Marketing 368 WSU Exam #1

Stratified Random Sampling:

-Break up population into meaningful groups then sample within each "strata"//combine -More representative, can compare strata -Can be hard to figure out what to base strata on though AKA: Race, Gender, Ethnicity, Political Preference

Disadvantages of Quantitative Methods:

-Difficulty of developing accurate survey questionnaires. -Limits to the in-depth detail of data -Limited control over timeliness, and potentially low response rates -Evaluating whether respondents are responding truthfully -Misinterpretation of data and inappropriate use of analysis procedures

Probability Sampling:

-Each sampling unit has a known probability of being included in the sample AKA: I know I'm 1/63 students in this class

Test-Retest Reliability:

-Extent to which scores are stable over time -Have people complete questionnaire twice and correlate scores

Quantitative Methods:

-Facts, estimates, relationships, predictions -Descriptive // Casual design -Structured -Longer time frames -Larger samples -Good representation of target populations -Statistical -Descriptive -Subjective interpretive skills -Very good, inferences about facts, estimates

Decision-Makers, who are they:

-Focus on symptoms of problem -Want info to confirm decision -Want quick information -Not as likely to pay -Dislike surprises -Decision/Results orientated -Interested in future performance

Researchers, who are they:

-Focus on underlying problems -Like to explore new questions -Tolerate long investigations -No cost concerns -Enjoy surprises -Speak in probabilities -Interested in past behavior

Threats to internal validity:

-History Effect: Course of history happens during study that effects dependent variable (9/11) -Maturation Effect: Something happens over time that changes affects of DV (taking 1 test --> gain knowledge --> take second test better) -Testing Effect: Pre-test // post-test design, effect time 2 DV by pretesting at time 1. (simple act of measuring the DV at the time 1 changes the DV at time 2) -Instrumentation Effect: Fact that you measure something such as observing behavior changes -Statistical Regression: Select groups based on extreme scores, regress toward mean, changing groups -Selection Bias: When groups (control) differ before experimental manipulation (creates unequal group) -Mortality: Some drop out, or die. Drop-outs change scores in condition, those who stick around may be different then the dropouts

When is marketing research needed?

-Is problem of strategic importance? -Is secondary data inadequate for addressing the problem? -Is there enough time to collect data for managerial decision? -Are there enough resources ($,people) to carry out the study? -Does value of research outweigh the costs of research?

Disadvantages of Qualitative Methods:

-Lack of generalizability -Inability to distinguish small differences -Lack of reliability and validity -Difficulty finding well-trained investigators, interviewers, and observers

What is marketing research?

-Links organization to its market. -Information allows for the identification and definition of market-driven opportunities and problems and allows for the generation, refinement and evaluation of marketing actions. Another way to put it: Market research is the process of collecting valuable information to help you find out if there is a market for your proposed product or service.

Quota Sampling:

-Sample fixed number of people from each of the X categories, possibly based on their relative prevalence in the population Can ensure that certain groups are included Because you aren't using random sampling generalizability may be questionable!

Simple Random Sampling:

-Sampling approach in which each sampling unit in a target population has a known and equal probability of being included -Good generalizability/unbiased estimates -But must be able to identify all sampling units within a given population (often not feasible)

Systematic Random Sampling:

-Similar to random sampling, but work with list of sample units that is ordered in some way (alphabetically) -Select starting point at random, then survey each "nth" person where the "skip interval" = (population size/desired sample size) -Quicker and easier than SRS -May be hiding patterns in the data unfortunately

Sampling error:

-Statistically speaking, the difference between the sample results and the population parameter -Assuming perfect survey, sampling frame, execution, and respondents, we will still have error due to sampling -Sampling error becomes smaller with larger groups (Goes down if you have more people, because its going to be larger / more information / more represented scale)

Snowball Sampling:

-You contact one person, they contact a friend Can make it easier to contact people in hard to reach groups There may be bias in the way recruit others

List of dependent variables:

Behavior: -Awareness -Knowledge -Liking -Preference -Intent to buy -Purchases (Measure of performance) AKA: Sales, Market Share, Profit, ROI, Image

Step 7 in Confidence Interval:

CI = Mean +/- (Sx) (z) Mean = 3 Standard Dev = 0.7 Confidence = 1.96 because above certain percent (he will tell you which number to use) therefore: 3 +/- (0.7)(1.96) aka: 3 +/- 1.37 3 + 1.37 = 4.37 3 - 1.37 = 2.63 So confidence interval = 2.63 ~ 4.37

Perform the Research:

Collect and analyze the data

Internal Reliability:

Extent to which items on a scale "hang together" or are correlated with one another -Cronbach's alpha: a measure of internal consistency, that is, how closely related a set of items are as a group. It is considered to be a measure of scale reliability. A "high" value for alpha does not imply that the measure is unidimensional. -Split-half reliability (split measures into halves, correlate)

List of independent variables:

Marketing Mix: Controllable things -Pricing -Promotion -Product -Distribution

Construct Validity:

Measuring what you're tending to measure, capturing all the factors that go along with it // what makes you brand loyal as a person? and what goes along with being brand loyal? Extent to which your constructs of interest are accurately and completely identified.

Types of Scales:

Nominal: Has assignment only // ex, political party Ordinal: Has assignment, order // ex, rank order of finish in a race Interval: Has assignment, order, equal intervals (temperature) Ratio: Has assignment, order, equal intervals, absolute zero (number of cars)

Step 5 in Confidence Interval:

S = Sample Standard Deviation to find this, we must find the square root of the SS/N-1 S = Square root of 10 / N -1 Because 5 people were surveyed is would be S = Square root of 10 / 5-1 10/4 = 2.5, we square root that and it ends up being (1.58)

Choose an Appropriate Research Design:

Secondary? Focus Group? Survey? Experiment? Sampling?

Cluster Sampling:

Similar to stratified random sampling, but with SRS the strata are thought to possibly differ between strata AKA: Men vs. Women. Cluster sampling you divide overall population into subpopulations (like SRS) but each of the subpopulations (Clusters) are assumed to be mini representations of the population

Semantic differential scale:

Unpleasant --> Pleasant Flimsy --> Sturdy Male --> Female This may suggest that males are to be evaluated negatively? You must be careful in designing scales so as not to bias results

Quantitative:

Use formalized standard questions and predetermined response options (yes, no) in questionnaires or surveys administered to large numbers of respondents

Qualitative:

Used in exploratory designs to gain preliminary insights into decision problems and opporotunities

Determine the Research Problem:

What does client want to know and what is the value of certain information?

Internal Validity:

When researchers can clearly identify cause and effect relationships (no confounds)

Common Pitfalls - Incomprehensible:

When you consider a new stadium, is it possible that part of your determine factors might rest on the fact that you sometimes choose to forgo entertainment in favor of more pedestrian activities like walking your dog?

Step 4 in Confidence Interval:

Take the X-X"With line on top" and Square them! (x-x"with line on top")^2 -2 is actually 4 -1 is actually 1 2 is actually 4 1 is actually 1 0 is actually 0 So the SS or "finding the some of squares" is actually (10)

Origin (absolute zero point):

Zero "means something" (absence of a given quality)

(Variant): optimal allocation //

here you use smaller sample sizes for strata within which there is low variability (as the lower variability will give you more precision with lower N).

Proportionate stratified sampling:

sample based upon size of populations (i.e, sample more from the bigger strata)... such as Caucasians

Unforced scale:

(Gives people a neutral option) 1: Very Unlikely 2: Unlikely 3: Somewhat Unlikely 4: Neither 5: Somewhat Likely 6: Likely 7: Very Likely

Balanced scale:

(Same number of positive and negative options) 1. Extremely dissatisfied 2. Dissatisfied 3. Somewhat dissatisfied 4. Neither 5. Somewhat satisfied 6. Satisfied 7. Extremely satisfied

Step 1 in Confidence Interval: Lets assume we want to know how many times people would go use the bowling facility in one month?

Lets ask 5 people: responses below 1 2 5 4 3

Unbalanced scale:

(ALL the options are positive) (Can give biased results; unless distribution is naturally skewed to one side of the scale, should use balanced scale) 1. Somewhat satisfied 2. 3. 4. 5. 6. 7. Very satisfied

Moderation:

(CHANGES) -Under what conditions is a relationship stronger/weaker Service Failure --> NWOM and Trait hostility is in the middle of that ^ Connecting an arrow to the middle of the two

Mediations:

(EXPLAINS) -Like a combination shot in pool... -Effect of one IV on the DV occurs through an intermediary variable -EX: Person experiences service failure --> infer negative attitude --> feel angry --> spread NWOM What it looks like: Inference of Negative Motive --> Anger --> Negative Word of Mouth

Moderations:

-When the effect of one IV (service failure) on the DV (negative word of mouth) depends on the level of another IV (trait hostility)

Non-Probability Sampling:

-When the probability of selecting each sampling unit is unknown AKA: Unfortunately you can have certain bias's associated with this

Independent Variables:

Things you manipulate

Key function of Marketing Research:

Understanding relationship between IV's and DV's

Common Pitfalls - Leading:

Wouldn't you agree that it is a great idea to build a new multiuser facility in Boise?

Communicate the Results:

Write the report

Step 3 in Confidence Interval:

X-Mean AKA: X-X"with line on top" 1 - 3 = -2 2 - 3 = -1 5 - 3 = 2 4 - 3 = 1 3 - 3 = 0 This totals up to: 0

Distance (equal intervals):

The distance between adjacent points on the scale is identical

Validity (in general):

The extent to which conclusions drawn from a study are true

External Validity:

The extent to which what you find in your study can be generalized to your target population (survey has better external validity then focus group) "Want to have lots of participants"

Forced Scale:

(Even number of options forces the respondent to lean one way or the other) 1: Very Unlikely 2: Unlikely 3: Somewhat Unlikely 4: Somewhat Likely 5: Likely 6: Very Likely

Stapel Scale:

-5 -4 -3 -2 -1 Fast Service 1 2 3 4 5 This kinda scale ^^^ Can draw comparative profile analysis (ex various shoe stores) as we did semantic differential scales Probably because positive number if you feel accurate, and negative numbers if you feel inaccurate

When are descriptive designs appropriate?

-Describing current characteristics of a market (attitude towards existing product) -Want to understand your target markets characteristics (demographics, psychographics) -Want to understand relationships between variables (price & purchase) or differences between groups (attitudes towards water filters from hikers vs backpackers)

When marketing research is not needed?

-Not a lot of money or time. -Sufficient information for a decision already exists -Must make immediate decision -Insufficient resources for research -Cost outweigh the benefits

Convenience Sample:

-Survey based on convenience Fast and Easy May not be representative

Qualitative Methods:

-Understand of feelings, ideas, objects -Normally exploratory design -Open-Ended, semistructured, unstructured, deep probing -Short time frames -Small samples -Subjective, content, interpretive, debriefing -Interpersonal Communications -Observations -Limited (preliminary insights)

Judgement Sampling:

-Use your judgement about who is best to survey Can be better than convenience is judgement is right But if judgement is wrong, may not be representative/generalizable

Four broad stages of a research project:

1. Determine research problem 2. Choose an appropriate research design 3. Perform the research 4. Communicate the results

Dependent Variables:

Any outcome measures

Area Sampling:

Clusters based on geographic region

Common Pitfalls - Double Barreled:

Do you think the old stadium needs to be replaced and a new stadium should be built downtown?

Step 2 in Confidence Interval:

Find the Mean "X" with line on top of 1 2 5 4 3 Add up all your data, divide it by what you got AKA : 3

Disproportionate stratified sampling:

Sample the same number of units from each strata, regardless of the strata's size in the pop

Step 6 in Confidence Interval:

Sx <-- lowercase x with line over it = Standard Error of Mean Sx = s / square root of "n" AKA: 1.58 / square root of "5" = 0.7

Confidence Intervals:

The statistical range of values within which the true value of the target population parameter is expected to lie.

Direction of scale:

Typical direction (lower values, negative connotations on the left side) Strongly disagree --> .. --> ... --> Strongly agree

Situational Factors:

UNCONTROLLABLE things.... -Demand -Competition -Legal/Political -Economic Climate -Technology -Government Regulation

Common Pitfalls- Unanswerable:

Will buying a new stadium in Boise cost too much?

Assignment:

You can assign objects to categories

Order (magnitude):

You can order objects in terms of having more or less of the some quality