data

AmericanMoviesByYear.csv

• Movie titles from Wikipedia's Lists of American Movies, 1900-2017

Billion-Dollar Distasters.csv

• Slightly cleaned data from Billion-Dollar Weather and Climate Disasters: Table of Events
• Referenced by 538 story, Hurricane Harvey’s Impact — And How It Compares To Other Storms
• Added derived columns that attempt to create shortened event names for graph point labeling.
• Examples:
• "Hurricane Matthew October 2016" -> "Matthew 2016"
• "Western Drought 2015" -> "Drought 2015"
• "Colorado Flooding September 10-16 2013" -> "Flooding 2013"
• "Colorado Hail Storm and Central Severe Weather May 2017" -> "Severe Weather 2017"
• "South/ Southeast Severe Weather March 2017" -> "Severe Weather 2017 2"

Cape Town rainfall averages

• capetownrainfallsince1977.csv is data from the CSAG JSON file: http://cip.csag.uct.ac.za/monitoring/gsod/data/688160_gsod_pr_day_1977-2018.json
  See graphs at http://www.csag.uct.ac.za/current-seasons-rainfall-in-cape-town/
• capetown_rainfallsince.csv scraped from the graph in Guardian story on 3 year drought
  So the values are only approximate.
  I did a couple remakes on Twitter

crossword_times.csv

• Four years of my own results for the New York Times daily crossword. Includes, time spent and whether I successfully completed the puzzle or not.
• I made some graphs in a Twitter thread
• Latest graphs with all four years of data

Du Bois Georgia Map 1899

• Data shown on the map by W.E.B. Du Bois of land ownership by blacks in 1899.
• Note that the counties are a little different from today. You can find historical shapefiles at the Atlas of Historical County Boundaries at Newberry Library.
• My reproduction is on Twitter at https://twitter.com/xangregg/status/1094401138353418240.

greenland_surface_melt.csv

• Data behind National Snow and Ice Data Center's Greenland surface melt chart.
• Missing days have been interpolated and a separate column indicates which dates were interpolated.
• A few graphs tweeted at https://twitter.com/xangregg/status/1141133285088735233.

internet_speeds_summer_2018.csv

• Results from downloading a 10MB file 3 times an hour over 5+ weeks on my "up to 60Mb/s" connection.
• A few graphs tweeted at https://twitter.com/xangregg/status/1028665493689913344.

journal.pone.0216362.s003.csv

• Data from Battle for the thermostat: Gender and the effect of temperature on cognitive performance
• Converted form Stata DTA from to CSV
• Added a few derived fields
  • Added rounded version of temp field, so "19.1000003814697" is "19.1". I'm guessing the Stata file was compressed to use single precision floating point numbers.
  • Computed wordscore for the word problem, since the participants were rewarded by total score rather than number complete.
  • Added Gender=F|M from Male=0|1
  • Added "Temp Category" to combine indicator variables: cool, normal, warm, hot

MosquitoTrendsData.csv

• Google Trends mosquito search data used for state grid blog post

pubdelays100.csv

• 100 publication delays for research with positive (70) and negative (30) results. Approximated from CDF graph in JAMA paper Delays in the Publication of Important Clinical Trial Findings in Oncology

scented candle reviews.csv

• Selected features from 50,000 scented candle reviews across 20+ products
• Features: 0 or 1 indicating whether review title or text included selected terms
  • no smell narrow: no smell|scent
  • no smell wide: no|zero|0|nothing smell|scent|fragrance|aroma
  • bad smell: bad|awful|terrible smell|scent|fragrance|aroma
  • broken: broken|shattered
  • overwhelm: overwhelm|overpower
  • artificial: artificial|synthetic
• Related tweeted charts
• Note: Amazon merges reviews for some candles.
  • Yankee Candle Balsam & Cedar includes White Christmas
  • Yankee Candle Vanilla Cupcake includes Apple Pumpkin
  • Yankee Candle Pineapple Cilantro includes Alfresco Afternoon
  • Yankee Candle Sage & Citrus includes Honey Clementine
  • Yankee Candle Pink Sands includes Calm & Quiet Place
  • Chesapeake Bay Balance+Harmony includes Strength+Energy

shirt_sponsorship.csv

• Inferred data on Premier league shirt sponsorship over time from this FT article. Use for experimenting with alternates to streams.

travel_motivation_recoded_stacked.csv

• Recoded subset of survey responses from /r/travel.
• Includes a few demographic fields and responses to the question "What motivates you to travel?"
• This was a free-form text field. I recoded the responses into common terms and expanded multiple responses into multiple rows (with same respondent id).
• For instance, "new cultures" could have been originally "different cultures", "see new cultures", "other lifestyles", ...
• Surely, some bias or error was introduced during the recoding process.
• I made a slopegraph of the responses by gender.

tukey_EDA_7.8_coal.csv

• Coal production data from John Tukey's 1977 book Exploratory Data Analysis, for time series smoothing
• Includes two other overlapping sources: coal census -1956 and EIA 1949-.
• The "combined" column uses data from the other sources for years where Tukey's table doesn't match his graphs

vaccine_safety_expressing, vaccine_safety_tcell_secretions

• Data shown in Figure 2 of Covid-19 vaccine safety experiment paper: https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)31208-3/fulltext
• scraped from PDF after converting to SVG, so should be high resolution and complete, but...
• For the secretions scatter plots, each combination should have 36 dots, but the Day 0 groups are missing a few. Most values for those groups are clustered at the bottom of the scale so possibly they are clipped or omitted for space. Or maybe I made a mistake.
• The pies show 7 of the 8 possible combinations of expressions. Not sure if the all negative category is omitted because it's empty or because it's uninteresting.
• Proportions multiplied by 36 do not generally match integers. Seems like they should.

vitamin d survival.csv.csv

• Data digitized from scatterplot in vitamin D study: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0239799
• Some make-over charts I posted to Twitter

wiseparatext.csv

• Timing test results for log scale bars makeover blog post

World rail line colors

• Rail colors shown in Nicholas Rougeux's Color of Rails project
• With changes as documented in world_rail_line_colors.md