EXAM 2 Flashcards
What are the different types of measures?
1) Nominal
2) Ordinal
3) Interval
4) Ratio
Nominal
Nominal - scales label objects (responses cannot be differentiated) EX:1 Male, 2 female, 3 prefer not to say
Ordinal
Ordinal - relative size differences between objects (responses ranked/scaled)
EX: How many influencers do you follow? 1= 1-10 influencers, 2=11-30 influencers, 3=31-50 influ.
Interval
Interval: (no true 0 origin)
EX: 1=strongly disagree, 2= somewhat disagree, 3= neither agree/ disagree, 4= somewhat agree
Ratio
Ratio: (true 0 origin)
EX: how many times have you spent $20 or more on an influencer-inspired purchase 0, 1, 2, 3, 4
Composite Measure
combination of two or more surveys measuring the specific construct
- semantic differential
- Likert
EX rate agreement to the following statements
Be able to calculate the composite measure
The average (add all numbers up and divide by those numbers; mean)
What is scale reliability?
The degree to which an instrument consistently measures a construct across items
Indicator of a measure’s internal consistency
Be able to interpret the internal consistency based on the Cronbach’s alpha table.
- Represents a measure of homogeneity/ extent to which an indicator converges on a common meaning
- Measured by correlating scopes of items making up a scale
- Cronbach’s alpha= most commonly applied estimate of multiple item’s scale reliability
>0.9 - excellent
>0.8 - good
>0.7 - acceptable
>0.6 - questionable
>0.5 - poor
What is descriptive analysis? What are descriptive statistics?
Descriptive analysis: used by marketing researchers to describe sample dataset, reveals general patterns of responses
Descriptive statistics: mean, median, mode, frequency, variability, range, standard deviation
What is inferential analysis?
Use statistical procedures to generalize the results of the sample to the target population it represents
What is a null or alternative hypothesis?
Hypothesizes the result of an alternative answer (Gen Z consumers will not buy a product promoted by a social media influencer.)
What are testing differences? What is a testing relationship?
Testing differences: on outcome variable between two or more groups (EX: do consumers behave differently on certain outcomes between 2 or more market segments). Determines the degree to which real and generalizable differences exist in the population
Be able to interpret t-test results.
- comparing the differences between 2 groups (ex. female and male) and their significant difference from 0
- Look at the MEAN when seeing which group is higher in the outcome variable
- when looking at the TWO TAILED TEST: if the p-value is less than 0.5 it means our null hypothesis is rejected
Be able to interpret the ANOVA table,
- used when comparing the means of
three or more groups - ANOVA will “signal” when at least one pair of means has a statistically
significant difference - when looking at the AVERAGE you can see how much each variable differs from one another, dependent on how far apart they are spread
- when looking at P VALUE you can see if the p-value is above 0.05 then the null hypothesis is accepted
Be able to interpret the regression statistics table, and coefficient table from the regression results.
Regression= look at r squared (interprets % of variance in dependent variable)
look at F and significance F = P value of F statistic and shows weather independent variable significantly predicts dependent variable
coefficients= largest amounts (- or +) are most signifiant relationships
P value= which ones are significant P> 0.05 null hypothesis
What is a testing relationship?
Tests relationship between 2 or more variables, and how they’re related. Insight on multiple relationships between variables.
relationship must be meaningful and actionable
Be able to propose null hypothesis regarding testing differences or testing relationship.
Testing differences: Light users, regular users, and heavy users of social media in Gen Z consumers do not differ from their purchase intention on the product promoted by a social media influencer.
Testing relationships: The expertise of a social media influencer is not related to (will not affect) Gen Z consumers’ purchase intention on the product promoted by this influencer.
What are the steps in hypothesis testing process?
1) Null hypothesis related to research question
2) Select appropriate statistical test (t-test, anova, regression)
3) Specify significance level (0.05)
4) Run test and get P value
5) Interpret P value and compare to significance level (0.05)
6) P value > 0.05 null hypothesis
rejected = sig difference
Be able to select the right statistical tests (e.g., t-test, ANOVA test, regression) given the scenario.
T-test = difference between two groups
ANOVA = test of variance = three or more groups
Regression = tests relationship between single dependent variable and multiple independent variables
