#3:  What is the median of column Promotion in DATA1? Copy/paste into the Google Form.

```{r setup, include=FALSE}

```

#4:  DATA1 contains data on customers to 3 different stores.  Customers were asked to rate Satisfaction with a store on a scale of Low, Medium, and High. Study the association between Satisfaction and Store.  Copy/paste into the Google Form the estimated p-value (not the range) for the test of association that is found after 500 permutation datasets.  Add the argument seed=2016 to associate.

```{r setup, include=FALSE}

```

#5a:  Let us predict Price (a relative measure of how satisfied people are with the prices at the store) from Promotion (a relative measure of how satisfied people are with the promotions at the store in DATA1 using transformed variables. Fit a simple linear regression model predicting Price from Promotion, then determine the best fitting set of transformations. Copy/paste into the Google Form the power to which we should raise x (Promotion). Note: if we the log transformation is suggested, enter 0.

#5b:  Report the order of the best polynomial model for predicting Price from Promotion using 0.005 criteria. Note:  consider a maximum order of up to 6 (the default)

```{r setup, include=FALSE}

```

#6:  Create a subset of DATA2 where q6 is strictly less than 56.74 . Copy/paste into the Google Form Spearman's rank correlation between q6 and q1 on this subset. Note: you may ignore all warnings output by R if they occur- they refer to ties in the data.

```{r setup, include=FALSE}

```

#7:  Find the variable that is most correlated with grade in DATA2, then fit a simple linear regression predicting grade from that variable. Copy/paste the intercept of this model into the Google Form

```{r setup, include=FALSE}

```

#8:  Fit a multiple regression model predicting grade from all variables in DATA2. Copy/paste the p-value of the test of equal spread into the Google Form.

```{r setup, include=FALSE}

```

#9:  Conduct a partial F test to determine if the two least significant predictors (i.e. with the largest p-values) in your model from #8 can be dropped. Refer to the a previous activity for guidance and an example for the partial F test. Copy/paste into the Google Form the number in the column labeled F during the anova() step of the analysis.

```{r setup, include=FALSE}

```

#10:  Using the regression model from #8, copy/paste into the Google Form the upper bound of a 94% prediction interval when each predictor is equal to 80. For guidance on creating the TO.PREDICT dataframe, refer to an in-class activity on the matter.

```{r setup, include=FALSE}

```

#11:  Examine whether the prediction from #10 requires extrapolation Copy/paste the value under Percentile after running the relevent command Note: review the chapter 4 part 2 notes.  This number tells us the percentage of points used to build the model that are CLOSER to the center of the predictor datacloud than this new point

```{r setup, include=FALSE}

```

#12:  Fit a regression model on DATA2 predicting grade from q2 and q1 , including the interaction. Copy/paste into the Google Form the coefficient of the interaction term.

```{r setup, include=FALSE}

```