• Shuffle
    Toggle On
    Toggle Off
  • Alphabetize
    Toggle On
    Toggle Off
  • Front First
    Toggle On
    Toggle Off
  • Both Sides
    Toggle On
    Toggle Off
  • Read
    Toggle On
    Toggle Off

Card Range To Study



Play button


Play button




Click to flip

Use LEFT and RIGHT arrow keys to navigate between flashcards;

Use UP and DOWN arrow keys to flip the card;

H to show hint;

A reads text to speech;

95 Cards in this Set

  • Front
  • Back
Research Design
The framework or plan for a study that guides the collection and analysis of the data
Once research problem(s) are formalized, the next question to ask is
“Exactly what information is needed to solve the research problem?”
The research design:
(1) is driven by the research problem
(2) depends upon how much is known about the problem
Types of Research Design
Qualitative research

Quantitative research
Qualitative research
Goal: Explore
Quantitative research
Descriptive Research
Goal: Describe
Causal Research
Goal: Establish Cause and Effect
Qualitative (Exploratory)
For example: if you want to know more about Venezuelan culture ask a Venezuelan
But the Venezuelan you ask will not perfectly represent ALL Venezuelan people
That’s why you need Quantitative research
Appropriate for seeing if your understanding is correct for the entire group that you want to study
Uses statistics to make guesses about an entire group from asking only a small part of the group
For example:
Your interview with Joey from the University of Texas has shown that he loves driving his truck in the mud.
You want to see whether the majority of people from UT also tend to love driving in the mud.
It takes too long to ask everyone attending UT!
So you just ask a few people and make an educated guess about everyone else—using statistics
Types of Research Design
Exploratory Research Design
What might be important to consider?
Descriptive Research Design
What are the characteristics of…?
Causal Research Design
If we do this, what will
be the impact?
Why conduct exploratory research?
Better formulate the manager’s decision problem
Increase researcher’s familiarity with the problem
Clarify concepts
Develop hypotheses
Exploratory research works to develop a “theory” regarding the research problem at hand
Literature Search
A search of statistics, trade journal articles, other articles, magazines, newspapers, and books for data or insight into the problem at hand (i.e., qualitative secondary data)

One common—but time intensive—methodology: content analysis
Ex: Collaboration in business
Depth Interviews
Open-ended interviews with people knowledgeable about the general subject being investigated

Some possibilities:
those who work with it (e.g., employees, consultants)
those who study it (e.g., researchers, analysts)
those who live it (e.g., consumers)

We will talk more about this next lecture
Focus Group
Group interview facilitated by a moderator
Participants playing off each other Typically 8 – 12 people
Homogeneous within group
1.5 to 2 hours in length
Sessions recorded and transcribed
Case Analyses
Intensive study of selected examples of the phenomenon of interest
Can include observation
Used for benchmarking, reality checks

Ethnographic study in stores video (RV1)
Especially effective with cases reflecting...
...recent change
...extremes of behavior
...the “best” and “worst” situations
Projective Methods
Indirect method of getting someone’s opinion
Useful when respondents do not want to talk about a subject
For example:
When respondents do not fully trust the researcher
For studying illegal or unethical behavior
For studying taboo or embarrassing topics

Content analysis or a similar technique is then used to find underlying meaning
Projective techniques
Role playing

Word association
What words best describe how you feel about each logo?

Sentence completion
People who ride bikes are ________.
Why bother with quantitative research?
To make guesses about a larger population
Cheaper and easier than conducting a census
Provides evidence
Exploratory research is very flexible; descriptive research is MUCH more rigid
Descriptive research requires the clear specification of…
-- before data collection can begin.
Longitudinal Analysis
Continuous Panels
Discontinuous Panels
Continuous Panels
Measurements taken on same items from same people over time.
Discontinuous Panels
Measurements taken on different items from same people over time.
Essentially it is the same group of respondents over a variety of projects.
Sample Survey

(Cross-Sectional Analysis)
Study in which the sample is selected to be representative of the target population and in which the emphasis is on the generation of summary statistics such as averages and percentages.
Causal Research
The purpose of causal research is to test cause and effect relationships:
X --> Y
condition X causes event Y
Casual: Why questions (does x cause Y)
Do people who are in love buy more candy?
More love More candy
Does using fear tactics in commercials actually reduce smoking?
More fear Less smoking
Do younger people prefer sweeter food?
Lower age More preference for sweet food
Do people with bad credit refinance their homes more often?
Worse credit More refinancing
Hypothesis: A statement that specifies how two or more measurable variables are related.
(Look at slide for example.)
Consistent variation
evidence of the extent to which X and Y occur together or vary together in the way predicted by the hypothesis
Time order
evidence that shows X occurs before Y
Elimination of other explanations
evidence that allows the elimination of factors other than X as the cause of Y

X—the cause (independent variable)
Y—the effect (dependent variable)
Experimental design is the “correct” way of testing causality
It is often not practical to actually conduct in business
Therefore, we often use surveys instead
A test under controlled conditions that is made to examine the validity of a hypothesis
Assign people into groups
Condition group vs. Control group
Internal validity:
Internal validity: extent to which observed effect can be attributed to experimental variable and not to other factors. Lab experiments tend to have higher levels of internal validity.
External validity:
External validity: extent to which observed effect can be generalized to particular populations and situations. Field experiments tend to have higher levels of external validity.
Secondary Data
Information not gathered for the immediate study at hand but for some other purpose.
Primary Data
Information collected specifically for the investigation at hand.
The Balancing Act with Secondary Data

Assessing the Accuracy of Secondary Data
Primacy of Source
Primary Source
Secondary Source
Purpose of Publication
Who sponsored the research?
General Evidence of Quality
Ability of researchers
Process of data collection
Types of Secondary Data
Internal Data
External Data
Internal Data
Data that originate within the organization for which the research is being done.

Most studies should begin with a search for internal secondary data.
External Data
Data that originate outside the organization for which the research is being done.
Types of Data Collection
Secondary Data
Primary Data
Internal Secondary Data Sources
Cash register receipts
Sales invoices
Financial records
Credit memos
Salespersons’ call reports
Salespersons’ customer/prospect records
Warranty cards
Salespersons’ expense accounts
Previous marketing research reports
Analyzing “thick” data
If the data consists of numbers, you can analyze it using statistics
If the data consists of words, you should:
“Convert” it to numbers
Try to draw conclusions directly
But your conclusions will be biased
Initial coding
Focused coding
Initial coding
the first stage in classifying and assigning meaning to pieces of information for data analysis.
Numerous codes are generated while reading through responses without concern for the variety of categories.
Focused coding
the second stage of classifying and assigning meaning to pieces of information for data analysis.
Coding categories are eliminated, combined, or subdivided, and the researcher identifies repeating ideas and larger underlying themes that connect codes.
Content Analysis
What is measured? Any or all of the following can be measured:
how many times does something occur, eg. a particular word, theme, stereotype?
the direction of the messages, e.g. positive or negative; supporting or opposed
the strength or power of the direction, e.g. strongly supporting  strongly opposed
how much space does a particular message occupy (e.g. paragraphs in an article, duration in a film or advertisement)
Content Analysis - quantitative
Codes need to be clearly defined so categories are comprehensive and mutually exclusive

Pilot studies are useful to check coding categories

If multiple coders are used, coding must be cross-checked to ensure consistency

Common types of coding categories
Setting/Context codes provide background information on the setting, topic, or subjects.
Defining the Situation

Common types of coding categories
Defining the Situation codes categorize the world view of respondents and how they see themselves in relation to a setting or your topic.
Respondent Perspective

Common types of coding categories
Respondent Perspective codes capture how respondents define a particular aspect of a setting. These perspectives may be summed up in phrases they use, such as, "Say what you mean, but don't say it mean."
Respondents' Ways of Thinking about People and Objects

Common types of coding categories
Respondents' Ways of Thinking about People and Objects codes capture how they categorize and view each other, outsiders, and objects. For example, a dean at a private school may categorize students: "There are crackerjack kids and there are junk kids."
What to do with the coded data?
Try to create hypotheses my connecting the coded data
For example:
You notice a theme about going to parties in your interviews
Marketing majors tend to have a positive attitude towards parties while Finance majors do not
One hypothesis would be:
H1: marketing majors like going to parties more than finance majors.
Why Use Observation Research?
Observation is often the best method for generating valid data about individuals’ behavior.
Structured Observation
The problem has been defined precisely enough so that behaviors that will be observed can be specified beforehand, as can the categories that will be used to record and analyze the situation.
Unstructured Observation
The problem has not been specifically defined, so a great deal of flexibility is allowed the observers in terms of what they note and record.
disguised observation
With disguised observation, subjects are not aware that they are being observed (e.g., mystery shoppers).
Ethical considerations  Debriefing
undisguised observation
With undisguised observation, subjects know that they are being observed.
Natural Setting
Subjects are observed in the environment where the behavior normally takes place

Shopping in a store
Using or consuming a product at home
Contrived Setting
Subjects are observed in an environment that has been specially designed for recording their behavior

“fake” store
computer simulation
human observation
With human observation, individuals are trained to systematically observe a phenomenon and to record on the observational form the specific events that take place.
mechanical observation
With mechanical observation, a mechanical device observes the phenomenon and records the events that take place.
Examples of Mechanical Observation
Video cameras
Bar code scanners
Response latency
Voice-pitch analysis
Eye camera
Purpose of Interviewing: to capture the other person’s perspective
Main concerns:
Advantages of using interviews
Enable us to learn about things that cannot be directly observed
Add an inner perspective to outward behaviors
Allows for probing
Respondents can raise concerns
Enable modification to lines of inquiry
Refining questions
Disadvantages of using interviews
Very time consuming
Usually sample size is small
Quality of information obtained is largely dependent upon the interviewer
Subject to interviewer bias
Subject to respondent bias
Before the interview
Research blueprint:
Sample size
Interview structure
Interview medium
Number of interviews per person
Interview analysis method
How big should an interview sample size be?
Remember, interviews are not generalizable to a greater population
Therefore, sample size can be determined either by:
Interviewing until information becomes redundant
Best for looking for generalizable trends
2 interviews per relevant category
Best for looking for differences between groups or exceptions to the norm
Mediums for interviewing
Approaches to Qualitative Interviewing:
Informal conversational interview (unstructured)
General interview guide approach
Standardized open-ended interview (structured)
Unstructured interview:
Resembles a conversation
informants may sometimes forget that they are being interviewed
Most of the questions asked will flow from the immediate context
Useful for exploring interesting topics for investigation
Typical of ‘ongoing’ participant observation fieldwork
Semi-structured interview (commonly called guided interview)
Basic checklist of questions are prepared
To make sure all relevant topics are covered.
The interviewer is still free to explore, probe and ask questions.
Useful for eliciting information about specific topics.
Allows for in-depth probing
WHILE still keeping the interview within the parameters of the study.
The standardized open-ended interview
A set of open-ended questions
Carefully worded and arranged
For the purpose of minimizing variation in the questions posed to the interviewees.
Useful when two or more researchers are involved in the data collecting process.
Less flexibility
Probing is less possible
How many interviews per participant?
Participants can be interviewed multiple times
Increase rapport between respondent and interviewer
Prevents boredom when a lot of information is needed
More difficulty securing permission
Different interviews can be different styles based on your needs
Multiple interviewers
Taping interviews
Content analysis
During the interview...
Encourage a free rein but maintain control
Don’t lose sight of your research interest
Gently guide respondents through probing questions if they get off track
Try not to interrupt
During the interview...
1. Ask clear questions
easy to understand
devoid of jargon
2. Ask single questions
3. Ask truly open-ended questions
Do not pre-determine the answers
Satirical example of predetermining an answer (from the Colbert Report):
“George Bush: Great president or the greatest president?
During the interview
Ask experience/behavior questions before opinion/feeling questions
Sequence the questions
Using funneling
Ex: All babies cry, of course. Some mothers feel that if you pick up a baby every time it cries, you will spoil it. Others think you should never let a baby cry for very long.
Broad: How do you feel about this?
Specific: How about the middle of the night?
More specific: What about on nights when you have to work the next day?
Probe & follow-up questions
The purpose of probing is to deepen the response to a question
Interpret questions
Researchers should clarify interviewee’s statements to avoid misinterpretations.
Ex: “Is it correct that you feel that……?";
During the interview
Probe & follow-up questions
Interpret questions
Follow up on Non-verbal cues such as laughter
Tolerate silence
5 probing techniques
Procuring details
Faking puzzling
Interject understanding
Procuring details
Procuring details
Asking further questions to see if more information can be obtained.
Faking puzzling
Faking puzzling
Pretending to be confused, indicating elaboration is needed.
Giving compliments to encourage the informants to carry on.
Interject understanding
Interject understanding
Making the informant know that his/her comments are understood and treasured + allowing him/her time for further comments.
Repeating the informant’s answer to show attention.
Ethical considerations of interviewing
Be careful to ensure sensitive information won’t hurt the respondent
Get permission of guardian when interviewing children
Institutional review board (IRB)
For research that is intended to be published
MUST get IRB approval