Question 1(3 points)

ListenFocus

Logistic and multiple regression have the same type of outcome variables

Question 1 options:

	True
	False

Question 2(3 points)

Listen

What transformation does this represent?

{"version":"1.1","math":"$\log \left(\frac{p}{1-p}\right)$"}

Question 2 options:

Question 3(2 points)

Listen

Question 3 options:

	Parabolic
	Negative
	No relationship
	Positive

Question 4(2 points)

Listen

An assumption of linear regression is that the relationship betweenxandycan be modeled by a straight line

Question 4 options:

	True
	False

Question 5(3 points)

Listen

You find an outlier than influences your regression line. The first thing you should do is...

Question 5 options:

	Remove it to assure your regression line is strong
	Investigate the point further to see if there is a reason to worry about it
	Log-transform it
	Leave it in

Question 6(3 points)

Listen

Question 6 options:

are observations that fall far from the "cloud" of points

Question 7(2 points)

Listen

The line that minimizes the sum of squared residuals is commonly called the

Question 7 options:

	minimal squared residual
	line of lowest sum
	maximum likelihood
	least squares line

Question 8(2 points)

Listen

Which of the following describes the amount of variation in the response that is explained by the least squares line?

Question 8 options:

	Line of best fit
	Variability
	{"version":"1.1","math":"$R^2$"}
	{"version":"1.1","math":"$\delta$"}

Question 9(2 points)

Listen

The line of best fit in linear regression will be the one that has the maximizes the sum of the squared residuals

Question 9 options:

	True
	False

Question 10(2 points)

Listen

Which of the following are assumptions when fitting a least squares line in linear regression? (Select all that apply)

Question 10 options:

	Linearity: The data should show a linear trend
	Nearly normal residuals: generally the residuals must be nearly normal
	Constant variability: Variability of points around the least squares line must remain roughly constant
	Symmetry: Lines of best fit should fit symmetrically along the median value

Question 11(2 points)

Listen

Question 11 options:

	Thet-value was obtained by dividing the estimate by standard error
	The 'unemp' coefficient is significant
	At-value of -1.23 is significant
	None of these

Question 12(2 points)

Listen

You run a regression forxpredictingyand find thep-value to be .04. Which of the following can you say most comfortably?

Question 12 options:

	The finding indicatesxstrongly predictsy
	Given the assumptions have been met, y significantly predictsx, but we can't be sure if it is important
	None of these
	Only 4% of the time would you expect to find a test statistic as extreme or more, assuming the null hypothesis is true

Question 13(4 points)

Listen

You can actually improve model fit by removing (a) predictor(s) in multiple regression.

Question 13 options:

	True
	False

Question 14(3 points)

Listen

Question 14 options:

Question 15(3 points)

Listen

The proper estimate of variance explained in multiple regression is

Question 15 options:

	All of these
	{"version":"1.1","math":"$R^2$"}
	Adjusted{"version":"1.1","math":"$R^2$"}
	>50% significant predictors

Question 16(3 points)

Listen

In multiple regression, you can have nominal and continuous predictors

Question 16 options:

	True
	False

Question 17(3 points)

Listen

Question 17 options:

	Both
	Neither
	'pubs'
	'time'