regression and plotting regression

General
3

Here is a more concise way actually:

library(olsrr)
library(tidyverse)

data(surgical)

surgical %>% 
  pivot_longer(
    cols = c(-pindex, -y),
    names_to = "predictor"
  ) %>% 
  group_by(predictor) %>% 
  group_modify(
    ~broom::tidy(lm(y ~ pindex + value, .x))
  )
#> # A tibble: 21 x 6
#> # Groups:   predictor [7]
#>    predictor term        estimate std.error statistic  p.value
#>    <chr>     <chr>          <dbl>     <dbl>     <dbl>    <dbl>
#>  1 age       (Intercept)   267.      310.       0.861 0.393   
#>  2 age       pindex          9.77      2.97     3.29  0.00183 
#>  3 age       value          -3.55      4.52    -0.786 0.436   
#>  4 alc_heavy (Intercept)   -65.2     173.      -0.377 0.707   
#>  5 alc_heavy pindex         10.8       2.60     4.15  0.000125
#>  6 alc_heavy value         461.      112.       4.12  0.000138
#>  7 alc_mod   (Intercept)   133.      191.       0.696 0.490   
#>  8 alc_mod   pindex         10.6       2.91     3.64  0.000634
#>  9 alc_mod   value        -187.       97.7     -1.92  0.0611  
#> 10 bcs       (Intercept)  -328.      241.      -1.36  0.180   
#> # … with 11 more rows
1 Like