This function takes an input dataset and normalizes observations
Arguments
- data
input dataset to be plotted (required)
- y_var
response variable, Y (required)
- grouping_var
select grouping variable to normalize by (required)
- ref_values
add reference (nominal) values. takes a string of values with values appearing in the same order as in grouping variable. string length must be equal to unique values in grouping variable (required)
Examples
library(dplyr)
polar_small_multiples_data %>%
filter(ID_Measurement_Angle %in% c(0, 45, 90, 135)) %>%
normalize_observations(y_var = ID,
grouping_var = Tip_Bottom,
ref_values = c(0.2075, 0.2225))
#> Joining, by = "Tip_Bottom"
#> # A tibble: 64 × 7
#> ID_Measurement_Angle Mold_Cavity_Number Tip_Bottom ID ID_2 ref_value
#> <dbl> <dbl> <chr> <dbl> <dbl> <dbl>
#> 1 0 1 Bottom ID 0.207 0.103 0.208
#> 2 45 1 Bottom ID 0.207 0.103 0.208
#> 3 90 1 Bottom ID 0.207 0.103 0.208
#> 4 135 1 Bottom ID 0.208 0.104 0.208
#> 5 0 2 Bottom ID 0.207 0.103 0.208
#> 6 45 2 Bottom ID 0.207 0.103 0.208
#> 7 90 2 Bottom ID 0.207 0.104 0.208
#> 8 135 2 Bottom ID 0.207 0.104 0.208
#> 9 0 3 Bottom ID 0.207 0.103 0.208
#> 10 45 3 Bottom ID 0.208 0.104 0.208
#> # … with 54 more rows, and 1 more variable: ID_normalized <dbl>