Albert Einstein Quote
The FORMULATION of the problem is often more essential than it’s solution.
The Process of Problem Definition
Understand the Business Situation – Identify Key Symptom =>
Isolate the problem from the symptoms
Write Managerial Decision Statement and corresponding research purpose
Determine the unit of analysis?
Determine Relevant Materials
State research questions / objections / hypothesis
Limitations of Observations in General
– Observations can describe the event that occurred but cannot EXPLAIN why that event occurred.
– Observations over long periods is EXPENSIVE or even impossible
Researcher Needs to Understand
the decision makers objectives for conducting research.
the business context in which the decision maker operates objectives
informal gathering of background information to familiarize the researcher and manager with the decision-making environment situation.
1) Develop many alt problem statements
2) think about potential solutions
3) Use free-associated techniques to generate lists of ideas
4) be open minded
the sign (indicator) that something is wrong
the actual reasons (possible causes) for the discrepancy between what is (current situation) and what is desired.
Process of Problem Definition
1) Understand the Business Decision
2) Isolate Symptoms from Problem
3) Write Managerial Decision Statements and Corresponding Research Objectives
4) Determine Unit of Analysis
5) Determine Relevant Variables
6) State Research Questions and Objectives
capture the researchers understanding of the decision makers objectives (reasons) for seeking the research.
Unit of Analysis
The target population from whom data needs to be collected to serve the research objectives. (individuals, customers, competitors, etc..)
Anything that may assume different values among members of the target population.
Categorical of Continuos
Dependent or Independent
an unproven statement about the nature of relationships between variables or possible solution to a problem.
can take on a range of **quantitative** values.
Indicates membership in some group.
Also called CLASSIFICATORY variable.
a process outcome or a variable that is predicted and / or explained by other variables.
A variable that is expected to influence the dependent variable in some way.
Goals that the research is designed to achieve.
Questions for which the research is designed to provide answers.
The researchers translation of the managerial (marketing) problem into a specific inquiry.
A WRITTEN statement of the research design (strategy blueprint) that oftens also includes.
Contents of the Research PROPOSAL
Purpose of the research
Presentation of Results
For Management (the Client) // Functions of a Research Proposal
1) Evaluation: it helps the client to evaluate the ability of the research to accomplish its objectives.
2) Choosing research suppliers: client judges quality of different research suppliers from their proposals
3) Judgement Standard
It is a standard for determining if the research was conducted as planned.
Functions of a Research Proposal // FOR THE RESEARCHER
Thinking Ahead: it forces the researcher to think through the different stages of the research process.
Bidding Document: it serves as the researchers bid to offer a specific service.
Functions of a Research Proposal // For Both Parties
Communication: it serves as a means of communication between the researcher and the client (decision maker)
Written Contract: it serves as a written contract that can always be referred to.
the systematic process of recording actual behavioral patterns of people, objects, and events as they happen.
LIMITATIONS on Observation of Human Behavior
– Cognitive phenomena cannot be observed.
– Interpretation of data may be a problem.
– Not all activity can be recorded.
– Only short periods can be observed.
Observer BIAS possible.
– Possible invasion of privacy.
Nature of Observation Studies // VISIBLE OBSERVATION
observation in whiche the observers presence is known.
Nature of Observation Studies // HIDDEN Observation
observation in which the subject is unaware that observation is taking place.
Advantages of Observation over Surveying:
Data are free from distortions, inaccuracies, or other response biases
Data are recorded when actual and nonverbal behavior takes place.
Benefits of Observing Human Behavior
– Communication is not necessary
– Data not distorted by self-report bias
– No need to rely on memories
– nonverbal behavior data may be obtained
– certain data may be obtained more quickly
– environmental conditions can be recorded
– may be combined with survey to provide complementary evidence.
observation provides an additional source of information that helps explain other research findings.
the amount of time it takes to make a choice. Used as a measure of preference.
straightforward attempt to observe and record what naturally occurs.
Observation in which the investigator creats an artifical environment in order to test a hypothesis.
Env. may increase the frequency of certain behavior patterns to be observed.
a distortion of measurement resulting from the cognitive behavior or actions of a witnessing observer.
Record events subjectively
Record events inaccurately
Interpreting Observation data incorrectly.
Ethical Issues of Human Observation
Respondent’s right to PRIVACY
Contrived observation as ENTRAPMENT
Researchers feel comfortable collecting obersvation data if:
– commonly performed in public
– setting assures anonymity
– observed person has agreed to be observed
Observation of Physical Objects
Artifacts: made and consumed withing a culture that signals something meaningful.
Inventories: count and record physical inventories.
Content Analysis: systematic observation and quantitative description of manifest content of communication.
TV / Radio
Website Traffic (Click through rate)
Scanner based consumer panel
Measuring Physiological Reactions
Eye Tracking Monitor
Pupilometer (measures pupil)
Psychogalvanometer: measures involuntary changes in electrical resistance of skin. Assumes physiological changes accompany emotion.
Voice Pitch Analysis.
Easier to collect, many sources, and cheap to get
Secondary Data (What is it?)
recorded prior to and for a purpose other than the current research project.
Disadvantages of Secondary Data
– Maybe outdated.
– Units of Mesaurement might need to be converted.
– Lack of accuracy
primary data method based on communication with representative of sample of individuals.
people who verbally answer an interviewers questions
Survey that emphasized contacting respondents who are a repre. sample of target pop.
Census vs Sample
Type of Data in Market Segmentation?
Type of Data in Customer Satisfaction?
Type of Data in Advertising Testing
Advantages of Surveys
Disadvantages of Survey
Classifying Survey Research Methods
– By Method of Communication (Personal v Telephone)
– By Degree of Structure (Unstructured Disguise)
– By time Frame (Cross Sectional vs Longitudinal)
Limit on responses (MC)
No limit on response
assume respondent is willing to answer
Assume the purpose of the study must be hidden (drugs)
Undisguised (Direct) Unstructured
Survey with open ended questions to discover new answers
Cross Sectional Studies
data collected at single point in time (only once, but could be a six month window)
Data collected at different points in time using:
– Successive (different samples in a tracking or cohort study)
– same sample in panel study
Random Sampling Error
Systematic Error (BIAS)
Sample biases that result from respondent action or inaction
Error caused by improper admin (execution) of research tasks
stat differences between the results of a survey in which the sample includes only those responses and a survey that would include those who failed to respond for various reasons (not home, refusal, self selection)
deliberate falsification or unconscious misrepresentation
Acquiescence Bias (Response BIAS)
tendency to agree with everything interviewer says
Extremity Bias (Response BIAS)
tendency to use extremes in response
Interviewer Bias (Response BIAS)
interviews presence affect response
Auspices Bias (Response BIAS)
knowledge of sponsor of research affecting response
Social Desirability bias
acceptable answers rather than truth
Categories of Admin Error
Sample Selection Error: improper sample design or procedure
Data Processing Error: incorrect into or bad programming
Door to Door
Mall Intercept Interviews
Advantages of Personal Interviews
2) probe complex questions
3) more willing respondents, time wise
4) reduction in rate of nonresponse (completing questionnaire
5) possibility of visual aide
6) High response rate
Disadvantages of Personal Interviews
Respondents cannot be anon
maybe need for several callbacks
most widely used commercial survey method
Telephone Interviewing Systems
– Central Location Interview: normally through WATS (wide area telecom service). allows supervision and control
– Computer Assisted: (CATI): usually includes: phone number selection, auto dialing, sample selection, auto callback, reporting on # of completed interviews
– Computerized voice activated: computer assumes all responsibility for interviewr
Advantages of Telephone Interview
– high speed of data collect
– inexpensive compared to personal
– better anon
– releatively higher respondent cooperations and lower nonresponse compared to personal interview
DisAdvantages of Telephone Interview
– problems in getting good samples
– answering machines
– respondents ability to easily hangup
– inability to use visual aide
– need for shorter forms of questioning
– DO NOT CALL LISTS
respondent reads and fills out questions.
Distribution of Self Admin
– mail or postal
– drop off
Advantages of Self Admin
– geographic flexibility
– low cost
– respondent convenience
– respondent anon
DisAdvantages of Self Admin
– absence of interviewer (loss of control, misinterpretation, no visual aid)
– standardized questions
– takes long time to get data
– short questionnaire
low response rate
– no assurance that the person the survey was meant for actually took it
Response Rate Issues in Postal
number of completed and returned surveys divided by number of eligible people contacted
prestige of sponsor
color of paper
type of postage
Increase in response rate over control group:
Prior tele calls (19%)
Pocket Knife (15%)
Anon (in company) (20%)
Anon external (10%)
Follow ups (12%)
eh. bad cause people dont really have fax anymore but quicker and cheaper than trad mail
Electron Self Admin
2) Internet (Web HTML)
4) On site comp
5) Disk by Mail
Advantages of Electron Self Admin
1) Speed of Distribution
2) Low Cost
3) Faster response
4) greater flex
5) less paper
DisAdvantages of Electron Self Admin
2) lack of anon
3) different types of computers and users
4) difficulties for email novices
Advantages of Internet Survey
– cost effective
– visual appealing / interactive
– accurate real time data
– easy callback
– flexible / personalized
DisAdvantages of Internet Survey
– non widespread use of internet in general pop
– difference in capabilities / computer skills
– lack of security of personal info
Protesting Survey Instruments
– Pre-Testing: screening procedures that involves trial run with a group of responds to iron out problems
– Basic Way to PreTest:
>screen with other professionals
>have client review final questionnaire
>collect data from small pool
– protect public from misrepresentation and exploitation
– protect privacy
– no deception
– inform respondents of purpose
– matain honesty / confidentiality
– maintain objective