Term
| Analysis of Variance (ANOVA): |
|
Definition
Used to determine whether there are any significant differences between the means of 2 or more independent groups. ANOVA compares the means between groups relative to the variability within groups and determines whether any of those means are significantly different from one another.
It tests the null hypothesis that all groups are simply random samples of the same population --> the null hypothesis is rejected and the alternative hypothesis is accepted when there are at least 2 group means that are significantly different from one another |
|
|
Term
|
Definition
| Used to check for an association between 2 categorical variables. For example, chi-square test could be used to determine if the distribution of gender and smoking status is random or if there is a difference between the sexes regarding smoking status |
|
|
Term
|
Definition
| Involves the poling of data from several studies to perform an analysis having greater statistical power than the individual studies themselves. For example, individual studies assessing the effects of aspirin on certain cardiovascular events may be inconclusive. However, analysis of data compiled from multiple clinical trials may reveal a significant benefit. |
|
|
Term
| Multiple linear regression: |
|
Definition
| Method used to model the linear relationship between a dependent variable and 2 or more independent variables. For example, this test could be used to quantify the effects of alcohol use, tobacco smoking, and charred food consumption on the incidence of gastric cancer. |
|
|
Term
| Pearson correlation coefficient: |
|
Definition
| Measure of the strength and direction of a linear relationship between 2 variables. For example, a study may report a correlation coefficient describing the association between HbA1c levels and average blood glucose levels |
|
|
Term
|
Definition
| Can be used when 2 group means need to be compared. This test could have been used for the example given in the question if the study participants were divided only into smoking and nonsmoking groups |
|
|
Term
|
Definition
| Likelihood ratio of any clinical finding is the probability of the finding in patients with the disease divided by the probability of that same finding in patients without the disease |
|
|
