Research Methods f5j42

Research Design and Methodology

Moses Mutua

Introduction to research What is research?  What types of research exist  What are the current debates in research? 

Definitions of Research The process of arriving at dependable solutions to business problems through planned, systematic collection, analysis and interpretation of data (Orodho and Kombo, 2002).  Kerlinger (1973) defines research as the systematic, controlled and critical investigation of al hypothetical propositions about the presumed relations among natural phenomenon. 

Definitions of research Tuchman (1978) : systematic attempt to provide answers to questions  Mouly (1978)defines it as a process of arriving at effective solutions to problems through systematic collection, analysis and interpretation of data.  To research is to carry out a diligent inquiry or critical emanation of a given phenomenon. It implies exhaustive study, investigation or experimentation following some logical sequence (Mugenda & Mugenda, 1999) 

Four components in definitions 

1. It is systematic: Research solves problems in all areas in a systematic way. It involves identification of a problem, review of related literature and data collection. Data collection requires proper organization and control in order to be able to make valid decisions about the problem. Data analysis, conclusion and recommendations follow.

Key words continued 2. It is objective: Research attempts to find objective unbiased solution to problem. It involves use of primary (first hand) and secondary data (using existing data). It is aimed at finding objective unbiased solution to the problem.  3. It is based on observable experience or empirical evidence: It demands accurate observation and description.  4. It employs carefully designed procedures and rigorous analysis. 

Types of research 

The centrality of the purpose of research in making methods decisions becomes evident in examining the types of research. Patton (2002) says that there are four types of research  1.

Basic research: this is based on fundamental knowledge and theory. It is applied in areas such as physics, chemistry. Research is meant to contribute to theory and knowledge.  2.Applied research: it is aimed at illuminating a society‟s concern. The purpose is to contribute knowledge that will help people understand the nature of a problem.

Research types 3. Action research: It is research meant to solver a particular problem in society. it aims at solving a problem. It is research carried out to facilitate change process.  4. Evaluation research: This is broken down in to two. 

 i)

Formative evaluation: aimed at improving an on going project implementation  Ii) Summative Evaluation: carried out at the end of a project to determine how effectively the stated goals were achieved.

Types of Research Kothari (2003) defines research in of the continuums  Descriptive vs. analytical: Descriptive includes surveys and fact finding enquiry of different kinds. Major purpose is to describe the state of affairs as they exist. One of the common designs is the ex post facto design „After the fact‟.  This requires no control and manipulation of the variables. E.g. frequency of shopping, fashion preferences. 

TYPES OF RESEARCH (KOTHARI) 

There are also  causal

comparative research design of descriptive

type.  Correlation research

ANALYTICAL  The researcher uses the facts or information already available and analyses to make conclusions.

2. Applied Vs Fundamental (basic) Applied is aimed at finding solutions to a problem facing society such as industry, business, e.g.. A problem to do with fluctuation in the share price at NSE with the announcement of new IPO.  Fundamental is also basic research. Asks questions like what happens when you mix Sodium (Na) and calcium (Ca). 

3. Quantitative VS Qualitative 

 





Quantitative is based on quantity while qualitative is based on the quality of a phenomena. These are the two main extremes of the continuum of data analysis. Quantitative deals with the quantities, the counts and statistical analysis are used while qualitative deals with the emerging themes fro the work and other cues that are gathered from the respondents. Techniques used in qualitative research are in-depth interviews, word association test, sentence completion test and other projective tests. Current debates revolve around this mode of research.

Qualitative vs quantitative research Qualitative research includes: ethnographies, phenomenology, case studies,  Quantitative research relies on the verifiability. This means confirmation, proof, corroboration and substantiation. Knowledge emerges from what can be proven by direct observation.  The researchers values, interpretations and feelings are not considered. Objectivity is reinforced.  The researcher is detached from the subject of study or the respondent. 

Applicability of Quantitative research When the research incorporates the statistical element (how many?)  When frequencies are sought to explain meaning  When control of approach is needed to allow for in-depth investigation of particular phenomenon. Rigid methods are applied and all procedures are specified.  It is a deductive process.  When the scenario is artificial e.g. In the laboratory. 

4. Conceptual VS Empirical Conceptual research is related to abstract idea or theory. Used mainly by philosophers or thinkers to develop new concepts.  Empirical research relies on experience or observation alone and often without due regard for system or theory. It uses data to come up with conclusions that are verifiable through observation or experiments. It is sometimes called experimental research.  This type of research tests hypothesis.  The researcher manipulates and controls are exercised. 

Others Longitudinal VS Cross section research  Field setting Vs laboratory setting research.  Historical research  Conclusion oriented Vs Decision result  Case studies  Action research 

Purpose of Research   



 

1. Discover new knowledge: this is discovery of new facts. This is the main purpose of research. 2. Describe a phenomenon: description provides knowledge. 3. Enable prediction: this is the ability to estimate a phenomenon. We are able to predict the markets through research. We also predict eclipses. 4. To enable control: this is the ability to regulate phenomenon under study. We Control one variable in order to regulate another. 5. To enable explanation of a phenomenon 6. Development of theory. This enables generalization. Research is also conducted to validate or falsify a theory.

Sources of Knowledge  

  

Research: This is still the most important source. It is objective and involves systematic procedures. Experience: This is a common mode of obtaining knowledge. This is subject to personal interpretations. Tradition: Transmission is done through cultural avenues and values. This is objective too. Authority: this takes the form of expert in a specialized area. Intuition: This is perception into a phenomenon through instinct.

Research and Theory 

 

 

A theory is a system of explaining a phenomenon by stating constructs and laws that interrelate these constructs to each other. It is a scientific prediction or explanation Kerlinger (1979) defines a theory as a set of interrelated constructs (variables), definitions, and propositions that presents a systematic view of a phenomena by specifying relations among variables, with the purpose of explaining natural phenomenon. A construct is a concept, a variable, abstraction, or idea drawn from the specific. Theory and research are interrelated. Theory guides research and research guides theory.

Basic in research 

Population: The entire group of individuals, events of objects having the common observable characteristic. Examples:  All

business people in the room  All indigenous trees in a natural forest.  All patients in a given hospital

A researcher must define the population that he/ she wants to generalize the results.  Ideally the research may wish to deal with the absolute population. This is the universe or the target population. 

defined 

  



The target (universe) population is not easy access. This may be due too logistical issues such as time , money and personnel that is required to locate ALL. Researchers draw samples from the accessible population. This is manageable although it may lack the ability for generalization. It may not meet the criteria for population validity. This is ensuring that the accessible population is in itself representative of the universe or target population. Once confirmed then generalizations may be made with confidence.

defined Sample: this is the selected number of cases from the accessible population. Sample should be selected in such a way that it is representative of the population. Each member of the sample may be called respondent or interviewee ( Subject was formally being used)  Sampling: this is the process of selecting a sample from the population. Sampling is a requirement for generalization, because it facilitates representativeness. There are broadly two types of sampling. Each of these two has its own variants. 

defined 

Variable: Is a measurable characteristic that assumes different values among the subjects.  Some

variables are quantitative while others are categorical. E.g. Age, height, weight are quantitative while Gender, profession are categorical. There are different types of variables (To be covered later). Variables are for the sample.



Parameter: This is the equivalent of a variable but for the population.

defined Descriptive statistics: These are indices that describe the sample. Examples are measure of central tendency (mean mode, median), and measures of dispersion( Range, standard deviation, variance) and measures of relationships (correlations)  Inferential statistics: This involves drawing inferences about the given phenomenon nto the population. Such inferences are based on the values obtained from the sample. Hypothesis are tested to enable the researcher to generalize the results from the sample to the population. 

continued Operational definition of : This refers to the measurement of a variable. It is the description of the operation that will be used in measuring a variable. May be stated quantitatively or categorically. E.g. Type of business may be categorized in different Manufacturing, communication, education etc.  certain variables can only be operationalised by use of indicators : Presence or absence of the indicator. Examples are intelligence, work performance, feeling of satisfaction etc. 

defined 

Data: Refers to all information a researcher gathers in the study. Data is in two forms  Primary

data: information obtained first hand from the respondents in the field.  Secondary data: information obtained from other sources such as journals, articles, books, etc.

Data may also be classified as being either qualitative (words and phrases) or quantitative (Numerical).  Statistics: it is the science of organizing data. May also refer to the indices derived from the data. 

Objectives Is the desired condition or end.  In research objectives are the specific aspects of the phenomenon under study that the researcher desires to bring out at the end of the research study. E.g. in a study of the factors that contribute to business failure, a researcher may have objectives such as: 

 Establish

the management related factors that hinder achieving of business goals.  To establish the characteristics of businesses that have failed in the recent past.  State others

Literature review Involves locating, reading and evaluating reports of previous studies, observations and opinions related to the planned study.  Literature review assists researchers in coming up with the problem, identifying the gap in information that locates the research in context, Helps in shedding light on the variables for the research and their interrelations, facilitates informed discussion of the research findings.  Besides literature provides the link between the present and the already existing body of knowledge. 

Statement of the problem It is a specific statement that clearly conveys the purpose of the research.  It answers the question: What exactly are you intending to in this research?  It focuses on the phenomenon that the study desires to describe, predict, control or explain.  The statement of the problem may be stated in question form.  It is best stated in one sentence. If you cant say it in a sentence, you still do not know it.  It is not easy to get it right but once gotten, the research is clear and easy to undertake. 

Examples of problem Statements What is the effect of design parameters on the maintenance cost of residential buildings in Nakuru?  The study seeks to establish the human and technical factors that contribute to road accidents in Kenya.  The study will compare the business leadership styles and performance in the Nairobi Stock exchange 



State a problem related to your proposed area of study.

Units of analysis Also called units of statistical analysis.  Are the units that are described for the purpose of aggregating their characteristics in order to describe some larger group of abstract phenomenon.  To describe a business society, we may study the management, the istration and the other workers.  Social artifacts may be units of analysis. Examples are books, paintings, buildings, jokes, dances, and sculptures. 

Units of observation

Is the subject, object item or entity from which the measure of the characteristic or the data required in the study is obtained.  May be a person, a group, a business or even a write up.  In most cases, unit of analysis is also the unit of observation.  However in other cases, units of observation may be different from units of analysis. Eg, Observe managers and supervisors to get information about the business. 

Hypotheses Is a researcher‟s anticipated explanation or opinion regarding the results of the study.  It is an educated guess of the explanations being sought.  It is a hunch. It is an opinion based on literature and existing body of knowledge  Some studies will have hypothesis while others will not.  It all depends on the nature of the study.  Hypothesizing may limit the information that would otherwise have come from the respondents. 

Theory 

 

 

Is a set of concepts or constructs and the interrelations that are assumed to exist among these concepts. A theory provides the basis for establishing the hypothesis to be tested in the study. A good theory can logically be broken down in a set of hypothesis which can be verified through experiments or observations. Theory and research are related to logic. There are two main types of logic: Inductive and deductive logic. Deductive is from general to specific while inductive is from specific to general.

Methods in research Generally, there are 3 methods in research according to Haring and Lounsbury (1983):  Experimental research method: Used in natural sciences. The effect of the independent variable to the dependent can be controlled. (Fertilizer vs. yield)  Normative research method: Control of variable is not possible. The relationship between the independent and dependent is observed. USED In social sciences. E.g.. Studying voting patterns in a given country.  Historical research methods: uses data from the past to learn about the causes, effects and trends of past events to explain the current situations. Past data is analyzed. 

Steps in Research There are seven steps in research:  1.Formaulation of a research problem: What causes…..  2.Definition of hypothesis: Reading from theory may hint on some possible guesses to the solution  3. Research design: Researcher how to move from “here” to “there”  4. Determination of data to be collected and how to collect it. May be Primary, secondary, qualitative quantitative. 

Steps in Research 5 Data collection: Data is collected using the appropriate instruments.  6. Analyzing and processing of collected data: Data is analyzed, hypothesis are tested. Decisions are made to reject or fail to reject the hypothesis.  7. Generalization: the established relationships are explicitly explained. This is stating the findings in form of a theory.  The generalization may be a new theory, a modification of an existing theory or a refutation of a theory. 

Chapter in proposal and thesis report A proposal has the following chapters.  Chapter I: Introduction 

Overview  Background information  Statement of the problem  Purpose o the study  Research objectives  Research questions  Hypothesis (Optional)  Justification of the study  Limitation of the study  Scope of the study  Operational definition of ****  Conceptual framework Or Theoretical framework 

Sections continued 

Chapter II: Literature review.  Contains reading and citations of other peoples works.  All works must be acknowledged  In text sources must appear at the reference section.  Correct citations must be followed  Citations may be author dominant or information dominant.  Literature reviewed must be recent and must inform the work being done.  Lack of citing is academic plagiarism. It punishable by disqualification.  Discussion in Chapter II is based on conceptual framework

Chapter III: Research Design and methodology  Key areas in this chapter are  Research design  Study area  Respondents  The population  Selection processes (Sampling)  Data collection  Instruments of data collection  Validity and reliability of instruments  Proposed methods of data analysis  All the sections in this chapter and the choices MUST be ed with relevant literature.  This is the end of Proposal Chapters.

REFERENCES It is a list of all cited works.  Must be written according to the style that the university proposes  Must be exhaustive to include all cited works  Must be written in Alphabetical manner  This is not an attachment but is part of the work.  Thorough reading and citing other people work only serves to improve on the quality of your work.  Read, Cite and improve yourself. 

APPENDICES These are documents that are useful in the research but are not really part of the main text.  They include: 

 Questionnaires

 Interview

schedules  Research permit and authorization  Transmittal letters  Detailed data analysis charts, tables and others.

Preliminaries These are details that appear before Chapter I.  They are numbered in I, ii, iii, ---- They include: 

 Declaration

both by the student and supervisor(s)  Acknowledgement  Dedication  Research abstract  Table of contents (This must be generated)  List of tables  List of figures  List of abbreviations and acronyms used in the write up.

Chapter I : Introduction The first step in in research is the identification of a problem. Without being clear about what you want to do, it is difficult to plan how you are going to do it.  “If you don‟t know where you are going, all destinations are correct.”  It is important that you know what you want to do.  This requires thought, reading and spending time wondering what the research should be all about. 

Attributes of a good topic  



 



Must meet the requirements of the examining body Must be within the abilities of the researcher to handle the issues (Available and possibly developable skills should suffice). Must excite the researcher: You must be able to sustain your interest in the topic for at least the period of study. Must be researchable with the available resources and within the stipulated time frame. Must require data that can be obtained or can be availed. Do not plan to utilize information that may be regarded state secrets. Mostly, should be related to some theory. This contextualizes your research. It helps in formulating research questions, objectives and hypothesis.

Attributes of a good topic      

A god topic should link to and lead to clearly defined research questions and objectives. If given a research idea by your organization, ensure that questions and objectives reflect the idea. The topic must have symmetry of possible outcomes. This is to do with the value of the findings. Must provide enough a scope for the project report to cover. Topic must be related to the purpose of the research. Must relate to the future career aspiration of the research. You are what your research is.

Identifying a research problem There is no limit to the number of good research problems. Knowledge is infinite and so are research areas.  First step is identifying the broad are that one wants to research. (Finances, management, human resource etc).  Second step is narrowing down to a specific issue within the area.  The criteria for choosing this area is: Must be important and researchable.  What exactly is an important area? 

A good problem should: Lead to findings with widespread applications  Challenge commonly held truism  Review the inadequacies of existing laws, theory  Cover reasonable scope 

 Scope

is determined by: Time available, money available, Equipment and availability of respondents.

Sources of research problem Existing theories- Validation or refutation  Existing literature- For identification of gaps  Discussion with experts-Topical areas mentioned  Previous research studies- Areas of further study in chapter V  Replication: to check if findings hold with time.  The media: Public interest expressed thro media  Personal experiences: Observations and reflection on intriguing issues is a source of research. 

Stating the problem The background builds the foundation for the statement of the problem.  Background is based on existing literature in order to identify the gaps.  It must ed with citations, well referenced  Starts from general to specific. A poor background leads to difficulties in stating the problem.  Past researches with their emphases may be used to the case at hand. 

A good statement of problem: IS clearly written to capture the readers interest.  Is objectively researchable  Indicates the scope of the research  Is justified: its importance to knowledge generation is clear  Must give the purpose of the research.  Purpose of the study is the broad focus of the study in 1 or 2 sentences.  Purpose crystallizes researchers inquiry into a particular area of knowledge.  Purpose centralizes the study to enhance focus. 

Purpose of the study   

 

 

Should not be too general. Its narrower than a topic but broader that statement of the problem. Requires sufficient planning, reading and inquiry in to the topic. Should be unambiguously stated in declarative manner. Should indicate the variables in the study. Relationship between the variable should also be stated where possible Target population must be stated in purpose. Consistency of variables between chapter I and III must be maintained.

Examples The purpose of the study is to investigate resource management skills among the leading business CEOs in Kenya.  The purpose of the study is to determine the effect of skilled training on business performance in Kenya.  The purpose of the study is find out the effect of alcohol on reaction time of truck drivers in Kenya.  State others based on your area of interest. 

Verbs in stating purpose 

Biased To  To  To  To  To  To  To  To  To  To 

show prove Confirm check Demonstrate indicate validate Explain illustrate



Neutral To Determine  To compare  To investigate  To Differentiate  TO EXPLORE  To find out  To examine  To inquire  To establish  To test. 

Stating the objective Objectives are the ends to be achieved.  Objectives build on the purpose of the study.  Objectives are evidence of a researchers clear sense of purpose and direction  Objectives are related to the purpose of the study.  Objectives limit the scope of the study  Several objectives should be generated from the purpose of the study.  Neutral should be used in stating the objectives. 

Hypotheses This is a researchers prediction regarding the outcome of the study.  Hypothesis states the possible difference , relationships or causes between variable or concepts.  Hypotheses are derived from existing theories, previous research, personal observations or experiences.  Multiple variables study should have multiple hypotheses.  Hypothesis without strong from theory should be avoided. 

HYPOTHESIS CONTINUED Exploratory research does not require hypothesis.  It is hypotheses that are tested.  Data either rejects or fails to reject hypotheses.  Purposes of hypothesis are: 

 Provide

direction by bridging the gap between the problem and the evidence.  Ensure collection of the necessary evidence to answer questions posed in the statement of problem.  Enable assessment of information collected in of relevance and organization.

Purposes of Hypothesis conti.  sensitize

investigator to certain aspects of the situation that are relevant regarding the problem at hand.  permit clear understanding of the problem by the researcher.  Guide collection of data and use the data collected to find solutions to problems.  Form the framework for ultimate conclusions

CHARACTERISTICS OF GOOD HYPOTHESES  

     

Must state clearly and briefly the expected relationship among variables Must be based on sound rationale derived from theory or previous research. Must be testable Must be consistent with common sense and generally held truth. Must be related to empirical phenomenon Variables must be consistent with variables in the Problem and chapter III Must be simple and concise as the complexity of variables allows. Must be stated in a way that implications may be deduced in the form of operations for validating or refuting relationships.

Types of hypotheses There are three main types of hypotheses  Null hypothesis: Also known as statistical hypothesis 

 States

that there is no real relationship, difference.  Any relationship or difference is due to chance.  Stated Ho: µ1=µ2 or µ1-µ2 =0 

Alternative non directional: Known as a research hypothesis 

States that there is a relationship of difference. Does not state the nature of the relationship or difference.



Stated H1: µ1≠µ2.



Hypothesis continued Alternative directional: Specifies the nature of the relationship or difference among variables.  Uses such as greater that, more than, less than, At least greater than, Increased, decreased, Higher than, lower than etc.  Stated: H1: µ1>µ2 or H1: µ1< µ2  The type of hypothesis determines if the study is going to be one tailed or two tailed.  Discuss confidence intervals at this stage 

Assumptions and limitations Assumptions: Are issues and facts that the researcher takes to be true without precisely ing them.  Limitations: Are aspects that may negatively influence the results of the study, but the researcher has no control. A common limitation may be lack generalizations of the study.  Others aspects that may fit here are Theoretical perspective, Conceptual framework, and Operational definition of 

Chapter II: Literature review 

The purpose of literature review  To

determine what is already done related to the research problem being studied. This is important because in helps in avoid unnecessary and unintentional replication, Forms a framework for interpretation of results, Increases familiarity with the existing body of knowledge.  Lit Review reveals strategies, procedures and measuring instruments that have been found useful in investigating the problem in question. This reduces mistakes.  Lit review suggests other approaches  Enables interpretation of the research findings

Literature review Helps the researcher to limit the research problem and define it better.  Stimulates new ideas and determines new approaches.  Divulges approaches that have been found futile.  Is a source of research based on suggestions for further research.  Puts together, integrates and summarize what is known in the area. Thus literature exposes gaps in literature and information. 

Sources of Literature Sources are broken down in two. Primary sources and secondary sources.  Primary source is those people who experienced the phenomenon. Direct witnesses, researchers of other reports.  Secondary: Include publications by other authors.  Sources: scholarly journals, Thesis and dissertations, Government documents, conference papers, Books, References quoted in books, international indices, Abstracts, Periodicals, reference sections of the library, computer search, the internet 

TIPS for literature review Do not be in a hurry. You may overlook important studies  Do not over rely on secondary information  Read all details of other works not just the findings. Read, methodology and methods.  Check daily newspapers to your other sources.  KEEP your reference list correct and up to date. Use end note program to make work easy.  Cite correctly and keep references all inclusive. 

Referencing within text Two methods: Content dominant and author dominant.  It has been found that variable A relates to variable B in direct proportions (Kamau & Otieno, 2006).  Kamau and Otieno (2006) found that variable A relates to variable B in direct proportions.  Discuss the APA style of referencing for different materials.  See Mugenda and Mugenda (1999, p.39) 



  



Chapter III: Research Design and methodology

A research design is the plan of action of moving from “here” to “there”. Here is the problem situation and there is the solution situation (Yin, 2002). Design encomes all issues that are planned to enable the research to achieve the desired ends. Design is informed and informs the key ingredients of the research process onion The onion has five layers that are interrelated and inform each other . The five layers of the general design process are shown next.

The research Onion Research philosophy

Positivism Deductive Experiment Cross-sectional Case study

Sampling

Research approaches

Survey

Secondary data

Grounded theory

Observation

Realism Research strategies

Interviews

Time horizons

Questionnaires

Longitudinal

Ethnography

Action research Inductive Interpretivism

Data collection methods

The research process onion 

Research philosophy: What do you believe to be the true way of generating knowledge? Positivism, Realism, interpretivism. Research is governed by what we think is true knowledge. The three views are different and close to mutually exclusive. 

Positivism adopts the approaches of natural sciences. Works with observable social realities and ends with law like generalizations similar to what physical sciences produce. Research is an objective analyst detached from the process and making data interpretations value free. Methods are highly structured to facilitate replication. Observations are quantifiable for statistical analysis. Researcher is independent of the research process. Is neither affected nor does he/ she affect the process and/or the subjects/ respondents of the research.

Philosophy cont. Interpretivism Argues that the world is too complex to be fitted in to laws as in physical sciences. Rich insights are lost when the research just observes for the sake of generalizations.  The world is complex and intertwined. Circumstances change and research should consider the prevailing contexts.  Remenyi et al (1998) says that there is a need to discover details of the situation in order to understand the reality or the reality working behind them.  This is associated with constructivism or the social constructionism 

Interpretivism continued It is necessary to explore the subjective meanings motivating peoples actions in order to understand the actions.  Situations may find different interpretations depending on the circumstances.  People not only interact with the environment but also make sense of it through interpretations of the events in the contexts of the social constructions.  Interpretivist makes sense of the circumstances by reaching out for the motives, actions and intensions in a way that is meaningful for the research participants. 

Realism Social constructionists also recognizes that people are likely to share interpretations of socially constructed environment.  Realism is based on the belief that reality exists that is independent of the human thoughts and beliefs.  There are large scale social forces and processes that affect people without necessarily being aware of the existence of such influence on the interpretations and behavior. 

Realism Realism shares some philosophical aspects with positivism such as external objective nature of some macro aspects in society, it recognizes that people are not objects to be studied like in natural sciences.  It recognizes the need for understanding the socially constructed interpretations and meanings (subjective reality) within the context of seeking to understand the broader social forces, structures or processes that influence the nature of people‟s views. 

Discussions of philosophies    



One can easily be trapped into saying one approach is better than another. They are better at different aspects of research. It all depends on the questions one seeks to answer. Research rarely falls neatly into only one of the philosophical inclinations. Most researches fall mid way between interpretivism and positivism reflecting the realism stance. More flexible research approaches are better than rigid and non flexible ones. Philosophical govern and protect the other decisions in the research process. This is the outer shell of the onion that determines all the rest.

Research approach This is the second layer. Has four strands namely: Deductive, inductive, abductive or Retroductive.  It governs the ways of logic that the research utilizes.  Governs the way of making conclusions and advancing arguments in the research.  It expounds on the link between the current research and theory.  Depending on the approach, different ends are achieved. 

Logic continued Deductive: Based on existing theory, hypotheses are tested. The research design is to test hypothesis about the current situation based on the already established. From general to specific. Related to positivism. This is using theory to test a particular situation.  Inductive: data is collected and trends are analyzed with a view of generating theory (Generalization). From specific to general. Related to interpretivism. This is building a theory from specific cases. 

 Read

and report about retroduction and abduction. Use the internet.



Implications of Deduction and Induction can be found in Saunders et al 3rd Edition (2003) Pages 86-90.

Layer 3. Research strategies      





This is the general plan of how you will go about answering the research questions. It will contain clear objectives derived from the research questions and purpose. It will contain sources from which data will be collected. will consider constraints that must be overcome such as time, access, money, location and ethical issues. Should reflect the fact that you have thought carefully about why you are employing particular strategy. All decisions on this must be justified. E.g. Why the organization? Why the department, why the respondents. Best justification is based on the research questions and objectives. Governs the tactics to be used for data collection.

Research strategies Strategy is concerned with the overall approach you adopt while a tactics about the finer detail of data collection and analysis methods.  Strategies: Experiments, survey, case study, action research, cross sectional and longitudinal.  Others are Exploratory, descriptive and Explanatory studies  Methodologies: Ethnography, grounded theory, Phenomenology, heuristic research etc. 

Research strategies 

Experiment: is a classical form of research. It owes much to natural sciences although features much also in Psychology related studies. It involves;  Definition

of theoretical hypothesis  Selection of samples of individuals from known population.  Allocation of samples to different experimental conditions  Introduction of planned change on one or more of the variables  Measurement on a small number of the variables  Control of other variables

Survey Is associated with deductive study and is common with social and business studies.  Allows collection of large amount of data from sizable population in a highly economical way.  Often uses a questionnaire (to a point that questionnaires are called surveys).  Is perceived authoritative by people in general because it is easily understood.  Answers mostly questions dealing with What? and How? 

Survey Continued Gives more control of the research process to the researcher.  Takes time in deg the questionnaires and piloting.  Data analysis is also time consuming.  Computer packages are now in use e.g. SPSS.  Data collected by survey may not be as wide ranging as other methods.  The numbers of questions can not be unlimited without exhausting the goodwill of the respondents. 

Case study Is a strategy for doing research which involves an empirical investigation of a particular contemporary phenomenon within its real life contexts using multiple sources of evidence.  Provides for in-depth study in order to gain rich understanding of the context of the research and the processes being enacted.  It has the capacity of generating answers to the question why? as well as what? and how?  Applies various methods of data collection.  Gives an “unscientific” feel to research.  It is worthwhile in exploring an existing theory. 

Time related studies Cross sectional studies: studies a given phenomenon at a give time.  Often employ survey strategy  May be seeking to describe evidence of a phenomenon or compare factors in different organizations.  They may also be qualitative. Many case studies are conducted over a short period of time hence are cross sectional.  Give relevant example. 

Longitudinal studies It research dealing with change and development  Takes a long period of time studying a phenomenon.  One can easily be affected by the research process.  Analyses trends of events over a long period of time.  Basic question is: what kind of change or has there been change in a given phenomenon over time? 

PURPOSE BASED STRATEGIES There are three main purposes of research: Explorative,, Descriptive, and explanatory studies.  Research purpose may be more than one. Purpose may change with time. 

 Exploratory: For finding out what is happening, to seek new insights to ask questions and assess phenomenon in a new light.  Useful in clarifying an understanding of a problem.

Exploration continued 

Three principal ways of conducting exploratory research: Search of literature, talking to experts Conducting focus group interviews.

Exploration is about all that is found on the way.  Its is flexible and adaptable to change.  New data may change direction of research.  This flexibility does not mean absence of direction of inquiry (Adams and Schvanelvelt, 1999) 

Descriptive 



  

Objective of this kind of study is to portray an accurate profile of persons, events or situations. It may be an extension or a forerunner to exploratory research. Its necessary to have a clear picture of the phenomenon on which you wish to collect data prior to the collection of data. One should not stop at describing the situations but should be able to draw conclusions. Description research has a clear place in business, social studies and management. Should be thought of as a means and not an end on itself.

Explanatory Research These are studies that establish the causal relationship between variables.  Emphasis here is to explain the relationship between variables  Example is where the research may want to establish the relationship between individual performance and pay systems applied.  Based on these three, where does your proposed research topic fall??  Why do you answer that way??  Is that the best approach that you can give to the issue at hand??? 

Practitioner research All the above strategies may involve you carrying out research in your own organization, thus adopting the role of a practitioner researcher.  Most part time students conduct research at their place of work. They may even deal with problems that are identified by the employer. Such times renders the research a practitioner researcher.  It has the advantage that the researcher is aware of all the complexities within the organization.  This advantage may come with the disadvantage the researcher may not see all the details since they are part of daily life to him or her. 

Practitioner research There is also the problem of status. Juniors will feel intimidated by seniors while conducting the research.  The issue of time cannot be ignored. The researcher is also supposed to be working at the establishment. These are heavy expectations.  Practitioner researchers should negotiate their time to get the correct proportions of work and research.  Being too close to the setting may influence the validity of the data collected. 

Triangulation This is the use of different aspects in the research as a way of ensuring that the research captures what it should.  There are different types of triangulation: 

 Methods

triangulation  Data analysis triangulations  Theory triangulations  Researcher triangulation  Data analysis triangulation 

Each method, tool or technique has its uniques strengths and weaknesses.

Research Designs 

Cohen, Manion, et al. (2000) calls these research styles. There are several styles:  Naturalistic  Historical  Case

studies  Correlational research  Ex post facto research  Experiments, quasi Experiments and single case studies  Action Research

Naturalistic research Elements of naturalistic research  Humans actively construct their own meanings of situations  Meanings arise from social situations and its handled through interpretive processes  Behavior and thus data is socially situated, context related, context dependent and context rich. Context must be understood first to understand behavior.  Realities are multiple, constructed and holistic 

Naturalistic continued Knower and known are interactive and inseparable  Only time and context bound hypothesis are possible  All entries are in a state of mutual simultaneous shaping hence cause and effect is not possible to establish.  Inquiry is value bound  Inquiries are influenced by the inquirers values as expressed in the choice of a problem, evaluation or policy option. 

Naturalistic continued 

    

Inquiry is influence by the choice of a paradigm that guides the investigation into the problem Inquiry is influenced by the theory utilized that guides data collection and analysis. Inquiry is influenced by values that are inherent in the context Inquiry is either value resident (Reinforcing) or value dissonant (conflicting) Research must include thick descriptions of contextualized behavior Attribution of meaning is continuous and evolving over time

Naturalistic conti.        

People are deliberate intentional and creative in their actions. We create our own futures but not necessarily in situations of our own choosing. Social research needs examine the situations through their eyes of the participants Researchers are instruments of the research Researchers generate rather than test hypothesis Researchers don not know in advance what they will see or what they will look for Humans are anticipatory beings Generalization is limited to particular settings

Implications of these axioms  Lincoln

and Guba say that based on these axioms,  Studies must be set in natural setting as context is heavily implicated in meanings  Humans are the research instruments  Qualitative methods sit more comfortably than quantitative  Purposive sampling enables more full scope of issues to be explored.

Implications conti  Data

is analyzed using inductive rather that priori and deductive theory emerges rather than pre-ordinate.  Research design emerges over time  Outcomes of the research are negotiated  Natural mode of reporting is case study  Applications are tentative and pragmatic  Trustworthy and its components replace more conventional views of reliability and validity.

Stages in naturalistic research   

  

  

  

1. Locating the field of study 2. addressing ethical issues 3. Deciding the sampling 4. Finding the role and managing entry into context 5. finding informants 6. developing and maintaining relations 7. data collection in situ 8. data collection inside the field. 9 data analysis 10 Leaving the field 11. writing the report Details can be found Cohen, Manion et al (2000 p.141-157

Historical research Is defined as the systematic and objective location, evaluation and synthesis of evidence in order to establish facts and draw conclusions about past events.  It is an act of reconstruction in the spirit of critical inquiry aimed to achieve a faithful representation of the past age.  Uses data from personal experiences, observations of others, documents and records.  Leads to new understanding of the past and its relevance to the present and the future. 

Historical research continues Values of historical research are as follows:  Enables solutions to contemporary problems to be sought in the past  Throws light to present and future trends  Stresses the relative importance and the effects of various interactions that are found within the cultures  Allows for the reevaluation of data in relation to selected hypothesis, theories and generalizations that are presently held about the past.  History is employed to predict the future and the present is used to explain the past thus it is useful research. 

CREDIBILLITY of RESEARCH FINDINGS This has to do with reducing the possibility of getting the answer wrong.  Has two particular emphasis in research design. These are reliability and validity.  Reliability: can be addressed by posing the following questions: 

 Will

the measures yield the same results on other occasions?  Will similar observations be reached by other observers?  Is there transparency in how sense was made from raw data?

Threats to reliability  







There are four threats to reliability. 1. Subject or participant error: this is closely related to the conditions under which people are responding to the questions. Monday is not like Friday. Selecting neutral tines helps. 2. Subject or respondent bias: sometimes people answer questions not to give the correct answer but to preserve their relationship with the person asking. (anonymity helps in this case). 3. Observer error: Different researchers may interview people differently depending on their personal attributes Structuring and education helps). 4. Observer Bias: Researchers may interpret same response differently depending on their inclinations.

VALIDITY 

There are two types in validity  Internal

validity: How well the target data is captured from the research sample.  External validity: How well the data collected from the sample can be generalized to the whole population. Can the values obtained from the sample suitably represent the values for the population? There are threats to both internal and external validity.

Threats to internal validity

Campbell and Stanley (1963) outlined 8 factors that affect internal validity. They are:  i) Attrition or exptal Mortality  Ii) Maturation  Iii) Statistical regression  Iv) Differential selection  V) Instrumentation  Vi) History  Vii) Interaction between Selection and maturation  Viii) Testing effects 

VALIDITY

Deciding on research approach and choosing a research strategy RESEARCH “ONION”

The onion Research philosophy

Positivism Deductive Experiment Cross-sectional Case study

Sampling

Research approaches

Survey

Secondary data

Grounded theory

Observation

Realism Research strategies

Interviews

Time horizons

Questionnaires

Longitudinal

Ethnography

Action research Inductive Interpretivism

Data collection methods

Introduction Before one can decide on the research strategy to be used in the research e.g. whether to ister a questionnaire or conduct interviews, you will be required to know the different approaches to the research.  This can be more easily explained through the research „onion‟, where there are important layers of the onion that need to be peeled away. These layers can be categorized in 5. 

1. Research Philosophy - Layer 1 

When you adopt this philosophy, it will depend on the way you think about the development of knowledge, the way we go about doing the research, which can be viewed as follows:-

1.1

POSITIVISM

The end product of this kind of research is law-like generalizations  There is detached interpretations about data collected  Emphasis is on a highly structured methodology  The researcher is independent of and neither affects nor is affected by the subject of research 

1.2. Interpretivism Persuasive in a case of business or Management Research  It is necessary to explore the subjective meanings motivating peoples actions in order to be able to understand those that they study the motives, actions and intentions of research participants 

1.3. Realism 

Reality exists that is independent of human thoughts and beliefs.

2. Research Approach - Layer 2 A research project will involve the use of theory. Theory becomes explicit in the presentations of the finding of conclusions of the research.  At the beginning of the research one must be clear of the design of your research project. Whether to adopt deductive or inductive approach. 

Deductive approach is where you develop a theory and hypothesis and design a research strategy to test the hypothesis.  Inductive approach is where you collect data and develop theory as a result of the data analysis. 

3. Research Strategy - Layer 3 Some of the research strategies that will be discussed can be deductive and some inductive.  The strategy that is adopted should be appropriate for your particular research questions and objectives should be thought of as being mutually exclusive.  The strategies to be considered are: 

3.1. Experiment Definition of a theoretical hypothesis  Selection of samples of individuals from known population  Allocation of samples to different experimental conditions  Introduction of planned change on one or more variables  Measurement on a small number of variables  Control of other variables 

3.2. Survey It is the deductive approach (where you develop the theory & hypothesis first)  It is a common strategy for business and management research  Allow collection of large amount of data from a sizable population in an economical way  Data is obtained through questionnaires  Data is standardized and easy for comparison  Survey strategy is easily understood e.g. news 

3.3. Case Study An empirical investigation of a particular phenomenon  It enables researcher to gain rich understanding of the context of the research  It has ability to generate answers to the questions  Data collection methods used can be various  It can enable you to challenge an existing theory and also provide a new source of new hypothesis. 

3.4. Grounded Theory Best approach for inductive approach (where data is collected and then theory is developed)  The data collected leads to generation of predictions  The predictions are further tested to confirm or otherwise the predictions  There is continual reference to the data 

3.5. Ethnography Firmly rooted to inductive approach  Emanates from the field of anthropology  It is time consuming  Not a dominant strategy for business 

3.6. Action Research It has explicit focus on action Management of change (objective) commences with an initial idea for change intervention.  The next step is fact finding and analysis of the change intervention to generate an overall plan.  The next step is to carry out the plan and implement planned action steps 

Finally, there is the monitoring of the implementation of the plans and evaluating to ensure the intervention for change meets the needs of the organization.  The change interventions can be revised and the same cycle applied again and again. 

4.  (i) (ii)

Time Horizons - Layer 4

This can be decided depending on your research question. Cross sectional - “snapshot” – when the research is taken at a particular time Longitudinal - “diary” - when the research is a representation of events over a given period.

5. Data Collection Methods - Layer 5 5.1. Documents  Historical literature review meta analysis Diaries content analysis secondary data  Identify trends in leisure research and practice  Diaries, journals are kept & researcher conducts content analysis of studies, reports and diaries.

•5.2. Observations Interpretive Ethnographic Participant observer case study  Observe how people behave and interact in public  Observe systematically 

5.3. Survey Questionnaire Interview Standardization Scales/Instruments  To learn what people think about leisure motivation  To identify relation between motivation and satisfaction 

5.4. Experimental True designs quasi designs  Obtain information under controlled conditions about leisure attitudes and experience with virtual reality – subject are chosen randomly. 

5.5.



Other Fields Methods - Normal Group Technique - Delphi

To identify trends and issues about leisure services management and delivery systems

5.6. Multi methods approach 

Combination of methods above

Chapter 6 Developing Research Instruments LET US RECALL THE DEFINATION OF RESEARCH Let us use the definition by Mouley(1978) - Research is the process of arriving at effective solutions to problems through a systematic collection, analysis and interpretation of data. Explanation - Before we can have the data/information, we must have a means of obtaining the data.

OBJECTIVES OF THE CHAPTER    

Identify the various instruments used in research and how to develop them, to collect data/information. To know the advantages and disadvantages of such instruments. To know and understand the rules that govern developing each instrument. To know and understand how to present and ister such instruments in collecting information/data.

MOST COMMONLY USED INSTRUMENTS - Questionnaires- Questions … - Interview schedules - Standardized tests - Observation forms

QUESTIONNAIRES - Commonly used to obtain important information about the population. -

Items in the questionnaires are developed to address a specific objective, research question, hypothesis of the study.

-

It‟s important to know how information obtained in Q will be analyzed.

Disadvantages of a question which is not well thought out -

May confuse respondents on the name of information Q. May discourage respondents to the extent of discarding the question and therefore leave out important information. RQD in the study. Kinds of questions used in questionnaires Are two broad categories 1. Structured or closed ended 2. Unstructured or open ended

a) -

-

Structured or closed ended

Such questions are accompanied by a list of all possible alternatives. How the respondents select the answers that best describes their situation. It is always impossible to exhaust all the categories Such questions therefore include a category called “other” to summarize what may not fit in the given list.

Example: What type of a car can u drive? 1. Toyota 2. Mercedes 3. Nissan

4. Other

Advantages of closed ended questions a) b) c)

Are easier to analyze since they are in an immediate usable form Easier to ister because each item is followed by alternative answers. Are economical to use

Disadvantages closed ended questions a) b)

Are more difficult to construct Responses are limited

b) Unstructured or open ended questions These questions are the respondent complete freedom of response - The free response permits an individual to respond in his or her own words - Amount of space produced is a good indicator whether a brief or lengthy answer is desired. - In some cases only a phrase or figure is needed. -

Example Question How often do you go shopping in a month? __________________________________________ __________________________________________ __________________________________________ __________________________________________

c) Contingency questions/filter question    

Are subsequent/follow questions which are asked after the initial questions. Are targets to contain groups of respondents. The purpose is to probe for more information. Simply the respondents task

Formats of formulating contingency questions. 1. Can be closed ended- alternative 2. Answers given can be open ended-personal response

c) Matrix Questions Are questions which show the same set of response categories Are common whenever scales are being used.

Example Please indicate whether you are 1. Extremely dissatisfied 3. Neutral 2. Dissatisfied 4. Satisfied 5. Extremely satisfied Question – How satisfied are you with your a) Financial status 1 2 3 4 b) Quality of life 1 2 3 4 c) Level of assets 1 2 3 4 d) Ability to save 1 2 3 4

5 5 5 5

Advantages of matrix questions Questions in matrix forms are complete hence respondent is not put off. 2. Space is used efficiently. 3. Easy to compare responses given to different items. Disadvantages 1. Because they are easier to construct they are sometimes used when other formats could have been more appropriate. 2. Some respondents may form a pattern of agreeing or disagreeing with statements. 1.

Scales used in questionnaires and interviews - The … type of scale is the most commonly used rating scale. Rating scales - Used to: - Measure perception - Attitude - Values - And behaviour - They consist of numbers and descriptions - The numerical scale helps to minimize the subjectivity - Therefore quantitative analysis is possible.

 

Like… scales comprises of 5-7 response categories The numbers here are ordered such that they indicate presence or they indicate presence or dispense of the … being measured.

Rules for constructing questionnaire and questionnaire items 1. List the objectives you want the questionnaire to accomplish. 2. Determine how information obtained from each questionnaire will be analyzed. 3. Clarity is essential 4. The intended meaning of a concept with several meanings must be defined. 5. Short questions are preferred to long ones.

1.

2. 3. 4. 5. 6. 7. 8.

Items should be stated positively as much as possible. Double … items should be avoided. Leading or biased questions should be avoided. Very personal and sensitive questions should be avoided. Simple words should be used when constructing. Questions that assume facts with no evidence should be avoided. Avoid psychologically threatening questions. Include enough information in each item.

Ordering the items in a questionnaire Items in a questionnaire should be arranged in a logical sequence. Tips on how to organize your questionnaire 1. Begin with non-threatening, interesting the … 2. Its advisable not to pot important questions at the end of a long questionnaire. 3. Have some logical sequence when potting them together. 4. Arrange questions according to themes being studied. 5. If the questionnaire is arranged into content sub-sections, introduce each section with a short statement concerning its content and purpose. 6. Socio-economic questions can be placed either at the beginning or towards the end but its advisable to start with personal interaction … report. 7. Each questionnaire should be given an identification number I.A.

Pre-testing of the questionnaire Make the questionnaire Organize and lay out the questions for ease of completion. All pages and items in the questionnaire should be numbered. Brief and clear instructions must be included 1. General instructions 2. Specific instructions 5. Make your questionnaire short 1. 2. 3. 4.

Pre-testing the questionnaire Pre-testing – Is trying a questionnaire out in the field once it is finalized. The questionnaire should be pre-tested to a selected sample which is similar to the actual sample the researcher plans to use in the study. Procedures which will be used in actual data collection are the same procedures that should be used in pre-testing.

Kinds of questions used in questionnaires 1. 2. 3. 4. -

Structured or closed-ended Unstructured or open-ended Contingency questions Matrix questions Pre-testing sample are normally between 1% and 10%

The importance of pre-testing a questionnaire 1. 2.

3. 4.

Reveals vague questions Comments from respondents in the pretest when considered and incorporated help to improve the questionnaire. Deficiencies in the questionnaire will be revealed by pre-testing. Helps the researcher to analyze and see if the methods of analysis are appropriate.

Ways of istering questions 1. Self istered questionnaires Researcher istered questions Use of internet

The letter of transmittal    

Also called a cover letter This accompanies every questionnaire The letter must be brief but must contain an adequate … about the research. A letter signed personally instead of a photocopy adds a personal touch.

What information should a cover letter contain? 1. Letter should explain purpose of the study 2. Should also briefly explain the importance and significance of the study. 3. Advisable to commit yourself to sharing results when study is complete.

4. 5.

6. 7.

The letter should include a brief assurance of confidentiality. If the study is d to a certain institution oil organization, its advisable to have an endorsement from such an institution, or organization. Assure the anonymity of respondents if the research is sensitive. The letter should contain specific deadlines by which the completed questionnaire should be returned.

Follow-up techniques Various follow up procedures 1. Sending a follow-up letter 2. A a questionnaire and follow-up letter Response rate and non-respondents Response rate This refers to the percentage of subjects who respond to questionnaires Students show that the first mailing yields 70% response rate. Acceptable response rate of 50% is adequate for analysis and reporting. A response rate of 60%is good and 70% and over very good.

Non-respondents Non respondents are subjects who do not respond to the questionnaire  It is a concern when 30% more of respondents do not return questionnaires since this could affect the results of the study.  Non-respondents can be significantly different from those who have responded.  Non respondents may also not have any differences. 

Interviews as a research instrument 

 

An interview is an oral istration of a questionnaire or an interview schedule. Interviews are face to face encounters. To obtain accurate information through interview a researcher needs to obtain maximum co-operation from respondents.

Advantages of interviews 1. They obtain in-depth data which is not possible to get using a questionnaire. 2. Interviews make it possible to obtain data required to meet specific objectives of the study. 3. Interviews guard against confusing the questions. 4. Are more flexible than questionnaires

5. 6. 7. 8. 9.

Very sensitive and personal information can be extracted from the respondent. The interviewer can clarify and elaborate the purpose of the research. The interviewer can get more information by using probing questions. The interviewer is able to get the respondent‟s negative side through interaction. Interviews yield higher response rates.

Disadvantages of interviews 1. Are more expensive 2. Interviews tend to be misused to get factual responsessince they are easier to use 3. Interviews require a high level of skill-communication and interpersonal skills

4. 5. 6. 7.

Interviews need to be trained to avoid bias. Interviews often introduce bias and subjectivity into the study. Generally involve smaller samples Responses may be influenced by the respondents reaction to the interviewer.

Rules pertaining to interviews 1. The interviewer must be pleasant 2. Interviews must show genuine interest in getting to know the respondents. 3. Interviewer must find out … what kind of person the respondent would like to interact with.

4.

5. 6.

7. 8. 9. 10. 11.

Interviewer should be relaxed and friendly. Interviewer should be very familiar with the questionnaire or the interview guide. Interviewer should have a guide of the questions and their order. Interviewer should interact with respondent as an equal. Should pretest the interview guide before using it. Should inform the respondent about confidentiality of the information given. Interviews should not ask leading questions. Should remain neutral in an interview situation.

An interview schedule -

Is a set of questions that the interviewer asks when interviewing Are also used to standardize the interview situation.

Types of interview schedule   

Structured interview schedule- closed Unstructured interview –open Semi-structured interviews both the above two are used.

Note taking during the interview -

This is the method of recording the respondent responses during the interview as the respondents‟ talk. Answers should be recorded exactly.

Advantages of note taking 1. 2.

It facilitates data analysis No information is left out owing to forgetfulness or omission

Disadvantages of note taking 1.

2. 3.

4. 5.

May interfere with the communication between the respondent and interview. Sometimes report is not maintained. Note taking may upset the respondent. It‟s answers are personal and sensitive. If delayed, important details may be forgotten. Make the interview lengthy and boring.

Tape recording - Here, a tape recorder or a video tape is used.

Advantages of tape recording

1. 2. 3. 4. 5. 6.

No unconscious selection of data in the course of recording. The tape can be played back for a thorough study. Makes it possible to re analyze the data in order in order to test objectives or hypothesis that were not there originally Another person can use the data to establish reliability. A tape recorder speeds up the interview. Communication is not interrupted.

Disadvantages 1. 2.

3.

The presence of a tapes recorder changes the interview situation considerably. Respondents may be reluctant to give sensitive information on tape. Transcribing the tape and then analyzing the information is time consuming and tedious.

 1.

2. 3. 4. 5. 6. 7.

Communication during interviews Some rules of communication Language used must ensure effective communication Technical must be avoided. The interview should explain purpose of the study. The interview should first establish report. Interviewer should monitor the interview situation. Avoid harsh words and discriminatory remarks. When respondent gets carried away, the interview should tactfully bring them back to the point.

Training of interviews - For objectivity and reliable information interviews must be trained.

Tips on interviewer training 1. 2. 3.

4. 5. 6. 7. 8.

9.

Should be done in groups rather than with individuals. Trainer should explain the study, its purpose, general guidelines and procedures to the interviewer. The groups should go through the interview schedules. Prepare specifications to accompany the interview guide. Trainer should demonstrate how to carry out interviews. Interviewer should be paired off to practice on each other. Pretest the interviews During field work, interviews should check the researcher for . Interviews should be given enough latitude to deal with unforeseen circumstances during fieldwork.

Probing questions during an interview Rules on probing questions 1. Must be … 2. Should be thought out before hard and jotted down. 3. Should be appropriate. Telephone interviews Advantages 1. Sometimes cheaper then to travel 2. Little costs be incurred when no one answers and this makes call back possible. 3. Can be easier to interview subjects on phone than home visits.

Standardized and non-standardized tests. Standardized

Can be divided into: 1. Norm referenced and criterion referenced tests.

Non-standardized tests

Examples are classroom tests Observation forms, schedules or checklist - A researcher utilizes an observation checklist to record what he or she disserves during data collection. - Researcher must define the behaviors to be observed. 

Points to bear in mind when using observation forms 1. 2. 3.

4. 5.

It must be tested once it has been developed. Not too many behaviors should be observed. Number of subjects must be controlled. The behavior to be observed should be defined in sufficient detail. Training is necessary when more than one observer is used in study a) Master the observation form b) Standardize observation procedures

DATA ANALYSIS

DATA ANALYSIS  Successful data analysis, whether quantitative or qualitative, requires  (1) Understanding a variety of data analysis methods,  (2) Planning data analysis early in a project and making revisions in the plan as the work develops;

Requirements cont.  (3) Understanding which methods will

best answer the study questions posed, given the data that have been collected;  (4)

Once the analysis is finished, recognizing how weaknesses in the data or the analysis affect the conclusions that can properly be drawn. The study questions govern the overall analysis.

DATA ANALYSIS But

the form and quality of the data determine what analyses can be performed and what can be inferred from them. This implies that the evaluator should think about data analysis at four junctures:

DATA ANALYSIS  i) When the study is in the design phase,  ii) When detailed plans are being made for data collection,  iii) After the data are collected, and  iv)As the report is being written and reviewed

DATA ANALYSIS  Involves

converting raw data from the field into a form that the researcher can make sense of.  It Involves – data coding, data entry and other statistical procedures

Pre-analysis of data: After istering the measuring instrument e.g. a questionnaire- the raw data collected must be systematically organized in order to facilitate analysis.  If empirical or quantitative analysis is anticipated, the response in the questionnaire will have to be assigned numerical values e.g. for Yes- 1 , No – 0 or vice versa. Usually easily accomplished if the item in the questionnaire are close ended.  If they are open ended the researcher must try to categorize all the responses given and assign numbers to them. In such questions subjects are free to give their own Reponses. 

Coding and entering data It involves conversion of data into numerical codes representing attributes on measurements of variables. • Coding should include as much information as possible because once the coded data is entered into a computer it is impossible to recover any details which were initially omitted. • Code categories must be exhausted and mutually exclusive e.g. only one code should be assigned to each response category • Start with the preparation of a code book – a document that describes in specific details the coding scheme to be followed. Code book describes a code assignment for each response category for each item in the questionnaire. Also indicates the location column and width of each variable in the spread sheet. •

Coding and Entering contd. A code sheet is designed to correspond to the number of columns in the spreadsheet.  An option is the use of a bubble sheet where the researcher / subject records the responses by darkening the appropriate circles using pencil.  The complete sheet is then directly scanned into the computer spreadsheet.  When data is already in the computer, the method of analysis will depend on: Proposed hypothesis and objectives  The research design used  Type of measurement scale used in the measurement of variables. 

Qualitative analysis This is non empirical analysis where the study does not require quantifiable data e.g. in such case studies as content analysis and historical studies.  The researcher analysis information in a systematic order to come to useful conclusion and recommendations by establishing patterns, trends and relationships. 

Quantitative analysis Descriptive statistics -Usually the first step in data analysis is to describe/ summarize data using descriptive statistics.  It enables the researcher to meaningfully describe a distribution of scores using statistics. 

Main types of descriptive statistics (measure of central tendency) 1. The mode 

It is the measure that appears most in a particular variable among a sample of subjects i.e. most frequent. e.g. variables:- number of people in a family 3 4 5 6 6 6 7 9 10 12,the mode is 6

Characteristic  A set of scores may have more than one mode e.g. 3 4 4 4 5 6 7 7 7 8 10 11  The mode are 4 and 7 such a set is „bimodal‟.  The mode tends to be unstable e.g. equal sized samples randomly selected from the same are likely to have different modes although they may have similar characteristics.  It is possible for a set of scores/ measurements not to have any mode – when all the scores in a group occur with the same frequency.

Types of descriptive statistics contd.    



2. The median i.e. the middle score or the 50th percentile in a group of scores. The point below and above which 50% of the score fall e.g. 75 85 82 84 87 Median = 84 Therefore scores must be arranged in ascending order in order to determine media. The score are even the median of the score is the average of the two middle scores e.g. 21 23 24 25 27 30, the Median = (24+25) /2 = 24.5 Characteristics Median does not take into the the extreme values in a distribution. It does not reflect the very low or the very high values in the distribution.

Types of descriptive statistics contd 



3. The mean The average of a set of scores/ measurements. Add up all the scores then divide by the total no of scores e.g. scores of students in a test 10 12 12 20 15 10 18 16 8 5 the Mean 126/10 = 12.6 The Formula is expressed as; X =  x/n Where X = mean  = sum of the scores X = Each score n = № of score

Types of descriptive statistics contd 





Characteristics of mean If equal sized samples are randomly selected from the same population their means tend to be more similar to each other than either median or modes therefore said to be „stable‟. Mean is the only measure that takes into each score in the distribution therefore effects of very high or very low values is reflected in the mean. A weakness of a mean is that it is pulled towards an outlier . Outlier are extreme scores. e.g. if scores are between 20 -40 a score of 90 is an outlier and would abnormally raise the mean .

Measures of variability Variability is the distribution (dispersion of scores around the mean of the distribution).  Purpose – help the researcher to see how spread out of score/measures for each variable are.  It gives information on extent of individual differences on a given variable.  Type of measures of variables 1. The range  This is the difference between the highest score and the lowest score in a distribution. Its determined by subtracting the lowest score from the highest. e.g. 78 79 80 81 82 85 the range 85 -78 = 7 Advantage  It gives a quick rough estimate of variability. Disadvantage  It only involves two scores therefore not sensitive to the totals distribution. 

Measures of variability contd. 2. Standard deviation and the variance  Standard deviation – is the extent to which scores in a distribution deviate from their mean/average. It involves subtracting the mean from each score.  Variance – determined by squaring each deviation and then divide this total by the degrees of freedom.  Square root of the variance gives the standard deviation.  A large value denotes greater variability. A small standard deviation denotes less variability of score in the distribution.         

Variance S2 1 Where S2 = S = Xi = X = N = N-1 =

=

 (X1 – X) 2 n–1

sample variance sample standard deviation each value of score sample mean sample size degrees of freedom

S=

 (X1 – X)

2

n–

Properties of standard deviation • Takes into all scores and responds to the exact position of every score relative to the mean of that distribution. • It is very sensitive to extreme score/ outlier

Measures of variability contd. 3. Frequency distribution tables Shows the distribution of scores in a sample for a specific variable. e.g. frequency distribution table of test scores Grouped frequencies-Where scores are combined into smaller categories for reasons such as:• When scores are in such a wall that contain scores are not obtained by any subject. • When the sample is very big • When information sought is sensitive e.g. annual income and response categories in the questionnaire are given in intervals. • Class intervals start with the lowest score as a multiple of the intervals width. So frequencies are given for each class interval. Class interval should be 10 -15 in №. •

Graphic representation of a frequency distribution A graph enables the reader to see the trend of the distribution more easily.  Graph should have;  Two labeled axis – score on horizontal axis and frequency on vertical axis  Appropriate scale – width: height be 3:5  Title Types of graphs commonly used to present data research 

1) Histograms; Consist of a series of adjacent bars whose height (Y axis) represents frequency and score are on horizontal axis

Figure 1: histogram showing frequency against scores

Graphic representation of a frequency distribution contd. 2. Frequency polygons  First establish the mid point of the class interval. i.e total of upper and lower class limit divide by 2 e.g. lower class limit 20.5 for class 20-25,the Upper class limit is 25.5 and Midpoint 20.5 +25.2)/2 = 23  Plot the mid points against the frequencies and the points with a straight line. e.g. distribution of scores from a test

Fig 2: Frequency polygon

Graphic representation contd. 3. Bar charts  Is like the histogram except that spaces are left between the bars to signify a lack of continuity of flow between the categories.  Usually preferred when data is discrete/categorically or when the scale in nominal/non ordered e.g. faculty of education enrolment in .K.U. 

Fig 3: Bar graph

Graphs contd. Which graph is best?  Depends on data to be represented e.g. Bar graphs are better used for discrete variables  Frequency polygons has advantage because several can be superimposed for comparisons.  Histograms are easier to understand especially when only one distribution is being represented. 

Inferential Statistics these are hypothesis testing techniques • Deal with inferences about population based on results obtained from samples i.e. it determines how likely it is for the results obtained from a sample to be similar to results expected from the entire population. Statistical procedures used in inferential statistics include; Parameter techniques – where certain assumptions obtain the data are made and Non parameter techniques where no assumptions about data are made. •

parameter techniques Correlation • Used to analyze the degree of relationship between two variables Pearson product moment correlation (r) • Used when both variables that the researcher wishes to study are measured at a ration on interval scales and one continuous e.g. age, income, years of education etc. Correlation coefficient (r) ranges from -1 to +1 • It tells the researcher:- The magnitude of the relationship between two variables the bigger the coefficient, the stronger the association between the two variables • The direction of the relationship between the two variables e.g. if ( r) is positive therefore there is a positive relationship i.e. the two variables vary together in the same direction if variable X1 increases X2 also increases. • If (r) is negative therefore as X1 decrease. Variable X2 increases- they vary together in opposite directions i.e. inverse relationship. • It determines how various variables are related

2. Chi-squire test •

•

• • •

•

Chi squire (X2) is a statistical technique which attempts to establish relationship between two variables both of which are categorical in nature e.g. you may want to test the hypothesis that there is a relationship between gender and the number of road accidents caused by drives. The variable „gender‟ is categorized as male and female. While the variables „number of accidents‟ categorized as „none‟ few and many. The technique compares the proportion observed in each category with what would be expected under the assumption of independence between the two variables. There is a need to establish the significance level of the test before hand. The S.L. is the probability of obtaining similar results through chance. It should be set as 0.01. A chi squire test is a non parametric technique therefore no assumptions about the data or the parameters in the population are made. It is therefore a less powerful tool for establishing relationships when compared to correlation techniques.

3. Regression Regression analysis is a type of analysis used when a researcher is interested in finding out our whether an independent variable predicts a given dependent variable. Two categories: Simple regression • Used when the research is dealing with only one independent variable and one depend variable. e.g. whether education influences financial status of house holds. •

Multiple regression • Attempts to determine whether a group of variables together predict a given dependent variable. E.g. whether age, education, household size and marital status influence financial status of households.

4). The analysis of variance (ANOVA) •

•

•

This is a data analysis procedure that is used to determine whether there are significant differences between 2 or more groups / samples at a selected probability level. One way analysis of variance Where groups are being compared on one variable but at different levels. Two way analysis of variance Where two or more groups are being compared in more than one variable

ANOVA contd. Chi- squire  

Deals with frequencies in different groups Researcher aims at testing whether the frequencies in each cell is significantly different from expected results if the variables are independent of each other

ANOVA

Deals with sample statistics mainly mean (x)  Researcher aims at finding out whether the means for various groups are significantly different at a given probability level  The dependent variable is measured at either ratio 

ANOVA contd. T-TEST  Its a special case of ANOVA  used to test whether there are significant difference between two means derived from two sample groups at a specified probability level e.g. researcher comparing 1 Q performance from rural and urban children. He may get 30 subjects from each, ister the 1 Q test and get the mean of each.

ANOVA contd. Types of T-Test T - Test for independent samples • In randomly formed samples where of one group are not related to of the other group in any systematic way, other than being in same population, It is assumed that the two groups are same at beginning of the study – any difference thereafter is due to the treatment istered to the groups. T- Test for non independent samples • non independent samples are those formed by some type of matching Tests of significance • Used to decide whether the results are statistically significant. It helps the researcher to decide whether they can reject the null hypothesis or not. Significant level normally 0,05, 0.01 and 0.001.

Type I and type II errors • Type I error refers to a situation where the null hypothesis is rejected when it is true. • Type II error – where the null hypothesis is not rejected when it is false.

THE RESEARCH REPORT An overview. A proposal is a document which details an intended activity. It demonstrates the researchers ability to think clearly about the intended research or proposal.

Research proposal. 

In a research proposal a researcher proposes to undertake a piece of research on a pertinent issue. A research proposal may be academic research leading to MBA .Other research proposal may not be academicoriented but are still aimed at undertaking research to

solve a problem,i.e action oriented research

.

Project proposal. 

This refers to a proposal which is not research –oriented and aimed at undertaking certain activities to solve a specific problem. However ,it is common to find a project proposal with a research component especially when baseline and evaluation data are required.



The differences between a project proposal and research proposal focus on expected outcome and an implementation process.



In a research project the outcome will be data which can be used as a basis for recommendations for action.



In a project proposal ,proposed strategies and activities are implemented and eventual evaluation will be conducted to ensure that the project objectives were met.

Components of a research proposal. An overview.

The Mandatory main sections of a proposal are, 1.An introduction section which includes the background to the problem ,the statement of the problem, the purpose ,objectives and justification of the study. 2.A literature review section where the researcher reviews literature related to the topic under investigation. 3.A methodology section which includes the description of the research design ,population ,sample and sampling techniques ,data collection procedures and method of analysis. 4.A time schedule in which the researcher explains the time period needed to carry out the research. 5.A budget. Non Mandatory section in the introduction include–limitations ,assumptions, hypothesis and theoretical or conceptual frameworks.

Components. Abstract. • • •

A proposal abstract is a one paragraph summary of what the researcher intends to do. It should be brief ,precise and to the point. An abstract gives the reader an overview of what the problem is and how the researcher intends to solve the problem through the proposed research.

Introduction. 

The introduction tells us a lot about what is to follow in the rest of the proposal.



It includes the background of the problem, the statement of the problem, the purpose of study, the objectives or research questions ,hypotheses, the rationale or justification of the study, the limitations and underlying assumptions ,the theoretical framework or conceptual model and the theoretical definitions of .

The background to the problem The researcher should broadly introduce the topic under investigation. This is where the global,regional,and national overview of the research topic is briefly discussed. For example if the topic is female education in Kenya the research should give an overview of the status of female education globally ,regionally – Africa and nationally –the Kenyan experience. This enables the reader to have an idea of what is happening regarding the area

The problem statement. In this section the researcher should narrow down the focus and state the problem under investigation  





The researcher states the problem under investigation e.g. High drop out rates among female pupils in primary schools. In addition the researcher describes factors that make the stated problem a critical issue to warrant the study-the researcher makes a case for the research. In the above example the researcher could elaborate on the consequences of girls dropping out of school, for example marginalization in the formal sector, a high fertility rate, child labor ,poor family nutritional status and a high child mortality The problem statement should be brief and precise and range from half paragraph to two pages.

The purpose of the study. 

A purpose is a broad statement indicating what the researcher intends to do about the problem.



Two formats; 1, One is to state the purpose of the study at the end of the problem statement section. In this case the researcher writes:……..based on the problem stated, the purpose of this study to…..in this format ,the researcher states the problem ,and narrows down to the purpose statement towards the end of the section. The purpose statement is declarative and uses verbs to describe the intended task.

2,The second format ,the researcher could decide to have a separate sub section on purpose statement. The purpose is stated in a declaratory not exceeding a paragraph. It is in order to follow the purpose statement with a brief rationale of the stated purpose.

The objectives of the study. The objectives should be stated clearly and should be testable.  The difference between the purpose statement and the objectives is that the purpose statement is broad, while the objectives are derived from the purpose and are more specific.  Objectives are very crucial because , 1,Objectives determine the kind of questions to be asked .The questions should address the objectives stated. 2,Objectives determine the data collection and analysis procedures to be used. 

Research questions. Refers to questions which a researcher would like answered by undertaking the study.  The difference between research questions and objectives is that a research Question is stated in a quiz form while an objective is a statement.  If the research quiz's and objectives are referring to the same phenomenon ,then one set becomes redundant-0nly one set is included in the set. 

Hypothesis. A Hypothesis may be directional (if positive or negative relationship is suggested) or nondirectional (if no direction is suggested).  A researcher is only able to state a directional hypothesis when some information on the phenomenon under study is available. Such information is derived from past experience ,literature review or existing theories. 

Theoretical background or framework. In many fields, theories and propositions about concepts and relationships have formulated. The researcher may be interested in ascertaining or testing a particular theory or framework. Such a theoretical framework should be Cleary explained in the proposal. The researcher must then show how the study in question is related to the theoretical background.

Conceptual Framework.  Refers to when a researcher conceptualizes the relationship between variables in the study and shows the relationship graphically or diagrammatically. It is a hypothesized model identifying the concepts under study and their relationship. The purpose of a conceptual model is to help the reader to quickly see the proposed relationships. Rationale or Justification and significance of the study. • This section highlights the reasons for conducting the study as well as the importance of carrying it out. This section will often address questions like;  What gaps in knowledge will the study address ?  Why is the study important ?

The section on significance address questions like;  How will the results be used ?  Who will benefit from the results ?  What is the expected end product ?  Is the study worth it?  The rationale or justification and significance must be strong enough to warrant the use of time ,energy and money in carrying out the research.

Limitations and Assumptions.  A Limitation . Is some aspect of the study that the researcher knows may negatively affect the results or generalizability of the study but over which he/she probably has no control. In other words most common limitations have to do with the sample sample size, length of the study, or data collection procedures . Assumptions of the study.  An assumptions is any important fact presumed to be true but not actually verified .

Data collection procedures.  This section describes data collection procedures. The data collection instrument is indentified,defined and its relevance discussed. Information about the organization of the instrument can also be discussed. Data collection procedures.  This section describes data collection procedures. The data collection instrument is indentified,defined and its relevance discussed. Information about the organization of the instrument can also be discussed.

Literature review .  It enables the researcher to know what has been happening in that particular field of study. Makes one aware of what achievements have been made and what challenges remain.  Gives good suggestions on the variables and procedures that could be used.  The literature review should lead logically to objectives and hypothesis of the study.  An analysis of the literature review must be done to reveal the gaps that need the attention and also to show how the existing literature ties in with objectives of the proposed study,

Methodology.  The methodology component of a proposal mainly includes the proposed research design ,population and sample, data collection procedures, data analysis procedures and sometimes measurements of variables. Research design.  A good proposal should discuss the type of research design to be used. For example ,through survey,experimental,correlational study etc. Population and sample.  Describes the population from which the sample will be drawn. The sample size and sample selection methods are also discussed in this section.

Data analysis.  The techniques intended for use in analyzing data must be described.  The method of analysis chosen depends on the type of research, the objectives and hypotheses to be tested.  Most studies utilize selected descriptive and inferential statistics.  Descriptive statistics summarize data and describe the sample and inferential statistics enable the researcher to infer the sample results for population. The researcher should state what he or she hopes to achieve by using the data analysis technique used.

Operational definitions of variables.  The dependent and the independent variables should be defined operationally. This means that the researcher should state how the variables will be measured in a particular study. Time schedule.  This is how long the researcher will take to finish the research. A time schedule includes listing of major activities left of the page) and the corresponding anticipated time period it will take to accomplish that activity. top of the page). Time schedules in months.  Importance of a time schedule. It enables the researcher to access the feasibility of conducting a study within the existing time limits.It helps the researcher to stay on schedule as schedule as the research progress.

Budget.  List of items that will be required to carry out the research and approximate cost. Should be well thought to avoid asking for too much or too little. The budget should be detailed enough to and precise on items needed ,prices per unit and the total cost. References.  A list of cited references or bibliography. This list will show the sources of the literature reviewed. An up to date bibliography shows the researcher or donor that one is aware of new information in that particular discipline Appendix.  A short curriculum vitae or any other document ie budget notes giving more details on budget lines which may add weight to the proposal

The qualities of a good research proposal. 1.Clearly written. A vague one will discourage funders or may not be accepted by the advisors. 2.Precise.A proposal is a research plan and the researcher should stick to it during the entire study. 3.Reasonable length. As long as the key components are included and adequately explained the no of pages do not matter. should not be too wordy. 4,Worth the time and money being proposed. The rationale and the significance of the study should be carefully thought out.

A project proposal.  Proposal aimed at undertaking a certain project to solve an existing problem in a community. Project proposals are action –oriented. Preliminary activities before writing a project proposal.  Identifying source of research problems,  Sources include: Past experience,indentified needs, follow up projects National policies and strategic plans, sudden occurencies,change in behaviour,review of literature and Networking.

Writing a project proposal. Basic characteristics.  Fundable proposals-finding out whether ones ideas fits in the organizational general operating framework and objectives. 

Collaborative proposals- Bringing together experts from various disciplines thus making the proposals richer and the project more viable.



Gender sensitive proposals,-aims at addressing gender concerns in education and other social activities.

Components of a project proposal. A cover sheet.  Gives crucial and important information at a glance. A prospective funder would like to see key issues which may determine funding clearly written, A project title. A proposed grant period. The amount of funds requested. The implementing agency. The funding agency. A credibility file.  This is a brief description of the implementing agency where the past accomplishments ,available resources, records of collaborative work ,institutional , work experience are clearly explained.

An abstract.  should give a brief outline of what the proposal is all about. An abstract should state the problem,objectives,implementation strategies and expected outcome very briefly. This is the selling point of the proposal and salient issues in the proposals should be clearly written. The introduction/background.  Gives a background of the proposed project and starts building a case for the problem statement. The problem statement.  Articulates the problem being resolved and provides evidence of the existence of the problem-specifies the problem. A problem statement should give the overall goal of the project and demonstrate how the proposed project will help in resolving the stated problem

The rationale /justification of the project.  Need and the usefulness or benefits of the proposed project. Objectives.  The acronym SMART is a quick way of ing the qualities of a good objective. Strategies and activities.  A Strategy refers to the approach or mechanism of implementing the proposed project, while activities are the specific tasks to be carried out. Strategies and activities emanate from objectives and the proposal should clearly show the relationship btn objectives, strategies and activities. An example of such relationship is shown below.

Objective.

Strategy.

Activities.

To increase the participation and performance of girls in science and mathematics by changing the girls attitudes.

Use role models to change the girls attitude towards mathematics and science

identify role models from uni.and other inst to talk to and encourage the girls.

Expected outcomes.  The expected outcome must be related to the objectives. Monitoring and evaluation.  Monitoring refers to the continuous assessment process in the implementation of the project, and is arrived at by continuously measuring discrepancies between the objectives, expected outcomes and what has been accomplished. This allows early revision of project objectives.  Evaluation. is amid term or final assessment of the project Important for assessing original objectives against outcomes. Internal evaluation or external evaluation.

Time frame,calender of activities, time schedule.  They suggest detailed list of activities necessary to implement the project.

Budget. budgetary notes clarifying more information about a certain budget line. Sustainability. Measures to sustain the project either through the community or other initiatives should be clearly spelled out. Institutional . This is the coming from the implementing institution i.e. premises, time etc.

References.  Its important to always acknowledge authors that are refereed to in the project proposal. Logical framework.  This shows the relationship between various components of a proposal referred to as logframe. A logframe is a table showing either all or some of the following components:objectives,strategies,activities,expected outcomes, indicators of outcomes, budget and critical assumptions. The following is an example of a logframe layout.

Logical framework Stra. Acti. Appro task ach.

Expe. Indicat Time Budge Critica outco or of frame. t line l me. outco assum me. ptions.