This function creates a column that contains a single observation for each unique value in the key variable. For each feature, a count corresponding to the number of times that feature is identified in a cluster for the give category is also provided.
qm_summarize(ref, key, clusters, category, count, geometry = TRUE, use.na = FALSE)An sf object that serves as a master list of features
Name of geographic id variable in the ref object to match input values to
A tibble created by qm_combine with two or more clusters worth of data
Value of the CAT variable to be analyzed
How should clusters be summarized: by counting each time a feature is included
in a cluster ("clusters") or by counting the number of respondents
("respondents") who associated a feature with the given category.
A logical scalar that returns the full geometry and attributes of ref
when TRUE (default). If FALSE, only the key and count of features is
returned after validation.
A logical scalar that returns NA values in the count variable if a feature
is not included in any clusters when TRUE. If FALSE (default), a 0 value
is returned in the count variable for each feature that is not included in any clusters. This
parameter only impacts output if the geometry argument is TRUE.
A tibble or a sf object (if geometry = TRUE) that contains a count of the number
of clusters a given feature is included in. The tibble option (when geometry = FALSE) will only
return valid features. The sf option (default; when geometry = TRUE) will return all
features with either zeros (when use.na = FALSE) or NA values (when use.na = TRUE)
for features not included in any clusters.
qm_combine
# load and format reference data
stl <- stLouis
stl <- dplyr::mutate(stl, TRACTCE = as.numeric(TRACTCE))
# create clusters
cluster1 <- qm_define(118600, 119101, 119300)
cluster2 <- qm_define(119300, 121200, 121100)
# create cluster objects
cluster_obj1 <- qm_create(ref = stl, key = TRACTCE, value = cluster1,
rid = 1, cid = 1, category = "positive")
#> old-style crs object detected; please recreate object with a recent sf::st_crs()
cluster_obj2 <- qm_create(ref = stl, key = TRACTCE, value = cluster2,
rid = 1, cid = 2, category = "positive")
#> old-style crs object detected; please recreate object with a recent sf::st_crs()
# combine cluster objects
clusters <- qm_combine(cluster_obj1, cluster_obj2)
# summarize cluster objects
positive1 <- qm_summarize(ref = stl, key = TRACTCE, clusters = clusters, category = "positive",
count = "clusters")
#> old-style crs object detected; please recreate object with a recent sf::st_crs()
class(positive1)
#> [1] "sf" "data.frame"
mean(positive1$positive)
#> [1] 0.05660377
# summarize cluster objects with NA's instead of 0's
positive2 <- qm_summarize(ref = stl, key = TRACTCE, clusters = clusters, category = "positive",
count = "clusters", use.na = TRUE)
#> old-style crs object detected; please recreate object with a recent sf::st_crs()
class(positive2)
#> [1] "sf" "data.frame"
mean(positive2$positive, na.rm = TRUE)
#> [1] 1.2
# return tibble of valid features only
positive3 <- qm_summarize(ref = stl, key = TRACTCE, clusters = clusters, category = "positive",
count = "clusters", geometry = FALSE)
class(positive3)
#> [1] "tbl_df" "tbl" "data.frame"
mean(positive3$positive)
#> [1] 1.2
# count respondents instead of clusters
positive4 <- qm_summarize(ref = stl, key = TRACTCE, clusters = clusters, category = "positive",
count = "respondents")
#> old-style crs object detected; please recreate object with a recent sf::st_crs()
mean(positive4$positive)
#> [1] 0.04716981