Writing a Quantitative Research Thesis

1. RESEARCH PROBLEM AND TOPIC

There are basically two types of problems:that whose aim is to increase our knowledge andthat whose aim is to make our life better (Selamat2008). Research topic emanates from a detailedand insightful analysis of a research problem asone of the possible ways through which a solutioncould be contributed to the problem. Suchanalysis might involve experiences gained fromliterature review. Generally, research topic shouldreflect three components: the independentvariable, the dependent variable and the popula-tion under study. The dependent variable is theproblem variable while the independent variableis that whose influence or relationship to theproblem variable is to be established, and whichif effectively manipulated is highly likely to bringabout desirable changes in the problem variable.

For example the problem of deterioratingquality of performance of mathematics by seniorsecondary students in Botswana, was analysed,and several possible contributory factors wereidentified. These included: quality of textbook,teaching methods, teachers’ experience, teachers’motivation, parental involvement, study time, etc.These are called the independent variables, whilethe problem variable is called the dependentvariable. A contribution to the solution of theproblem could be done by manipulating any ofthese independent variables in a direction thatwould enhance desirable changes in the problemvariable - quality of students’ performance inmathematics. But first of all, we must determinethe amount of influence, if any, which such

© Kamla-Raj 2009 Int J Edu Sci, 1(1): 19-32 (2009)

Writing a Quantitative Research Thesis

H. Johnson Nenty

Educational Foundations, P. Bag UB 0702, University of Botswana, Gaborone, BotswanaE-mail: [email protected] or [email protected]

KEYWORDS Research Problem and Topic. Review of Literature. Research Methodology and Design. Data Analysis.Research Findings

ABSTRACT Research is an exciting adventure which if properly carried out adds richly to the student’s experience,to the school academic prestige and to the society through the new knowledge it creates which could be applied insolving related problems and in other services. Young researchers always encounter problems designing and carryingout their first study which usually is their project, thesis or dissertation. Some who are not properly guided orsupervised get frustrated and drop out of their programmes because of these problems. The ideas in this paper whichmetamorphosed over 25 years of teaching and supervising research, represents an attempt to contribute to thesolution of such problems especially for graduate students. It presents elaborately, in very simple language and in fivesections, the practical steps that should guide beginning researchers on how to carry out their study and report it.

manipulation would have on the problem. Hencethe first step in tackling this problem is to carryout a study to determine the amount of influenceany of the independent variables has on theproblem variable, and based on the result of suchstudy, we make recommendations or come outwith a model through which contribution couldbe made towards the solution of the problem.

Since we may not, in one study, handle all thevariables that could possible influence ourproblem variable, we have to choose one orsometimes few of them to study, and the chosenones(s) become(s) the independent variable inour study. Let us say that in our own case, wechoose teacher motivation, hence our topic wouldbe “Teacher motivation and quality ofperformance in mathematics by seniorsecondary students in Botswana” Here we have‘the level of teacher motivation’ as ourindependent variable; ‘quality of performance inmathematics’ as our dependent variable; and‘senior secondary school students in Botswana’as our population under study. Each of these hasto be defined or operationalized during theprocess of the study. Topics should be stated asconcisely as possible. Irrelevant phrases like “AStudy of . . .” “A Comparison Between . . . .”should be avoided. The maximum length, innumber of words, of a typical topic should be 20(Graduate Studies, College of Education,University of Florida [GS/COE-UFL] 2005).

The Problem

The first chapter in a thesis research report is

20 H. JOHNSON NENTY

often labelled as “The Problem” becauseeverything that goes into it has to do with thepresentation, analysis and validation of theresearch problem. Some authors title it as“Introduction” but each of the other chapters hasa section on introduction hence it is sometimesconfusing when it is so titled. If the first chapterof the research report is titled “Introduction”,there will be no need for any section or sub-section to be titled “Introduction”

Generally, the chapter presents, analyses, andvalidates the problem for which a solution issought through the study. It validates the problemthrough presenting and analysing:(1) current observations, experiences, views,

opinions that reflect or bear on the existenceof the problem;

(2) background (historical, cultural, economic,social, etc.) information on the problem or onits development;

(3) rationale or theoretical/conceptual back-ground/foundation information that providesthe knowledge so far available to human onthe area of the problem. This is the ‘majorpremise’ of the study. This provides the basisor a foothold for speculating the possiblesolution to the problem at hand. Solution toproblems emanates from putting what isknown or theoretical knowledge to work. Onewho does not have the theoretical knowledgecannot speculate and follow through possiblesolution to a practical problem. It is what isknown so far in the problem area that providesthe foundation for practice or for an attemptto solve any related problem.

(4) The actual problem to which one intends tocontribute a solution.

(5) The purpose of the study and if possible, itsobjectives based on which researchquestions could be derived.

(6) Research hypotheses which, informed bytheory, speculate possible solution to theproblem.

(7) The significance of the study.In a very elaborate form, the chapter as pre-

sented has the following sections: introduction,background, rationale or theoretical/conceptualbackground, statement of the problem, purposeof the study, research objectives, researchquestions, research hypotheses, significance ofthe study, assumptions, limitations, delimitations(or scope) and definition of terms and sometimesa summary of the chapter.

1.1 Introduction

This section presents and analyses currentobservations, experiences, views, opinions, etc.that reflect or bear on the existence of the problem.It argues with evidence for the existence of theproblem. It serves to present the existence of theproblem to potential readers in such a way as toinvolve them as stakeholders and capture andkeep their attention and whet their appetite toread the work. (For large funded studies, richmaterials for this section can be generated duringa stakeholders’ consultation or need assessmentstudy.) The aim is to create the existence of theproblem to which you are trying to contribute asolution in their minds, their feelings and make itreal to them and thus arouse their interest in theproblem at hand. Try to make it their problem too,let them see the problem with you, and to feelthat the solution to the problem is important tothem. What brought this as a problem to yournotice? Introduction is a ‘public relation’ sectionof the study. The questions you should try toanswer here are:1. What observations, experiences, etc., caught

your attention and raised interest and/orquestions in your mind and hence gave riseto the study?

2. In other words, what did you observe, hear,see, or experience that prompted you to thinkthat this is a problem worth spending yourtime and funds on trying to contribute asolution? or

3 What prompted your curiosity or interestwhich you are trying to satisfy?Present and analyse these in such a way as to

whet other person’s appetite to read your workby raising similar interest or curiosity in theirminds. Your problem should intrigue and motivatenot only you but also your audience. Othersshould find your problem interesting and worthattention (Selamat, 2008). Explain your and otherrelated observations, experiences, views andrelated opinions in such a way as to make theproblem obvious to everybody. Current viewsand opinions such as those expressed innewspapers, magazine on issues related to theexistence of the problem at hand are good at thispoint. Get and present, if possible, data andfigures that highlight the existence of the problem.Avoid theory or review of empirical literature herebut use current information and data thathighlight your problem, curiosity or interest.

21WRITING A QUANTITATIVE RESEARCH THESIS

1.2 Background to the Problem

Sometimes it is necessary to present andanalyse the historical, cultural, social, etc.background to the problem of your study withinthe population under study. Place the problemunder study in a historical, sociological,traditional or cultural etc perspective. Does theproblem have any history, or does it have anycultural or sociological root in the tradition orculture of the population under study? Presentand analyse such historical, cultural orsociological roots here.

1.3 Rationale, Theoretical or ConceptualBackground/Foundation

Research, especially quantitative research isbasically a theory validation procedure hence theresearch process is cyclic in nature. It starts fromtheory and ends in theory. It is a process throughwhich theories are validated, refined orinvalidated. Though this may not always beobvious, especially to beginning researchers,effective solution to any human problem is basedon an application of one existing theory or theother or an application of “what is known so far”in the search of what could be known in theproblem area. Hence, especially with appliedresearch one needs a theory related to the problemto provide a foundation and a sort of guidancefor the search of a solution to the problem. ToHair and his colleagues (2001), theory serves threebasic purposes in a research study. These are:the provision of a conceptual framework for aresearch study; serving as a vehicle forincorporating prior knowledge; and linkingresearch to the larger body of knowledge in thearea of the problem (p. 2).

Any attempt at the solution of any problem isalways founded on what is known so far in thearea of the problem. What is known so far is oftenenveloped in one theory or the other. Ourreasoning toward a solution to the problem startsdeductively from here and ends inductively atadding to the theory or increasing our confidenceor lack of confidence in the theory. It is based onwhat is known that a possible solution to theproblem is speculated. For example, one cannotspeculate a valid solution to a faulty carburettorunless he/she knows the theory of the workingof a carburettor. So the logic is from theoreticalunderstanding to observations underlied by that

understanding; and from a consensus orconvergence or these observations back totheory. Generally therefore, from a body ofacceptable knowledge, a hypothesis is deducedand tested inductively based on the result of anempirical observation of a large enough typicalcases. Tested hypothesis are used inductively tobuild, revise or validate a generalization or theory.Hence at this point the researcher should presentand analyse an existing knowledge or theoriesthat relate to or underlie his/her problem orinterest. What is the theoretical foundation, orwhat do humans know so far, which provides thebasis for speculating possible solution to theproblem at hand, or which provides the basis forany attempt to solve this or related problems?Such theoretical or conceptual foundation shouldprovide an explanation for the relationship amongvariables involved in your problem. What is thetheory (major premise) that provides thefoundation to your study and to which yourfindings will provide a contribution? Hitch yourwagon to theory (Aleman 1999), or fit yourobservation or experience into the framework ofwhat is already known in this area, and from itderive definitions, orientations, questions, anddirections to your thinking; and speculate, basedon relationships among variables as postulatedby the theory, the possible solutions to yourproblem (Kerlinger and Lee 2000; Mouly 1978).Your speculations (hypotheses) must standtheoretical test by being compatible andconsistent with existing knowledge. Theoreticalbackground should precede the researchquestions or hypotheses which are speculatedbased on it.

1.4 Statement of the Problem

Here the researcher presents and analyses theproblem to which he/she is trying to contribute asolution through his/her research. Or exactly whatis the curiosity or the interest which you are tryingto satisfy? According to Selamat (2008), “the heartof any research project is the problem” and“researchers get off a strong start when theybegin with an unmistakably clear statement ofthe problem” (Slide 25). Unlike the public-relationminded ‘Introduction” which presents andanalyses current observations, experiences,views, opinions, etc. that reflect or bear on theexistence of the problem or interest, here astatement of the actual problem or interest is

22 H. JOHNSON NENTY

presented and elaborated on. Ideas from thetheoretical foundation are conscripted and usedto ‘paint the problem red.’ The purpose of thestudy is not its problem, the two are different.Present and analyse, with a high level of clarity,the problem of the study and then specificallyfocus it by narrowing it down to the aspect yourstudy will contribute solution. Relevant questionsto be answered here are:1. What exactly is the problem to which you are

trying to contribute a solution? Present theproblem explicitly and elaborately, clarify anddelineate it. Specify the nature of the problemand, if necessary limit, the scope of theinvestigation. You might delineate and definesub-problems. Present and analyse problemsin education, for example, to which your studyis going to contribute solutions.

2. Specifically, what problem in education, or inany area, or in the society/community, or inyour job, etc, are you intending to, orinterested in, contributing a solution throughyour research? ‘Paint the problem red.’

3. How and why is this a problem? What are thecurrent adverse consequences of the prob-lem? What are the impacts of the problem?What are likely to be the consequences ofthis problem if a solution for it is not sort fornow? In other words, why is it important toaddress the problem?You should not be writing a thesis only for

the sake of meeting graduation requirements butyou should make a significant contribution toknowledge or to the solution of a significantproblem. Avoid trivial problems as such problemsmay not sufficiently “enlarge our wisdom” andmay not reach findings that are useful to a largeaudience (Selamat 2008). You should enlarge ourwisdom in the area of your problem or curiositythrough your study. Information from literatureshould help you delineate, clarify and enrich yourproblem statement; it should rarely be the sourceof your problem since you should aim atcontributing solution to a real life problem.

1.5 Purpose of the study

The statement of the purpose of the study isoften confused with the statement of the problemof the study. While the statement of the problemof the study is a presentation and analysis of theproblem to which a solution is being soughtthrough the study, the purpose of the study is

the intention which the study hopes to realize atthe end of the attempt to contribute a solution tothe problem at hand. Some guiding questions are:1. What are the general goals/purposes or aims

of the study?2. What are the emanating objectives that the

study hopes to achieve?3. Following from these, when the research is

completed, what are the research questionsto which reasonable answers can beexpected?

1.6 Research Questions

Research concerns are operationalizedthrough research questions to which validanswers are sought for during and are expectedby the end of the study. Research questionsshould be clear, concise, and as simple aspossible, focused and empirically answerable.They should not be questions that require a yesor no answer (Selamat 2008). They should beframed to provide the guide for the conduct ofthe study. For descriptive empirical studies theyare answerable through descriptive analysis ofdata, for inferential study, they are answerablethrough testing hypotheses that emanate fromthem. A study can have some research questionsthat are not convertible into hypotheses, and thenboth research questions and hypotheses areaccommodated in such cases. This might call fora mix-method approach involving bothquantitative and qualitative methods. Theresearch question should be framed in such away that it will not provoke “yes” or “no” answer.For example, based on the sample topic selectedearlier, the research question “to what extent doesteachers’ satisfaction influence students’performance in mathematics?” is more appropriatethan “does teachers’ satisfaction influencestudents’ performance in mathematics”

1.7 Statement of the Hypotheses

To Kerlinger (1968) “hypothesis is the mostpowerful tool man has invented to achievedependable knowledge” (p. 27). It is a powerfultool for the advancement of knowledge becauseit serves as the working instrument or thetentacles of theory and a means for seekingsolution to human problems. They are tentative,intelligent guesses posited for the purpose ofdirecting one’s thinking and action towards the


solution of a problem (Selamat 2008). Hypothesesare derivable from theory and they serve asmeans through which theories are validated,revised or invalidated through research (Glassman2007). Once stated, they guide the rest of the studyespecially in terms of the literature to be reviewed,research design and methodology, data analysisprocedure, arrangement of research sections,discussions of research findings, etc. Thequestions here are:1. What are your intelligent or informed (by

analysis of related theory) guesses orspeculations as to the solution to yourproblem?

2. What are the scientific research speculationsthrough which you hope to find solutions toyour problems? Be very concrete and clearand state each hypothesis in terms ofobservable behaviour allowing for empiricaltesting and evaluation of the implications?While research hypotheses are never express-

ed in the question form, research questions areconvertible to research hypotheses. Hypothesisgenerally always indicates a possible influenceon the problem variable by another variable – theindependent variable. Underlying the establish-ment of influence is the determination of therelationship among variables. As a means ofcontributing a solution to the problem, suchinfluence or relationship, if significant, points tothe fact that if the independent variable ismanipulated, there could be desirable changeson the problem variable, leading to a solution ofthe problem under consideration.

A speculation is a scientific hypothesis if andonly if it is testable. That is, it must be amenableto being confirmed or disconfirmed through thecollection and analysing of empirical data. Theconvergence or consensus among severalempirical evidence must be that which determinesthe tenability of the statement. A researchhypothesis statement necessarily involves orimplies the fact that the observed difference orrelationship is over and above that which couldbe accountable for by chance or random error;hence the word ‘significant’ is always used. Adifference or a relationship can be observed, butif such difference or relationship is not aboveand beyond that which random error could bringabout then it is not significant. If the observeddifference or relationship is not bigger than thatwhich random error could account for, then thenull hypothesis of no significant difference or

relationship is retained, but if it bigger, then thenull hypothesis is rejected and the researchhypothesis retained. Retaining the null hypothesisis like acquitting a defendant based on evidencebefore the judge. There may be other evidencenot before the judge which if available would alterthe decision of the judge, hence an appeal isalways allowed (Churchill and Iacobucci 2004).

Sometimes the level to which one is willing toreject a true null hypothesis, that is, significanceor alpha level is specified generally for allhypotheses or indicated in each statement of thenull hypothesis. For example for our study onteachers’ motivation and quality of performancein mathematics by senior secondary students inBotswana, one of the statistical null hypothesiswith the alpha level indicated in the statement ofhypothesis might be: “there is no significantrelationship, at the .05 alpha level, betweenteachers’ level of intrinsic motivation to teach andthe performance by his/her students” (GS/COE-UFL 2005).

The logic of quantitative research does notencourage the idea of accepting, but of retainingthe null hypothesis, because failure to reject thenull hypothesis still has some probability that itmight be incorrect. So instead of being accepted,such a speculation is retained because furtherevidence might come to show it false and hencerejected. This is especially so as we did not studythe entire population but studied a sample fromit. It is therefore not logical to accept that forwhich there is some likelihood that it might beincorrect as the truth but to retain it as the truthso far pending further evidence.

Descriptive studies make do with researchquestions only, but inferential studies mightinvolve both research questions and hypotheses.The problem variable, or some aspect of it, whichis the dependent variable for the study is oftendirectly or indirectly implied in each researchquestion/hypothesis.

1.8 Significance of the Study

What contributions do you think your studyintend to make to the advancement of knowledgeand/or to the solution of some practical ortheoretical problem(s)? What are the intrinsicand extrinsic benefits of the study? What is thepotential usefulness of the findings of the studyto all stakeholders? How will the findings assistor be of benefit to each stakeholder? How would

24 H. JOHNSON NENTY

the solution to the problem influence educationaltheory and practice? What are the educational,social, theoretical, cultural, etc. significance ofthe study? Convince the stakeholders of yourstudy of the potential impact of the study forthem. Formulate and present the anticipatedimplications or possible applications of the resultsof your study. Indicate significance to all relatedstakeholders of the study including: (1) theory oradvancing accumulated knowledge; (2) currentliterature; (3) policy development and imple-mentation; (4) practices in the field; (5) the study’ssubjects or participants; (6) teachers and otherpractitioners; (7) researchers; (8) parents,government and other stakeholders, etc.(GS/COE-UFL 2005)

1.9 Assumptions

What assumptions have you made, or whathave you taken for granted about the nature ofthe behaviour you are investigating, about theconditions under which the behaviour occurs andabout your methods and measurements (Isaacand Michael 1990). Unlike hypotheses,assumptions are not testable but are statementsabout observations and experiences related tothe study that are taken for granted or areassumed to be true. They are statements that helpto remove/reduce doubts on the validity of thestudy and are accepted in faith, or taken to betrue without proof or verification (Assumption2007; GS/COE-UFL 2005; Hungler and Polit 1999).For example, if your study involves scores fromor subjects’ performance in public examination,then your study would assume that theexamination was reliable and valid. The questionis what are you taking for granted in the conductof your study?

1.10 Limitations

What are the unavoidable limitationsconditions (not within your control) surroundingyour study and within which conclusions to thestudy must be confined? What limitations exist inyour method, design or approach, samplingrestrictions, uncontrollable variables, faultymeasurements, and other compromises to internalvalidity? (Isaac and Michael 1990). For exampleconstraining characteristics of methodology/design that tend to limit the interpretation andapplication of the research findings. For example

the constraint in, or lack of simple random samplingplaces limitation on generalizability of the findings.What are those constraints that would be facedby anybody who tries to conduct or replicate thestudy? To what extent is the study limited andwhy? Limitations become clearer or more obviousat the conclusion of the study hence sometimestheses could not be fully specified until after thestudy, and could be placed elsewhere possibly atthe interpretation or discussions stages of thereport (GS/COE-UFL 2005). They serve as warningor precaution to the readers or consumers of theresearch findings on the extent to which they canapply or generalize the findings.

1.11 Delimitations

What are your self-imposed limitations in thestudy? How have you narrowed the scope of thestudy? Such scope includes geographical andconceptual scope (often presented underdefinition of terms). State precisely what you did/what you did not do, or what you intend to do /what you do not intend to do, or define theboundary for your study with regard to the scopeof the study. Have you decided to focus only onselected aspects of the problem, certain areas ofinterest, a limited range of the subjects, etc.?(Isaac and Michael 1990). You might, throughsampling, narrowed the population as in the topic.This will jeopardize external validity of the study.What are those constraints imposed by, and maybe unique to, you in carrying out this study?

1.12 Definition of Terms

List and define conceptually unusual termsor acronyms to avoid misinterpretationsparticularly where they have different meaningsto different people. (If such definitions are basedon new instruments developed for the study, theoperational definitions should come after thedevelopment of such instrument in section/chapter three). Make sure you do not repeatdefinitions implied in your theoretical foundationshere.

2. REVIEW OF LITERATURE

2.1 Introduction

What is the content of the chapter on reviewof literature, and how are you going to arrange


the materials in it? Unlike the theoretical materialspresented under the section for theoretical/conceptual background/foundation in the firstchapter, here the researcher should present andcritically analyse the empirical experiences of otherresearchers like him/her who had tried tocontribute solution to problems similar or relatedto his/her problem. In quantitative research,literature reviewed under this chapter is seen asimportant data or input into the research processas it provides the researcher a source of materialswith which to revise one’s problem, topic, andhypotheses and an informed source for writingone’s methodology. It also provides importantinputs into the discussion of the researchfindings.

2.2 Literature Review

Review other researchers’ empirical experience(theoretical review belongs to the section ontheoretical background) with your, related orsimilar problems? Critically and analyticallyreview the experiences, findings, and the design,method, analysis used and such results. Whatare others’ views based on the results/findingsof these relevant studies? Organize and presentthese reviews in a systematic way in relatedsections, in terms of relationships betweenvariables as implied by your hypotheses. Thisshould not be merely descriptive or narrative, butcritical and analytic. When you review others’studies, pinpoint or indicate flaws, and othertechnicalities that may seriously distract fromtheir validity (Mouly 1978, p. 92). The guidingquestions here are: When other researchers likeyou tried to contribute a solution to this, relatedor similar problems:

What did they do?How did they do it?What population, strategy, method, and

measures did they use?What did they do wrong?What did they find out?What have they or others said about what

they did and what they found out?What are the inherent weaknesses/strengths

of the strategies, methods? measures?How (if at all) did the researchers mitigate/

address the weaknesses? (Did they acknowledgethe weaknesses?). Think of these questions whenyou are planning your literature review.

To maintain the systematic nature of research,

present your reviewed materials in sectionsaccording to your research questions or hypo-theses. For example, if your first research questionwas: “To what extent does level of teachers’extrinsic motivation influence students’ perfor-mance in mathematics?” The first hypothesesmight be “level of teachers’ extrinsic motivationhas a significant influence on students’mathematics performance.” Then the first sectionin the literature review following the introductionshould be headed: “level of teachers’ extrinsicmotivation and Mathematics Performance.”Under this heading, relevant and current empiricalexperiences of other researchers like you on thisarea should be critically reviewed, analysed andarranged systematically to show a trend in thedevelopment of empirical knowledge in the area.Have a specific reason for reviewing a particularresearch study. This might be its importance givenyour problem; its unique methodology, dataanalysis procedure and findings (GS/COE-UFL2005). While empirical literature from studiesundertaken within the last 10 years are generallyaccepted as current, literature from classic/important studies in the area should also bereviewed

2.3 Reviewing Literature

The most important skill for efficient reviewof literature is the ability to read fast, comprehendand summarize material quickly and put it in aform that can easily be retrieved and used in thefuture. During a review, a researcher’s activitiesare quoting, paraphrasing, summarizing andevaluating. Each of these should be done in amanner as to protect the right of the author andavoid plagiarism.

2.4 Summary of Literature Reviewed andJustification (Rationale) for the Study

Provide a comprehensive summary of thereviewed materials. How are you going to usethem in your study? In other words, what is theplace of the reviewed materials in your study?Present a brief appraisal of the need or justificationfor your study in the light of the reviewedmaterials. Generally, what is known so far in thearea of your problem, what still remains to bedone? (Selamat 2008). These should be taken intoconsideration at the end of the report whenwriting your recommendations. What in the

26 H. JOHNSON NENTY

materials you have reviewed in terms of coverageby the studies reviewed, methodology used,types of analysis done, findings, etc. justifies theneed for your research study? Given other relatedstudies reviewed, why is this study necessary?What are the gaps, inconsistencies, inadequacies,or deficiencies in existing literature which youare trying to fill or correct through your study?

3. RESEARCH DESIGN ANDMETHODOLOGY

3.1 Introduction

Introduce the chapter, and define your prob-lem operationally, and on the basis of these,establish an appropriate and suitable design forthe study. You may restate your hypotheses inthe statistical form and for each or for all indicatethe level of significance at which you want totest the hypotheses.

Note that problem for which we are seekingsolution through research exists or occursbecause the relationship between or amonginteracting variables or phenomena involved in asituation, an experience, an observation or eventresults in an undesirable or intriguing conse-quences. Seeking any solution to such problemcalls technically for disentangling such relation-ship or identifying the variables or phenomenainvolved and analyzing the relationship devoidof extraneous influences. The more we can dothis devoid of extraneous influences, the morevalid will be our findings (Best and Kahn 1996).So research design involves procedures throughwhich we can explore and analyze the relationshipamong the variables involved in our problem withminimum error while controlling for sources ofextraneous variability. In all aspect of the researchdesign, we want to know specifically the reasonwhy the particular procedure used was preferredover the others. So all choices or preference ofany particular procedure or method over othersmust be justified preferably by referring to otherpublished studies which had used orrecommended the use of the method or procedure.Each decision about any aspect of the designand methodology procedures used must bedefended. Is the preferred procedure consistentwith the acceptable practice in the problem areaor field? Each procedure under the researchdesign and methodology section/chapter mustbe valid or appropriate given the research

problem, research question and/or hypotheses.Anticipated method- or design-related problemsshould be highlighted and steps taken to preventthem from occurring should be indicated andexplained. Procedures should be vividly explainedto allow for valid replication. Fundamentally,quantitative research is an inferential endeavourwhich seeks to uncover universal truths andprinciples in the form of relationship amongvariables or phenomena. Philosophically andmethodologically it is designed to enableinference from a sample to a larger population(The Validity and Reliability of QualitativeResearch, n.d.).

3.2 Research Design

Besides indicating and justifying the type ofresearch – qualitative, quantitative or mixed model– and the underlying philosophy or paradigm,clearly describe the design of the study and defendyour choice of type and design for your study.Accepting research as a search for the truth, howwill you ensure that your findings reflect the truthas much as possible? What precautions will yoube taking under every aspect of the study to inhibiterror from carrying the findings of the study away?How are you going to minimize error and managethe variation in the variables under the study inorder to avoid the results being significantlyinfluenced by extraneous sources? How do youcontrol for probable sources of error? Researchdesign is fundamentally a variance managementprocedure. What is, in an analytic form, thefundamental objective of the study? Is it an attemptto reconstruct and describe the past objectivelyand accurately, that is to describe what was? Inthat case it is historical, ethnographical, etc. Is itan attempt to discover relationships that existbetween non-manipulated variable that is todescribe, explain or infer some aspects of thepresent - what is? In that case it is either a survey-descriptive, or a survey-inferential, evaluationstudy, etc. Is it an attempt to find out what will bewhen variables of interest are carefully manipulatedunder controlled condition, that is, to describe -what could be? In that case it is quasi- or pure-experimental study (Van Dalen 1979)

3.3 Population of the Study

Describe your population explicitly in termsof geographical limits, demographic and


sociological descriptors, time period covered, etc.,as dictated by the problem. Have you delimitedthe population as specified in the topic, if yes,how and why? Describe the population so wellthat you leave no doubt in the readers’ mind as towho is qualified or not qualified to be members ofthe population of the study (Isaac and Michael1990). If possible, what is the number of membersin the population? Note that there is the need tofirst define or describe vividly the researchpopulation in your study; for it is not in all casesthat the research population will be human beings.Research population could be certain productsin the market or certain textbooks in the schoolsystem, etc.

3.4 Sampling Procedure

We can only get from the population to thesample through sampling; therefore describe thesampling procedure that you used to select thesample from the population described above. Beexplicit to allow for replicability. Do not just say“simple random sampling was done.’ Whatexactly did you do? What are the possibleconstraints that may be encountered in anattempt to carry out a scientific selection, andhow did you overcome these? The main concernhere is, given the sampling method you are using,how well does the sample that results from itrepresent the population? This is one of the twosources of random error which research designmust minimize. It is the cornerstone for ensuringgeneralizability or external validity of the study.

3.5 The Sample

Describe the resulting sample in details interms of each level of the independent variablesimplied in the study. Present this in a tabularform if possible. Describe, if possible, the sample-to-population ratio? Evaluate this ratio in termsof the anticipated or desirable precision of yourstudy.

3.6 Instrumentation

What instrument(s) are you going to use tocollect data for the study? If the instrument to beused is constructed by you, describe clearly, withaccurate details, how the set of items designed toillicit responses for each variable wereconstructed, or how you came by the set of items

which you are using as an operational definitionof each of the variables. How was the instrumentvalidated? Describe the validation process, andindicate and evaluate the results. If the instrumentwas adopted, why it had to be adopted? If theitems were adapted from existing instrumentsummarize the procedure that was used in itsconstruction and its measurement properties.What method did you use in selecting or on whatbasis did you select the items? How well is thevalidation population of the existing instrumentwhich you are using comparable with thepopulation of your study? What changes oradaptations did you make to the originalinstrument? If the validation population issignificantly different from your population forthe study, or if the adaptations or changes madeon the original instrument are significant, describeyour revalidation process, indicate and evaluatethe results. Validity is at the heart of bothmeasurement and research.

Here again, we are talking about represen-tativeness that sustains generalization. Thequestion here is to what extent will your instru-mentation procedure or methods producemeasures or scores that validly represent thebehaviour or variable under measurement for everysubject or participant? To enhance external validityavoid under-representation of the subjects orparticipants, of the research setting, of thevariables or measures. This is the deficiency whereyour instrument fails to include all items that arerepresentative of and from all aspects of theconstruct or variable under measurement. On theother hand avoid measure contamination in whichitems that are irrelevant to the construct or variableare included in the instrument. Items mustsuccinctly represent the behaviour undermeasurement. Here the question is, how well dothe set of items you have designed, selected orused to measure your research constructs orvariables represent all possible indicants oruniverse of all indicators of the behaviourrepresented by your research variable or construct?

In all cases, the reliability of the instrumentdesigned to measure each of the variables shouldbe presented at the proposal stage. Besidessampling, instrumentation and data collectionprocedures are the other sources of random errorwhich research must be designed to minimize. Wecan limit this by using instruments that have highreliability. While validity represents the accuracyof your measure, reliability indicates the precision

28 H. JOHNSON NENTY

of your measure. Here we are talking about howwell the items of the instrument combine tomeasure precisely whatever it is measuring. It canbe measuring precisely whatever it is measuringbut not accurate at measuring what it wassupposed to measure. Reliability has to do withstability, repeatability and consistency of theresults of our measurement or observation.

3.7 Data Collection

Under what condition(s) were the datacollected? Be explicit to allow for replication.What did you do to limit the error that might havebeen as a result of data gathering procedure?What special precautions, instructions, etc., didyou take or give during the administration of theinstrument or interview or during the observationto minimize the data-collector effect on the resultof the data collection procedure? This is anothersource of measurement error which if systematiccomes in to adulterate the results of our researchstudy and reduce the internal validity of suchfindings. Hence we want to know specifically howthe research data were collected.

3.8 Ethical Considerations

What type of and how is each ethical issuelikely to manifest during the study? What is likelyto be consequences of each of these to thesubject? What procedures have you put in placeto address each of these issues and their probableconsequences? For example, what precautionsand safeguards have you incorporated in thestudy design to protect the rights of humansubjects? (see American Educational ResearchAssociation [AERA] 2005)

3.9 Data Preparation

Describe how you will score, in respect ofeach variable, the responses by the subjects toeach item of your data collection instrument. Notethat in your instrument some of the items may benegatively worded while some are positivelyworded and to add these up you need to transformall of them to the same polarity. How did youcode, roster, or transform the emanating data inpreparation for analysis? There may be need foryou to transform the data into a numerical formmore suitable for appropriate statistical analysis.

3.10 Operational Definition of ResearchVariables

How did you measure or manipulate eachof the variables or constructs in your study? Howdid you get, for each subject, the data thatreplaced or took the place of each variable forevery subject or participant in your study? Howdid you turn the participants’ behaviour into ascore that validly represents the behaviour? Beexplicit to allow for replicability. Give meaning toyour variables by spelling out what you did (andhow you did it?) to measure or manipulate it. Thiscan be through experimentation or throughmeasurement.

Quantitative research is conceptually madeup of three sections: (1) a section based comple-tely on verbal discourse; followed by (2) a sectionbased on numerical or quantitative discourse; andfinally concludes with another (3) section basedagain on verbal discourse. In the first section theresearch problem is presented, analysed andvalidated; literature is reviewed and reported; andthe procedure for carrying out the study ispresented and justified, all verbally. Towards theend of this section, research variables, conceptsand constructs are replaced with numbers. It isthrough the operationalization process that this isdone in such a way that the resulting quantificationwith a score or normative qualification representsthe variables, concepts or construct validly. Aconstruct is an imaginative or abstract entity,purposely created to provide a guide, a foothold,for further study and understanding of the conceptit represents (Kerlinger and Lee 2000).

During the next quarter of the process thediscourse is purely numerical and technical logic,involving hypothesis testing and the attainmentof results. Owing to lack of understanding ofnumerical logic by many readers, the report oftenloses some audience within this section. It isthrough the interpretation process that this logicis again converted from numerical to verbal. It isat this point that results, which are expressed intechnical terms, are translated into findings. Thelast section which includes summary of the study,discussion of the results, presentation of theimplications of the findings, conclusion andpresentation of the recommendations is purelyverbal and non-technical (Nenty 1999b) and youshould avoid bothering the readers with technicaljargons and numerical expressions in thesesections.


3.11 Procedure for Testing Hypotheses: orAnswering Research Questions

Restate each of the hypotheses in the nullform and for each identify the variables(independent and dependent variables) involved.And given the nature and type of variablesinvolved, indicate and defend your choice of thetype of statistical analysis you will be doing totest the hypothesis. This may be presented in atabular form.

4. ANALYSIS OF DATA ANDINTERPRETATION OF RESULTS

4.1 Introduction

What intends to be the focus of this chapter?How will you arrange the contents of the chapter?Data analysis is a means of determining andtesting for the extent of convergence, commona-lity or divergence among data collected duringthe study, and hence the relationship amongvariables which these data represent. The ques-tion here is, for example, to what extent do thedata collected from several subjects and throughdifferent sources converge to support the existentof a significant influence or relationship betweenor among the research variables? Present andinterpret the results without discussion orcomments here (Morley 2006).

4.2 General Description of Data

Give a general descriptive analysis includingthe number of cases, mean and standard deviationfor each dependent variable given each level ofeach of the independent variables in your study.This is a descriptive summary of the data andshould be presented in tabular and figural forms.

4.3 Answering Research Questions

For those research questions that were notconvertible to hypothesis answers to them mustbe sought through descriptive analyses ofrelevant data. These might involve frequenciesand/or percentages; graphs; and other descriptiveanalyses including determining the relationshipamong relevant variables. The results of descrip-tive data analysis should be presented in appro-priate tables and figures. Tables must be numberedand given titles (Morley 2006)

4.4 Hypothesis-by-Hypothesis Presentation andAnalysis of Data, and the Interpretation ofResults

Taking each hypothesis in turn, state in thenull form and indicate, and if necessary, explain,the variables involved. Analyse the data usingappropriate quantitative (statistical) techniques,and indicate and present the tables and figures ifnecessary. Prepare and present the tables andfigures neatly in such a way that they can standalone in describing the outcomes of the study.Tables and figures must always be referred to inthe text at least once before they are inserted.Interpret the results, that is, read appropriatemeanings into the results and present this in non-technical language. Interpretation transforms theresults of data analysis into the findings of thestudy, in other words, interpreted data analysisresults become the findings of the research study.Selamat (2008) calls interpretation an “analyticthinking that squeezes meaning out of the mereaccumulation of facts”. The question here is“What do the results of the data analysis mean ina ‘lay person’s’ language? (Slide 9) Specificallythe reporting steps in this section should include:(a) State the statistical version of the hypothesis

in the null form;(b) Present and describe exactly what you did, in

terms of data analysis procedures, to test thehypothesis. How did you do these? Whatwas the result? End this up by referring tothe table on which you present the results.

(c) present the table that has been referred to in(b) above. Avoid using words like above” or

“below” to refer to the table or to a figure orany diagram in your paper

(d) what does what you found out imply for theresearch question or hypothesis under test?For each research question or hypothesis,

what does the finding mean, in a layman or non-technical language? (Nenty 1999b).

To ensure validity in the interpretations ofthe results of data analysis, it is pertinent to askyourself “to what extent ‘does your data say whatyou say it is saying?’” (MacLean and Mohr 1999,p.117).

4.4 Summary of Findings

Summarize all the findings in a paragraph ortwo, and with it, prepare the reader’s mind for theensuring discussion of the findings that follows.

30 H. JOHNSON NENTY

5. DISCUSSIONS, SUMMARY,CONCLUSIONS AND RECOMMENDATIONS

5.1 Introduction

What is going to be the content of thischapter, and what is its thrust? How are yougoing to present or arrange materials in thechapter?

5.2 Discussions

Discuss each of the findings exhaustively perresearch question or hypothesis. All discussionsshould be done in the light of the limitations tothe study. Discussion is a means of presentingand analysing the convergence or divergence ofideas from the findings of your study, the findingsfrom the literature reviewed, and from the theoriesthat underlie the problem under consideration. Itis an integrative analysis of information from thesesources. It is mainly an attempt to answer thequestion “why?” and “how come?” Why doboys do significantly better than girls in mathe-matics in the population under study, for example?Advance reasons to support and/or explain yourfindings. Discussion is never within one personbut between or among persons, ideas, etc. Inresearch, especially in quantitative research,discussion is among four ‘persons,’ these are:the findings from your study; theories related tothe problem at hand (these were reviewed andanalysed in the first chapter); empiricalexperiences and findings of other researcherswhen they were trying to contribute solution tothe problem at hand or to related or similarproblems (these were reviewed and analysed inthe second chapter), and the researcher with allhis/her expertise and experiences as an impartialand insightful moderator and recorder of theresults of the discussions (Nenty 1999a, 2001).

Discussions should be analytic, logical andcomprehensive and should bring together yourprofessional knowledge, the findings of yourstudy, the findings of related studies reviewedearlier and you should argue in the light of relatedtheories presented in the first chapter (SeeKerlinger 1968, pp. 135-136). To what extent doyour findings support the findings of most relatedstudies? How? If not, what are the possiblereasons for the discrepancy in the findings? Doesthe discrepancy or similarity result fromdifferences or similarities in design, methodology,

data analysis methods, etc.? Is it as a result of aflaw in yours or other’s design, methodology, dataanalysis methods, etc? Do your results tend tosupport others in arguing for the related theory?If so, how? If not, why? Argue towards the relatedtheory presented in chapter one. Discussion tieseverything up? Do not re-report, the findings butdiscuss based on them. Answers to the followingconcerns should guide and be integrated into thediscussions of the research findings:(1) What are the findings of other empirical

studies that support or fail to support yourfindings in arguing for the theories on whichyour study is based and what are the probablereasons for this?

(2) How has your findings altered, strengthenedor weaken the theory on which your study isbased?

(3) What theoretical assertions has your findingsbeen able to account for, or what theoreticalclaims has your findings reinforced? (Aleman1999)Explain unanticipated and statistical insigni-

ficant findings. Bring out and discuss in the lightof the shortcomings, limitations, delimitations andassumptions of this study. Integrate and theorizeby trying to make your findings part of acomprehensive body of the knowledge/theoryeither by working within an existing theory or bygenerating pointers to an original theory(Tuckman 1978). Discuss in the light of thelimitations of your research methodology andhighlight the possibility of generalizing yourfindings.

Discussion is the most complex (Morley 2006)and important section of a research report as it isintegrative, synthesizing the theory, and allfindings reviewed and reached given severalmethodological procedures, etc. It is the mostintellectually demanding section, and the onlyone that calls on the researcher’s creativity,synthesis and evaluation skills. Inputs to all othersections have been borrowed from literature andfrom available physical resources and facilities,here the thinking is original and contribution toknowledge, if any, is made at this point.

5.3 Implications

It is important to remember that the study setout or was designed to contribute a solution to agiven specific problem, or to satisfy a specificinterest or curiosity, so, to complete the logical


loop of the research process, such problemshould be restated here, and the implications ofthe findings of the study to the solution of thisproblem highlighted.

What is the implication of the findings of thestudy to the various stakeholders? What do thefindings of the study suggest or what doevidences from the study suggest to eachstakeholder? These include theory, policy andpeople involved in education, for the classroom,administration, other researchers in the area, etc?What implications have these results or findingsto existing related theories? What are theoretical,practical and policy implications of the findingsof your study? Given now the findings of thestudy how well is each of the significancerealized? What and how have the findings of thestudy assisted each stakeholder to given theproblem of the study? What are possible practicalfallouts from the study? What possibleintervention towards solving the researchproblem do the findings of your study imply?

5.4 Conclusion

Present, in a comprehensive form the conclu-sion based on the findings of the study. Summarizeyour thought and present your last words on theproblem and the significance of the findings.According to Morley (2006), conclusions servetwo purposes: to summarize and bring togetherthe main area covered in the report, and thesecond is to give a final comment or judgment ofthe report. Just like the introduction in the firstchapter which in trying to make every reader astakeholder in the study gave them the ‘breathein’ type of feeling and whetted their appetite toread the work, conclusions should give thereaders the ‘breathe out’ feeling of having learnedsomething new and worthwhile and giving themthe closure thus making them feel it worth whileto have read the work.

5.5 Recommendations

Recommendations must be such that wouldfacilitate the solution to the problem for which thestudy was out to contribute a solution. Presentappropriate recommendations that follow directlyfrom the findings of the study. How can the resultsof your study be used? Having found acontribution to the solution of the problem at handwhat role can each stakeholder play in implementing

this solution? Be very clear, specific and practicalin your recommendations. Specifically, to whomare you directing each recommendation? Hereagain the recommendation should target each ofthe stakeholders of the study.

5.6 Suggestions for Further Studies

What other related studies should be carriedout in order to add to the contribution to thesolution to this problem?

Writing an Abstract or Summary of the Work

An abstract, a short, concise and compre-hensive summary of ones work is that whichrepresents ones study in many on-line andabstract publications. Package your entire studyin an envelop that can carry only at most 250words. It should be written in such a way as toprovide a motivation and whet potential readers’appetite to want to read the entire work. It normallyconsists of four parts: (1) the motivation partwhich should reflect briefly the problem of thestudy, the purpose and possibly the theoretical/empirical gap the study is addressing. (2) Thedesign, method and procedures part shouldindicate briefly and precisely what was actuallydone to arrive at the findings of the study. (3)The result and findings section should brieflypresent a summary of the findings arrived atthrough the implementation of the proceduresindicated. (4) The last part should providesummarizing materials on the discussion,conclusion, implications, and recommendations.

REFERENCES

Aleman AM 1999 Literature Review. Boston CollegeSchool of Education. Retrieved April 12, 2006 fromhttp://www2.bc.edu/~alemanan/text.html

American Educational Research Association [AERA]2005. Ethical Standards. Retrieved April 23, 2009from http://www.aera.net/aboutaera/?id=222

Assumptions (2007). Assumptions. Retrieved April 20,from http://web.utk.edu/~wrobinso/540_lec_assume.html

Best JW, Khan V 1996. Research in Education. 7th

Edition, New Delhi: Prentice Hall of India.Churchill GA, Iacobucci D (2004). Marketing Research

with Infortrac: Methodological Foundations 9th

Edition. N.Y.: South-Western Pub.Glassman J 2007. Conjecture, Hypothesis, Theory, Law.

Crossfit Journal, 6. Retrieved April 16, 2009 fromhttp://library.crossfit.com/free/pdf/64_07_Conjecture_to_Law.pdf

32 H. JOHNSON NENTY

Graduate Studies, College of Education, University ofFlorida [GS/COE-UFL] (2005). Suggestions forDissertation/Thesis Proposal. Retrieved April 20,2009 from http://www.coe.ufl.edu/web/?pid=86

Hair J, Money A, Samouel P, Kalafatis S, Katsikeas C,Pitt L, Jorgensen JE 2001. The role of theory indoctoral research in business: Global perspectives.Doctoral Consortium Abstract. Retrieved February5, 2009 from http://www.sbaer.uca.edu/research/sma/2001/106.pdf

Hungler BP, Polit DF. 1999. Nursing Research Principlesand Method. 6th Edition. New York: J. B. LippincottCompany.

Isaac S, Michael WB 1990. Handbook in Research andEvaluation. San Diego, California: Edits Publishers.

Kerlinger FN 1968. Foundations of BehavioralResearch. 1st Edition, New York: Holt, Rinehartand Winston.

Kerlinger FN, Lee HB 2000. Foundations of BehavioralResearch. 4th Edition, New York: Holt, Rinehart andWinston.

MacLean MS, Mohr MM 1999 Teacher-researcher atWork. Berkeley, CA: National Writing Project.

Morley J 2006. The academic phrasebank. TheUniversity of Manchester. Retrieved April22, 2009from http://phrasebank.manchester.ac.uk/index.htm

Mouly GJ 1978. Educational Research: The Arts andScience of Investigation. Boston: Allyn and Bacon,Inc.

Nenty HJ 2001. Methodologies and procedures for

reviewing education sector studies. South-SouthJournal of Culture and Development, 3(1): 1-31

Nenty HJ. 1999a. Discussions of research findings forreview of education sector analysis. Paper presentedat a WGESA, UNESCO sponsored capacity-buildingWorkshop on the Review of Education Sector Studiesin Lesotho in National University of Lesotho, Roma,Lesotho, October 11, 1999.

Nenty HJ 1999b. Interpretation of the results of dataanalysis for review of education sector analysis.Paper presented at a WGESA, UNESCO sponsoredcapacity-building Workshop on the Review ofEducation Sector Studies in Lesotho at NationalUniversity of Lesotho, Roma, Lesotho, October 11,1999.

Sax G 1979. Foundations of Educational Research.Englewood Cliffs, New Jersey: Prentice-Hall, Inc.

Selamat MH (2008). SAK 5090. Retrieved April 18, 2009from http://www.csnotes.upm.edu.my/kelasmaya/web.nsf/de8cc2082fc4d31b4825730e002bd111/.../$FILE

The Validity and Reliability of Qualitative Research(VRQR), n.d. Ppt. Slide 4 of 40. Retrieved February26, 2009 from http://www.omni.org/docs/validity_and_reliability.ppt

Tuckman BW. 1978. Conducting Educational Research.New York: Harcourt Brace Jovanovick, Inc.

Van Dalen DB. 1979. Understanding EducationalResearch. 4th Edition, New York: McGraw Hill, Inc.

Documents

Writing a Quantitative Research Thesis