Statistics with SAS on Coursera: Week 5 Flashcards
How do you interpret an odds ratio?
When the odds ratio is 1, there is no relationship between the predictor variable and the outcome variable.
If the odds ratio is > 1, the group in the numerator is more likely to have the outcome.
If the odds ratio is < 1, the group in the denominator is more likely to have the outcome.
How can you measure the strength of a binary predictor variable and a binary response variable?
You can use an odds ratio to identify whether or not a certain event or outcome is more likely in the occurrence of one group relative to its occurrence in another group.
What statistic can you use to formally test the association for statistical significance?
Chi-square
What does chi-square measure?
Chi-square measures the difference between the observed cell counts and the expected cell counts of the null hypothesis is true and there is no association between the variables.
How do you interpret chi-square?
The larger the difference between the observed and selected cell counts, the more likely your data are to have a statistically significant association between variables.
Does the chi-square statistic and its p-value describe the magnitude of an association between variables?
No. They indicate how confident you can be that an association exists between the variables. The chi-square and p-value reflect the sample size.
What conclusion should you draw from a significant chi-square statistic?
You can reject the null hypothesis that there is no relationship between variables
What is the Cramer’s V statistic used for?
Cramer’s V statistic may be used to measure the strength of an association between two categorical variables
How do you interpret the Cramer’s V statistic?
The closer a Cramer’s V statistic is to zero, the weaker the association between two variables. Values further away from zero indicate stronger relationships.
What statistic considers the ordinal relationship between two variables?
The Mantel-Haenszel test considers ordinal associations
What is an ordinal association between two variables?
As the value of one variable goes up, it is associated with an increase or decrease in the other variable
How can you measure the association between two ordinal variables?
You can use the Spearman correlation statistic
What is a rank correlation?
A rank correlation provides a degree of association between the ranks or levels of the ordinal variables.
How do you interpret Spearman’s correlation statistic?
Values range from -1 to 1. Values close to zero have a weak relationship. Values close to 1 indicate high degrees of positive correlation. Conversely, values close to -1 have a high degree of negative correlation.
When you use honest assessment, which of the following would be considered the best model?
The best model is the simplest (the most parsimonious) model that has the best performance on the validation data. The training data is used to fit the model and generate the possible models to be assessed.
