Students Struggled With…

• Remembering what needed to be checked for each of the five conditions. 

• Understanding the difference between a residual plot and a dotplot of the residuals.

• Mixing up the slope and y-intercept from computer output. 

 

Remedies…

• (Condition) -  how to check it --> SO WHAT?

• (Linear) - scatterplot linear or residual plot with no leftover pattern --> so there is a fairly linear relationship between x and the mean of y.

• (Independent) - 10% condition --> so sampling without replacement is OK

• (Normal) - dotplot of residuals is approximately normal --> so for each x, the distribution of y is approximately normal.

• (Equal Variance) - residual plot shows similar size residuals for each x-value --> so for each x, the standard deviation of y is the same.

• (Random) - random sample or random assignment --> so we can generalize to the population or we can show causation.

• Remind students that statisticians use y = a + bx rather than y = mx + b. It makes sense that the computer output would give us the y-intercept first. ​hey should calculate a z-score and find the area above it.