resolve conflict Merge branch 'main' of https://github.com/RSGInc/tra…

…velSurveyTools

DESCRIPTION

@@ -26,7 +26,8 @@ Suggests:
     knitr (>= 1.45),
     rmarkdown (>= 2.25),
     ggplot2 (>= 3.4.4),
-    bit64
+    bit64,
+    tigris
 Config/testthat/edition: 3
 Depends: 
     R (>= 2.10)

data-raw/create_test_data.R

@@ -26,6 +26,16 @@ ids_to_keep = sample(hts_data$hh$hh_id, size = 1000)
 
 hts_data = lapply(hts_data, function(dt) dt[hh_id %in% ids_to_keep])
 
+# Create fake home counties and lat/lng for geographic analysis
+hts_data$hh[, home_county := sample(1:3, size = nrow(hts_data$hh), replace = TRUE)]
+
+hts_data$hh[, home_lat := sample(seq(from = 33.00000, to = 40.00000,
+                                     length.out = nrow(hts_data$hh)),
+                                 size = nrow(hts_data$hh), replace = TRUE)]
+
+hts_data$hh[, home_lon := sample(seq(from = -100.00000, to =-80.00000,
+                                     length.out = nrow(hts_data$hh)),
+                                 size = nrow(hts_data$hh), replace = TRUE)]
 
 ## Choose a subset of columns --------------------------------------------------
 keep_cols = c(
@@ -41,6 +51,9 @@ keep_cols = c(
   'income_followup',
   'num_people',
   'residence_type',
+  'home_county',
+  'home_lat',
+  'home_lon',
 
   # Person variables:
   'race_1',
@@ -125,7 +138,13 @@ setDT(value_labels)
 variable_list = variable_list[variable %in% keep_cols]
 value_labels = value_labels[variable %in% keep_cols]
 
+# add fake home_county labels to value_labels and remove real ones
+value_labels = value_labels[variable != 'home_county']
+
+county_labels = data.frame(variable = rep('home_county', 3), value = 1:3,
+                            label = c('Arike County', 'Clark County', 'Moore County'))
 
+value_labels = rbind(value_labels, county_labels, fill = TRUE)
 ## Subset to minimum required columns ------------------------------------------
 variable_list = variable_list[, c(
   'variable',

vignettes/geographic_summaries.Rmd

@@ -0,0 +1,50 @@
+---
+title: "Creating user-specified geographic summaries"
+description: >
+output: html_vignette
+vignette: >
+  %\VignetteIndexEntry{Creating user-specified geographic summaries}
+  %\VignetteEngine{knitr::rmarkdown}
+  %\VignetteEncoding{UTF-8}
+---
+
+```{r setup, include=FALSE}
+knitr::opts_chunk$set(echo = TRUE, message = FALSE)
+
+library(travelSurveyTools)
+library(data.table)
+library(tigris)
+```
+
+### Creating summaries by geographic variables using hts_summary
+
+`hts_summary` can summarize geographic variables.
+
+```{r, county_income, echo=TRUE, eval=TRUE}
+
+DT = hts_prep_data(summarize_var = 'income_detailed', summarize_by = 'home_county', data = test_data)
+
+output = hts_summary(prepped_dt = DT$cat, summarize_var = 'income_detailed', summarize_by = 'home_county')
+
+factorize_df(output$summary$unwtd, vals_df = value_labels, value_label_colname = 'label')
+
+```
+
+### Using spatial_join to join together separate shapefiles
+
+`spatial_join` can link together multiple geographies.
+
+```{r, spatial_join, echo=TRUE, eval=TRUE}
+
+states = states()
+
+hh = join_spatial(
+    hh,
+    dplyr::select(states, home_state = NAME),
+    id_col = 'hh_id',
+    lon_col = 'home_lon',
+    lat_col = 'home_lat',
+    largest = TRUE)
+
+hh[, .N, home_state]
+```