R R E E S S E E A A R R C C H H M M E E T T H H O O D D O O L L O O G G Y Y
Students, in our introduction to the subject lecture we had Students, in our introduction to the subject lecture we had conquered the areas where research is used as a tool for conquered the areas where research is used as a tool for decision-making
making. Also we know that all business operat. Also we know that all business operat es in thes in the world of e world of uncertainty. Research methodology minimizes the degree of uncertainty. Research methodology minimizes the degree of uncertainty involved in management decisions. Research lays the uncertainty involved in management decisions. Research lays the structure for decision-making
structure for decision-making
Let us recapitulate what we have stud
Let us recapitulate what we have studied in thied in the last lecture. We sawe last lecture. We saw that research plays a dominant role in the field of
that research plays a dominant role in the field of
• • MarketingMarketing • • ProductionProduction • • BankingBanking • • MaterialsMaterials •
• Human Resource DevelopmentHuman Resource Development •
• GovernmentGovernment
You can classify Research in one of three categories: You can classify Research in one of three categories: 1.
1. ExplExploratororatory resy reseaearchrch 2.
2. DescDescriptiriptive resve reseaearchrch 3.
3. CaCausausal rel reseasearchrch
These classifications are made according to your objective of the These classifications are made according to your objective of the research. In some cases the research will fall into one of these research. In some cases the research will fall into one of these categorie
categories, but in os, but in other cases different phther cases different phases of the same researchases of the same research project will fall into different categories.
project will fall into different categories. No
Now we ww we will discuss these categories in dill discuss these categories in d etail:etail:
E x p l o r a t o r y R e s e a r c h
E x p l o r a t o r y R e s e a r c h
Exploratory research has the
Exploratory research has the goal of formulating problems moregoal of formulating problems more precisely, clarifying concepts, gathering explanations, gaining precisely, clarifying concepts, gathering explanations, gaining insight, eliminating impractical ideas, and forming hypotheses. insight, eliminating impractical ideas, and forming hypotheses. Exploratory research can be performed using a literature search, Exploratory research can be performed using a literature search, surveying certain people about their experiences, focus groups, surveying certain people about their experiences, focus groups, and case studies.
and case studies.
When you will be surveying people, exploratory research studies When you will be surveying people, exploratory research studies would not try to acquire a representative sample, but rather, seek would not try to acquire a representative sample, but rather, seek to int
to interview those who are knowledgeable and who might be ableerview those who are knowledgeable and who might be able to provide you the insight concerning the relationship among to provide you the insight concerning the relationship among variables.
variables.
Case studies can include contrasting situations or benchmarking Case studies can include contrasting situations or benchmarking against an organization known for its excellence. Exploratory against an organization known for its excellence. Exploratory research may develop hypotheses, but it does not seek to test research may develop hypotheses, but it does not seek to test them. Exploratory research is characterized by its flexibility. them. Exploratory research is characterized by its flexibility.
Descriptive Research Descriptive Research
Descriptive research is more rigid than exploratory research and Descriptive research is more rigid than exploratory research and seeks
seeks to describe users of a product, determine the propto describe users of a product, determine the prop ortion oortion of f the population that uses a product, or predict future demand for the population that uses a product, or predict future demand for a product or describes the happening of a certain phenomenon. a product or describes the happening of a certain phenomenon. As opposed to exploratory research, if you are doing descriptive As opposed to exploratory research, if you are doing descriptive research you should define questions, people surveyed, and the research you should define questions, people surveyed, and the method of analysis prior to beginning data collection.
method of analysis prior to beginning data collection. In o
In other words, tther words, the who, what, where, whe who, what, where, when, why, ahen, why, and hnd how aspectsow aspects of the research should be defined. Such preparation allows you of the research should be defined. Such preparation allows you the opportunity to make any required changes before the costly the opportunity to make any required changes before the costly process of data collection has begun.
process of data collection has begun.
There are two basic types of descriptive research: There are two basic types of descriptive research: L
Lonon gitudgitud inal Studiesinal Studies
Longitudinal studies are time series analyses that make repeated Longitudinal studies are time series analyses that make repeated measurements of the same individuals, thus allowing you to measurements of the same individuals, thus allowing you to monitor
monitor behaviour such as behaviour such as brand brand swiswitching. However, longitching. However, longitudinaltudinal studies are not necessarily representative since many people may studies are not necessarily representative since many people may refuse to participate because of the commitment required. refuse to participate because of the commitment required.
C
Cross-sectional ross-sectional StudiesStudies
Cross-sectional studies sample the population to make Cross-sectional studies sample the population to make measurements at a specific point in time. A special type of measurements at a specific point in time. A special type of cross-sectional analysis is a cohort analysis, which tracks an aggregate of sectional analysis is a cohort analysis, which tracks an aggregate of individuals who experience the same event within the same time individuals who experience the same event within the same time interval over time.
interval over time. You caYou can use Cohort analysen use Cohort analyses for long-s for long-forecasting of product demand.
forecasting of product demand. Causal Research
Causal Research Casua
Casual Reseal Research seeks trch seeks to find o find cause and affect rcause and affect relationships betweenelationships between variables. It accomplishes this goal through laboratory and field variables. It accomplishes this goal through laboratory and field experiments.
experiments.
Research process involves important Research process involves important
steps-•
• ProbProblem definilem definitiontion •
• ResResearch prearch proposaloposal •
• Research DesignResearch Design •
• Data Data CollCollectionection •
• Data AnalysiData Analysis & interprets & interpret ationation •
• Report writingReport writing •
• Interpretation of ResearchInterpretation of Research
Refer diag
Refer diagram below to unram below to un derstand each stderstand each steps clearlyeps clearly The Research Process
The Research Process
LESSON 2:
LESSON 2:
STEPS IN RESEARCH PROCESS
STEPS IN RESEARCH PROCESS
R R E E S S E E A A R R C C H H M M E E T T H H O O D D O O L L O O G G Y Y
P r o b l e m D e f i n i t i o n
P r o b l e m D e f i n i t i o n
First of all one should be clear as what exactly the problem is. First of all one should be clear as what exactly the problem is. If you are working in a
If you are working in a compancompany, y, the pthe probroblem is lem is assign by assign by the tthe topop manage
management; usually ment; usually you get broyou get bro ad ideas regarding the pad ideas regarding the problem.roblem. Then with the broad concept from the top management you Then with the broad concept from the top management you define the specific problem statement.
define the specific problem statement. Or if you are doing some research project th
Or if you are doing some research project then you have to identifyen you have to identify y
your pour problem statement oroblem statement o f your own. Remef your own. Remember your probmber your prob lemlem statement should be specific.
statement should be specific. Let us suppose th
Let us suppose th at we wat we want to knant to know wow which of thhich of th e two methodse two methods can be employed
can be employed Metho
Metho d I d I or Methor Meth od IIod II
The researcher should be clear with the alternative choices he The researcher should be clear with the alternative choices he has.S
has.Should othould ot her Methods bher Methods b e also considered?e also considered?
Let us assume that the alternative choices are clearly Let us assume that the alternative choices are clearly specifie
specified: either II d: either II or I or I be employed.Wbe employed.We have identifiede have identified the alternative choices but not completely specified the alternative choices but not completely specified the problem.
the problem. The complete prob
The complete prob lem ilem is concerned with the criterions concerned with the criterion that will determine the superiority of the two that will determine the superiority of the two methods.
methods.
The criterion could be: The criterion could be:
•
• CostCost
•
• Efficiency of materialsEfficiency of materials •
• AvaiAvailabililability of ty of resourresources, etc.ces, etc.
There are three aspects of
There are three aspects of researesearch prorch problemblem a.
a. The speciThe specificafication of units tion of units to be studieto be studiedd b.
b. The identifiThe identificatication of thon of the particue particular unilar units wits withinthin the scope of study
the scope of study
cc.. The specThe specifiificatication of the kind of informaon of the kind of information totion to be sought.
be sought.
W
What would you like to knohat would you like to kno w if infw if informatormat ion ision is free and without error?
free and without error?
A complete answer to this question defin
A complete answer to this question defines the initiales the initial research problem. It can be redefined later if some research problem. It can be redefined later if some difficulty arises.
difficulty arises.
R
Re s e a
e s e a r c
r ch
h P
Pr o p o
r o p o s a l
s a l
Research proposal are necessary for all business Research proposal are necessary for all business research, it may be the internal proposal or it may be research, it may be the internal proposal or it may be the external proposal. But research proposal is not the external proposal. But research proposal is not required in case of resea
required in case of research studies for Prch studies for P. hd., or . hd., or paperpaper presentation as concerned.
presentation as concerned.
A proposal is known as a work plan, prospectus, A proposal is known as a work plan, prospectus, outline, statement of intent, or draft plan. The outline, statement of intent, or draft plan. The pro
proposal tells us what, why, posal tells us what, why, how, where, and to how, where, and to whomwhom it wil
it will be dol be done.ne.
The proposal of research is: The proposal of research is: 1.
1. To present the To present the manamanagemgement quesent question to betion to be researched and its importance
researched and its importance 2.
2. To discTo discuss the uss the researesearch efforts of rch efforts of others whoothers who have worked on
have worked on related managrelated management questions.ement questions. 3.
3. To suggTo suggest the data necesest the data necessary sary for solving for solving thethe manageme
management qnt question and uestion and how thhow the data wile data will bel be ga
gathered, trthered, treated, and interpreted.eated, and interpreted. We w
We will diill discuss the scuss the research prresearch propooposal in detail in lessonsal in detail in lesson 3.
3.
Research Design Research Design
Data
Data CCollollection -Tection -Types ypes and and Sources:Sources:
On
Once the ce the researcher has decided tresearcher has decided the ‘Rhe ‘Research Design’esearch Design’,, the next job i
the next job is of data s of data collecollection. ction. For data to beFor data to be useful, our observations need to be organized so useful, our observations need to be organized so
R R E E S S E E A A R R C C H H M M E E T T H H O O D D O O L L O O G G Y Y
that we can get some patterns and come to logical conclusions. that we can get some patterns and come to logical conclusions. Statistical investigation requires systematic collection of data, so Statistical investigation requires systematic collection of data, so that all relevant groups are represented in the data.
that all relevant groups are represented in the data. To determine th
To determine the potente potential ial market for a new product, for example,market for a new product, for example, the r
the researcher migesearcher might ht study 500 consumers in a certain geographicalstudy 500 consumers in a certain geographical area
area. . It must be ascIt must be ascertained ertained that the group contains peoplthat the group contains peoplee representing variables such as income level, race, education and representing variables such as income level, race, education and neighborh
neighborh oodood . . The quality The quality of data wof data will ill greatly greatly affect theaffect the conclusions and hence, utmost import
conclusions and hence, utmost import ance must be givance must be given to en to thisthis process and every possibl
process and every possible precaution should be precaution should b e taken to e taken to ensureensure accuracy, while gathering and collecting data.
accuracy, while gathering and collecting data. Depending upon
Depending upon the sources utilized, the sources utilized, whether the data has comewhether the data has come from actual observ
from actual observations or from ations or from records that are kept for records that are kept for normalnormal purposes, statistical data can be classified into two categories, purposes, statistical data can be classified into two categories, primary and secondary.
primary and secondary.
Pri
Primama ry Datary Data
Primary data is one, which is collected by th
Primary data is one, which is collected by th e investigae investigator tor himself himself for t
for the purphe purpose of a specific iose of a specific inquiry or study. nquiry or study. Such data is origSuch data is originalinal in character and
in character and is generated by surveyis generated by surveys conducts conducted by individualed by individualss or r
or reseaesearch institutions.rch institutions.
Some common types of primary data are: Some common types of primary data are:
•
• Demographic and socioeconomic characteristicsDemographic and socioeconomic characteristics •
• Psychological and lifestyle characteristicsPsychological and lifestyle characteristics •
• Attitudes and opinionsAttitudes and opinions •
• Awareness and knowledge:for example, brand awarenessAwareness and knowledge:for example, brand awareness •
• IntIntentions:for example, purchase intentions. While usefulentions:for example, purchase intentions. While useful,,
intentions are not a reliable indication of actual future intentions are not a reliable indication of actual future behavior.
behavior.
•
• MotivMotivation:a person’ation:a person’s motives are more stable than hs motives are more stable than h is/is/ herher
behavi
behavior, so or, so motive is a better predictor omotive is a better predictor o f future bf future behaviehavioror than is past behavior.
than is past behavior. Behaviour
Behaviour Primary da
Primary data can be ta can be obtained bobtained by:y: Communication
Communication
Involves questioning respondents either verbally or in writing. Involves questioning respondents either verbally or in writing. This method is versatile, since you need only to ask for the This method is versatile, since you need only to ask for the information; however, the response may not be accurate. information; however, the response may not be accurate. Communication usually is quicker and cheaper than observation. Communication usually is quicker and cheaper than observation. Observation
Observation
Involves the recording of actions and is performed by either a Involves the recording of actions and is performed by either a person or some mechanical or electronic device. Observation is person or some mechanical or electronic device. Observation is less ve
less versatilrsatile than e than communicaticommunication since some attron since some attributes of a pibutes of a personerson may not be readily observable, such as attitudes, awareness, may not be readily observable, such as attitudes, awareness, knowledge,
knowledge, intentintentions, and motivation. Oions, and motivation. O bservation also mightbservation also might take longer since observers may
take longer since observers may have to have to wawait for it for appropriate eventsappropriate events to
to occur, though ooccur, though o bservatibservation on using scusing scanner data might be quickeranner data might be quicker and more cost effective. Observation typically is more accurate and more cost effective. Observation typically is more accurate than communication.
than communication. P
Personaersona l Il Intente rvirviewew ss
Have an interviewer bias that mail-in questionnaires do not have. Have an interviewer bias that mail-in questionnaires do not have. For example, in a personal interview the respondent’s perception For example, in a personal interview the respondent’s perception of the interviewer may affect the responses
of the interviewer may affect the responses
Questionnaire Questionnaire The question
The questionnaire is naire is an importanan important toot tool for gal for gatherthering primary ing primary data.data. Poo
Poorly rly constconstructed questructed questions can result in lions can result in large earge errorrors and invalirs and invalidatedate the research data, so significant effort should be put into the the research data, so significant effort should be put into the Questionnaire
Questionnaire S
Secondecond ary Dataary Data
When an investigator uses th
When an investigator uses the data, which has already e data, which has already been collectedbeen collected by others, such data is call
by others, such data is called secondary data. ed secondary data. This data is primaryThis data is primary data for t
data for the agency he agency that cothat collellects it and bcts it and becomes secondary daecomes secondary data forta for someone else
someone else who uses who uses this data for his this data for his own purposes. own purposes. TheThe secondary da
secondary data can be ta can be obtained from obtained from journals, reports, governmentjournals, reports, government publications, publica
publications, publication otion of prof professional afessional and research ond research organizarganizationtion and so on.
and so on. For example, iFor example, if a f a researcher desires to analyresearcher desires to analyze theze the weather conditions of different regions, he can get the required weather conditions of different regions, he can get the required information or data from the records of the meteorology information or data from the records of the meteorology department.
department.
There are several criteria that you should
There are several criteria that you should use to use to evalevaluate secondaryuate secondary data.
data.
•
• Whether the data is useful in the research study.Whether the data is useful in the research study. •
• How current tHow current t he data is and whether it appliehe data is and whether it applies to ts to t ime periime periodod
of interest. of interest.
•
• Errors and accuracy - whether the data is dependable and canErrors and accuracy - whether the data is dependable and can
be
be verifiverified.ed.
•
• Presence of bias in the data.Presence of bias in the data. •
• Specifications and methodologies used, including dataSpecifications and methodologies used, including data
collection method, response rate, quality and analysis of the collection method, response rate, quality and analysis of the data, sample size and sampling technique, and questionnaire data, sample size and sampling technique, and questionnaire design.
design.
•
• Objective of the original data collection.Objective of the original data collection. •
• Nature Nature of tof the data, including defihe data, including definition of nition of variavariables, unitsbles, units
of measure, categories used, and relationships examined. of measure, categories used, and relationships examined.
D
Da t
a t a
a C
Co l l
o l le c t i o n P r o c e d u
e c t i o n P r o c e d u r e f o
r e f o r P r i
r P r im
m a r y
a r y
D a t a
D a t a
P
Pllannann iing ng the the SStudtud yy
Since the quality of results gained from statistical data depends Since the quality of results gained from statistical data depends upon the quality of information collected, it is important that a upon the quality of information collected, it is important that a sound investig
sound investigativative process be established to ensure te process be established to ensure t hat thhat the datae data is hig
is highly hly represenrepresentative and unbiased. tative and unbiased. This requires a higThis requires a high degreeh degree of skill and also certain precautionary measures may have to be of skill and also certain precautionary measures may have to be taken.
taken. Mode
Mode s of Data s of Data CCollollectionection Following are w
Following are widely idely used methused methods foods for collection or collection of primary data:f primary data:
• • ObservationObservation • • ExperimentationExperimentation • • QuestionnaireQuestionnaire • • InterviewingInterviewing •
• Case Study MethodCase Study Method
Observation Process Observation Process Observing the pro
Observing the pro cess at wcess at work collects iork collects information. nformation. The followiThe followingng are a few examples.
are a few examples. i.
i. ServService ice Stations:Stations:Pose as a customer, go to a service stationPose as a customer, go to a service station and observe
R R E E S S E E A A R R C C H H M M E E T T H H O O D D O O L L O O G G Y Y ii
ii. To . To evalevaluate the uate the effectiveffectiveness of display of the Deness of display of the D unlopunlop cushions in a departmental store, an observer not cushions in a departmental store, an observer not es:es:
•
• How many pass by?How many pass by? •
• How many stopped to look at the display?How many stopped to look at the display? •
• How many decide to buy?How many decide to buy?
ii
iii.i. Super Super MaMarket Wrket What is hat is the best locatithe best location in the shelf?on in the shelf? HiddenHidden cameras are used.
cameras are used.
iiv.v. Concealed Concealed tape recorder with the investigtape recorder with the investigator helps toator helps to determine typical sales arguments and find out sales determine typical sales arguments and find out sales enthusiasm shown b
enthusiasm shown b y variy various salesmen. By ous salesmen. By this meththis method,od, response bias is eliminated.
response bias is eliminated.
The method can be used to study sales techniques, customer The method can be used to study sales techniques, customer movements, customer response, etc.
movements, customer response, etc. However, However, the customer’s/ the customer’s/ consumer’
consumer’s state s state of mind, of mind, their btheir buyiuying motives, their images areng motives, their images are not reveale
not revealed. d. Their income Their income and educatiand education is alon is also not known. so not known. ItIt also takes time for the investigator to wait for particular sections also takes time for the investigator to wait for particular sections to take place.
to take place.
E
Ex p e r i
x p e r im
m e n t
e n t a l M e t h o
a l M e t h o d :
d :
Many of the important decisions facing the marketing executives Many of the important decisions facing the marketing executives cannot b
cannot be settled by secondary reseae settled by secondary research, obrch, observation or bservation or by surveyy surveyinging the opinions of customers or experts.
the opinions of customers or experts.
Experimental method may be used in the following situations. Experimental method may be used in the following situations. i.
i. WWhat is hat is the best methe best method for trainithod for training sang saleslesmen?men? ii
ii.. What What is is the best remuneration plan the best remuneration plan for salefor salesmen?smen? ii
iii.i. What is the best shelf arrangement for displayWhat is the best shelf arrangement for displaying a product?ing a product? iiv.v. What is the effectiveWhat is the effectiveness of a pointness of a point-of-purch-of-purchase displaase display?y? v
v.. WWhat packahat package desige design should be used?gn should be used? vi.
vi. WhiWhich copy is the most effective?ch copy is the most effective? vii
vii.W.What media are that media are t he most he most effective?effective?
viii.Which version of a product would consumers like best? viii.Which version of a product would consumers like best? In a marketing experiment, the experimental units may be In a marketing experiment, the experimental units may be consumers, stores, sal
consumers, stores, sales territories, ees territories, etc.\tc.\ Factors or marketingFactors or marketing vari
variables under the ables under the control ocontrol of thf the researcher whie researcher which can be ch can be studiedstudied are pr
are price, packagice, packaging, displaying, display, sales incentive p, sales incentive plan, flavor, color lan, flavor, color shape,shape, Competitor’s acti
Competitor’s actions, ons, weweather ather changes, in cooperative dealers, etc.changes, in cooperative dealers, etc. are environmental factors.
are environmental factors.
To study the effect of the marketing variables in the presence of To study the effect of the marketing variables in the presence of environmen
environmental factors, a sufficiently ltal factors, a sufficiently large sample should be used.arge sample should be used. Or sometimes a
Or sometimes a contrcontrol gol group is set up. roup is set up. A CA Contontrol group is rol group is aa group equival
group equivalent tent t o to t he experimental group and he experimental group and differing only idiffering only inn not receiving any treatment.
not receiving any treatment. The result/
The result/ response of a marketing eresponse of a marketing experimexperiment wilnt will be in thel be in the form of sales, attitudes or behaviour.
form of sales, attitudes or behaviour. Data Analysis-Preliminary Steps Data Analysis-Preliminary Steps
•
• Before the analysis can be performed, raw data must beBefore the analysis can be performed, raw data must be
transformed into the right format. transformed into the right format.
•
• Then, it must be edited so that errors can be corrected orThen, it must be edited so that errors can be corrected or
omitted. omitted.
•
• The data must then be coded; this procedure converts thThe data must then be coded; this procedure converts th ee
edited raw data into numbers o
edited raw data into numbers o r symbols. A r symbols. A codeboocodebook isk is crea
created to ted to document how thdocument how th e data was coded.e data was coded.
•
• FinallFinally, the data is tabulay, the data is tabulated to ted to count tcount t he number he number of samplesof samples
fall
falling into ing into various categories.various categories.
Simple tabulations count the occurrences of each variable Simple tabulations count the occurrences of each variable independently of the
independently of the other other varivariableables.s. Cross tabulation
Cross tabulations, als, also known as contingency tables or cross tabs,so known as contingency tables or cross tabs, treats two or more variables simultaneously. However, since the treats two or more variables simultaneously. However, since the variables are in a two-dimensional table, cross tabbing more than variables are in a two-dimensional table, cross tabbing more than two variables is difficult to visualize since more than two two variables is difficult to visualize since more than two dimensions would be required. Cross tabulation can be perfo dimensions would be required. Cross tabulation can be perfo rmedrmed for nominal and ordinal variables. Cross tabulation is the most for nominal and ordinal variables. Cross tabulation is the most commonly utilized data analysis method in marketing research. commonly utilized data analysis method in marketing research. Ma
Many studies take the analyny studies take the analysis no fursis no further ther than than cross tabulation.cross tabulation. This technique divides the sample i
This technique divides the sample into sunto sub-groups tb-groups to show howo show how the dependent variable varies from one subgroup to another. A the dependent variable varies from one subgroup to another. A third variable can be introduced to uncover a relationship that third variable can be introduced to uncover a relationship that initially was not evident.
initially was not evident.
Conjoint Analysis Conjoint Analysis
The Conjoint Analysis is a powerful technique for determining The Conjoint Analysis is a powerful technique for determining consumer preferenc
consumer preferences for es for product attributes.product attributes. H
Hypotypot hehe sisis Testings Testing A basic fac
A basic fact about t about testing hypotheses is that testing hypotheses is that a hypothesis may bea hypothesis may be rejected but that the hypothesis never can be unconditionally rejected but that the hypothesis never can be unconditionally accepted until all possible evidence is evaluated.
accepted until all possible evidence is evaluated.
In the case of sampled data, the information set cannot be In the case of sampled data, the information set cannot be complete. So if a
complete. So if a test using such data dotest using such data does not reject a hypothes not reject a hypoth esis,esis, the conclusion is not necessarily that the hypothesis should be the conclusion is not necessarily that the hypothesis should be accepted.
accepted.
The null hypothesis in an experiment is the hypothesis that the The null hypothesis in an experiment is the hypothesis that the independ
independent variable ent variable has no effect on thhas no effect on the dependent variable. Thee dependent variable. The null hypothesis is expressed as H0. This hypoth
null hypothesis is expressed as H0. This hypoth esis is asesis is assumed tosumed to be true unless proven otherwise. The alternative to the null be true unless proven otherwise. The alternative to the null hypothesis is the hypothesis that the independent variable does hypothesis is the hypothesis that the independent variable does have an effect on t
have an effect on t he dependent he dependent variavariable. This hypothesis is knowble. This hypothesis is knownn as the alternative, research, or experimental hypothesis and is as the alternative, research, or experimental hypothesis and is expressed as H1. This alternative hypothesis states that the expressed as H1. This alternative hypothesis states that the relationship observed between the variables cannot be explained relationship observed between the variables cannot be explained by chance alone.
by chance alone.
There are two types of errors in evaluating hypotheses: There are two types of errors in evaluating hypotheses:
•
• Type I error: occurs when one rejects the null hypothesis andType I error: occurs when one rejects the null hypothesis and
accepts the alternative, when in fact the null hypothesis is accepts the alternative, when in fact the null hypothesis is true.
true.
•
• Type II error: occurs when one accepts the null hypothesisType II error: occurs when one accepts the null hypothesis
when in fact the null hypothesis is false. when in fact the null hypothesis is false.
R R E E S S E E A A R R C C H H M M E E T T H H O O D D O O L L O O G G Y Y Bec
Because their names are noause their names are not very descriptive, these types of errot very descriptive, these types of errorsrs sometimes are confused. Some people jokingly define a Type III sometimes are confused. Some people jokingly define a Type III error to
error to occur when one confuses Type I and Type II. To illoccur when one confuses Type I and Type II. To illustrateustrate the difference, it is useful to consider a trial by jury in which the the difference, it is useful to consider a trial by jury in which the null hy
null hypothpothesis iesis is that s that the defendant the defendant is innocent. If tis innocent. If t he jury convihe jury convictscts a truly innocent defendant, a Type I error has occurred. If, on the a truly innocent defendant, a Type I error has occurred. If, on the other
other hand, thhand, the jury dece jury declares a truly lares a truly guilguilty defendant to ty defendant to be innocentbe innocent,, a Type II error has occurred.
a Type II error has occurred. Hypothesis testing involves th
Hypothesis testing involves th e following steps:e following steps:
•
• Formulate the Formulate the null and alternative hynull and alternative hypothpoth eses.eses. •
• Choose the appropriate test.Choose the appropriate test. •
• Choose a level of significance (alpha) - determine theChoose a level of significance (alpha) - determine the
rejecti
rejection on regiregion.on.
•
• Gather the data and calculate the test statistic.Gather the data and calculate the test statistic. •
• Determine the probability of the observed value of the testDetermine the probability of the observed value of the test
statistic under th
statistic under th e null hypothesis give null hypothesis given then the samplie samplingng distribution that applies to the chosen test.
distribution that applies to the chosen test.
•
• Compare the value of the test statistic to the rejectionCompare the value of the test statistic to the rejection
threshold. threshold.
•
• BaBased on tsed on t he comparison, rejhe comparison, reject or dect or do not o not rejereject the ct the nullnull
hypothesis. hypothesis.
•
• MaMake the ke the marketing research conclusion.marketing research conclusion.
In order to analyze whether research results are statistically In order to analyze whether research results are statistically significant or simply by chance, a test of statistical significance can significant or simply by chance, a test of statistical significance can be run
be run
Tests of Statistical Significance Tests of Statistical Significance The chi-square goodness-of-fi
The chi-square goodness-of-fit tt test is used to est is used to determine whetherdetermine whether a set of proportions have specified numerical values. It often is a set of proportions have specified numerical values. It often is used to analyze bivariate cross-tabulated data.
used to analyze bivariate cross-tabulated data.
Some examples of situations that are well suited for this test are: Some examples of situations that are well suited for this test are:
•
• A manufacturer of packaged products test markets a newA manufacturer of packaged products test markets a new
product and wants to know if sales of the new product will product and wants to know if sales of the new product will be in the same relative proportion of package sizes as sales be in the same relative proportion of package sizes as sales of existing products.
of existing products.
•
• A company’A company’s sales revenue comes frs sales revenue comes from Pom Product roduct A (50%)A (50%),,
Product B (30%), and Product C (20%). The firm wants to Product B (30%), and Product C (20%). The firm wants to know w
know whether recent hether recent fluctuafluctuations in thtions in th ese proportions ese proportions areare random or whether they represent a real shift in sales. random or whether they represent a real shift in sales.
Defining k categories and observing the number of cases falling Defining k categories and observing the number of cases falling into each category perform the chi-square test. Knowing the into each category perform the chi-square test. Knowing the expected number
expected number of cases falliof cases falling in each category, one ng in each category, one can definecan define chi-squared as: chi-squared as:
??
22= SO
= SO
ii- E
- E
ii))
22/ E
/ E
ii ii Where Where OOii= = the number the number of obof observeserved cases in categd cases in category i,ory i,”” E
Eii= the number of expected cases in category i,= the number of expected cases in category i, k = the
k = thenumber onumber of categoriesf categories,, th
th ee summation runs from i = 1 to i = k.summation runs from i = 1 to i = k. Before cal
Before calculaticulating the ng the chi-square vachi-square value, one needs tlue, one needs to determine to determine t hehe expected frequency for each cell. This is done by dividing the expected frequency for each cell. This is done by dividing the number of samples by the number of cells in the table. number of samples by the number of cells in the table. To use the output of the square function, one uses a To use the output of the square function, one uses a chi-square table.
square table. To do To do so, one needs to so, one needs to know the number oknow the number o f degreesf degrees
of freedom (df). For chi-square applied to cross-tabulated data, of freedom (df). For chi-square applied to cross-tabulated data, the number of degrees of freedom is equal to
the number of degrees of freedom is equal to (number of columns - 1) (number of rows - 1) (number of columns - 1) (number of rows - 1)
This is equal to the number of categories minus one. The This is equal to the number of categories minus one. The conven
conventiontional critical al critical level of 0.05 level of 0.05 nonormally rmally is used. If this used. If the calculae calculatedted outp
output value from the function is greater than thut value from the function is greater than th e chi-square le chi-square look- ook-up table value, the null hypothesis is rejected.
up table value, the null hypothesis is rejected. Anova
Anova Another
Another test test of significaof significance is thnce is the Analye Analysis of Variance (ANOVA)sis of Variance (ANOVA) test. The primary purpose of ANOVA is to test for differences test. The primary purpose of ANOVA is to test for differences between multiple means. Whereas the t-test can be used to between multiple means. Whereas the t-test can be used to compare two means, A
compare two means, ANONOVA VA is needed to compare this needed to compare th ree or moreree or more means. If multiple t-t
means. If multiple t-tests were applied, the proests were applied, the probability of a bability of a TYPETYPE I error
I error (rejec(rejecting a true null hypothesis) increating a true null hypothesis) increases as the number of ses as the number of comparisons
comparisons increaseincreasess
One-way ANOVA examines whether multiple means differ. The One-way ANOVA examines whether multiple means differ. The test is called an F-test. ANOVA calculates the ratio of the variation test is called an F-test. ANOVA calculates the ratio of the variation between groups t
between groups to tho the variae variation within grouption within group s (the F ratio). Whiles (the F ratio). While ANOVA was desi
ANOVA was designed for gned for comparing several meacomparing several means, it also can ns, it also can bebe used to
used to compare tcompare two means. Two-wawo means. Two-way Ay ANONOVA aVA allowllows for s for a seconda second independent variable and addresses interaction.
independent variable and addresses interaction. To run a one-way Anova, use the following steps: To run a one-way Anova, use the following steps:
•
• IdentIdentify ify the independent the independent and dependent and dependent variavariables.bles. •
• Describe the variation by breaking it into three partsDescribe the variation by breaking it into three parts - the- the
total variation, the portion that is within groups,
total variation, the portion that is within groups,and theand the portion that is between groups (or among groups for more portion that is between groups (or among groups for more than two groups).
than two groups). The total variation (SS
The total variation (SStotaltotal) is the sum of the squares of the) is the sum of the squares of the differences between each value
differences between each valueand thand the grand mean of all the valuese grand mean of all the values in all the groups.
in all the groups. The in-group variation (SSThe in-group variation (SSwithinwithin) is the sum of ) is the sum of the squares of th
the squares of the differencese differencesin each element’in each element’s value and ths value and th e groupe group mean. The variation between group means (SS
mean. The variation between group means (SSbetweenbetween) is the total) is the total varia
variation tion minus the minus the in-group variation (SSin-group variation (SStotaltotal- S- SSSwithinwithin).). 1.
1. MeaMeasure the difference sure the difference between eabetween each group’ch group’s meas mean and then and the grand mean.
grand mean. 2.
2. Perform a sigPerform a signifinificanccance test on e test on the differethe differencesnces.. 3.
3. InterprInterpret the reset the resulults.ts.
This F-test assumes that the group variances are approximately This F-test assumes that the group variances are approximately equal a
equal and tnd that thhat the observations are independent. It e observations are independent. It alsalso assumeso assumes normally
normally distributed data; hodistributed data; ho wevwever, since this is a ter, since this is a t est on est on meansmeans the Centr
the Central Lial Limit Theormit Theorem holds as long as the sample size is notem holds as long as the sample size is not too small
too small
ANOVA is efficient for analyzing data using relatively few ANOVA is efficient for analyzing data using relatively few observation
observations and can be used with categorical s and can be used with categorical variablesvariables. Not. Note thate that regression can perform a similar analysis to that of ANOVA. regression can perform a similar analysis to that of ANOVA. Discriminant Analysis
Discriminant Analysis
Analysis of the difference in means between groups provides Analysis of the difference in means between groups provides information about individual variables, it is not useful for information about individual variables, it is not useful for determine their individual impacts when the variables are used in determine their individual impacts when the variables are used in combination. Since some variables will not be independent from combination. Since some variables will not be independent from one another, one needs a test that can consider them one another, one needs a test that can consider them simul
simultaneously itaneously in ordn order to ter to take iake into nto account their interrelationship.account their interrelationship. One such test is to construct a linear combination, essentially a One such test is to construct a linear combination, essentially a weighted sum of the variables. To determine which variables weighted sum of the variables. To determine which variables
R R E E S S E E A A R R C C H H M M E E T T H H O O D D O O L L O O G G Y Y
discriminate between two or more naturally occurring groups, discriminate between two or more naturally occurring groups, discriminant analysis i
discriminant analysis is used. Ds used. Discriminant analysis caiscriminant analysis can detn determineermine whi
which variablch variables are the best es are the best predictors of predictors of group membership. Itgroup membership. It determines which groups differ with respect to the mean of a determines which groups differ with respect to the mean of a variable,
variable, and thand then uses that variable to pren uses that variable to predict new cases edict new cases of groupof group membership. Essentiall
membership. Essentiallyy, th, the discrimie discriminant nant function function probproblem is alem is a one-way ANOVA problem in that one can determine whether one-way ANOVA problem in that one can determine whether multiple groups are significantly different from one another with multiple groups are significantly different from one another with respect to
respect to the the mean of mean of a particular variaa particular variable. A ble. A discriminant analysidiscriminant analysiss consists of the following steps:
consists of the following steps: 1.
1. FormulFormulate ate the problthe problem.em. 2.
2. Determine the disDetermine the discrimcriminant function coeffiinant function coefficiecients that resultnts that result in the highest ratio of between-group variation to in the highest ratio of between-group variation to within- within-group variation.
group variation. 3.
3. Test the sigTest the significnificance of the diance of the discrimscriminant function.inant function. 4.
4. InterpInterpret the reret the resulsults.ts. 5.
5. Determine the valDetermine the validity idity of the analyof the analysissis.. Discriminant analysi
Discriminant analysis analyzes analyzes ts the dhe dependency relationship, ependency relationship, whereaswhereas factor analysis and cluster analysis address the interdependency factor analysis and cluster analysis address the interdependency among variables.
among variables. Factor Analysis Factor Analysis
Factor analysis is a very popular technique to analyze Factor analysis is a very popular technique to analyze interdependence.Factor analysis studies the entire set of interdependence.Factor analysis studies the entire set of interrelationships wi
interrelationships without thout defining vadefining variableriables to s to be dependent be dependent oror independent.
independent. Factor Factor analyanalysis combines variablesis combines variables to s to create a smallercreate a smaller set of factors.
set of factors.
Mathematically, a factor is a linear combination of variables. A Mathematically, a factor is a linear combination of variables. A factor is not directly observable; it is inferred from the variables. factor is not directly observable; it is inferred from the variables. The techn
The technique identifies underlyique identifies underlying structure amoning structure among the variables,g the variables, reducing the numb
reducing the number of variables to a more manageable set. Factorer of variables to a more manageable set. Factor analy
analysis groups variables according to tsis groups variables according to t heir corrheir correlationelation
The factor loading can be defined as the correlations between the The factor loading can be defined as the correlations between the factors and their underlying variables.
factors and their underlying variables.
A factor-loading matrix is a key output of the factor analysis. An A factor-loading matrix is a key output of the factor analysis. An example matrix is shown below.
example matrix is shown below.
Factor
Factor 1 Facto1 Factor 2r 2 Factor 3Factor 3 Variable 1 Variable 1 Variable 2 Variable 2 Variable 3 Variable 3
Column's Sum of Squares: Column's Sum of Squares:
C
Coo lluu mm nn ’’s s SSuu m m oo f f SSqq uu aa rree ss
Each cell in the matrix represents correlation b
Each cell in the matrix represents correlation betweeetween thn the variae variableble and th
and the factor associated wie factor associated with tth that cell. The square of this correlationhat cell. The square of this correlation represents th
represents th e propoe proportion ortion of the variation in the variaf the variation in the variable explble explainedained by the factor. The sum of the squares of the factor loadings in by the factor. The sum of the squares of the factor loadings in each column is called an eigen value. A
each column is called an eigen value. An eigen value representn eigen value represents ths thee amount
amount of variance in thof variance in th e original varie original variables that ables that is associais associated withted with that factor
that factor. The communality i. The communality is the amount s the amount of thof the variae variable variable variancence explained by common factors
explained by common factors
A rule of thumb for deciding on the number of factors is that A rule of thumb for deciding on the number of factors is that each included factor must explain at least as much variance as does each included factor must explain at least as much variance as does an average variable. In other words, only factors for which the an average variable. In other words, only factors for which the
eigen value is greater than one are used. Other criteria for eigen value is greater than one are used. Other criteria for determining the number of factors include the Scree plot criteria determining the number of factors include the Scree plot criteria and the percentage of variance criteria.
and the percentage of variance criteria. To facili
To facilitate interprtate interpretation, thetation, the axe axis cis can be rotated. Rotation oan be rotated. Rotation of thef the axis
axis is equivais equivalent to flent to forming linear combinations oorming linear combinations of the factof the factors. Ars. A common
commonly ly used rotused rotation stration strategy ategy is the varimax rotation. Varimaxis the varimax rotation. Varimax attempts to fo
attempts to fo rce the column entries to be either clrce the column entries to be either close to zero orose to zero or one.
one.
Cluster Analysis Cluster Analysis
Market segmentation usually is based not on one factor but on Market segmentation usually is based not on one factor but on multiple factors. I
multiple factors. Initiallnitially, eacy, each variable represents h variable represents its own cluster.its own cluster. The ch
The challenge iallenge is to fins to find a way d a way to coto combine variables so thmbine variables so that relativelyat relatively homogenous clusters can be formed. Such clusters should be homogenous clusters can be formed. Such clusters should be internally homogenous and externally heterogeneous. Cluster internally homogenous and externally heterogeneous. Cluster analysis is one way to accomplish this goal. Rather than being a analysis is one way to accomplish this goal. Rather than being a statistical test, it is more of a coll
statistical test, it is more of a collection of algorithms foection of algorithms for groupingr grouping objects, or in the case of marketing research, grouping people. objects, or in the case of marketing research, grouping people. Cluster analysis is useful in the exploratory phase of research when Cluster analysis is useful in the exploratory phase of research when there are no a-priori hypotheses.
there are no a-priori hypotheses. Cluster Analysis Steps Cluster Analysis Steps Formulate th
Formulate the problem, collecting data ae problem, collecting data and chond choosing the variablesosing the variables to analyze.
to analyze.
Choose a distance measure. The most common
Choose a distance measure. The most common is the Eis the E ucliuclideandean distance. Ot
distance. Other poher possibilissibilities include the squared Euclidean distance,ties include the squared Euclidean distance, city-block (Manhattan) distance, Chebychev distance, power city-block (Manhattan) distance, Chebychev distance, power distance, and percent disagreement.
distance, and percent disagreement.
Choose a clustering procedure (linkage, nodal, or factor Choose a clustering procedure (linkage, nodal, or factor procedures).
procedures). Det
Determine the numbermine the numb er of clusters. They should be weler of clusters. They should be well separatedl separated and ideall
and ideally they should be distinct enoy they should be distinct enough to give them descriptiveugh to give them descriptive names such as professionals, buffs, etc.
names such as professionals, buffs, etc. Profile the clusters.
Profile the clusters. A
Assess tssess t he he VValidialidity oty o f thf th e Cle Clustust eringering The for
The format of the marketing research report varies wimat of the marketing research report varies with thth the needse needs of the organization.
of the organization. The report
The report often contoften cont ains the followains the following seing sections:ctions: Authorization letter for the research
Authorization letter for the research
•
• Table of ContentsTable of Contents •
• List of illustrationsList of illustrations •
• Executive summaryExecutive summary •
• Research objectivesResearch objectives • • MethodologyMethodology • • ResultsResults • • LimitationsLimitations •
• ConclusiConclusions and recommendationons and recommendationss •
• Appendices containing copies of the questionnaires, etcAppendices containing copies of the questionnaires, etc
Data is then p
Data is then p rocessed irocessed in ordn order to summarise the resuler to summarise the resultsts
It seeks to determine how units covered in the research project It seeks to determine how units covered in the research project respond to the items under investigation.
respond to the items under investigation. It could be It could be • • UnivariateUnivariate • • BivariateBivariate
R R E E S S E E A A R R C C H H M M E E T T H H O O D D O O L L O O G G Y Y • • MultivariateMultivariate
IIntente rpretarpreta tion of Rtion of Resultsesults
It
It is the “so what”? of research.is the “so what”? of research. Research is w
Research is wasted if it is not uasted if it is not used in decision making or influencingsed in decision making or influencing actions. Not only the results should be interpreted into action actions. Not only the results should be interpreted into action recommendations but the recommendations must also be recommendations but the recommendations must also be communicated in an understandable manner. Results should be communicated in an understandable manner. Results should be presented in
presented in as simple maas simple manner nner as possible.as possible. B
Basic Rasic Reqeq uiremeuireme ntnt 1.
1. ReReseasearcher musrcher must understand the proper interpretat understand the proper interpretation of tion of research results and the assumptions embodied in them. research results and the assumptions embodied in them. 2.
2. ReReseasearcher shoulrcher should understand the kinds of questid understand the kinds of questionsons
research can handle and the type of structure required making research can handle and the type of structure required making a prob
a problem “researchabllem “researchable”e” 3.
3. ReseaResearcher must be capable of appraising rcher must be capable of appraising the feasibilthe feasibility ity of of resea
research rch proposalsproposals Spe
Spe cicial Caseal Case
In management maximum research is done in the field of In management maximum research is done in the field of marketing.
marketing.
We conceptualize the marketing research process as consisting of We conceptualize the marketing research process as consisting of six steps.
six steps. Each and Each and every every step in market research we wistep in market research we will be doingll be doing in detail. First let us discuss these briefly now.
in detail. First let us discuss these briefly now.
Step 1: Problem Definition Step 1: Problem Definition
The first step in any marketing research project is to define the The first step in any marketing research project is to define the problem.
problem. In defining In defining the problem, the ythe problem, the you should take ou should take intointo account
account
•
• The purpose of the study,The purpose of the study, •
• The relevant backgThe relevant background round information,the informationinformation,the information
needed, and needed, and
•
• How it How it wiwill be used ll be used in decision-making.in decision-making.
Problem definition involves discussion with the Problem definition involves discussion with the decision makers,
decision makers,
•
• Interviews with industry experts,Interviews with industry experts, •
• Analysis of secondary data, andAnalysis of secondary data, and •
• Perhaps, some qualitative research, such as focus groups.Perhaps, some qualitative research, such as focus groups.
Once the problem of investigation has been precisely defined, the Once the problem of investigation has been precisely defined, the research can be designed and conducted properly
research can be designed and conducted properly
Step 2: D
Step 2: Development evelopment of an of an AApproach to pproach to the the PProblemroblem
Development of an appro
Development of an appro ach to thach to the problem includee problem includess
•
• Formulating an objective or theoretical framework,Formulating an objective or theoretical framework, •
• analytical models,analytical models, •
• ResResearch questions, and hypotheses and earch questions, and hypotheses and identifyiidentifying thng thee
information needed. information needed. This pro
This process is gcess is guided byuided by
•
• Discussion with managemeDiscussion with management nt and industrand industr y exy experts,perts, •
• Analysis of secondary data,Analysis of secondary data, •
• Qualitative research, andQualitative research, and •
• Pragmatic considerationsPragmatic considerations
Step 3: R
Step 3: Reseese arch Design Farch Design Formorm ulationulation
A research design is a framework or blueprint for conducting the A research design is a framework or blueprint for conducting the marketing
marketing research project. research project. It detIt detailails the pros the procedures necessacedures necessary forry for obtaining the required information, and its pur
obtaining the required information, and its pur pose is to design apose is to design a study that wil
study that will test thl test the hypotheses of inte hypotheses of int erest, determine possibleerest, determine possible answers to the research questions, and provide the information answers to the research questions, and provide the information needed for decision making.
needed for decision making.
Conducting exploratory research precisely defines the variables, Conducting exploratory research precisely defines the variables, and designing appropriate scales to measure them are also a part and designing appropriate scales to measure them are also a part of the research design
of the research design
The issue of how the data should be obtained from the The issue of how the data should be obtained from the respondents (for example, by conducting a survey or an respondents (for example, by conducting a survey or an experim
experiment) must be addressed. ent) must be addressed. It is also necessaIt is also necessary to desigry to design an a questionnaire and a sampling plan to select respondents for the questionnaire and a sampling plan to select respondents for the study.
study.
More formally, formulating the research design involves the More formally, formulating the research design involves the followi
following stng steps:eps: 1.
1. DefiniDefinition of the information of the information needetion neededd 2.
2. SeSecondary dacondary data analyta analysissis 3.
3. QualiQualitativtative ree reseasearchrch 4.
4. Methods of collecMethods of collecting quantitative ting quantitative data (survedata (survey, y, observation,observation, and experimentation)
and experimentation) 5.
5. MeMeasureasurement and scalment and scaling proceing proceduresdures 6.
6. QuestiQuestionnaironnaire dese desigignn 7.
7. SaSampling mpling process and saprocess and sample simple sizeze 8.
8. PlaPlan of n of data adata analnalyysisiss
Step
Step 4: F4: Fieldwoieldwo rk or Datrk or Dat a Colla Collectionection
Data collection involves a fi
Data collection involves a field force or steld force or staff that oaff that operates either inperates either in the
the fielfield, as in d, as in the the case of pcase of personal interviewersonal interviewing (in-home, malling (in-home, mall intercept, or computer-assisted personal interviewing), from an intercept, or computer-assisted personal interviewing), from an office by telephone (telephone or computer-assisted telephone office by telephone (telephone or computer-assisted telephone interviewing), through mail (traditional mail and mail panel interviewing), through mail (traditional mail and mail panel surveys with pre-recruited households), or electronically (e-mail surveys with pre-recruited households), or electronically (e-mail or I
or I nternet)nternet)..
Proper selection, training, supervision, and evaluation of the field Proper selection, training, supervision, and evaluation of the field force help
force help minimiminimize data-collectize data-collection on errors.errors.
Step 5: D
Step 5: Data ata PreparatPreparat ion and ion and AAnalysinalysiss
Data preparation
Data preparation inclincludes the udes the editingediting, coding, transcription and, coding, transcription and veri
verificafication of data. tion of data. Each questionnaire or observation form isEach questionnaire or observation form is inspected or edited and, i
inspected or edited and, if necessaf necessary, ry, corrected. corrected. Number or letterNumber or letter codes are assigned to represent each response to each question in codes are assigned to represent each response to each question in the questionnaire.
the questionnaire. The data from the questionnaire aThe data from the questionnaire are transcribedre transcribed or key punched onto magnetic tape or disks, or input directly into or key punched onto magnetic tape or disks, or input directly into
R R E E S S E E A A R R C C H H M M E E T T H H O O D D O O L L O O G G Y Y the comput
the computer. er. The data are analThe data are analyzed to derive yzed to derive information relatedinformation related to th
to the components of te components of t he marketing he marketing research problem and, thus,research problem and, thus, provide input in to the management decision problem.
provide input in to the management decision problem.
Step 6:
Step 6: RRepoepo rt Prt Preparepa ration anration an d Prd Presenesen tattat ionion
The entire project should be do
The entire project should be do cumented in a wricumented in a written rtten report teport thathat addresses the specific research questions identified, describes the addresses the specific research questions identified, describes the approach, the research design, data collection, and data analysis approach, the research design, data collection, and data analysis procedures adopted, and presents the results and the major procedures adopted, and presents the results and the major findings.
findings. The findings should be presented in a The findings should be presented in a comprehensiblecomprehensible format so
format so that that managmanagement can readily ement can readily use them in use them in the decisionthe decision making
making process. process. In addition, an In addition, an oral presentation should beoral presentation should be made to management using tables, figures,
made to management using tables, figures, and graphs to enand graphs to en hancehance clarity and impact
clarity and impact
Errors in Research Process Errors in Research Process There are 2 types of errors There are 2 types of errors aa.. SSamamplipling errong errorr
b
b.. Non samNon samplipling eng errorrror
Sampling error is the difference between sample value and Sampling error is the difference between sample value and corresponding po
corresponding po pulation value.pulation value. No
Non sampling error are errors in thn sampling error are errors in the difference stages of research.ase difference stages of research.as a.
a. DefectiDefective ve problem problem desidesigngn b.
b. DefecDefectivtive populatie population on defidefinitinitionon cc.. FraFrame me sesett
d.
d. SSurrogaurrogate informte informatiation erroron error ee.. Non reNon responssponse ee errorrror
f.
f. MMeaeasursuremement eent errorrror g
g.. ExperiExperimentamental l errorerror h.
h. Poor questiPoor questionnaionnaire desire designgn i.
i. InterInterviview ew bibiasas j.
j. Data Data proceprocessissing ng errorerror k.
k. Data aData analnalyysisis errorss errors l.
l. Interpretation Interpretation errorserrors
R
Re f e r e n
e f e r e n cce s a
e s a n d
n d F
Fu r t h
u r t h e r R
e r Re a
e a d i n g s
d i n g s
1.
1. BoyBoyd, wesd, westfall, atfall, and stasch, “Marketing Rnd stasch, “Marketing Research Text andesearch Text and Cases”, All India Traveller Bookseller, New Delhi.
Cases”, All India Traveller Bookseller, New Delhi. 2.
2. Brown, Brown, F.E. “Marketing F.E. “Marketing ReseaResearch, a rch, a structure for decisionstructure for decision making
making”, Addison ”, Addison – Wesley publishing company– Wesley publishing company 3.
3. Kothari, C.RKothari, C.R. “Res. “Research Mearch Methodology-Methodology-Methods andethods and Techniques”, Wil
Techniques”, Wiley Eastern ey Eastern Ltd.Ltd. 4.
4. SStockton and Clatockton and Clark, “Introduction to Busirk, “Introduction to Business andness and Econo
Economic Statismic Statistics”, Dtics”, D.B.B.Taraporeval.Taraporevala Sons a Sons andCo. PrivateandCo. Private Limited, Bombay.
Limited, Bombay. 5.
5. DunDunn On Olive live Jean and VJean and Virginia irginia A CA Clarck, larck, “Applied Statistics”“Applied Statistics” John Wiley and Sons.
John Wiley and Sons. 6.
6. Green Paul E and DoGreen Paul E and Do nald Snald S. Tull. Tull, “Res, “Researcearch for Markeh for Marketingting Decisions” Prentice Hall of India, New Delhi
Decisions” Prentice Hall of India, New Delhi
Ac
Activity tivity 11
The definition of the problem comprise The definition of the problem comprise
Ac
Activity tivity 22
Explain the importance of interpretation of results? Explain the importance of interpretation of results?
Ac
Activity tivity 33
Name and briefly discuss the five steps of research process? Name and briefly discuss the five steps of research process?
Ac
Activity tivity 44
Explain with the help of a suitable example the need for Explain with the help of a suitable example the need for introducing two types of environmental conditions in a research introducing two types of environmental conditions in a research problem?
problem?
Clue Clue
The environmental conditions specified in the research problem The environmental conditions specified in the research problem are of two types:
are of two types: i.
i. Those Those beybeyond fiond firm’rm’s cs controlontrol ii
ii.. Those wiThose within the firm’thin the firm’s controls control
Ac
Activity tivity 55
A local supermarket has experienced a decline in unit _sales and A local supermarket has experienced a decline in unit _sales and little change in rupee value sales. Profits have almost vanished. little change in rupee value sales. Profits have almost vanished. The chief executive in searching for ways to revitalize the operat The chief executive in searching for ways to revitalize the operat ionion wa
was adviss advised to ed to increaincrease the nse the number of umber of hours thours the market is openhe market is open for b
for business. He comes to you for usiness. He comes to you for _advic_advice in structure in structuring a reseaing a researchrch problem that will provide relevant information for problem that will provide relevant information for decision-making.
making.
Define the problem, taking care to: Define the problem, taking care to: a.
a. SState the reltate the relevaevant questint questionon b.
R R E E S S E E A A R R C C H H M M E E T T H H O O D D O O L L O O G G Y Y
cc.. CleClearly arly define the units of analydefine the units of analysis asis and characteristics of nd characteristics of interest.
interest.
What are the relev
What are the relevant “ stant “ states of nature” which would lates of nature” which would lead to thead to th ee selection each
selection each alternative answalternative answer?er?
Ac
Activity tivity 66
A s
A shampoo manufhampoo manuf acturing acturing company wcompany wishes to test two types ishes to test two types of of makes i
makes in order tn order t o determine which io determine which is the best os the best o ne .ne . A.
A. Propose and defend a preciPropose and defend a precise defise definition of “best”nition of “best” B
B.. WhaWhat is the set of t is the set of hypothesihypothesis that should be tested?s that should be tested? C.
C. WhaWhat action would be associated wit action would be associated with th each hypothesis?each hypothesis?
Activity 7 Activity 7
Define and state the research problem for the following case: Define and state the research problem for the following case: “Why i
“Why is the ps the productivity in Jroductivity in Japan so much hapan so much h igigher thher th an in Inan in India”?dia”? Think about problem in a broader sense and narrow down the Think about problem in a broader sense and narrow down the research problem.