Textbook Notes (290,000)
CA (170,000)
U of G (10,000)
MCS (700)
Chapter 1

MCS 3030 Chapter Notes - Chapter 1: Postpositivism, Multidimensional Scaling, Cluster Analysis

Marketing and Consumer Studies
Course Code
MCS 3030

of 7
Chapter 1: Foundations
Background on Research:
Examples of Applied Questions:
Politics- “What issue is most important to my
Small business owner- “Where is the best
location for me to open my daycare?
Large Corporation- “If we close this factory will
it hurt demand for our products?”
Deontological Approach- philosophy regarding obligation rather than in the maximization of some good
Teleological Approach- the perception of purposeful development toward an end
Research method idiosyncrasies
1.The language of research - “Language shapes the way we think and determines what we can think
Mixed methods- any research that uses multiple research methods to take advantage of the unique
advantages each method offers (ie. Study that combines interviews and experimental design)
Theory- A plausible or scientifically acceptable general principle or body of principles offered
to explain the phenomena
Unit of analysis
-unit of analysis- entity your analyzing in analysis (individuals, groups, social interactions (divorces,
arrests) geographical units (town, census tract, state))
-hierarchical modeling- incorporation of multiple units of analysis at different levels of hierarchy within a
single analytic model (comparing student performance and teacher expectations requires collecting data at teacher
and student level)
-helpful in sorting out the effects and relationships in a more comprehensive way that
reflects the ‘big picture’
Constructs (concepts) and variables
Variables- what we use to measure the phenomenon (any entity that can take on different
--an observation where the values can vary
-ie. Salary, sales, gender, age, agreement, trust, convenience, likability, paranoia, satisfaction
-third-variable problem- an unobserved variable that accounts for a correlation between
two variables
-Quantitative-measured using numbers
-Qualitative- non numerical (ie. Gender)
-Attribute- specific value of a variable (gender has two attributes male and female)
-there are two traits variables should always be achieved:
1) exhaustive- property of a variable that occurs when you include all possible
answerable responses
2) mutually exclusive no respondent should be able to have two attributes
simultaneously (gender)
1) Measurement characteristics:
2) Causal role the variables play
-Independent (predictor or explanatory) variable that you manipulate
-Dependent (outcome) variable affected by independent variable
-Key words: causes, increases, is related to
-ie. “smoking (IV) causes cancer (DV); Loyalty’s (IV) related to profitability (DV)
*four possible relationships: 1. None, 2. Positive, 3. Negative, 4. Curvilinear
Time in Research:
-time is important element in any research design
-cross-sectional study- a study that takes place at a single point in time
-longitudinal study- a study that takes place over time
-repeated measures- at least 2 waves of measurement over time
-time series-many waves of measurement over time (typically at least 20 waves)
Hypothesis- specific statement of prediction
-what we think will happen in general based on theory or observation
-what we think will happen between the IV and DV variables
-testable hypothesis has 2 parts:
1. null hypothesis (H0) “a hypothesis may be rejected but can never be
accepted except tentatively because further evidence may prove it wrong”
-All remaining possible explanations
2. Alternative hypothesis (H1 or HA) my prediction that I personally support
(what you expect to will happen in your study)
one-tailed hypothesis- hypothesis that specifies direction (the pill will
decrease depression)
two-tailed hypothesis- hypothesis that doesn’t specify direction (pill
will impact depression)
Hypothetical-deductive model- model in which two mutually exclusive hypothesis that
together exhaust all possible outcomes are tested, such that if one hypothesis is
accepted, the second must therefore be rejected
Reject or Fail to Reject:
-Scientific method is about ruling out alternatives (deductive reasoning model)
-We don’t directly prove our hypothesis
-We reject other alternatives in favor of our hypothesis
-We reject the null hypothesis or we fail to reject the null
-if you reject the null you accept (temporarily) the alternative
-if you “fail to reject the null” then you are saying you cannot prove it is incorrect”
Examples of Null vs. Alternative:
-HN: There is no difference between X and Y
-HA: There is a difference between X and Y
 
 
Types of studies
-feasibility assessment involves careful examination of what it will take to conduct a study
successfully and whether the necessary resources are available
-Descriptive-designed primarily to describe what’s going on or what exists (public opinion
polls seeking to describe proportion of people who hold various opinions are (ie. Want to know
percent of population that would vote Democrat or Republican next election)
-Relational- look at relationships between two or more variables (public opinion poll
comparing what females vs. males would vote Democrat or Republican next election)
Causal- studies designed to determine whether one or more variables causes or affects
one or more outcome variables
-probably most demanding of the three studies
-two major variables of interest: the cause and the effect
Research Fallacies
-fallacy-an error in reasoning, usually based on mistaken assumptions
-ecological fallacy- faulty reasoning that results from making conclusions about
individuals based only on analyses of group data (ie. You shouldn’t assume any one student
is a math whiz because they’re in the highest scoring math class)
-exception fallacy- faulty conclusion reached as a result of basing a conclusion on
exceptional or unique cases
-at core of a lot of sexism and racism (ie. Women are terrible drivers because one
woman makes a driving error)
2. The process of research - “Good process is always the key”