Overview

Workshop Goals

• Demonstrate how to work with CTPP data in R
• Derive some comparable variables and estimates from the NHTS
• Visualize estimates from multiple sources
• Create reproduceable workshop materials to publish online

Topics

1. Setting Up an R Analysis Environment
2. Accessing CTPP Tables
3. Visualizing CTPP Estimates
4. Merging CTPP over Time
5. Merging Estimates with the NHTS
6. Extending CTPPr Visualizations
7. Producing a Travel Analysis Report

“House Cleaning”

• 90 minutes!?
• Hold on to big questions
• This presentation is reproducible
• Take it with you, download it later, go at your own pace

This presentation is intended for re-use

• Focus on the presentation itself today
  ↪ Later, follow along in RStudio at your own pace

Presentation hotkeys

Setting Up an R Analysis Environment

Click here for latest installation instructions

Instructions provide explicit download links for

• R
• R Tools
• RStudio

Once installed, open RStudio

Setting Up an R Analysis Environment (cont.)

Make sure you have the “CTPPr” R software package installed

install.packages("devtools")
devtools::install_github("Westat-Transportation/CTPPr")

Some portions of this workshop will use NHTS data and will also require the “summarizeNHTS” package

install.packages("devtools")
devtools::install_github("Westat-Transportation/summarizeNHTS")

OK, we’re ready!

• Your R environment for CTPP analysis is set up
• You only need to set up an environment once
• From now on, just open RStudio and load CTPPr package

library(CTPPr)

Setting Up an R Analysis Environment: RStudio + Workshop Materials

RStudio Document Outline

Accessing CTPP Tables

• Browse available tables
• Request tables
• Review additional options
• Summarize and subset the output of a request

Just a reminder, let’s make sure the CTPPr package is loaded in R.

library(CTPPr)

Accessing CTPP Tables: Browsing Available Tables

• ctpp_tables is a function that lists the CTPP tables available for request
  • The default output is an interactive HTML page
  • Optionally specify a traditional table output

ctpp_tables()

Accessing CTPP Tables: Requesting Tables

# TOTAL WORKERS (1) (WORKERS 16 YEARS AND OVER)
A102101 <- download_ctpp(A102101)

• You can download tables from the 2006-2010 or 2012-2016 dataset using the download_ctpp function.
  • Downloads directly from the Beyond 20/20 sites.
• This basic download_ctpp request is accepting default options
  • For example, '2012-2016' is the default dataset parameter
• A102101 is the new object we created.
  • Note: It can be named anything to include more detail, e.g. A101100_2016_OREGON_COUNTY
• The tables outputted by download_ctpp always return:
  • The Residence or Workplace geography
  • Any categorical variable group, labelled
  • The Estimate and its Standard Error

Accessing CTPP Tables: Requesting Tables with more Detail

# TRAVEL TIME TO WORK (12) (WORKERS 16 YEARS AND OVER)
A102110 <- download_ctpp(
    id = "A102110", 
    geography = "County", 
    state = "Oregon", 
    dataset = "2016", 
    output = "Name"
)

• download_ctpp parameters
  • Required:
    • id - Table ID
  • Important:
    • geography - Geography level of results
    • state - State(s) to limit results by
    • dataset - Dataset year
    • output - Naming output type

Accessing CTPP Tables: Documentation

• Refer to documentation within R using

?download_ctpp

R Documentation for download_ctpp

Accessing CTPP Tables: Review the Output

• View data directly in R console

A102110

• Use the summary or str function to get an overview of the data structure

summary(A102110)
str(A102110)

• View with interactivity

View(A102110)

• If you need to remember what was in your request, review the attributes:

attributes(A102110)[c("id","name","report_type","dataset","output","geography")]

Accessing CTPP Tables: General Subsetting

• Select specific columns

# By position
A102110[, c(1, 2, 3)]

# By name (single variable)
A102110$Estimate

# By name
A102110[, c("Travel Time 12", "Estimate")]

• Select specific rows

# By row numbers (first 5 rows)
A102110[1:5, ]

# By condition
A102110[RESIDENCE == "Multnomah County, Oregon", ]

# By condition (multiple values)
A102110[`Travel Time 12` %in% c("Less than 5 minutes","90 minutes or more"), ]

Accessing CTPP Tables: Subsetting Total Records

• Total number records in tables are convenient but require our attention before analysis

• For example, A102110 (Travel Time to Work), contains multiple total levels

  • All workers
  • All workers - Worked at home / Did not work at home

• By subsetting, we can separate this request into 3 tables for analysis

A102110_total <- A102110[`Travel Time 12` %in% c("All workers")]
A102110_subtotal <- A102110[`Travel Time 12` %in% c("Worked at home", "Did not work at home:")]

• Now remove the appropriate total variables from our original request

A102110 <- A102110[!`Travel Time 12` %in% c("All workers","Did not work at home:"), ]

Accessing CTPP Tables: DIY

Using the documentation for download_ctpp, request the following estimate from CTPP

• Estimate: Means of transportation to work by age
• Dataset: Most recent
• Geography: State
• Labelled output: Names

B102203 <- download_ctpp(
    id = "B102203", 
    geography = "State",
    dataset = "2016",
    output = "Name"
)

Visualizing CTPP Estimates

• Demonstrate 3 core visualization functions
  • make_table - Create report-ready, formatted tables.
  • make_chart - Create interactive bar charts.
  • make_map - Create interactive choropleth maps.
• Quick and easy steps for visuals:
  1. Assign the output of download_ctpp to a new object.
  2. Feed this new object to one of the visualization functions to review results.

Visualizing CTPP Estimates: Tables - Intro

• Let’s look at default table outputs for the two tables we just requested

make_table(A102101)

make_table(A102110)

Visualizing CTPP Estimates: Tables - Details

• Default options make a best guess based on type and length of data

• Use different options to improve your desired output

• make_table parameters

  • Required:
    • tbl - Table object
  • Important:
    • row_vars - variable(s) to put at the row level
    • col_vars - variable(s) to put at the column level
    • title - output with a title
    • confidence - Confidence level for margin of error

• Using the A102110 table from the previous slide, we can provide a title, specify a confidence level, and flip default row/column output

make_table(
    tbl = A102110,
    row_vars = "RESIDENCE",
    title = "Travel Time to Work for Counties in Oregon",
    confidence = 0.95
)

Visualizing CTPP Estimates: Charts - Intro

make_chart(A102101)

Visualizing CTPP Estimates: Charts - Details

• make_chart parameters
  • Required:
    • tbl - Table object
  • Important:
    • x - variable for x-axis
    • y - variable for y-axis
    • fill - variable to categorize by color
    • facet - variable to categorize smaller multiple charts by
    • flip - rotate chart by axis
    • order - order values by frequency vs default
    • pallette - a color palette to use
    • ..and more
• Using the A102110 table from the previous slide, let’s experiment with different outputs…

Visualizing CTPP Estimates: Charts - Example

• Remove the legend

make_chart(
    tbl = A102101,
    legend = FALSE
)

Visualizing CTPP Estimates: Charts - Example 2

• Order the values

make_chart(
    tbl = A102101,
    legend = FALSE,
    order = TRUE
)

Visualizing CTPP Estimates: Charts - Example 3

• Rotate it

make_chart(
    tbl = A102101,
    legend = FALSE,
    order = TRUE,
    flip = TRUE
)

Visualizing CTPP Estimates: Maps - Intro

• make_map is our geographic visualization helper
• Intro example: Let’s map median travel time to work for bike commuters across each state

# B206203 - MEDIAN TRAVEL TIME (1) BY MEANS OF TRANSPORTATION (18)(WORKERS 16 YEARS AND OVER WHO DID NOT WORK AT HOME)
B206203 <- download_ctpp(
    id = "B206203",
    geography = "State",
    dataset = "2016",
    output = "FIPS Code"
)

# Subset to include just bicycle commuters
bike_commuters <- B206203[`Means of Transportation 18` == "Bicycle", list(WORKPLACE, Estimate, SE)]

Visualizing CTPP Estimates: Maps - Intro Example

map <- make_map(
    tbl = bike_commuters,
    tbl_geo_id = "WORKPLACE",
    geo_layer = "resources/shapefile/composite_state_layer/composite_state_layer.shp"
)

map

Visualizing CTPP Estimates: Maps - Details

• make_map parameters
  • Required:
    • tbl - Table object
    • tbl_geo_id - Column name of table with geography id
    • geo_layer - Location of shapefile
  • Important:
    • geo_layer_id - Column name of shapefile with geography id
    • geo_layer_name - Column name of shapefile with geography name
    • tbl2 - Optional second table with same geography level for including as tooltip
• The values in tbl_geo_id and geo_layer_id will need to be in the same format to match

Visualizing CTPP Estimates: Maps - More

Include a second table grouping by the original geography plus one variable.

tbl <- B206203[`Means of Transportation 18` == "Total, means of transportation", list(WORKPLACE, Estimate, SE)]
tbl2 <- na.omit(B206203)

map <- make_map(
    tbl = tbl,
    tbl2 = tbl2,
    tbl_geo_id = "WORKPLACE",
    geo_layer = "resources/shapefile/composite_state_layer/composite_state_layer.shp"
)

map

Formatting

• All 3 visualization function support the same value formatting options:
  • digits - Number of decimal places to use
  • percentage - Treat proportions as percentages
    • TRUE or FALSE
  • scientific - Use scientific notation
    • TRUE or FALSE
  • multiplier - A value multiplier
    • Ex: To display a value “In Thousands”, use multiplier = 1000
• Formatting example with make_table

# The top 10 states with the most workers
top_10 <- head(A102101[order(-Estimate), ], 10)

make_table(
    tbl = top_10,
    title = "Top 10 - Total Workers (Millions)",
    digits = 2,
    multiplier = 1000000
)

Merging CTPP Datasets

• Use download_ctpp function to make requests to multiple datasets
• Organize tables and labels
• Merge estimates
• Visualize merged results over time in table, charts, and maps with make_* functions
• Perform significance test

Merging CTPP Datasets: Request

• Select a table available in both datasets

• Make both requests:

2010

B106202_2010 <- download_ctpp(
 id = "B106202",
 geography = "County",
 state = "Oregon",
 dataset = "2010",
 output = "FIPS and Name"
)

2016

B106202_2016 <- download_ctpp(
 id = "B106202",
 geography = "County",
 state = "Oregon",
 dataset = "2016",
 output = "FIPS and Name"
)

Merging CTPP Datasets: Organize

• We requested “FIPS and Name” in our output because having both is more convenient
  • The geography column is a | delimited column
• Let’s make copies so that we have a FIPS code version and a named version

2010

# Make a FIPS code copy
B106202_2010_FIPS <- data.table::copy(B106202_2010)
B106202_2010_FIPS[, RESIDENCE:=strsplit(RESIDENCE, " \\| ")[[1]][1], by=RESIDENCE]

# Named version
B106202_2010[, RESIDENCE:=strsplit(RESIDENCE, " \\| ")[[1]][2], by=RESIDENCE]

2016

# FIPS code version
B106202_2016_FIPS <- data.table::copy(B106202_2016)
B106202_2016_FIPS[, RESIDENCE:=strsplit(RESIDENCE, " \\| ")[[1]][1], by=RESIDENCE]

# Named version
B106202_2016[, RESIDENCE:=strsplit(RESIDENCE, " \\| ")[[1]][2], by=RESIDENCE]

Merging CTPP Datasets: Review

• We can merge our tables now or visualize independently

make_table(
  tbl = B106202_2010,
  row_vars = c("RESIDENCE","Means of Transportation 18")
)

make_table(
  tbl = B106202_2016,
  row_vars = c("RESIDENCE","Means of Transportation 18")
)