– 60 points Grade:_________
Name: ___________________________ (Also, click on the header and enter your full name)
Note: The word document of this assignment can be downloaded from CANVAS course page.
Please first read instructions and/or examples below and then answer the numbered questions. For all Labs, you many use any software tool you want (Excel, R, Python, SAS, SPSS, etc.). You are not required to provide screenshots or code.
For Download instructions, see the Statistical Software page in Canvas: https://unt.instructure.com/courses/67141/pages/the-statistical-software
In this lab, we are going to use the pokemon dataset to perform a simple linear regression.
Response variable: weight
Predictor variable: height
- Perform a scatterplot of the two variables (4 points)
- Does the scatterplot suggest a relationship between the two variables? (4 points)
- Perform a box and whisker plot of the response variable weight. (4 points)
- How many outliers are suggested by the box and whisker? (4 points)
- Perform two simple linear regressions with weight as the response variable and height as the single predictor. The first regression will use all the data and the second should remove outliers as defined by weight >= 6. Then fill out the following table. (32 points, 4 points each)
| Regression including Outliers | Regression without Outliers | |
| R-squared | ||
| Adjusted R-squared | ||
| Intercept Coefficient (B0) | ||
| Height Coefficient (B1) |
- Which model (with or without outliers) is better considering the table above? (4 points)
- If a new pokemon character was created with height = 0.3, what weight would we expect using:
- Regression including Outliers formula (4 points) _________________________
- Regression without Outliers formula (4 points) ___________________________
