Final Exam Flashcards
(37 cards)
What do cross-tabulations display?
The relationship between two categorical variables of nominal or ordinal scale.
What type of probability distribution do cross-tabulations show?
Joint probability distribution of two variables.
What can be calculated using cross-tabulations?
Marginal, joint, and conditional probabilities.
What is an odds ratio?
A measure that can be calculated from cross-tabulations.
Name three statistical tests that can be calculated from cross-tabulations.
- Chi-square test
- Fisher’s exact test
- McNemar test
True or False: Cross-tabulations can only be used for continuous variables.
False
Fill in the blank: Cross-tabulations can assess the statistical significance of unequal distributions found in the _______.
cross-tabulation.
What does each row in a cross-tabulation represent?
Each row represents the categories of one variable (e.g., some exposure of interest).
What does each column in a cross-tabulation represent?
Each column represents the categories of a second variable (e.g., some health outcome).
How can cross-tabulations be used in probability estimation?
Cross-tabulations can be used to estimate probabilities by calculating proportions using the numbers in the table.
What is the most common type of cross-tabulation?
The 2 x 2 table
What do the variables in a 2 x 2 table represent?
Two dichotomous (two categories) variables
What are conditional probabilities?
The probability of falling within a specific category for one variable given that the person already falls within a specific category on the other variable.
Conditional probabilities help in understanding the relationship between two variables.
What are marginal probabilities?
The proportion of the total sample with a specific characteristic.
Marginal probabilities provide insight into the overall distribution of a characteristic in the sample.
What is joint probability?
The probability of the co-occurrence of being in a specific category for one variable and a specific category for the second variable.
Joint probability helps in understanding the relationship between two categorical variables.
What statistic is commonly used to assess statistical significance for two categorical variables?
The chi-square statistic (X2).
The chi-square test helps determine if there is a significant association between the variables.
There are three assumptions underlying the chi-square test:
The data are frequency data (counts of the number of individuals falling within each category).
Adequate sample size—all expected cell counts must be ≥ 5
Measures independent of each other
What does the chi-square test compare?
The expected frequencies with the observed frequencies in each cell
What are ‘expected frequencies’ in the context of the chi-square test?
The number of observations in each cell of the cross-tabulation table that you would expect to see if the two variables are not associated with each other
True or False: Expected frequencies indicate a relationship between two variables.
False
What is the McNemar test?
A variation on the chi-square test used for nonindependent dichotomous variables
When is the McNemar test applicable?
When two dichotomous variables are nonindependent or if assumptions no 3 cannot be met
When do you use a fisher’s exact test?
If assumption no. 2 (adequate sample size—all expected cell counts must be ≥ 5) cannot be met
What is Sensitivity?
Proportion of positive results among all patients who have a certain disease.
Sensitivity measures the true positive rate of a diagnostic test.
