3.1 Scatterplots
- Response Variable: measures the outcome of a study (a.k.a. dependent)
- Explanatory Variable: tries to explain the observed outcome or response (a.k.a. independent)
- Scatterplots are most effective in comparing these two quantitative variables.
- When designing a scatterplot, one should be aware of patterns, striking deviations, form, direction, strength, or any potential outliers and maybe make mental note of these characteristics


 
 


3.2 Correlations

-Correlation: measures direction and strength of the linear relationship in a scatterplot and is written as "r"
-When r>0 there is a positive association & r<0 there is a negative association

-A positive correlation means that high values of one variable are associated with high values of a second variable. The relationship between height and weight, between IQ scores and achievement test scores, and between self-concept and grades are examples of positive correlation.

- A negative correlation or relationship means that high values of one variable are associated with low values of a second variable. Examples of negative correlations include those between exercise and heart failure, between successful test performance and feelings of incompetence, and between absence from school and school achievement.
    Positive Correlation 
                                                                        WEAK                                                                              STRONG
     Negative Correlation 
                                                                          WEAK                                                                            STRONG


3.3 Least-Squares Regression

- The regression line is a straight line that describe how y changes as x changes which can also help us predict y given x
- LSRL: a line that makes the sum of the squares of the vertical distances form a line that is as small as possible

  LSRL
-Residual plot: a scatter plot of the regression of the residuals against the explanatory that helps us assess the fit of the regression line.
-Outliers: lie outside the overall pattern of other observations.

  One Can clearly see this outlier at (390,14)

-Influential point: different than an outlier because if you remove an influential point from the plot they would significantly change the results of your calculations.

Formulas you'll need for Chapter Three