%	N	%	N	%	N	%	1
20.0%	30	20.0%	44	20.0%	75	20.0%	1
20.0%	30	20.0%	44	20.0%	75	20.0%	1
20.0%	30	20.0%	44	20.0%	75	20.0%	1
20.0%	30	20.0%	44	20.0%	75	20.0%	1
20.0%	30	20.0%	44	20.0%	75	20.0%	5
100.0%	150	100.0%	220	100.0%	375	100.0%

· Sample response

Questionnaire returned								CIO
Managers		End-users		N		%	N
%	N	%	N	%
33.3% 33.3% 33.3%	24 18 19	39.3% 29.5% 31.1%	29 21 23	39.7% 28.8% 31.5%	54 40 43	39.4% 29.2% 31.4%	3
100.0%	61	100.0%	73	100.0%	137	100.0%

· Response rate

Response rate
CIO	Managers	End-users		Response per respondent group
60%	46.66%	33.18%		Total response rate
0.8%	16.26%	19.46%	36.53%

5.8 Data analysis

The different measuring instruments of mean, mode, standard deviation and so on were used to analyse the data collected (see appendices G and H). Furthermore, statistical analysis through chi-square was conducted to test the relationships between the hypotheses formulated (H0 and H1) in cross-tabulation with the different questions formulated. Fisher's exact test exact significance (one-sided) value was used as the p-value in this study. In cases where the p-value was less than 0.05, the hypothesis was rejected. If the p-value was greater than 0.05, the hypothesis was accepted.

5.9 Conclusion

This chapter dealt with the various methods and techniques used to collect the data in this study. An overview of the self-administered questionnaire as the specific measuring instrument was also provided. Reference was made to the sample size, the population of this study, as well as the detail of the interdependence of the questionnaire and the primary and secondary objectives and hypotheses of this study for the empirical part of the research.

The chapter to follow (chapter 6) will provide the qualitative and quantitative findings of this study relating to the various hypotheses proposed in chapter 1.