General Topics #1
General Topics #2
Statistical Tests #1
Statistical Tests #2
Interpretations
100
What are the goals of statistical analysis?
- Summarize data
- Show basic patterns
- Interpret patterns
- Generalize patterns to population
100
What is the first step of hypothesis testing?
Formulation of the null hypothesis and the alternative hypothesis
100
What influences our test results?
Sample size
Variance (dispersion of sample values)
How certain we want to be about the conclusion
100
If you want to compare two variables for each respondent in the sample, which test should be used?
Paired Sample T-Test
100
What is the impact of a larger sample size?
Easier to find statistical significance
200
What is the goal of inferential statistics?
Use characteristics of the sample to estimate characteristics of the population
In other words, use sample statistics to estimate population parameters
200
A false hypothesis that is not rejected is what Type of Error?
Type II Error
200
When should a One Sample T-test be used?
When we want to compare the mean of one sample to a set threshold
200
When testing the relationship between two variables, what tests should be used if our goal is to test difference in means?
Paired sample t-test
Independent sample t-test
200
Considering a binary “dummy” variable, variables are defined as 0 or 1. What would a 0 variable mean?
0 = condition is false.
1 = condition is true.
300
What are two examples of tests of associations?
Correlation (interval and ratio)
Proportions (ordinal and nominal) (Chi-Square tests)
300
What is the difference between inferential statistics and descriptive statistics?
Inferential statistics: quantities from sample data calculated to make inferences about the population being studied
Descriptive statistics: summary measures that describe your sample data
300
When should an Independent Sample T-Test be used?
When we want to compare the mean of one variable across two sub-groups in your population
300
When testing the relationship between two variables, what test should be used if our goal is to test associations with frequency distributions?
Bivariate Chi-Square Test
300
What does the R-squared value indicate?
What percentage of the variability in Y is explained by X
400
What is the difference between parameter and statistic?
Parameter is a number describing the whole population
Statistic is a number describing a sample
400
What is the correct interpretation of a p-value that is below alpha?
Rejection of the null hypothesis
400
What test should be used if we want to compare the frequency distribution of a sample variable to a hypothesized population distribution?
Chi-Square Test
400
What does an F-Test tell us?
The F-test of overall significance tells us whether our linear regression model provides a better fit to the data than a model that does not contain independent variables
400
According to this output, when “TR price in $” increases by $1, holding all else constant, what is the predicted increase in MM OJ sales?
+ 114.9
500
What are the two basic properties that we calculate for our sample?
Central tendency (mean, median, mode)
Dispersion (standard deviation, absolute frequencies, relative frequencies)
500
What are the two steps of the independent t-test?
Hypothesis test of equality of variance of 2 samples
Hypothesis test of difference of means
500
What are the assumptions of a Chi-Square Test?
The expected counts (Ei) has to be greater than 5 for at least 80% of the cells in the table
No cell can have an expected count below 1
These assumptions check if there is enough data for the test to be accurate
500
When does the adjusted R-squared increase?
When the new term improves the model more than it would be expected by change
But, the adjusted R-squared is always lower than the R-squared
500
According to this output, when there is a Feature Ad that week, holding all else constant, what is the predicted increase in MM OJ sales?
+ 270.33