MARKETING RESEARCH

1
MARKETING RESEARCH

• 5th edition
• Alvin C. Burns Ronald F. Bush

2
Because Marketing Research is part of Marketing
we should understand

• What is marketing?
• What is the marketing concept?
• What is marketing strategy?

3
What is Marketing Strategy?

• A Marketing Strategy consists of selecting a
  segment of the market as the companys target
  market and designing the proper mix of the
  product/service, price, promotion, and
  distribution system to meet the wants and needs
  of the consumers within the target market.

4
Learning by Doing Lets Apply Marketing to a
Restaurant

• Target market segment?
• Marketing strategy
• Location?
• Menu?
• Prices?
• Type?
• Advertising?

5
Key Point

• To practice marketing to implement the marketing
  concepts to implement marketing strategy,
  managers must make decisions.
• Many decisions require additional information and
  marketing research is needed in order to supply
  that information.

6
What is Marketing Research?(Burns and Bush
Definition)

• Marketing research is the process of designing,
  gathering, analyzing, and reporting information
  that may be used to solve a specific marketing
  problem.

7
The Marketing Research Process 11 Steps

• Step One Establishing the Need for Marketing
  Research
• Step Two Defining the Problem
• Step Three Establishing Research Objectives
• Step Four Determining Research Design
• Step Five Identifying Information Types and
  Sources
• Step Six Determining Methods of Accessing Data

8
The Marketing Research Process 11 Steps

• Step Seven Designing Data Collection Forms
• Step Eight Determining Sample Plan and Size
• Step Nine Collecting Data
• Step Ten Analyzing Data
• Step Eleven Preparing and Presenting the
  Final Report

9
Steps in the Marketing Research Process

• Steps
• 1 Establish Need
• 2 Define Problem
• 3 Research Objectives
• 4 Determine Design
• 5 Identify Information Sources
• 6 Decide Data Collection Method
• 7 Design Questionnaire
• 8 Determine Sample Plan Size
• 9 Collect Data
• 10 Analyze Data
• 11 Write and Present Report

Figure out what to research (Chapters 2 4)
Design the way to do the research (Chapters 5-13)
Gather data from respondents (Chapter 14)
Generate findings and interpret them (Chapters
15-20)
10
The Marketing Research Process

• An 11-step process there is nothing
• Not all studies use all 11 steps
• Few studies follow the steps in order

magic
about 11 steps
11
Step 1 Establish the Need for Marketing Research

• Is there a real need for marketing research?
• Research takes time and costs money.
• Marketing research is not always needed.

12
Step 1 Establish the Need for Marketing Research

• When is marketing research not needed?
• The information is already available.
• Decisions must be made now.
• We cant afford research.
• Costs outweigh the value of marketing research.

13
Step 2 Define the Problem

• This is the most important of the 11 steps.
• If the problem is incorrectly defined, all else
  is wasted effort.
• Problems may be either specific or general.

14
Step 2 Define the Problem

• Problems stem from gaps between what is supposed
  to happen and what did happen and gaps between
  what did happen and what could be happening.

15
Step 3 Establish Objectives

• Research objectives, when achieved, provide the
  information necessary to solve the problem
  identified in step 2.
• Research objectives state what the researchers
  must do.

16
Step Four Determine Research Design

• Exploratory Research collecting information in
  an unstructured and informal manner.
• Descriptive Research refers to a set of methods
  and procedures describing marketing variables.
• Causal Research (experiments) allows isolation
  of causes and effects.

17
Step 5 Identify Information Types and Sources

• Primary information information collected
  specifically for the problem at hand
• Secondary information information already
  collected

18
Step 6 Determine Methods of Accessing Data

• Secondary data is relatively easy to access
  primary data is more complex.
• Three main choices for primary data
• Have a person ask questions
• Use computer assisted or direct questioning
• Allow respondents to answer questions themselves
  without computer assistance

19
Step 7 Design Data Collection Forms

• Questionnaire must be worded objectively,
  clearly, and without bias in order to communicate
  with respondents.
• Software programs are available to assist
  marketing researchers in preparing forms.

20
Step 8 Determine Sample Plan and Size

• Sample plan refers to the process used to select
  units from the population to be included in the
  sample.
• Sample size refers to determining how many
  elements of the population should be included in
  the sample.

21
Step 9 Collect Data

• Data collection is very important because,
  regardless of the data analysis methods used,
  data analysis cannot fix bad data.
• Nonsampling errors may occur during data
  collection.

22
Step 10 Analyze Data

• Data analysis involves entering data into
  computer files, inspecting data for errors, and
  running tabulations and various statistical tests.

23
Step 11 Prepare and Present the Final Research
Report

• The last step is one of the most important phases
  of marketing research.
• Its importance cannot be overstated because it is
  the report, or its presentation, that properly
  communicates the results to the client.

24
Data Analysis andDifference Tests

• Difference Tests are used a lot especially in
  segmentation. There are several types of
  difference tests
• One Sample T Test
• Independent Sample T Test
• ANOVA
• Paired Sample T Test

25
Summary

• Difference testing can be done with variables
  that are measured either categorically (nominal
  or ordinal) or metrically (interval or ratio).
• SPSS can do the stats if the measures are metric
• If the variable you want to analyze is measured
  categorically, you need to get the basic data
  from SPSS and then use the formulas from the book
  and your calculator
• Our book does NOT provide formulae for the manual
  calculation of statistical differences for
  Paired Sample T Tests or ANOVA. These formula
  can be found in other books, though.
• Our book DOES provide formulae for the manual
  calculation of One Sample and Independent Sample
  T Tests .

26
Summary

• When to use what
• Identify how the question is measured nominal,
  ordinal, interval, ratio
• If nominal or ordinal, use the formulae and your
  calculator once you get the descriptive stats out
  of SPSS
• If interval or ratio, use SPSS alone

27
Summary

• When to use what
• 2. Identify What Test To Use
• If you are testing a hypothesis (like, Patrick
  believes the average amount of money people
  expect to pay for an entree is 18), then use the
  One Sample T Test approach and test if the sample
  mean is different (enough) from Patricks
  hypothesis.
• If you are testing for differences between how
  TWO different groups answer a question (like if
  males gave different answers than females), then
  use the Independent Sample T Test.

28
Summary

• When to use what
• 2. Identify What Test To Use
• If you are testing how MORE THAN TWO groups
  answered a question (like how answers differ
  depending on what radio station people listen
  to), then use ANOVA which will tell you if ANY of
  the two-group differences are statistically
  significant. If they are, use the Duncan test to
  see what groups differ.
• If you are testing how ONE GROUP answers a pair
  of questions , then use a PAIRED SAMPLE T Test

29
Summary (or, So what?)

• If the Sig. column is .05, then there IS a
  Statistically significant difference between
  what (ever) you have tested
• Between what you hypothesized and found (One
  Sample T Test)
• Between two independent groups such as males and
  females answers to a question (Independent
  Sample T Test)
• Between the answers given by multiple groups like
  those who listen to Country, versus the answers
  of those who listen to Talk/News, versus Rock
  versus Jazz (ANOVA)
• Between the answers given by the same people
  concerning their preference for waterfront dining
  versus how much they prefer wait staff in tuxedos
  (Paired Sample T Test)
• A caution big samples (n) mean lots of
  statistically significant differences so be
  careful. Just because there is a statistically
  sig. difference doesnt mean that the marketer
  should necessarily do anything about it. It has
  to also be managerially significant.

30
Tests of Association

• We also want to know about what relationships
  exist between different variablesamong people
  who want a drive of less than 30 minutes, what
  menu/décor preferences are important?
• If variables are nominal, use cross tabs (and the
  Chi Square).
• If variables are metric, use correlations (and
  Pearsons r)