R | Methodology Shop

Basic code structure for Linear models

Linear models with numeric dependent variable and no clustering variable

	Linear model used	Relevant chapters	Code
Comparing two means from two independent samples	Linear model with Numeric dependent variable Dichotomous independent variable	Chapter 6	model < - lm ( y ∼ x , data = dataset ) model % >% tidy ()
Comparing more than two means from independent samples	Linear model with Numeric dependent variable Categorical independent variable	Chapter 6	If x is a factor variable model < - lm ( y ∼ x , data = dataset ) model % >% anova % >% tidy () If x is stored as a numeric variable model < - lm ( y ∼ as . factor ( x ) , data = dataset ) model % >% anova % >% tidy ()
Testing the interaction effect of two categorical variables on a numeric dependent variable	Linear model with Numeric dependent variable Two independent categorical variables	Chapter 9	If x and z are factor variables model < - lm ( y ∼ x + z + x :z , data = dataset ) model % >% anova % >% tidy ()
Testing the interaction effect of two numeric variables on a numeric dependent variable	Linear model with Numeric dependent variable Two numeric independent variables	Chapter 9	If x and z are numeric variables model < - lm ( y ∼ x + z + x :z , data = dataset ) model % >% tidy ()
Testing the interaction effect of one independent numeric variable and one numeric dependent variable	Linear model with Numeric dependent variable One numeric independent variable One categorical independent variable	Chapter 9	model < - lm ( y ∼ x + z + x :z , data = dataset ) model % >% anova % >% tidy ()
Regression	Linear model with Numeric dependent variable Numeric or dummy independent variable	Chapter 4 and 6.5	model < - lm ( y ∼ x + z , data = dataset) model % >% tidy ()

MODELS WITH A NUMERIC DEPENDENT VARIABLE AND A CLUSTERING VARIABLE (DUE TO REPEATED MEASUREMENTS)

	New approach	Relevant chapters	Code
Comparing two means from two related samples	Linear mixed model with Numeric dependent variable Dichotomous independent variable Clustering variable “id”	Chapter 12	library ( lme4 ) model < - lmer ( y ∼ x + (1\| id ) , data = dataset ) model % >% tidy ()
Comparing more than two means from related samples	Linear mixed model with Numeric dependent variable Categorical independent variable Clustering variable “id”	Chapter 13	library ( lme4 ) model < - lmer ( y ∼ x + (1\| id ) , data = dataset ) model % >% tidy ()

New approach

Relevant chapters

Code

Comparing two means from two related samples

Linear mixed model with

Numeric dependent variable
Dichotomous independent variable
Clustering variable “id”

Chapter 12

library ( lme4 )
model < - lmer ( y ∼ x + (1| id ) , data = dataset )
model % >% tidy ()

Comparing more than two means from related samples

Linear mixed model with

Numeric dependent variable
Categorical independent variable
Clustering variable “id”

Chapter 13

library ( lme4 )
model < - lmer ( y ∼ x + (1| id ) , data = dataset )
model % >% tidy ()

MODEL WITH A DUMMY OR A COUNT DEPENDENT VARIABLE

	New approach	Relevant chapters	Code
Logistic regression	Generalized linear model with Dummy dependent variable Numeric/dummy/categorical independent variable	Chapter 15	model < - glm ( y ∼ x , family = binomial , data = dataset ) model % >% tidy ()
Testing the independence of two categorical variables	Generalized linear model with Dependent variable that represents counts Independent categorical variables	Chapter 16	model < - glm ( y ∼ x + z + x :z , family = poisson , data = dataset ) model % >% tidy ()

New approach

Relevant chapters

Code

Logistic regression

Generalized linear model with

Dummy dependent variable
Numeric/dummy/categorical independent variable

Chapter 15

model < - glm ( y ∼ x , family = binomial , data = dataset )
model % >% tidy ()

Testing the independence of two categorical variables

Generalized linear model with

Dependent variable that represents counts
Independent categorical variables

Chapter 16

model < - glm ( y ∼ x + z + x :z , family = poisson , data = dataset )
model % >% tidy ()

Manuals & Websites

Please keep in mind that when working with R it is normal to search for the specific codes needed.

The UT provides a manual with different R codes for several types of analyses. It is mainly aimed at codes used in the statistics courses of several Bachelor programmes but can be a good starting point.

If you need information on how to analyse data using linear models you can have a look at this book.

If you have questions regarding Bayesian statistics please have a look at this book. It also provides information on how to set up R and how to install libraries from cran but also other sources.