Free Printable Simple/Multiple linear regression - Austin

1.

Match the following definitions with their

a)

When not maintained properly leads to bias. controlling for one variable while we change another

1.

Variable control

b)

A form of error. statistical terms/values that do not provide an accurate representation of the population

2.

Bias

c)

Situations in which researchers must accept that some of or all of the design without modification

3.

Nature experiments

d)

A statistical method that allows you to examine the relationship between two or more variables of interest

4.

Regression Analysis

e)

Any factor that prevents appropriate statistical interpretation of a results within practical context of a study

5.

Confounder

2.

Match the following definitions with their descriptions

a)

A specific and observable factor that is omitted from the analysis

1.

Omitted Variable

b)

Fundamentally different estimates then the population parameter

2.

Biased Samples

c)

The unadjusted results when adjustments lead to changes in any estimate of variation

3.

Inefficient

d)

A variable that, when accounted for, leads to a meaningfully different interpretation of relationship between the primary and independent variables and the dependent variables

4.

Confounding effect

e)

Any variable that produces a modifying effect

5.

Effect Modification

3.

Match the following definitions with their descriptions

a)

Statistical adjustments for the cofounder result in

1.

Unbiased Sample

b)

A intermediate between the primary independent variable and the dependent variable that can be used to analyze a relationship

2.

Mediating Effect

c)

Vertical relationship between each data point and trendline

3.

Residual

d)

Simplest process of estimation which assumes dependent variable is continous

4.

Ordinary least squares (OLS)

e)

A value between 0 and 1. As it approaches 0 regression explains very little variation in y. As it approaches 1 it explains it well.

5.

R²

4.

An R² close to 0 indicates a strong correlation and linear relationship

a)

True

b)

False

5.

Both confounding variables and/or omitted variables can lead to bias

a)

True

b)

False

6.

For which of the following types of data would you use dummy variables?

a)

Patients height

b)

Time spent drinking coffee

c)

Hours spent traveling

d)

Visited another country

7.

You design a study to determine if the numbers of hours watching scrubs correlates to the grade range achieved in IPP3. You also consider other factors such as location, video speed used, and average time snacking. Which of the following is the independent variable

a)

Number of hours watching scrubs

b)

Time snacking

c)

Video speed used

d)

Grade acheived

8.

You design a study to determine if the numbers of hours watching scrubs correlates to the grade range achieved in IPP3. You also consider other factors such as location, video speed used, and average time snacking. Number of hours watching scrubs would be which of the following

a)

Confounding effect

b)

Effect modulator

c)

Mediating effect

d)

None of the above

9.

You design a study to determine if the numbers of hours watching scrubs correlates to the grade range achieved in IPP3. You also consider other factors such as location, video speed used, and average time snacking. Video speed would be which of the following

a)

Confounding effect

b)

Effect moderator

c)

Mediating effect

d)

None of the above

10.

You design a study to determine if the numbers of hours watching scrubs correlates to the grade range achieved in IPP3. You also consider other factors such as location, video speed used, and average time snacking. How would you conduct this study?

a)

Simple regression

b)

Multiple regression

c)

Logistical regression

d)

Survival analysis

11.

You noticed you had some bias in your research study. You choose to make adjustments to decrease the bias. The initial results after the adjustment are

a)

Unbiased

b)

More relliable

c)

inefficient

d)

dependent

12.

You design a study which analyzes the amount of caffeine used by pharmacy students. You randomly collect samples from different students. You notice the parameter estimates and the error terms can be represented by a normal distribution. Can you use OLS?

a)

Yes

b)

No

13.

Ordinal least squares (OLS) can be used for a study which does not use random sampling and the distribution of error term can not be explained by a normal distribution

a)

True

b)

False

14.

Match the following definitions and terms with their descriptions

a)

The omitted variables which if it can be incorporated into regression, the bias/inefficiency associated would be removed

1.

Variable Z

b)

The error term for linear regression. The residual

2.

ε

c)

Adds a penalty for each extra parameter that is added to a model. Might indicate presence of one or more superfluous regressors

3.

Adjusted R²

d)

Additional variables which have no bearing on the study

4.

Superfluous regressor

e)

Represent categories of a discrete variable that are binary in nature.

5.

Dummy Variable

15.

Match the following

a)

Common coding scheme where you are either in a category or not in a category

1.

α-1 coding

b)

Transforms variables to Z-scores allowing for direct comparison

2.

Standardized regression coefficient

c)

Raw metric form

3.

Unstandardized regression coefficient

d)

A variable that influences how the independent variable determines dependent variable

4.

Effect modification

e)

A factor that connects the independent and a dependent variable

5.

Mediation effect

16.

The R² gets smaller as you add in multiple variables

a)

True

b)

False

17.

In multiple linear regression modules we need to use an adjusted R² value to determine the model of fit

a)

True

b)

False

18.

We can represent a patients weight with dummy variables

a)

True

b)

False

19.

We can represent a patients favorite color with dummy variables

a)

True

b)

False

Simple/Multiple linear regression - Austin - IPP 3 Exam 2

Total questions: 19

Worksheet time: 15mins

Similar Resources on Wayground