diff --git a/scripts/phase2/transform_data_pca.r b/scripts/phase2/transform_data_pca.r
index 54cd5bb62979bf12c0a62b05460d203dc8301bdd..6a08dfb6adaf24fd5a683245a547bad32e89b12a 100644
--- a/scripts/phase2/transform_data_pca.r
+++ b/scripts/phase2/transform_data_pca.r
@@ -149,17 +149,15 @@ data <- data %>% mutate(LTD_INDICATOR = case_when(LTD_INDICATOR == "With LTD"
 data <- data %>% mutate(ACTIVE_TERMED = case_when(ACTIVE_TERMED == "Active"     ~ 1,
                                                   ACTIVE_TERMED == "Terminated" ~ 0))
 
-# keep numerics for PCA
-data <- data %>% select(#"REGION", "INDUSTRY",
-                        #"COVG_CODE",
-                        "AVG_SALARY", "AVG_AGE", "PCT_FEMALE",
+# numerics only for PCA
+data <- data %>% select("AVG_SALARY", "AVG_AGE", "PCT_FEMALE",
                         "TRUE_GROUP_VOL", "LTD_INDICATOR", "ACTIVE_TERMED",
                         "MAX_LIVES", "POLICY_DURATION", "PREM", "EST_ANNUALIZED_NET_PREM",
                         "RTN",
                         "PAID_COMMISSION", "PAID_CLAIMS", "IBNR",
                         "PERCENT_COMMISSION", "PREMIUM_TAX", "INTERNAL_EXPENSES", "PERCENT_PEPM")
 
-# use FactoMineR for PCA
+# use FactoMineR to compute PCA
 data.pca <- PCA(data, scale.unit = TRUE, ncp = 5, graph = TRUE)
 
 # create scree plot