Merge pull request #269 from fhdsl/cleaning_improvement

Cleaning improvement

data/microplastics_in_blood.csv

@@ -6,8 +6,8 @@ A,Bisphenol A,3,-0.5
 B,BPA,5,0.5
 B,dioxin,8,0.5
 A,bpa,6,2.5
-B,O,5,3.5
+B,Other,5,3.5
 B,dioxin,2,0.5
 A,bpa,1,1.5
 B,BPA,7,1.5
-B,O,3,2.5
+B,Other,3,2.5

modules/Data_Cleaning/Data_Cleaning.Rmd

@@ -460,10 +460,9 @@ Need quotes for conditions and new values!
 ```{r}
 plastics %>% 
   mutate(microplastic_recoded = case_when(
-                               microplastic == "O" ~ "Other",
                                microplastic == "Bisphenol A" ~ "BPA",
                                microplastic == "bpa" ~ "BPA",
-                               microplastic =="dioxin" ~"Dioxin"))  %>% 
+                               microplastic == "dioxin" ~"Dioxin"))  %>% 
   count(microplastic, microplastic_recoded)
 
 ```
@@ -477,10 +476,9 @@ We didn't specify what happens to values that were already `Other` or `Dioxin` o
 ```{r, eval = FALSE}
 plastics %>% 
   mutate(microplastic_recoded = case_when(
-                               microplastic == "O" ~ "Other",
                                microplastic == "Bisphenol A" ~ "BPA",
                                microplastic == "bpa" ~ "BPA",
-                               microplastic =="dioxin" ~"Dioxin"))  %>% 
+                               microplastic == "dioxin" ~"Dioxin"))  %>% 
   count(microplastic, microplastic_recoded)
 ```
 
@@ -496,39 +494,37 @@ Note that automatically values not reassigned explicitly by
 {data_input} %>%
   mutate({variable_to_fix} = case_when({Variable_fixing}   
              /some condition/ ~ {value_for_con},
-                         TRUE ~ {value_for_not_meeting_condition})
+                     .default = {value_for_not_meeting_condition}) # need this to avoid NAs
 
 ```
 :::
 
 {value_for_not_meeting_condition} could be something new 
 or it can be the original values of the column
 
-## case_when with TRUE ~ original variable name
+## case_when with .default = original variable name
 
 ```{r}
 plastics %>% 
   mutate(microplastic_recoded = case_when(
-                               microplastic == "O" ~ "Other",
                                microplastic == "Bisphenol A" ~ "BPA",
                                microplastic == "bpa" ~ "BPA",
                                microplastic =="dioxin" ~"Dioxin",
-                                TRUE ~ microplastic)) %>%
+                               .default = microplastic)) %>%
   count(microplastic_recoded)
 ```
 
-## Typically it is good practice to include the TRUE statement
+## Typically it is good practice to include the .default statement
 
 You never know if you might be missing something - and if a value already was an NA it will stay that way. 
 
 ```{r, eval = FALSE}
 plastics %>% 
   mutate(microplastic_recoded = case_when(
-                               microplastic == "O" ~ "Other",
                                microplastic == "Bisphenol A" ~ "BPA",
                                microplastic == "bpa" ~ "BPA",
-                               microplastic =="dioxin" ~"Dioxin",
-                               TRUE ~ microplastic)) %>%
+                               microplastic == "dioxin" ~"Dioxin",
+                               .default = microplastic)) %>%
   count(microplastic, microplastic_recoded)
 ```
 
@@ -558,8 +554,7 @@ plastics %>%
                               microplastic == "Bisphenol A" ~ "BPA",
                               microplastic == "bpa" ~ "BPA",
                               microplastic == "dioxin" ~"Dioxin",
-                              microplastic == "O" ~ "Other",
-                               TRUE ~ microplastic)) %>%
+                              .default = microplastic)) %>%
   count(microplastic)
 
 ```
@@ -571,10 +566,9 @@ plastics %>%
 ```{r}
 plastics %>% 
   mutate(microplastic_recoded = case_when(
-    microplastic %in% c("dioxin", "Dioxin") ~ "dioxin", 
+    microplastic %in% c("dioxin", "Dioxin") ~ "Dioxin", 
     microplastic %in% c("BPA", "bpa", "Bisphenol A") ~ "BPA",
-    microplastic %in% c("O", "Other") ~ "Other",
-                      TRUE ~ microplastic)) %>%
+                     .default = microplastic)) %>%
 
   count(microplastic, microplastic_recoded)
 
@@ -606,6 +600,40 @@ head(plastics)
 plastics %>% 
   count(Foods, Effect)
 ```
+## Note that if you change data classes this might impact .default
+
+```{r, eval = FALSE}
+plastics <- plastics %>% 
+      mutate(Effect = case_when(
+                    blood_level_change_nM > 0 ~ "Increase",
+                    blood_level_change_nM == 0 ~ "Same",
+                    blood_level_change_nM < 0 ~ "Decrease",
+                    .default = blood_level_change_nM)) 
+# this will give an error!
+
+plastics <- plastics %>% 
+      mutate(Effect = case_when(
+                    blood_level_change_nM > 0 ~ "Increase",
+                    blood_level_change_nM == 0 ~ "Same",
+                    blood_level_change_nM < 0 ~ "Decrease",
+                    .default = as.character(blood_level_change_nM))) 
+# this works!
+
+
+```
+
+## multiple conditions with `case_when` recoding
+
+```{r}
+plastics %>% 
+      mutate(Amt_change = case_when(
+                    blood_level_change_nM > 0 & blood_level_change_nM < 2 ~ "Small",
+                    blood_level_change_nM >= 2 ~ "Large",
+                    blood_level_change_nM < 0 & blood_level_change_nM > -2 ~ "Small",
+                    blood_level_change_nM <= -2 ~ "Large",
+                    blood_level_change_nM == 0 ~ "none")) %>%
+  head()
+```
 
 ## GUT CHECK: we need to use what function with `case_when()` to modify or create a new variable?
 
@@ -616,11 +644,11 @@ B. `select()`
 C. `mutate()`
 
 
-## GUT CHECK: If we want all unspecified values to remain the same with `case_when()`, how should we complete the `TRUE ~` statement?
+## GUT CHECK: If we want all unspecified values to remain the same with `case_when()`, how should we complete the `.default` statement?
 
-A. With the name of the variable we are modifying or using as source
+A. = the name of the variable we are modifying or using as source
 
-B. With the word "same"
+B. = "same"
 
 
 # Working with strings
@@ -721,8 +749,7 @@ plastics %>%
   mutate(microplastic_recoded = case_when(
     microplastic %in% c("Dioxin", "dioxin") ~ "Dioxin",
     microplastic %in% c("BPA", "bpa", "Bisphenol A") ~ "BPA",
-    microplastic %in% c("O", "Other") ~ "Other",
-    TRUE ~ microplastic))
+    .default = microplastic))
 ```
 
 ## `case_when()` improved with `stringr`
@@ -733,9 +760,8 @@ plastics %>%
 plastics %>% 
   mutate(microplastic_recoded = case_when(
     str_detect(string = microplastic, pattern = "^b|^B") ~ "BPA",
-    str_detect(string = microplastic, pattern = "^o|^O") ~ "Other",
     str_detect(string = microplastic, pattern = "^d|^D") ~ "Dioxin",
-    TRUE ~ microplastic)) %>%
+    .default =  microplastic)) %>%
   count(microplastic, microplastic_recoded)
 ```
 
@@ -791,7 +817,7 @@ plastics_comb
 ## Summary
  -  `case_when()` requires `mutate()` when working with dataframes/tibbles
 -   `case_when()` can recode **entire values** based on **conditions** (need quotes for conditions and new values)
-    -   remember `case_when()` needs `TRUE ~ varaible` to keep values that aren't specified by conditions, otherwise will be `NA`
+    -   remember `case_when()` needs `.default = varaible` to keep values that aren't specified by conditions, otherwise will be `NA`
 
 **Note:** you might see the `recode()` function, it only does some of what `case_when()` can do, so we skipped it, but it is in the extra slides at the end.
 
@@ -827,6 +853,13 @@ Image by <a href="https://pixabay.com/users/geralt-9301/?utm_source=link-attribu
 
 # Extra Slides
 
+## `n_complete_row()` evaluating how many columns are complete for each row
+
+```{r}
+head(plastics_comb)
+head(plastics_comb) %>% n_complete_row()
+```
+
 ## `recode()` function
 
 This is similar to `case_when()` but it can't do as much.
@@ -856,8 +889,7 @@ plastics %>%
   mutate(microplastic_recoded = recode(microplastic,  
                         "Bisphenol A" =  "BPA",
                                 "bpa" = "BPA",
-                             "dioxin" = "Dioxin",
-                                  "O" = "Other")) %>%
+                             "dioxin" = "Dioxin") %>%
   count(microplastic, microplastic_recoded)
 ```
 
@@ -869,8 +901,7 @@ plastics %>%
   mutate(microplastic_recoded = recode(microplastic,  
                         "Bisphenol A" =  "BPA",
                                 "bpa" = "BPA",
-                             "dioxin" = "Dioxin",
-                                  "O" = "Other")) %>%
+                             "dioxin" = "Dioxin")) %>%
   count(microplastic, microplastic_recoded)
 ```
 
@@ -884,12 +915,40 @@ plastics %>%
   mutate(microplastic = recode(microplastic,  
                         "Bisphenol A" =  "BPA",
                                 "bpa" = "BPA",
-                             "dioxin" = "Dioxin",
-                                  "O" = "Other")) %>%
+                             "dioxin" = "Dioxin")) %>%
   count(microplastic)
 ```
 
 
+## More complicated case_when
+
+```{r}
+ces <- read_csv(file = "https://daseh.org/data/CalEnviroScreen_data.csv")
+set.seed(123)
+ces %>% mutate(new_col_case_when = 
+                 case_when(Longitude < -121 & Latitude > 37.8 ~ "Distract A",  
+                           .default = "District B")) %>% 
+                 select(Longitude, Latitude, new_col_case_when) %>%
+                 slice_sample(n = 6)
+```
+
+## Don't need `case_when()` if just calculating new variables
+
+```{r}
+ces %>% mutate(num_col_mutate = Longitude * Latitude) %>% pull(num_col_mutate) 
+
+```
+
+## `case_when()` if you want NA values
+
+```{r}
+ces %>% mutate(num_new =  case_when(
+  Longitude < -121 & Latitude > 37.8 ~  Longitude * Latitude),
+  .defualt = NA) %>% 
+  pull(num_new)
+```
+
+
 
 ## String Splitting