Generate a mixed design from existing data

sim_mixed_df() produces a data table with the same distributions of by-subject and by-item random intercepts as an existing data table.

sim_mixed_df(data, sub_n = NULL, item_n = NULL, dv = "y",
  sub_id = "sub_id", item_id = "item_id", seed = NULL)

Arguments

data	the existing tbl
sub_n	the number of subjects to simulate (if NULL, returns data for the same subjects)
item_n	the number of items to simulate (if NULL, returns data for the same items)
dv	the column name or index containing the DV
sub_id	the column name or index for the subject IDs
item_id	the column name or index for the item IDs
seed	a single value, interpreted as an integer, or NULL (see set.seed)

Value

a tbl

Examples

sim_mixed_df(faceratings, 10, 10, "rating", "rater_id", "face_id")
#> # A tibble: 100 x 7
#>    sub_id item_id     y grand_i sub_i  item_i    err
#>    <fct>  <fct>   <dbl>   <dbl> <dbl>   <dbl>  <dbl>
#>  1 S01    I01      2.40    3.02 -1.49  1.11   -0.242
#>  2 S01    I02      3.19    3.02 -1.49  0.442   1.22 
#>  3 S01    I03      2.42    3.02 -1.49 -0.629   1.52 
#>  4 S01    I04      2.57    3.02 -1.49  0.236   0.801
#>  5 S01    I05      1.09    3.02 -1.49  0.446  -0.888
#>  6 S01    I06      2.32    3.02 -1.49  0.297   0.486
#>  7 S01    I07      3.97    3.02 -1.49  1.82    0.618
#>  8 S01    I08      3.02    3.02 -1.49  1.59   -0.103
#>  9 S01    I09      1.21    3.02 -1.49  0.0123 -0.337
#> 10 S01    I10      2.14    3.02 -1.49 -0.216   0.825
#> # … with 90 more rows