Drop or combine all-tx or all-co blocks — patch_singleton

Given dataset with some blocks that are all treated or all control, drop those blocks or replace the block ID with canonical new block ID shared by all such blocks.

Usage

patch_singleton_blocks(
  formula = NULL,
  data,
  drop_data = TRUE,
  pool_clusters = TRUE,
  warn_missing = TRUE
)

Arguments

formula: Notation for Y ~ Z | clusterID | blockID. If NULL, `data` is assumed to already be in canonical form.
data: Dataframe to patch.
drop_data: Drop the troublesome blocks if TRUE, pool them if FALSE.
pool_clusters: If pooling blocks rather than dropping them, also pool clusters in 100 cluster.
warn_missing: Say something if anything happens.

Value

data with the blockID column modified or troublesome rows of data missing.

Details

Missing values in the block ID are all considered singletons.

Also, depending on pool_clusters, pool the clusters in each of these identified blocks into single clusters.