Research Methodology and Research Design

For Pre PhD Workshop

R h M th d l

For Pre-PhD Workshop

Research Methodology

Kalyan Acharjya

Assistant Professor, Dept. of ECE

Jaipur National University, Jaipur

With due respect, few contents within this presentation arecopied or prepared from the help of internet, book, article, classnotes etc without prior permission from original owner.

O l O i i l O h f ll i h d f dOnly Original Owner has full rights reserved for usedimage stickers and used contents.

This PPT is only for fair academic use.

Kalyan Acharjya

12/9/2015© Kalyan Acharjya, Jaipur National University, India2

OutlinesWh t i R h?What is Research?Research Objective.Scope of Research.

Unit IResearch Process.Research Types.Research Limitations.

Unit-I

esea c tat o s.

Sampling Design Techniques.D t C ll tiData Collection.

Primary DataSecondary Data

D i i Q i i

Unit-II

Designing Questionnaire.Data Preparation.


Unit-I

About Research !


What is Research?“The definition of research includes any gathering of data, informationand facts for the advancement of knowledge”- Martyn Shuttleworth

"Research is a process of steps used to collect and analyze informationto increase our understanding of a topic or issue” -Creswellto increase our understanding of a topic or issue . -Creswell

“A studious inquiry or examination; especially: investigation orq y ; p y gexperimentation aimed at the discovery and interpretation of facts,revision of accepted theories or laws in the light of new facts, orpractical application of such new or revised theories or laws”practical application of such new or revised theories or laws

- Merriam-Webster Online Dictionary.


Research ObjectivesIt should reflect the aspirations and expectations of the researchtopic.To gain familiarity with a phenomenon or to achieve new insightsinto it.To portray accurately the characteristics of a particular individualTo portray accurately the characteristics of a particular individual,situation or groups.Hypothesis testing for the casual relationship between variables.Emphasize how aims are to be accomplished and must be highlyfocused and feasible.Obj ti h ld b dd th i di t j tObjectives should be address the more immediate projectoutcomes and make accurate use of concepts which to be sensibleand precisely described.


Why People does Research?

The motivation behind research may have followingreasons-reasons-

Interested to face the challenge in solving the unsolvedproblems.problems.To get intellectual joy of doing some creative work.Desire to be service to society.yCuriosity about new phenomenon or thing.Concerned about social thinking.gDesire to get a research degree along with itsconsequential benefits etc.


Scope of ResearchpFor Ph.D. students, the contribution expected at world level-e.g.g

Background investigation on all past work.Make meaningful addition to world knowledge.g gExpect new innovation and findings.


Research ProcessLiterature

Review

Problem

Formulation

Publish

Findings

Research

Design

Interpret

Findings Design

Data Collection

Findings

12/9/2015© Kalyan Acharjya, Jaipur National University, India

Collection

9

Research Process..contd

Initial idea or Concept.

Background investigation and review related work.

Refinement of idea.

Core Work-Investigation and Development.Documentation.Prototype (if applicable)Prototype (if applicable).

Evaluation.

Identification of Future WorkIdentification of Future Work.

Report, Publication and Presentation.


Research TypesD i ti A l ti l I D i ti h i l dDescriptive vs. Analytical: In Descriptive research includessurveys and fact findings enquiries of different kinds.In Analytical Research, the researcher has to use or analyze facts orinformation already available.

Applied vs Fundamental: In Applied (Action) Research aims toApplied vs. Fundamental: In Applied (Action) Research aims tofindings to a solution of immediate problem.On the other hand in Fundamental (pure) Research mainly

d b t f l ti f b i f tconcerned about formulation of basics facts.

Quantitative vs. Qualitative: Quantitative Research is based onQ Q Qthe measurements of quantity or amount.

Quantitative Research is relate with quality or kind.


Quantitative vs. Qualitative ResearchQuantitative Research Strategy

• Investigation aims to assess a pre-

Qualitative Research Strategy

•Investigation aims to create ag pstated theory (Deductive).

• Often involves hypothesis testing.

Investigation aims to create anovel theory (Inductive Reasoning)

•Ethnography: Researcher becomes• Attempts to minimise the

influence of the researcher on theoutcome

Ethnography: Researcher becomesan inherent part of the study.

•Qualitative data infers complexoutcome.

• Quantitative data infers statistics.

D ll i i l d

•Qualitative data infers complexstatements or opinions.

D t ll ti th f it• Data collection requires closedresponses.

•Data collection therefore permitsopen responses


Research Types..contd

Conceptual vs. Empirical: Conceptual Research is related withsome idea or theory based on philosophy or thinking to develop newconcepts or reinterpret the existing oneconcepts or reinterpret the existing one.

In Empirical Research based on experience and observation alonewithout due regard for system and theory.

Some Other Types of Research:OneTime Research vs. longitudinal research.Field Setting vs. Laboratory Research(Si l ti R h i L b t R h)(Simulation Research is Laboratory Research)Conclusion Oriented or Decision oriented.


Dimensions of Research

Theories

Deductive ReasoningConcepts

Research

Scientific M h d Inductive Method Inductive

ReasoningPropositions

EmpiricismHypotheses


Scientific Method in ResearchThe characteristics of the scientific method may havefollowing characteristics elements:g• Empirical approach.

• Observations.

• Questions.

• Hypotheses.yp

• Experiments.

• Analyses.y

• Conclusions.

• Replication- Repeating Same Work to re-check result.


p p g

15

Research Limitations

Data is sometimes not available or accessing data is very difficult.

Some case difficult to understand context of a phenomenon, difficult toanalyze and don’t fit neatly in standard categories.

Data may not be robust enough to explain complex issues.

Fi di ll t b li d t th t d l tiFindings usually cannot be generalized to the study population orobservation categories.

Some case shortage of manpower and finance is also a major issue, ifg p j ,researcher project is not financial granted by some organization.

In few case, the data collection is so rigorous and time consuming.


S li D i T h i

Unit-II-Part-I

Sampling Design Techniques


What is Sampling and Why?The process of selecting a number of individuals for a study insuch a way that the individuals represent the larger group fromwhich they were selectedwhich they were selected.

The process of obtaining information from a subset (sample) of alarger group (population).g g p (p p )

To gather data about the population in order to make anconclusion that can be generalized to the population.

Sampling can save time and money.


Population Vs. SamplePopulation Vs. Sample

Population of Interest

Population Sample

Sample

Parameter Statistic

p

•We measure the sample using statistics in order to drawp ginferences about the population and its parameters.


Sampling Terminologies•Population: Data set of interest from the researcher needs to obtain

information. One unit from a population called Element.

•Sampling Frame: Listing of population from which a sample is chosen.

•Census and Survey: A polling of the entire population, but surveyy p g p p y

means a polling of the sample. Parameter:The variable of interest.

•Statistics: The information obtained from the sample about theStatistics: The information obtained from the sample about the

parameter.

i i l i h l h f h•Critical Assumption : The sample chosen is representative of the

population.


Sampling Design ProcessDefine Population

Determine Sampling Frame

D t i S li P dDetermine Sampling Procedure

Probability Sampling: •Simple Random SamplingProbability Sampling: •Simple Random Sampling

Non-Probability Sampling•Convenience Sampling•Simple Random Sampling

•Stratified Sampling•Systematic Sampling•Cluster Sampling

•Simple Random Sampling•Stratified Sampling•Systematic Sampling•Cluster Sampling

•Convenience Sampling•Judgmental Sampling•Snowball Sampling•Quota Sampling

Determine Appropriate Sample Size

Execute Sampling Design


Goal of Sampling Process and IssuesTo be able to make inferences about the populationparameter from knowledge of the relevant statistic to draw

l l b h lgeneral conclusions about the entire population.

T o Important factors during Sampling DesignTwo Important factors during Sampling Design–Select the right data: Have to be selected scientifically so thatdata representative of the entire population.Select the adequate number of the right data: To minimizesampling errors that means choosing the wrong data by chance.

Sampling Error: Frame Error + Chance Error + ResponseError.


Sampling ProcedureP b bili li E l h f b i i l d d i• Probability sampling : Equal chance of being included inthe sample (random).

Simple random samplingSimple random sampling

Systematic sampling

Stratified samplingp g

Cluster sampling

• Non-probability sampling: Unequal chance of beingincluded in the sample (non-random).

Convenience sampling

J d liJudgment sampling

Snowball sampling

Quota samplingQuota sampling


Simple Random Sampling• An objective procedure in which the probability of selection is non zero

and is known in advance for each population unit.

E i f i i b i d f i l• Ensures information is obtained from a representative sample.

• Population members are selected directly from the sampling frame.

• Equal probability of selection for every sample (sample/population size)• Equal probability of selection for every sample (sample/population size)

• Normally Use random number table or random number generator.


Systematic Sampling• Systematic sampling involves a random start and then proceeds

with the selection of every kth element from then onwardswith the selection of every kth element from then onwards.

Here k=(population size/sample size).

• It i i t t th t th t ti i t i t t ti ll th fi t i• It is important that the starting point is not automatically the first in

the list, but is instead randomly chosen from within the first to the kth

element in the list.

• Order all units in the sampling frame based on some variable and

number them from 1 to N


Stratified Sampling The chosen sample is forced to contain units from each of thesegments or strata of the population. Its Equalizing "important"

i blvariables.Population is divided into mutually exclusive andexhaustive strata based on an appropriate populationpp p p pcharacteristic. (e.g. time, SNR, Frequencies etc.)

• Minimize the sampling error since a source of variation is eliminated.


Cluster Samplingh h l i f d i i f h f hThe chosen sample is forced to contain units from each of the

segments or strata of the population. Its Equalizing "important"variables.Population is divided into mutually exclusive and exhaustivestrata based on an appropriate population characteristic.(e.g. power level, SNR, Frequencies etc.)( g p , , q )

Divide the cluster based on parameters, whose information soughtforfor.


Sampling Design..choose smart way for best result and minimize cost.

• When to use Stratified Sampling-If primary research objective is to compare of variousp y j pgroups.Using stratified sampling may reduce sampling errors.

• When to use Cluster Sampling-If there are substantial fixed costs associated with eachdata collection.When there is a list of clusters but not of individualpopulation members.


Non-Probability Sampling• Subjective procedure in which the probability of selection for some

population units are zero or unknown before drawing the sample.I f ti i bt i d f t ti l f th• Information is obtained from a non-representative sample of thepopulation.

• Sampling error can not be computed.g• Survey results cannot be projected to the population.• Advantages:

h b b l• Cheaper and faster than probability.• Reasonably representative if collection in proper way.


Types of Non-Probability SamplingConvenience Sampling: The process of including whoever happensto be available at the time, also called “accidental” or “haphazard”sampling.

Judgment Sampling: A researcher exerts some effort in selecting asample that seems to be most appropriate for the study.p pp p y

Snowball Sampling: Selection of additional respondents is based onreferrals from the initial respondents and used to sample from lowp pincidence or rare populations.

Quota Sampling: The population is divided into cells on the basis ofQuota Sampling: The population is divided into cells on the basis ofrelevant control characteristics.

A quota of sample units is established for each cell.A i l i d f h ll til th t i tA convenience sample is drawn for each cell until the quota is met.


Probability Vs. Non-Probability SamplingNon-probability sampling is less time consuming and lessexpensive.

The probability of selecting one element over another is notknown and therefore the estimates cannot be projected to theknown and therefore the estimates cannot be projected to thepopulation with any specified level of confidence.Quantitative generalizations about population can

l b d d b bili lionly be done under probability sampling.

H k ti h l l t ti ti t t dHowever, marketing researchers also apply statistics to studynon-probability samples.


Sampling Errors R d E Th l l d i i f h• Random Error : The sample selected is not representative of thepopulation due to chance.

• The level of it is controlled by sample sizeThe level of it is controlled by sample size.

• A larger sample size leads to a smaller sampling error.

Non Sampling ErrorsNon-Sampling Errors • A response or data error is any systematic bias that occurs during datacollection or in data analysis or interpretation is considered a non samplingerror. e.g.

• Respondent error.

• I t i bi• Interviewer bias.

• Recording errors.

• Poorly designed questionnaires• Poorly designed questionnaires.


Unit-II-Part-II

Data Collection !


Data For Research

Data Collection begins after the Research Problem Defined andResearch Design carried out.g

There are two types of Data-

Primary Data: Original Data which are taken for the first time.

Secondary Data: Those are already been collected by someone elseand which have already been passed through the statistical process.

The procedure for data collection for primar and secondar data areThe procedure for data collection for primary and secondary data aredifferent.

Type Research as per Data Collection-Experimental Research-yp p pperforming experiment for data collection and Descriptive TypeResearch doing surveys and others.


Primary Data CollectionThere are several methods for collection the primary data-(For Surveys and descriptive research)

Observation Method: Specially used for studies relating to behavioral sciences.Advantage is subject bias is eliminated.This method is independent of Respondents.p pLimitation: It is very expensive and some times unforeseen factors may interfere withthe observation leads to error.

Interview Method: Collection of data involves presentation of oral verbal responses.Personal Interview: Personal interview is usually carried out in a structured way.There should be clear formulation for pre-requisites and basic tenets of p qinterviewing.Telephonic Interview: More FlexibleThis method also have some demerits.


Primary Data Collection..contd

Through Questionnaire: A questionnaire consists of numbers of question printed or typed on a forms.Also collection the data by mailing the questionnaire to respondents.

h l l h d l d dThere is low cost to implement the process and easily send to any respondents acrossthe globe.Respondents have adequate time to give well thought out answers.Limitation: Slow process, low rate of return of duly filled of questionnaires and it isp , y qonly used if respondents are educated and co-operating.

Through Schedules: Similar as Collection of data in questionnaire, but the differences is schedulesb i fill d b h i ll i d f hibeing filled by enumerators who are specially appointed for this purpose.Schedules are Performa containing a set of questionnaire.Enumerators should be intelligent and must posses the capacity of crossexamination in order to find the truth.Useful for extensive enquiries and fairly reliable results.Population census over the world is conducted through this method.


Secondary Data CollectionSecondary data means data are already available.

Those data which have already been collected and analyzed.

Various Publication across the globe

Technical Journals

Books, magazine and News papers

Reports of various association and organization

Theses and dissertation for scholarsTheses and dissertation for scholars

Public records and statistics.


Secondary Data CollectionResearcher must careful before using secondary data.

Researcher must make a scrutiny because it is possible that the secondaryd i bl i d i h l d hi h hdata may unsuitable or inadequate in the context related which theresearcher wants to study.

According to Dr. A.L Bowley it is never safe to take published statistics attheir face value without knowing the meaning and limitations.

Before using the Secondary Data, it must posses the followingcharacteristics-characteristics

Reliability of Data:Tested by findings out such things about the said data.

Suitability of Data: The object, scope and nature of the original enquiry must bestudied before using those secondary datastudied before using those secondary data.

Data Adequacy : Is the level of accuracy is consider for the present findings.


The Appropriate Methods for Data CollectionBased on following factors-

Nature, scope and Enquiry: The method will be such that it suits forrequirements of researcherrequirements of researcher.Availability of Funds: It determine to how large extents the methodto be used for the data collection. For those developing countriesthe finance is big constraint and researcher has to act within itslimitations.Timing Factor: The time requirements and availability alsoTiming Factor: The time requirements and availability alsodetermine the specific data selection.Precision Required: Precision is yet another important factor to beconsidered before selection the method of data collection.


Designing QuestionnaireResearcher must clear about the various aspects of his/her researchproblem to be dealt.Th tt f Q ti i d d th t f i f tiThe pattern of Questionnaire depends on the nature of informationsought.The type of question should be closed or open up, decided based onresult expected for.Questions should be simple and must be constructed with a view totheir forming a logical part of a well thought out tabulations plan.g g p g pBefore finalizing the questions, rough draft should be prepared.Pilot study should be undertaken for pre-testing the questionnaire.Questionnaire must contain simple but straight forward directionsfor the respondents.


Unit-II-Part-II

Data Preparations


Data PreparationData Preparation involves checking or logging the data in;checking the data for accuracy; entering or tabulation the data intothe computer; transforming the data; and developing andthe computer; transforming the data; and developing anddocumenting a database structure that integrates the various measures.

Why Data Preparations-Adequacy Check.A h l ibl / d bl ?Are the responses legible/readable?Are all important questions answered?Are the responses complete and acceptable?p p pIs all relevant contextual information included to my intended findings?


Process of Data PreparationsProcess of Data Preparations

Step-1Editing

Step-2Coding

Step-3pTabulation


EditingTh f h ki d dj ti i th l t dThe process of checking and adjusting responses in the completedquestionnaires or observation for omissions, legibility, and consistencyand readying them for coding and storage.

Purpose:For consistency among responses or observation.For completeness in responses to reduce effects of non-response items.p p pTo order the observation or answered and facilitate the coding process.

Types:Fi ld Edi i P li i di i b fi ld i hField Editing: Preliminary editing by a field supervisor on the sameday, check legibility of handwriting, and clarify responses that arelogically or conceptually inconsistent.In-house Editing: Editing performed by a head office; often domemore rigorously than field editing


CodingThe process of identifying and classifying each answered or observationwith a numerical score or other character symbol.

Th i l b l i ll d d d lThe numerical score or symbol is called a code, and serves as a rulefor interpreting, classifying, and recording data

Id tif i ith d i if d t i t bIdentifying responses with codes is necessary if data is to beprocessed by computer.

Coded data is often stored electronically in the form of a data matrix aCoded data is often stored electronically in the form of a data matrix - arectangular arrangement of the data into rows (representing cases) andcolumns (representing variables)

Field: A collection of characters that represents a single type of data.

Record: A collection of related fields i.e., fields related to the same case.

File: A collection of related records i e records related to the same sampleFile: A collection of related records, i.e. records related to the same sample.


TabulationTabulation is the final step of Data Preparation.

It normally counting the number of observation and responses iny g pvarious data categories.

For small size, manual tabulation certainly be the best.

Cross tabulation also a important process for finalizing tabulationprocess.

D T f i b h f b l iDataTransformation may be the part of tabulation.Data Transformation: Converting some of the data from the format inwhich they were entered to a format most suitable for particularw c t ey we e e te e to a o at ost su ta e o pa t cu astatistical analysis.


Conclusions:Research is a scientific method to used to collect and analyzeResearch is a scientific method to used to collect and analyzeinformation to increase our understanding or solve issue onparticular field.Research should reflect the aspirations and expectations of theresearch topicIt has been understood that Sampling Design and DataIt has been understood that Sampling Design and DataCollection are important part of research design.The last part of presentation covered up Data Preparation forresearch, it consists three steps-editing, coding and tabulation.

After conducting the study or gathering the data the next step involvesAfter conducting the study or gathering the data, the next step involvesAnalyzing the Data, which generally calls for the use of statisticaltechniques.The type of statistical techniques used by a researcher depends on theR h D i d T f d b i h d


Research Design and Type of data being gathered.

47

Home Assignment-I..Go through it and Submit.Sampling Distribution of Mean.Sampling Distribution

ProportionProportion.Students t distribution.F distribution.

Chi-square Distribution.Hypothesis Testing (Null Hypothesis) and Central Limit theorem.Scaling Techniques.g q

You must have understand and able to calculate the following terms-

Mean and ModeMean and ModeMedianVarianceS d d E d S d d D i i


Standard Error and Standard Deviation.

48

Home Assignment-II….Submit.1) Suppose the admission cell of Jaipur National University conducting the survey

among 500 HS students for their interest to take B.Tech admission in fourdepartments (EC,CS,EE,FT).The admission cell also prepared the expectedp ( , , , ) p p padmission in these departments as based on last year status, EC-35 %, EE-30 %, CS-20 % and FT-15 % .And survey results are as below;

Sl. No Department No. of Student

1 EC 150

2 EE 140

3 CS 120

4 FT 90

Based on Chi-square test whether you reject the null hypothesis or not? (Consider level of significance 0.05)


References

1. Text-Book: C.R. Kothari, “Research Methodology Methods andTechniques”, New Age International Publisher, Second Edition, ISBN-978-224-1522-3224 1522 3

2. Text-Book: Creswell, J. W. , Research design : Qualitative and QuantitativeApproaches. Thousand Oaks, Calif.; London : Sage Publications, ISBN0803952546, 19940803952546, 1994

3. Data Preparation, Zikmud Book, Chapter 19. Notes of Dr. Alhassan G. AbdulMuhmin, Department of Management and Marketing, King Fahd Universityof Petroleum & Mineralsof Petroleum & Minerals.

4. Case Study Notes from Deikin University.5. Research Design article from Queen University, Belfast6. Lecture notes of Dr. G. Singaravelu, Associate Professor, UGC-ASC,

Bharathiar University Coimbatore


Any Questions?y


Thank You

[email protected]


Research Methodology and Research Design

Engineering