Analysis Prep
| Term | Definition |
|---|---|
| Alpha(a) | Probability of rejecting the hypothesis incorrectly |
| Association | Relationship between two, or more, variables |
| Assumption | A condition that must be met before a certain analysis can be run |
| Case | One object or instance in a collection, often a person. One case can have many variables |
| Categorical | A variable that has a fixed number of values |
| Continuous | A variable that can take any value between its minimum or maximum |
| Dependent Variable (DV) | The variable being measured (outcome variable) |
| Homogeneity | Is Homoscedasticity for ANOVAs |
| Homoscedasticity | Making sure you have the same amount of variance across all variables |
| Independence | The probability of one even occuring has no impact on the probability of another event occuring |
| Independent Variable (IV) | The variable being maninpulated (experimental variable) |
| Level | Different stages or groupings or IV or DV |
| Linearity | The ability to graph the data in a straight line |
| Logit | The log of the odds (L=ln(p/(1-p)) where L is the logit and p is the probability) |
| Multicollinearity | When the correlations between two or more variables is high enough that one can predict the other. This skews the results in the regression model. |
| Normality | A normal distribution is a distribution of random data points that create a bell-curve with the most points around the mean |
| Outlier | A data point that is extremely disperate from other results enough to skew the over all data |
| Population | The entire group from which a sample is taken to be tested |
| Residual (e) | The difference between the dependent variable and the independent variable in a regression model. |
| Sample (n) | A sub-set of a population taken to represent the whole |
| Scale | A group of survey questions used to measure a particular concept |
| Sphericity | The condition if all variances of the differences between all combinations of levels are the same |
| Standard Deviation | Summary measure of variation or spread of a set of data. Shows the most common distance from mean |
| Variance | How far a dataset is spread out |