03_desain_riset

8/14/2019 03_Desain_Riset

1/30

Rahmad 2006 1

Pertemuan 5

Oleh :


2/30

Research Design

Conclusive

Research Design

Descriptive

Research

Casual

Research

Cross-Sectional

Design

Longitudinal

Design

Single Cross-

Sectional Design

Multiple Cross-

Sectional Design

Exploratory

Research Design

A Classification of Marketing Research Designs


3/30

Objective:

Character-istics:

Findings/Results:

Outcome:

To provide insights andunderstanding.

Information needed is definedonly loosely. Research process isflexible and unstructured. Sampleis small and non-representative.Analysis of primary data isqualitative.

Tentative.

Generally followed by furtherexploratory or conclusiveresearch.

To test specific hypotheses andexamine relationships.

Information needed is clearlydefined. Research process is formaland structured. Sample is large andrepresentative. Data analysis isquantitative.

Conclusive.

Findings used as input into decisionmaking.

ExploratoryExploratory ConclusiveConclusive

Difference between Exploratory and Conclusive Research


4/30

ObjectiveObjective::

CharacteristicsCharacteristics::

MethodsMethods::

Discovery of ideasDiscovery of ideasand insightsand insights

Flexible, versatileFlexible, versatile

Often the front endOften the front end

of total researchof total research

designdesign

Expert surveysExpert surveysPilot surveysPilot surveys

Secondary dataSecondary data

Qualitative researchQualitative research

Describe marketDescribe marketcharacteristics orcharacteristics or

functionsfunctions

Marked by the priorMarked by the prior

formulation of specificformulation of specific

hypotheseshypotheses

Preplanned andPreplanned and

structured designstructured design

Secondary dataSecondary data

SurveysSurveys

PanelsPanels

Observation and otherObservation and other

datadata

Determine cause andDetermine cause andeffect relationshipseffect relationships

Manipulation of oneManipulation of one

or more independentor more independent

variablesvariables

Control of otherControl of other

mediating variablesmediating variables

ExperimentsExperiments

ExploratoryExploratory DescriptiveDescriptive CausalCausal

A Comparison of Basic Research DesignsA Comparison of Basic Research Designs


5/30

Non-sampling

Error

Non-sampling

Error

Non-response

Error

Non-response

Error

Interviewer

Errors

Interviewer

Errors

Respondent

Error

Respondent

Error

Researcher

Error

Researcher

Error

Total Error

Total Error

Random Sampling

Error

Random Sampling

Error

Surrogate Information Error

Measurement Error

Population Definition Error

Sampling Frame ErrorData Analysis Error

Surrogate Information Error

Measurement Error

Population Definition Error

Sampling Frame Error

Data Analysis Error

Respondent Selection Error

Questioning Error

Recording Error

Cheating Error

Respondent Selection Error

Questioning Error

Recording Error

Cheating Error

Inability Error

Unwillingness Error

Inability Error

Unwillingness Error

Response

Error

Response

Error

Potential Sources of Error in Research DesignsPotential Sources of Error in Research Designs


6/30

Rahmad 2006 6

Exploratory Research Design:Exploratory Research Design:Secondary DataSecondary Data


7/30

Primary DataPrimary Data Secondary DataSecondary Data

Collection purpose For the problem at hand For other

problems

Collection process Very inoled Rapid ! easy

Collection cost "igh Relatiely lo#Collection time $ong Short

% Comparison o& Primary ! Secondary Data% Comparison o& Primary ! Secondary Data


8/30

CriteriaCriteria 'ssues 'ssues Remar(s Remar(s

Speci&ications! methodology

Error !

%ccuracy

Currency

Ob)ectie

*ature

Dependability

Data collection method+ responserate+ ,uality ! analysis o& data+

sampling techni,ue ! si-e+

,uestionnaire design+ &ield #or(.

Examine errors in approach+

research design+ sampling+datacollection ! analysis+ ! reporting.

/ime lag bet#een collection !

publication+ &re,uency o& updates.

0hy #ere the data collected1

De&inition o& (ey ariables+ units o&

measurement+ categories used+

relationships examined.

Expertise+ credibility+ reputation+

! trust#orthiness o& the source.

Data should be reliable+alid+ ! generali-able to

the problem.

%ssess accuracy by

comparing data &romdi&&erent sources.

Census data are updated

by syndicated &irms.

/he ob)ectie determines

the releance o& data.

Recon&igure the data to

increase their use&ulness.

Data should be obtained

&rom an original source.

Criteria &or Ealuating Secondary DataCriteria &or Ealuating Secondary Data


9/30

% Classi&ication o& Secondary Data% Classi&ication o& Secondary Data

Secondary Data

'nternal External

Ready

to 2se

Re,uires

Further

processing

Published

3aterials

Computeri-ed

Databases

Syndicated

Serices


10/30

StatisticalData

Published

Secondary Data

4eneral usinessSources

4oernmentSources

4uides Directories 'ndexes CensusData

Other4oernment

Publications

% Classi&ication o& Published Secondary Sources% Classi&ication o& Published Secondary Sources


11/30

ibliographicDatabasesibliographicDatabases

Computeri-ed

Databases

On6$ine O&&6$ine

*umericDatabases

Full6/extDatabasesFull6/extDatabases

DirectoryDatabasesDirectoryDatabases

Special6Purpose

Databases

Special6Purpose

Databases

% Classi&ication o& Computeri-ed Databases% Classi&ication o& Computeri-ed Databases

'nternet


12/30

2nit o&

3easurement

% Classi&ication o& Syndicated Serices% Classi&ication o& Syndicated Serices

"ouseholds7

Consumers

'nstitutions


13/30

Syndicated Serices: ConsumersSyndicated Serices: Consumers

"ouseholds 7

Consumers

Sureys

3ail Diary

Panels

Electronic

scanner sericesPurchase 3edia

Psychographic

& Lifestyles General

Advertising

Evaluation

olume

!rac"ing #ata

$canner #iary

Panels

$canner #iary

Panels %ith

a'le !


14/30

Syndicated Serices: 'nstitutionsSyndicated Serices: 'nstitutions

'nstitutions

Retailers 0holesalers 'ndustrial &irms

%udits

Direct

'n,uiries

Clipping

SericesCorporate

Reports

Corporate

Reports


15/30

Type Characteristics Advantages Disadvantages UsesSurveys Surveys conducted

at regular intervals

Most flexible way of

obtaining data;

information on

underlying motives

Interviewer errors;

respondent errors

Market

segmentation,

advertising theme

selection and

advertising

effectiveness

Diary

PurchasePanels

Households provide

specific informationregularly over an

extended period of

time; respondent

asked to record

specific behaviors as

they occur

Recorded purchase

behavior can belinked to the

demographic /

psychographic

characteristics

Lack of

representativeness;response bias;

maturation

Forecasting sales,

market share andtrends; establishing

consumer profiles,

brand loyalty and

switching;

evaluating test

markets, advertising,

and distribution

Diary

Media

Panels

Electronic devices

automatically

recording behavior,

supplemented by a

diary

Same as diary

purchase panel

Same as diary

purchase panel

Establishing

advertising rates;

selecting media

program or air time;

establishing viewer

profiles

Oerie# o& Syndicated SericesOerie# o& Syndicated Serices

$an)utan$an)utan


16/30

/ype Characteristics %dantages Disadantages 2ses$canner

olume

!rac"ing

#ata

(ousehold

purchases are

recorded through

electronic scanners

in supermar"ets

#ata reflects actual

purchases) timely

data* less e+pensive

#ata may not 'e

representative)

errors in recording

purchases) difficult

to lin" purchases toelements of

mar"eting mi+ other

than price

Price trac"ing*

modeling*

effectiveness of in,

store promotions

$canner

#iary

Panels

%ith &a'le

!

$canner panels of

households that

su'scri'e to ca'le

!

#ata reflect actual

purchases) sample

control) a'ility to

lin" panel data to

householdcharacteristics

#ata may not 'e

representative)

-uality of data

limited

Promotional mi+

analyses* copy

testing* ne% product

testing* positioning

Audit

services

erification of

product movement

'y e+amining

physical records or

performing

inventory analysis

Relatively precise

information at the

retail and %holesale

levels

&overage may 'e

incomplete)

matching of data on

competitive activity

may 'e difficult

.easurement of

consumer sales and

mar"et share*

competitive activity*

analy/ing

distri'ution patterns0

trac"ing of ne%

products

1ndustrial

Product

$yndicated

$ervices

#ata 'an"s on

industrial

esta'lishments

created through

direct in-uiries of

companies* clippingservices* and

cor orate re orts

1mportant source of

information on

industrial firms*

particularly useful in

initial phases of the

pro2ects

#ata are lac"ing in

terms of content*

-uantity* and -uality

#etermining mar"et

potential 'y

geographic area*

defining sales

territories* allocating

advertising 'udget

$an)utan8$an)utan8


17/30

'nternational

Organi-ations

DomesticOrgani-ations in

the 2nited States

DomesticOrgani-ations in

the 2nited States

'nternational Secondary Data

'nternationalOrgani-ations in

the 2nited States

Organi-ations in

Foreign Countries

4oernment

Sources

*ongoernment

Sources4oernments

/rade

%ssociations

% Classi&ication o& 'nternational Sources% Classi&ication o& 'nternational Sources


18/30

Rahmad 2006 13

Descriptive Research Design :Descriptive Research Design :

Survey and ObservationSurvey and Observation


19/30

SureySurey3ethods3ethodsSureySurey

3ethods3ethods

/elephone/elephone PersonalPersonal 3ail3ail

'n6"ome'n6"ome 3all3all

'ntercept'ntercept

Computer6%ssistedComputer6%ssisted

PersonalPersonal

'nterie#ing'nterie#ing

/raditional/raditional

/elephone/elephoneComputer6%ssistedComputer6%ssisted

/elephone/elephone

'nterie#ing'nterie#ing

3ail3ail

'nterie#'nterie#3ail3ail

PanelPanel

A Classification of Survey MethodsA Classification of Survey Methods

ElectronicElectronic

E6mailE6mail 'nternet'nternet

S D i i R l d hS D i i R l t d t th


20/30

Outgoing Envelope4utgoing envelope si/e* color* return address

Postage .ethod of addressing

Cover Letter

$ponsorship !ype of appeal PostscriptPersonali/ation $ignature

QuestionnaireLength $i/e Layout 5ormat

ontent Reproduction olor Respondent anonymity

Return Envelope!ype of envelope Postage

Incentives.onetary versus non,monetary Prepaid versus promised amount

Some Decisions Related to theSome Decisions Related to the

Mail Interview PackageMail Interview Package


21/30

A lassification of 4'servation .ethods

Obseration 3ethods

Personal

Obseration

3echanical

Obseration%udit

Content

%nalysis

/race

%nalysis

lassifying

4'servation

.ethods


22/30

Criteria Personal 3echanical %udit Content /race

Obseration Obseration %nalysis %nalysis %nalysis

Degree o& structure $o# $o# to high "igh "igh 3edium

Degree o& disguise 3edium $o# to high $o# "igh "igh

%bility to obsere "igh $o# to high "igh 3edium $o#

in natural setting

Obseration bias "igh $o# $o# 3edium 3edium%nalysis ias "igh $o# to $o# $o# 3edium

3edium

4eneral remar(s 3ost &lexible Can be Expensie $imited to 3ethod o&

intrusie communications last

resort

A Comparative Evaluation of Observation MethodsA Comparative Evaluation of Observation Methods


23/30

riteria !elephone Personal .ail Electronic

(igh sample control , ,#ifficulty in locating ,

respondents at home

naccessi'ility of homes ,

7navaila'ility of a large ,

pool of trained intervie%ersLarge population in rural areas , , ,

7navaila'ility of maps ,

7navaila'ility of current , ,

telephone directory

7navaila'ility of mailing Lists , Lo% penetration of telephones , ,

Lac" of an efficient postal system ,

Lo% level of literacy , , ,

5ace,to,face communication culture , , ,

Poor access to computers and nternet8 8 ,

9ote A denotes an advantage* and a , denotes a disadvantage:

A Comparative Evaluation of Survey Methods forA Comparative Evaluation of Survey Methods for

International Marketing ResearchInternational Marketing Research


24/30

Rahmad 2006 2;

Causal Research Design:Causal Research Design:ExperimentationExperimentation


25/30

% Classi&ication o& Experimental Designs% Classi&ication o& Experimental Designs

Experimental Designs

Pre6experimental/rue

Experimental

9uasi

Experimental Statistical

One6Shot Case

Study

One 4roup

Pretest6Posttest

Static 4roup

Pretest6Posttest

Control 4roup

Posttest: Only

Control 4roup

Solomon Four6

4roup

/ime Series

3ultiple /ime

Series

Randomi-ed

loc(s

$atin S,uare

Factorial

Design


26/30

Treatment GroupsBlock Store Commercial

Commercial Commercial

Number Patronage A BC

1 Heavy

2 Meium! "o#$ None

%n Example o& a Randomi-ed loc( Design


27/30

%nterest in t&e StoreStore Patronage Hig&

Meium "o#

Heavy B AC

Meium C BA

"o# an none A CB

%n Example o& $atin S,uare Design%n Example o& $atin S,uare Design


28/30

Amount o' Humor

Amount o' Store No MeiumHig&%n'ormation Humor HumorHumor "o#

MeiumHig&

%n Example o& a Factorial Design%n Example o& a Factorial Design


29/30

(actor "aboratory(iel

Enironment %rti&icial Realistic

Control "igh $o#

Reactie Error "igh $o#

Demand %rti&acts "igh $o#

'nternal Validity "igh $o#

External Validity $o# "igh

/ime Short $ong

*umber o& 2nits Small $argeEase o& implementation "igh $o#

Cost $o# "igh

$aboratory ersus Field Experiments$aboratory ersus Field Experiments

C i i & S i &


30/30

Criteria &or the Selection o& /est 3ar(etsCriteria &or the Selection o& /est 3ar(ets

/est 3ar(ets should hae the &ollo#ing ,ualities:

1< =e large enough to produce meaningful proections: !hey

should contain at least 2> of the potentialactual

population: 2< =e representative demographically:?< =e representative %ith respect to product

consumption 'ehavior:

;< =e representative %ith respect to media usage:

@< =e representative %ith respect to competition:

6< =e relatively isolated in terms of media and physical distri'ution:

< (ave normal

historical development in the product class 3< (ave

03_desain_riset

Documents