How to only print (adjusted) R-squared of regression model?

Question

I am a beginner with R. I have a data set on air pollution. The columns are site, measured concentration and 80 variables (v1-v80) that might influence the concentration. I want to make a model with forward stepwise regression based on R-squared/adj with my own code (so I do not want to use something like step() or regsubset()). The dependent variable is concentration and the variables v1-v80 as independent variables. I wrote the following code for the first step (data set is simplified):

  site concentration         v1         v2         v3
1    1   -0.84085548  1.7114409 -0.2857736 -1.0803926
2    2    1.38435934 -0.6029080  0.1381082 -0.1575344
3    3   -1.25549186 -0.4721664  1.2276303 -1.0717600

for (j in names(df)){
  model <- lm(concentration ~ df[[j]], data = df)
  print(j)
  print(summary(model))
}

This works well, but I am only interested in R-squared and adjusted R-squared. I tried to only have (adjusted) R-squared printed with:

for (j in names(df)){
  model <- lm(concentration ~ df[[j]], data = df)
  print(j)
  print(summary(model$r.squared))
  print(summary(model$adj.r.squared))
}

But then I get as output (this is only a part):

[1] "v1"
Length  Class   Mode 
     0   NULL   NULL 
Length  Class   Mode
     0   NULL   NULL
[1] "v2"
Length  Class   Mode 
     0   NULL   NULL 
Length  Class   Mode
     0   NULL   NULL

Etcetera.

How can I get as output only the name of the relevant variable and (adjusted) R-squared for every model that is produced in the for-loop?

Thanks!

A. Suliman · Accepted Answer · 2018-05-28T10:37:10.810

library(broom)
glance(model)[c(1,2)]

Input = ("site concentration         v1         v2         v3
          1    1   -0.84085548  1.7114409 -0.2857736 -1.0803926
          2    2    1.38435934 -0.6029080  0.1381082 -0.1575344
          3    3   -1.25549186 -0.4721664  1.2276303 -1.0717600")

df = read.table(textConnection(Input),header=TRUE)

for (j in names(df)){
    model <- lm(concentration ~ df[[j]], data = df)
    print(j)
    print(glance(model)[c(1,2)])
}

[1] "site"
    r.squared adj.r.squared
 1 0.02132635    -0.9573473
[1] "concentration"
    r.squared adj.r.squared
  1         1             1
[1] "v1"
  r.squared adj.r.squared
1 0.1717716    -0.6564568
[1] "v2"
  r.squared adj.r.squared
1 0.1482473    -0.7035055
[1] "v3"
  r.squared adj.r.squared
1 0.9762587     0.9525174
Warning message:
  In stats::summary.lm(x) :
  essentially perfect fit: summary may be unreliable

Using base R

summary(model)$adj.r.squared
summary(model)$r.squared

How to only print (adjusted) R-squared of regression model?

1 Answers1

Linked