📚 Topic Summary
High leverage points are data points that, if changed even slightly, can drastically alter the outcome of a regression analysis. They often lie far from the rest of the data in terms of predictor variables. Identifying and handling these points is crucial because they can disproportionately influence the regression model, leading to misleading conclusions about the relationship between variables. Best practices involve careful examination of these points, assessing their impact on the model, and deciding whether to retain them, transform them, or remove them based on a sound understanding of the data and the research question.
Ignoring high leverage points can result in a model that poorly represents the underlying relationships in the data. Addressing them appropriately ensures a more robust and reliable analysis.
🧮 Part A: Vocabulary
Match the term with its definition:
| Term |
Definition |
| 1. Leverage |
A. A measure of how far an observation's predictor values are from the average of the predictor values. |
| 2. Outlier |
B. A data point that significantly deviates from the overall pattern of the data. |
| 3. Regression Analysis |
C. A statistical method used to model the relationship between a dependent variable and one or more independent variables. |
| 4. Influence |
D. The degree to which a single observation affects the estimated regression coefficients. |
| 5. Cook's Distance |
E. A measure of the overall influence of a data point on the regression model. |
✏️ Part B: Fill in the Blanks
High leverage points can have a significant impact on __________. Identifying these points involves examining their __________ values compared to the average. Removing or transforming these points should be done cautiously, with a clear understanding of the potential __________ on the analysis.
🤔 Part C: Critical Thinking
Why is it important to not automatically remove high leverage points from a dataset? Give an example of a situation where a high leverage point might be a valid and important observation.