Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 37

Simon France

S182106

MARK0793-1
UER Management

Marketing Research
Master 1 SG
2022-2023

Madame Cécile Delcourt


Marketing Research

Table of contents
1. Introduction to marketing research.......................................................................................3
2. Problem definition and research approach development.....................................................4
2.1. Problem definition..........................................................................................................4
2.1.1. Environmental context.............................................................................................4
2.1.2. Marketing decision problem to marketing research problem.................................5
2.2. Research approach development...................................................................................6
3. Research design development................................................................................................7
3.1. Research design..............................................................................................................7
3.1.1. Exploratory research design.....................................................................................8
3.1.2. Descriptive research design.....................................................................................8
3.1.3. Causal research design.............................................................................................8
3.1.4. Predictive research design.......................................................................................8
3.2. Qualitative design...........................................................................................................9
3.2.1. Direct techniques...................................................................................................10
3.2.2. Indirect techniques.................................................................................................11
3.3. Quantitative design.......................................................................................................12
3.3.1. Survey and quantitative observation techniques..................................................12
3.3.2. Causal research design...........................................................................................16
3.3.3. Measurement and scaling......................................................................................18
3.3.4. Questionnaire design.............................................................................................24
3.3.5. Sampling.................................................................................................................27
4. Survey fieldwork...................................................................................................................30
5. Data integrity and analysis....................................................................................................31
5.1. Data integrity................................................................................................................31
5.2. Data analysis.................................................................................................................32
5.2.1. Qualitative data analysis........................................................................................32
5.2.2. Quantitative data analysis......................................................................................32
6. Communicating research findings........................................................................................35
6.1. Data visualization..........................................................................................................35

2
Marketing Research

1. Introduction to marketing research


Marketing research is the function that links the consumer, customer and public to the
marketer through information.

That information is used to :


 Identify and define marketing opportunities and problems
 Generate, refine and evaluate marketing actions
 Monitor marketing performance
 Improve understanding of marketing as a process

Marketing research :
 Specifies the information required to address these issues
 Designs the method for collecting information
 Manages and implements the data collection process
 Analyses the results
 Communicates the findings and their implications

3
Marketing Research

2. Problem definition and research approach development

2.1. Problem definition


The problem definition is a broad statement of the general problem and identification of the
specific components of the marketing research problem. (Malhotra et al. 2017)

2.1.1. Environmental context

The objective of meeting a decision-maker (DM) is to develop a problem audit (a


comprehensive examination of a marketing problem to understand its origin and nature) so
that the researcher can identify the underlying causes of the problem.

A problem audit involves discussions with the DMs about :


 The events that led to the decision that action is needed
 The alternative courses of action available (and the criteria used to evaluate them)
 What the DM perceives to be gaps in their knowledge
 The manner in which the DM will use each item of information in making the decision

An expert is an individual knowledgeable about the firm, industry, issue or topic investigated
that can help diagnosing the nature of the marketing research problem. It can be found
inside or outside the organisation investigated. (In-depth) interviews with a ready to use list
of topics/questions to cover with the expert is common practice.

4
Marketing Research

Secondary data are data collected for some purpose other than the problem at hand. They
are compared to primary data.

Primary data Secondary data


First hand data gathered by the Second hand data collected by
Meaning
researcher himself someone else earlier
Surveys, observations, Government publications,
Examples experiments, questionnaire, panels, websites, report, journal
personal interview, … articles, studies, …
Cost Relatively expensive Relatively economical
Collection time Relatively long Relatively quick
Always specific to the May or may not be specific to
Specific
researcher's needs the researcher's need

2.1.2. Marketing decision problem to marketing research problem

The marketing decision problem evaluates what the decision-maker needs to do. It is action
oriented and focus on symptoms (decreasing market shares, increasing absenteeism, …).

The marketing research problem evaluates what the researcher needs to know to support
the decision-making process. It is information oriented and focus on the underlying causes
(decreasing service quality, lack of customer orientation, …).

5
Marketing Research

2.2. Research approach development


The research approach aims to decide on how to approach the problem.

The objective theoretical framework is a theory (conceptual theme) based on foundational


statements or axioms that are assumed to be true.

The analytical model is an explicit specification of a set of variables and their


interrelationships designed to represent some real system or process in whole or in part. It
can be verbal (written), graphical (visual representation) or mathematical (equation form).

The research questions are refined statements of the specific components of the problem
and the hypotheses are unproven statements about a factor or phenomenon.

6
Marketing Research

3. Research design development

3.1. Research design


A research design is a framework or plan for conducting the marketing research project. It
details the procedures necessary for obtaining the information needed to structure or solve
marketing research problems.

Research design components :


 (Information needed)
 Type of design (exploratory, descriptive, causal or predictive)
 Sequence of techniques of understanding and measurements
 Research instrument for data collection :
 Interview guide (qualitative)
 Questionnaire (quantitative)
 Sampling process and sample size
 Pre-test of the research instrument

Qualitative Quantitative
Exploratory Descriptive Causal Predictive

7
Marketing Research

3.1.1. Exploratory research design

An exploratory research design is a research design characterised by a flexible and evolving


approach to explore marketing phenomena that are inherently difficult to measure or that
have been barely investigated.

Types of exploratory research designs :


 In-depth interviews (based on unstructured or semi-structured interview guides)
 Focus groups (based on unstructured or semi-structured research instruments)
 Unstructured observations

Exploratory designs are appropriate :


 When the nature of the topic cannot be measured in a structured or quantifiable manner
 When the problem needs to be defined more precisely
 When research questions or hypotheses need to be developed
 When variables need to be isolated and classified as dependent or independent variables

3.1.2. Descriptive research design

A descriptive research design is a research design that aims to describe a phenomenon


(usually market characteristics or functions) without explaining it.

Types of descriptive research designs :


 Markets studies (consumer perceptions, …)
 Market share studies
 Brand concept maps
 Polls

3.1.3. Causal research design

A causal research design is a research design that aims to explain the impact of an
independent variable on a dependent variable (causal relationship).

Types of causal research designs :


 Experiments

3.1.4. Predictive research design

A predictive research design is a research design that aims to predict what is more likely to
happen in the future based on data mining to best anticipate issues (prevent issues such as
customer defection, employee burn-out, employee turnover, …) by using customer or
employee analytics.

8
Marketing Research

3.2. Qualitative design


A qualitative design is an unstructured, primarily exploratory design based on small samples
intended to provide in-depth insights. (Malhotra et al. 2017)

9
Marketing Research

3.2.1. Direct techniques

A direct approach is a type of qualitative research in which the purposes of the project are
disclosed to the participants or are obvious given the nature of the interview.

A focus group is a discussion conducted by a trained moderator among a small group of


participants in an unstructured (or semi-structured) and natural manner. The moderator
(individual that conducts the focus group) plays a vital role by establishing rapport with the
participants and by setting the purpose of the interview, questioning, probing and handling
the process of discussion.

Advantages Disadvantages
Synergy Misjudgement
Snowballing Moderation
Stimulation Messiness
Security Misrepresentation
Spontaneity Meeting
Serendipity (“heureux hasard”)
Specialisation
Scientific scrutiny
Structure
Speed

An in-depth interview is an unstructured (or semi-structured), direct and personal interview


in which a single participant is questioned and probed by an experienced interviewer to
uncover underlying motivations, beliefs, attitudes and feelings on a topic. The emphasis
should be upon a full interaction to understand the meaning of the participant’s experiences
and lifestyle.

10
Marketing Research

3.2.2. Indirect techniques

An indirect approach is a type of qualitative research in which the purposes of the project
are disguised from the participants.

A projective technique is an unstructured research technique that encourages participants


to project their underlying motivations, beliefs, attitudes or feelings regarding the issues of
concern.

The Zaltman metaphor elicitation technique (ZMET) is a projective technique that creates a
“mental map” of the consumer’s view towards a particular topic (product, service, idea,
organisation, policy, brand, …) using images and metaphors. It lets participants’ own frames
of reference shape the in-depth interview without losing sight of research aims.

The ethnography is based upon the observation of the customs, habits and differences
between people in everyday situations. The aim is to see through the eyes of the people
being studied.

Ethnography is appropriate :
 High-intensity situations (bad news delivery)
 Behavioural processes (home-cleaning or driving a car)
 Memory of the respondent may be inadequate (blind people)
 Shame or reluctance (observation of vulnerable people)

The netnography is a type of ethnography based on the analyse of the free behaviour of
individuals in online environments (active or passive).

Advantages compared to ethnography :


 Faster
 Simpler
 Cheaper
 Unelicited and more natural
 Unobtrusive
 Can be passive or active

11
Marketing Research

3.3. Quantitative design


A quantitative design is a set of research techniques that seek to quantify data and apply
some form of measurement and statistical analysis. (Malhotra et al. 2017)

3.3.1. Survey and quantitative observation techniques

A survey method is based on a structured questionnaire administered to a sample of a


target population to elicit specific information (fixed-response alternative questions).

Different aspects can influence the order of questions :


 Independent variables before dependent variables
 Most important variables before least important variables
 Easy and engaging questions before complex and sensitive questions

12
Marketing Research

The flexibility of data collection is determined primarily by the extent to which the
participant can interact with the interviewer and the survey questionnaire.

The diversity of questions that can be asked in a survey depends on the degree of
interaction the participant has with the interviewer and the questionnaire, as well as the
ability to see the questions.

The ability to use physical stimuli such as the product, a product prototype, commercials or
promotional displays during the interview.

The sample control is the ability of the survey mode to reach the units specified in the
sample effectively and efficiently.

The ability to collect large amounts of data.

The response rate is broadly defined as the percentage of the total attempted interviews
that are completed. The non-response bias is a bias caused when actual participants differ
from those who refuse to participate (Ethias Assistance).
13
Marketing Research

The control of the data-collection environment is the degree of control a researcher has
over the environment in which the participant answers the questionnaire.

The control of field force is the ability to control the interviewers and supervisors involved in
data collection. Field force consists of both the actual interviewers and supervisors involved
in data collection.

The potential for interviewer bias involves that an interviewer can bias the results by the
manner he :
 Introduces himself
 Selects participants
 Asks questions
 Probes
 Records and transcripts answers
 …

The potential to probe participants is the potential of the interviewer to create bias in the
responses elicited from participants. It is balanced somewhat by the amount of probing that
can be done.

The potential to build rapport may be vital to communicate why the survey is being
conducted, with a corresponding rationale for the participant to spend time answering the
questions.

The speed is the total time taken for administering the survey to the entire sample.

The cost is the total cost of administering the survey and collecting the data.

The perceived participant anonymity refers to the participants' perceptions that their
identities will not be discerned by the interviewer or the researcher.

The social desirability is the tendency of participants to give answers that they feel to be
acceptable in front of others, including interviewers.

The sensitive information may mean an issue that is personally sensitive. It may be deemed
‘sensitive’ varies enormously between different types of participants.

The incidence rate refers to the rate of occurrence or the percentage of persons eligible to
participate in a study.

The participant control is a set of methods that allow participant control over the
interviewing process can solicit greater cooperation and engagement.

14
Marketing Research

The quantitative observation techniques involve recording and counting behavioural


patterns of people, objects and events in a systematic manner to obtain information about
the phenomenon of interest.

In structured observation (more quantitative), the researcher clearly defines the behaviours
to be observed and the techniques by which they will be measured.

In an unstructured observation (more qualitative), the researcher monitoring all relevant


phenomena without specifying the details in advance.

In a disguised observation, the participants are unaware that they are being observed.
Disguise may be accomplished by using two-way mirrors, hidden cameras or inconspicuous
electronic devices. Observers may be disguised as mystery shoppers or sales staff.

In an undisguised observation, the participants are aware that they are under observation.

In a natural observation, participants’ behaviour is observed in its natural environment.

In a contrived observation, participants’ behaviour is observed in an artificial environment.

In a personal observation, the researcher observes actual behaviour as it occurs. He does


not attempt to manipulate the phenomenon being observed but merely records what takes
place.

In an electronic observation, electronic devices record the phenomenon being observed. It


is used for continuously recording ongoing behaviour for later analysis.

In a trace analysis, data collection is based on physical traces or evidence of past behaviour.

15
Marketing Research

3.3.2. Causal research design

Causality applies when the occurrence of X increases the probability of the occurrence of Y.

Conditions (necessary but not sufficient) :


 Concomitant variation (X and Y vary together)
 Time order of occurrence of variables (X must occur before or simultaneously with Y)
 Absence of other possible causal factors (no extraneous variables)

The placebo effect refers to the phenomenon whereby individuals change with an inactive
substance in a direction in which they think they should change. (Davison & Valins 1969)

The Hawthorne effect refers to the tendency of people being observed in a research context
to behave differently from the way they would otherwise. (Eckmanns & al. 2006)

Independent variables are variables or alternatives that are manipulated and whose effects
are measured and compared (price levels).

Dependent variables are the variables which measure the effect of the independent
variables on the test units (sales, profits and market shares).

Extraneous variables are all other variables that affect the response of the test units (store
size, store location and competitive effort).

Test units are individuals, organisations or other entities whose response to the
independent variables or treatments is being examined (consumers or stores).

An experimental design is a set of procedures specifying :


 The test units and how they are to be randomly divided into homogeneous subsamples
 What independent variables or treatments are to be manipulated
 What dependent variables are to be measured
 How the extraneous variables are to be controlled

The internal validity (laboratory studies) refers to whether the manipulation of the
independent variables or treatments caused the observed effects on the dependent
variables. Control of extraneous variables is a necessary condition for establishing internal
validity.

The external validity (field studies) refers to whether the cause-and-effect relationships
found in the experiment can be generalised (in what populations, settings, times,
independent variables and dependent variables can the results be projected).

Extraneous variables :
 History refers to specific events that are external to the experiment but occur at the same
time as the experiment.

16
Marketing Research

 Maturation (MA) refers to changes in the test units themselves that occur with the
passage of time.
 Testing effects are caused by the process of experimentation :
 Main testing effect (MT) occurs when a prior observation affects a later observation.
 Interactive testing effect (IT) occurs when a prior measurement affects the test unit’s
response to the independent variable.
 Instrumentation (I) refers to changes in the measuring instrument, in the observers or in
the scores themselves.
 Statistical regression effects (SR) occur when test units with extreme scores move closer
to the average score during the experiment.
 Selection bias (SB) refers to the improper assignment of test units to treatment
conditions.
 Mortality (MO) refers to the loss of test units while the experiment is in progress.

Controlling extraneous variables :


 Randomisation refers to the random assignment of test units to experimental groups by
using random numbers.
 Matching involves comparing test units on a set of key background variables before
assigning them to the treatment conditions.
 Statistical control involves measuring the extraneous variables and adjusting for their
effects through statistical analysis.
 Design control involves the use of experiments designed to control extraneous variables.

17
Marketing Research

3.3.3. Measurement and scaling

Scale characteristics and levels of measurement (DODO) :


 Description refers to the unique labels or descriptors that are used to designate each
value of the scale. All scales possess description.
 Order refers to the relative sizes or positions of the descriptors. Order is denoted by
descriptors such as greater than, less than and equal to.
 Distance means that absolute differences between the scale descriptors are known and
may be expressed in units.
 Origin means that the scale has a unique or fixed beginning or true zero point.

An illustration of primary scales of measurement

A nominal scale (non-metric) is a figurative labelling scheme in which the numbers serve
only as labels or tags for identifying and classifying objects. When used for identification,
there is a strict one-to-one correspondence between the numbers and the objects. The only
permissible operation on the numbers in a nominal scale is counting. It only allows statistics
based on frequency counts (percentages and mode).

An ordinal scale (non-metric) is a ranking scale in which numbers are assigned to objects to
indicate the relative extent to which the objects possess some characteristic. It can
determine whether an object has more or less of a characteristic than some other object,
but not how much more or less. It allows statistics based on frequency counts and centiles
(percentile, quartile and median).

An interval scale (metric) permits comparison of the differences between objects. Both the
zero point and the units of measurement are arbitrary. It is not meaningful to take ratios of
scale values. It only allows statistics based on frequency counts, centiles and more

18
Marketing Research

(arithmetic mean, standard deviation and other statistics commonly used in marketing
research).
A ratio scale (metric) possesses all the properties of the nominal, ordinal and interval scales.
It has an absolute zero point. It is meaningful to compute ratios of scale values. All statistical
techniques can be applied to ratio data.

Primary scales of measurement

19
Marketing Research

In a comparative scale (non-metric), each object is compared to other stimulus objects (Do
you prefer theatre or cinema?). Comparative scale data must be interpreted in relative terms
and have only ordinal order properties.

Paired comparison scaling

Rank order scaling

20
Marketing Research

In a non-comparative scale (metric), each object is scaled independently of the others in the
stimulus set (To what extent do you like to go the Theatre? To what extent do you like to go
the Cinema?). The resulting data are generally assumed to be interval or ratio scaled. It is
widely used in marketing research.

Continuous rating scale

Likert scale

21
Marketing Research

Semantic differential scale

Development of a multi-item scale

22
Marketing Research

The true score model provides a framework for understanding the accuracy of
measurement.

XO = XT + XS + XR

where XO is the observed score or measurement, X T is the true score of the characteristic, X S
is the systematic error (error in the measurement) and X R is the random error (random
changes).

The reliability is the extent to which a scale produces consistent results if repeated
measurements are made on the characteristic. It can be defined as the extent to which
measures are free from random error (XR = 0).

Approaches :
 Test-retest reliability
 Coefficient alpha (or Cronbach alpha)

The validity is the extent to which a measurement represents characteristics that exist in the
phenomenon under investigation. It can be defined as the extent to which differences in
observed scale scores reflect true differences among objects on the characteristic being
measured, rather than systematic or random error. Perfect validity requires that there be no
measurement error (XO = XT, XR = 0 and XS = 0).

Approaches :
 Content validity
 Criterion validity
 Construct validity :
 Convergent validity
 Discriminant validity
 Nomological validity

The generalisability is the degree to which a study based on a sample applies to the
population as a whole.

23
Marketing Research

24
Marketing Research

3.3.4. Questionnaire design

A questionnaire is a structured technique for data collection consisting of a series of


questions that a participant answer. It must translate the information needed into a set of
specific questions that the participants can and will answer. It must uplift, motivate and
encourage the participant to become involved in the interview, to cooperate and to
complete the interview. It should minimise response error.

Specifying the information needed consists in ensuring that the information obtained fully
addresses all the components of the problem. Review components of the problem and the
approach, particularly the research questions, hypotheses and characteristics that influence
the research design. Prepare a set of dummy tables. Have a clear idea of the characteristics
and motivations of the target participants.

Specifying the type of interviewing method consists in reviewing the type of interviewing
method determined.

Determining the content of individual questions consists in avoiding unnecessary questions


and double-barrelled questions (questions that attempt to cover two issues).

25
Marketing Research

Overcoming the participant’s inability and unwillingness to answer consists in asking


questions that measure familiarity, product use and past experience (filter questions) before
questions about the topics to know if the participant is informed and avoid errors of
omission, telescoping and creation.

If the information needed is sensitive :


 Place sensitive topics at the end of the questionnaire
 Preface the question with a statement that the behaviour of interest is common
 Ask the question using the third-person technique
 Hide the question in a group of other questions which participants are willing to answer
 Use randomised techniques (if appropriate)

Choosing question structure consists in choosing the right question type to collect the
information needed.

A structured question is a question that pre-specify the set of response alternatives and the
response format used whenever possible.

An unstructured questions is an open-ended question that participants answer in their own


words used for exploratory research and as closing questions.

In a multiple-choice question, the response alternatives should include the set of all possible
choices and should be mutually exclusive.

In a dichotomous question (yes or no), if a substantial proportion of the participants can be


expected to be neutral, include a neutral alternative.

The split-ballot technique consists in randomly distributing different forms of questionnaires


for the same survey to reduce order bias in dichotomous and multiple-choice questions.

Choosing question wording :


 Use ordinary words (vocabulary level of the participants)
 Avoid ambiguous words (usually, normally, frequently, often, regularly, occasionally, …)
 Avoid leading questions (question that gives the participant a clue as to what answer is
desired) because it can lead to acquiescence bias (yea-saying)
 Avoid answers based on generalisations or computed estimates (complex questions)
 Use positive (and negative) statements

Arranging the questions in proper order :


 Use interesting, simple and non-threatening opening questions (qualifying questions)
 Place difficult, sensitive or complex questions at the end of the sequence
 Place general questions before specific questions (funnel approach)

Questions order :
1. Basic information (information related to the research problem)
2. Classification (socio-economic and demographic characteristics)
3. Identification information (name, address, phone number, …)

26
Marketing Research

A branching question is a question used to guide an interviewer (or participant) through a


survey by directing the interviewer (or participant) to different spots on the questionnaire
depending on the answers given. It should cover all possible contingencies, be placed as
close as possible to the question causing the branching and be ordered so that the
participants cannot anticipate the additional information.

Identifying the form and layout consists in dividing a questionnaire into several parts and
numbering each part and question. If hard copies of the questionnaire are used, coding
should be printed on the forms to facilitate manual data entry.

Reproducing the questionnaire :


 Design a visually engaging questionnaire
 Use vertical response columns
 Use grids for a several related questions that use the same set of response categories
 Place directions or instructions as close to the questions as possible
 Use booklet format for hard copies of long questionnaires.

Eliminating problems by pilot-testing consists in testing all aspects of the questionnaire


(question content, wording, sequence, form and layout, question difficulty, instructions and
rewards for taking part in the survey). The participants in the pilot-test should be few (15 to
30) and similar to those who will be included in the actual survey. After each significant
revision of the questionnaire, another pilot-test using a different sample should be
conducted. The responses should not be aggregated with responses from the final survey.

27
Marketing Research

3.3.5. Sampling

The objective of most marketing research projects is to obtain information about the
characteristics of a population. Information about population parameters may be obtained
by taking a census or a sample.

A census is a complete enumeration of the elements of the population or study objects.

A sample is a subgroup of the elements of the population selected for participation in the
study.

In a convenience sampling, participants are randomly selected to obtain a sample of


convenient elements. Often, participants are selected because they happen to be in the right
place at the right time.

In a judgemental sampling, participants are selected based on the judgement of the


researcher (form of convenience sampling).

In a quota sampling, participants are selected (convenience or judgemental sampling)


among control categories or quotas (two-stage restricted judgemental sampling).

In a snowball sampling, participants are selected by previous participants that belong to the
same target population.

28
Marketing Research

In a simple random sampling, participants are randomly selected with the same probability
of selection.

In a systematic sampling, participants are selected by using a random starting point and
then picking every ith element in succession from the sampling frame (to select a sample of 1
000 in a population of 100 000, you have to take one person every 100).

In a stratified sampling, participants are randomly selected (simple random sampling)


among subpopulations or strata (partitioned population). Every element is assigned to only
one stratum (mutually exclusive) and cannot be omitted (collectively exhaustive). The strata
are heterogeneous and the elements within are homogeneous.

29
Marketing Research

The confidence level is the probability that a confidence interval will include the population
parameter.

30
Marketing Research

The confidence interval is the range into which the true population parameter falls assuming
a given level of confidence.

The response rate is the percentage of the sample that has completed the interview.

Number of completed interviews


Response rate =
Number of eligible units in the sample

4. Survey fieldwork

Interviewers should be trained to make opening remarks that will convince potential
participants that their participation is important.

Asking the questions :


 Be thoroughly familiar with the purpose and structure of the research instrument
(interview guide or questionnaire)
 Ask the questions in the order of the questionnaire (more flexible with an interview
guide)
 Use the exact wording given in the questionnaire
 Read each question slowly
 Repeat questions that are not understood

31
Marketing Research

 Ask every applicable question


 Follow instructions, working through any filter questions and probe carefully

Probing (open or unstructured questions) :


 Repeat the question
 Repeat the participant’s reply
 Use a pause or silent probe
 Boost or reassure the participant
 Elicit clarification
 Use objective or neutral questions or comments

Recording the answers (unstructured questions) :


 Audio-Record responses during the in-depth interview
 Use the participant’s own words
 Do not summarise or paraphrase the participant’s answers
 Include everything that pertains to the question objectives
 Include all probes and comments
 (Repeat the response as it is written down)

The participant should be left with a positive feeling about the interview.

32
Marketing Research

5. Data integrity and analysis

5.1. Data integrity

A questionnaire returned from the field may be unacceptable for several reasons :
 Parts of the questionnaire may be incomplete
 The pattern of responses may indicate that the participant did not understand or follow
the instructions
 The responses may show little variance
 One or more pages may be missing
 (The questionnaire may be received after the pre-established cut-off date)
 The questionnaire may be answered by someone who is not qualify for participation

Editing consists in reviewing the (filled out) questionnaires with the objective of increasing
accuracy and precision.

Coding consists in assigning a code (usually a number) to each possible response to each
question (ID for each respondent).

The consistency checks identify data that are out of range, logically inconsistent or have
extreme values (outliers). Computer packages such as SPSS, SAS, SNAP, EXCEL and MINITAB
can be programmed to identify out-of-range values for each variable and print out the
participant code, variable code, variable name, record number, column number and out-of-
range value.

33
Marketing Research

5.2. Data analysis

5.2.1. Qualitative data analysis

The qualitative data analysis consists in analysing textual material or communications. The
text being analysed may come from the narrative held in brochures or advertising copy
through to dialogues held in interview data. It is considered as an objective, systematic and
quantitative description of the manifest content of a communication. The main objective is
to reduce the data (simplify) by summarising and structuring the data according to rules
derived from existing theory.

NVivo is a qualitative data analysis (QDA) computer software package produced by Lumivero
(formerly by QSR International) that helps qualitative researchers to organise, analyse and
find insights in unstructured or qualitative data like interviews, open-ended survey
responses, journal articles, social media and web content when deep levels of analysis on
small or large volumes of data are required.

5.2.2. Quantitative data analysis

The qualitative data analysis consists in analysing numerical data from surveys.

34
Marketing Research

A regression allows to predict an outcome variable (DV) from one (simple regression) or
several (multiple regression) predictor variables (IV).

A t-test allows to compare the means in 2 different conditions .

A between-subject design analyses different people in different conditions (independent t-


test).

A within-subject design analyses same people in different conditions (paired t-test).

A chi-square test allows to analyse the relationship between 2 nominal or ordinal variables
by testing the frequency (number of time they occur).

Conditions :
 Frequencies of each category > 5
 Independent design (no paired sample)
 2X2 contingency table (contains the frequencies of all categories)

35
Marketing Research

A cross-tabulation is a statistical technique that describes two or more variables


simultaneously and results in tables that reflect the joint distribution of two or more
variables.

Overall perceived risk (higher score = greater risk)


Product Catalogue Store
Shoes 58,60 50,80*
Pocket calculator 49,62 42,00*
Hi-fi 48,89 41,98*
Perfume 34,85 29,79*
* Significant at 0,01 level computed through paired t tests

An ANOVA (ANalysis Of VAriance) is a statistical technique for examining the differences


among means for two or more groups.

An ANCOVA (ANalysis of COvariance) is an advanced ANOVA procedure in which the


researcher statistically controls for the effects of other variables that are not of primary
interest, known as covariates (CV).

A multiple regression is a statistical technique that simultaneously develops a mathematical


relationship between two or more independent variables and a metric dependent variable.

A discriminant analysis (DA) is a statistical technique for classifying individuals or objects


into mutually exclusive and exhaustive groups on the basis of a set of independent variables.
It is used to anticipate customer churn and elaborate a profile of the relevant segments (low,
medium, heavy users).

36
Marketing Research

A conjoint analysis is a technique that attempts to determine the relative importance


consumers attach to salient attributes. It is used to best understand the criteria consumers
value when involved in a decision-making process.

A factor analysis is an interdependence technique in that an entire set of interdependent


relationships is examined without making the distinction between dependent and
independent variables. It is a general name denoting a class of procedures primarily used for
data reduction and summarisation.

A cluster analysis (CA) is a technique used to classify objects into relatively homogenous
groups called clusters. Objects in each cluster tend to be similar to each other and dissimilar
to objects in the other clusters. It is used to identify relevant criteria to segment a market.

A multidimensional scaling (MDS) is a class of procedures for representing perceptions and


preferences of participants spatially by means of a visual display. It is used to position
brands.

6. Communicating research findings

6.1. Data visualization


Research findings should be clearly communicated so they can be easily understood, and
managers are clear about the action they should take.

According to Edward Tufte (pioneer in data visualization), graphical displays should:


 Show the data
 Induce the viewer to think about the substance rather than about methodology
 Avoid distorting what the data has to say
 Present many numbers in a small space
 Make large data sets coherent
 Encourage the eye to compare different pieces of data
 Reveal the data at several levels of detail (from a broad overview to the fine structure)
 Serve a reasonably clear purpose (description, exploration, tabulation or decoration)
 Be closely integrated with the statistical and verbal descriptions of a data set

37

You might also like