CHAPTER research project together. Kerlinger (2003: 38), define a

CHAPTER3RESEARCHMETHODOLOGY3.0 IntroductionThepre-occupation of this study was to make an assessment Enterprise ResourcePlanning Post implementation sustainability factor analysis of NUST from an endusers’ perspective. This chapter will explore methods of data collection to beused in the study. Primary and secondary data collection methods will beadopted for the purpose of the research and will be explained in detail in thischapter. On primary collection of data, a case study was used through the useof self-administered questionnaires and structured interviews for NUSTemployees at the Bulawayo main campus administration offices. Issues discussedin this chapter include the concept of research design, population and samplesto highlight their significance to the study.3.

1 ResearchMethodology Kothari (2005) described research methodologyas the logical approach to answering the research problem. The logical process incorporatesan analysis of stages available to the researcher and the logic behind theirapplication. As postulated in his remarks, Kothari (2005), argued that researchmethodology is a journey of discovery not necessarily based on researchmethods, but an emphasis is made towards the logic behind the methods used incontext of the research study. Henning (2004), described methodology as alogical set of methods that can deliver data and outcomes that highlights theresearch objectives suitable for the problem under study.3.2 Research DesignDr.SueGreener (2008:40) postulates that, “a research design is a grand plan ofapproach to a research topic”. It also provides glue that holds the researchproject together.

Kerlinger (2003: 38), define a research design as a “plan,structure and strategy of investigation conceived so as to obtain answers toresearch questions and to control variance”. The quality of the researchproject is heavily influenced by the research design. Kothari (2005) propoundedthat a research design stipulates what information to be collected, from whichsource and by what producers. For the research study to be consideredoverwhelmingly done and meaningful, the study has to magnify the importantfeatures of the research design. A good design ensures that the informationgathered is necessary to the research problem under study and that it wascompiled by objective and economical procedures. Burns and Groove (2003:195)cited on

zw (27/03/2014:11:15am)states that a research design “is a blue print for conducting a study for withmaximum control over factors that may interfere with the validity of thefindings”.3.2.0 ExploratoryResearch According to Dr.

Sue Greener (2008),exploratory research is designed to allow the researcher to have an insight ofsome issues with the aim of developing suggestive ideas. The exploratoryresearch design is the most basic design meant to provide insight and ideasabout a problem. It explores trends in the industry with the notion of identifyingspecific ideas and insights to generate explanations as to certain behaviour inthe industry. The fundamental concept behind exploratory research is to exploreand interpret relationships between different concepts and build theory inrelation to the topic.

(Kothari, 2005). 3.2.

1 Descriptive ResearchS.KYogesh (2006:104) stipulates that “descriptive research is concerned with thepresent and attempts to determine the status of the phenomenon underinvestigation”.  The objective ofdescriptive research is to provide a description of various phenomena connectedto individuals, situations or events that occur. It is primarily concerned withthe gathering of numeric measurable data (Safaei Farzad, 2006).3.

3 Target PopulationAbdu(2009), states that the term population means the total number or the aggregateconcern which by virtue of the common characteristics that may lead to theobtainance of relevant information. Target population also refers to the samplepopulation on which the researcher is going to draw conclusion from. Thetarget population of this research is therefore the employees of NUST Bursars departmentat the main campus. The population ranges from the head of department, heads ofsections and accounting assistants’. 3.4 SamplingDrSue Greener (2008:51) defines sampling as a practical way of studying peopleand their activities, thoughts, attitudes, abilities, relationship in relationto business.

A sample can also be referred to a relatively small group or apart of the population selected for observation and analysis. According to Yogesh(2006:81) sampling is an indispensable technique of behavioral research; theresearch work cannot be done without use of sampling. Sampling is a necessityin researching because the study of the total population is impracticable.

Asample representing target population of NUST would be undertaken. A sampletherefore refers to the representation of the target population.Advantages of Samplingv  Itmakes data collection process fasterv  Reducesthe cost of performing the researchv  Followup on respondents becomes easier and cheaperv  Itensures homogeneity and improvement of data quality and accuracyv  Dataanalysis becomes easier and cheaperDisadvantages ofSamplingv  Thesample chosen may not reflect the whole population and may left out somecharacteristics which that material effect on the material findingsv  Determinationof the exact sample size might be practically difficultv  Inthe event of non-probability methods of sampling such as convenience and judgmentaltechniques not being used, interviewer bias is likely to be high. 3.

4.0 SamplingTechniquesTheresearcher opted to use a stratified random sampling technique to gather datafrom a range of NUST employees in Bulawayo. Focus was directed towards the useof stratified random sampling because the sample to which conclusion was drawndoes not consist of a homogeneous group.Twentyemployees were selected from NUST as a sample. The selected employees were fromdifferent sections of the Bursars Department of NUST namely, Students accounts,Buying section, Budgetary Control, Salaries and Benefits, Payments section.

 Thesample is represented diagrammatically in table 1 below:Table 3.1 Populationand the sample size Department sections(management  and  accounting staff members) Population Sample Size Students accounts                           5 Purchasing                 10          3 Budgetary Control                  8          2 Payments                 11          5 Salaries and Benefits                  6          5 Total                 49         20  3.5 Types of Data Theresearcher used a multi approach of data usage, thus, both primary andsecondary data were used in the research study.3.5.0 Primary Data Accordingto Kothari (2004:95), primary data refers to data that which is collectedafresh and for the first time and thus happen to be original in character.Primary data is information used in research that has been collected usingquestionnaires, interviews and observation by the researcher. Usuallyprimary data is of paramount importance because it provides information that iscurrent and up to date.

Data is always reliable since it originates directlyfrom the parties involved. Primary data can either be overt or non-overt(D.Cooper and W.Emory:1991).Overt primary data results from asking questions tointerviewees.

Non-overt primary data refers to data derived from the oral orwritten responses, for example data from participant observation. The researcherused both overt and non-overt data to accomplish the objectives of the study.3.5.1 Secondary DataAspostulated by Kothari (2004:95) secondary data refers to information which hasalready been collected by someone else and which already has passed through thestatistical process. Secondary data refers to information that is in existence,available on publications such as books, reports, journals, magazines and workby various scholars. Secondary data for this research was collected fromdatabases, magazines and journals.

Other information was sourced from publisheddissertations from other scholars, internet and textbooks to appraiseliterature arguments and recent developments on ERP post-implementation.Secondary data made a significant contribution in identification of literaturegaps and case study. These sources were used to justify the research problemand reducing the time and cost for gathering primary data.3.6 ResearchInstruments3.6.

0 QuestionnairesAquestionnaire is a list of written questions which can either be open or closedended, used to gather responses on particular issues from respondents.According to Wegner (2005), a questionnaire is in two basic ways; firstly therespondents can be asked to complete the questionnaire with the researcher notpresent (postal questionnaires), which basically means the respondents completethe questionnaire without the aid of the researcher. Secondly respondents couldbe asked to complete the questionnaire with the assistance of the researcher.The researcher used both open and closed questions, open ended questions wereused to increase the respondents’ contribution and closed questions were usedto guide the respondents to remain focused on the problem under study.Advantages ofquestionnairesv  Castillo(2009) postulates that questionnaires are one of the cheapest and most feasibleways of gathering data. A questionnaire is also an easy way to standardizeresponds since each respondent is asked the same questions in the same way.v  Questionnairesenables the respondents to provide as much information as possible without fearof victimization, since names are not a pre-requisite and they can complete theforms probably at home lunch time or anywhere else alone (Naoum 2007).

v  Questionnairesallow interviewees enough time to answer questions honestly and accurately andit also enables the researcher to reach some inaccessible areas moreeasily.   Disadvantages of questionnairesv  Wegner(2005) stipulates that the major drawback questionnaire is that the presentationof the questionnaire makes it cumbersome for the researcher to classify complexmatters and opinions, even if open ended questions are used.v  Thedepth of responses that the respondents provide tends to be more limited thanwith any other method of research. In circumstances where the researcher is notpresent, it is difficult to gain assurance respondent has understood a questionproperly.v  Theresearcher has no control over respondents. There is no assurance whatsoeverthat the questions were answered by the right respondents especially in caseswhere the questionnaires are answered in the absence of the researcher.v  Questionnairesare not flexible; all respondents can answer the same questions in the samemanner.3.

6.1 InterviewsHolborn(2002) postulated that an interview is a meeting between the interviewee andthe interviewer. An interview can also be defined as a face-to-face meeting inwhich an interviewer asks respondents questions designed to draw respondspertinent to the research hypothesis. It is used to collect factual informationas well as opinions. Interviews can be classified into three namely,structured, unstructured and semi-structured. Some studies may require only oneform of interview, while others require both. In this research, the researcheris going to use the structured interview to collect data.

A structured interview was scheduled togather data from Innscor Africa Ltd employees. The researcher used theinterview scheduled plan to gather data from the sample of selected employeeswithin Innscor Africa ltd employees. The interviews provided the researcherenough time to speak to the respondents giving their in depth ideas and viewsconcerning controllable and uncontrollable costs at Innscor Africa Ltd.

Advantages ofInterviewsv  Naoum(2007) postulated that interview gives high response rate, approximately 60%-70% and there is instant feedback.v  Interviewsprovide more accurate answers since the researcher is able to probe for moreaccurate answers.v  Answerscan be explored, finding out ‘why’ the particular answers are given.Disadvantages ofInterviewsv  Interviewsare time consuming; they require careful preparation in advance to arrangeappointments with respondents.

v  Interviewsare expensive to carry as their require travelling to the interviewees, in thisresearch the researcher travelled to Bulawayo.3.7 Pilot StudyApilot, or feasibility study, is a small experiment designed to test logisticsand gather information prior to a larger study, in order to improve thelatter’s quality and efficiency (www. validation of thequestionnaire, the researcher carried out the pilot feasibility study.

Beforedispatching the actual questionnaires a preliminary analysis was conducted. Thepilot study was carried out to establish reliability, clarity and validity ofthe research instruments deployed for the gathering of the final data. Thepilot study exposed misconceptions of the interviewees and the need torestructure some of the questions for relevance and clarity purposes.  Thefeasibility study was carried out with the assistance of six members, two ofthem were professionals with relevant working experience and four of them werefellow accounting students.

Draft questionnaires and interview guide questionswere delivered to these selected members for analysis and review before comingup with the final questionnaire and interview draft guide. Theresults of the pilot testing were enormous; interviewees helped the researcherto rephrase some of the questions both for interviews and questionnaires whichthey think were irrelevant. Above all, the results helped the researcher todetect possible flaws in measurement procedures including time limits andinstructions.Table 3.2 Results ofthe Pilot Study Sample Questionnaire and Interview questions distributed Questions approved Professional 1                         19                      18 Professional 2                         19                      18 Fellow Student                         19                      19 Fellow Student                         19                      19 Fellow Student                         19                      18 Fellow Student                         19                      18  Theresearcher found it necessary to remove all questions unapproved questionswhich were found to ambiguous and irrelevant.3.

8 Validity andReliability of Research Instruments 3.8.0 ValidityValiditycan be defined as the extent to which the instrument measures what it purportsto measure. Validity can be further classified into content validity,criterion- related validity (or predictive validity), face validity, constructvalidity, factorial validity, concurrent validity, concurrent and convergentvalidity. (www.michaeljmillerphd.com05/04/2014:10:46pm).

Validity ensures thatthe sample that has been adopted by the researcher adequately represent thetarget population under study.3.8.1 ReliabilitySaunders(2009:156) postulated that “reliability refers to the extent to which your datacollection techniques or analysis procedures will yield consistent findings”.It can also be defined as to the extent to which a questionnaire, test,observation, interview or any measurement procedure produces the same resultson repeated trials.

(www.michaeljmillerphd.com05/04/2014:10:46pm).Reliability also encompasses minimization of errors in thedata collected. The information gathered by the researcher was consistent to alarger extent hence it was considered reliable.

3.9 Chapter SummaryThischapter explored the research methods used in this research study. It addressedthe aspects of research methodology, research design, data sources, samples andresearch techniques. The researcher described the strengths and weaknesses ofall research instruments and methods used, which include the questionnaires andinterviews. The next chapter will explore the findings that were conducted withNUST bursars’ department employees. Focus will also be directed towardspresentation and analysis of data collected using tables, graphs, descriptionand diagrams.