Scatter Plot
A graph that shows a relationship between 2 quantitative (can be measured) variables.
Explanatory Variable
Explains or predicts change in the response variable (x).
Response Variable
Depends on the independent variable (y).
Correlated Data
Means there's a linear relationship between the data.
Describe the Data
#NAME?
Direction
Positive/Negative
Outliers
Data outside the normal scatter.
Form
Linear/Nonlinear
Strength/Correlation Coefficient
When r is 0 = no correlation
When r is </= 0.5, weak correlation
When r is > 0.5, strong correlation
When r is 1 = perfect correlation
Calculating Outliers
-Put data in order from least to greatest
-Median
-Q1 & Q3
-IQR = Q3 - Q1
-IQR x 1.5 = x
-Inner Fences
*Q3 + x =
*Q1 - x =
Line of Best Fit/Trend Line
The best line that can be drawn through a scatter plot that best represents the data such that there an equal amount of data points above & below the line respectively.
Interpolate Data
To answer a question within the data points that have been given.
Extrapolating data
To answer a question outside of the data points given.
Regression Line
-A straight line that describe how a response variable (y) changes as the explanatory variable (x) changes.
-Used to predict the values of (y) given an (x).
r
Tells us how close to a straight line the falls/is; correlation coefficient.
r2
Tells us the percentage of points that lie on or about the line, that attributes to the correlation.
Residual Plots
Residual is another word for "left over" or "difference". To find a residual value you take the actual value & subtract the predicted value. When you plot all of the residual values on to one graph scatter plot that shows the relationship of the residual to the x-value. If the regression model is the right or best model for the data, then the RESIDUAL plot will have a random scatter. If the residual plot creates a pattern then the regression model use is not a good model.