DDI_KEN_2012-2023_RTC_v01_M Development Data Group 2024-05-23 Metadata Editor KEN_2012-2023_RTC_v01_M Road Traffic Crashes 2012-2023 Derived from Crowdsourced Reports from Ma3Route RTC 2012-23 Sveta Milusheva World Bank Metadata Editor Sveta Milusheva Guadalupe Bedoya Robert Marty Amy Dolinger Arianna Legovini The datasets contain the time and location of road traffic crashes in Kenya (primarily Nairobi); crash information is derived from crowdsourced reports from @Ma3Route. Ma3Route is a mobile/web/SMS platform that crowdsources transport data and provides users with information on traffic, road traffic crash (RTC), matatu directions and driving reports. Users post RTC or traffic information to Ma3Route, where Ma3Route then publishes the post on Twitter. Tweets from @Ma3Route were queried using the Twitter API (tweets were no longer queried once Twitter rebranded to X). This project geolocated the location of road traffic crashes based on crowdsourced reports of crashes from Ma3Route, a mobile/web/SMS platform that crowdsources transport data Kenya Primarily Nairobi, Kenya Road traffic crashes Observation data/ratings [obs] All tweets from @Ma3Route from August 2012 to July 2023 Internet [int] Sveta Milusheva Guadalupe Bedoya Robert Marty Amy Dolinger Arianna Legovini Milusheva S, Marty R, Bedoya G, Williams S, Resor E, et al. (2021) “Applying machine learning and geolocation techniques to social media data (Twitter) to develop a resource for urban planning.” PLOS ONE 16(2): e0244317. https://doi.org/10.1371/journal.pone.0244317 The user of the data acknowledges that the original collector of the data, the authorized distributor of the data, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses. ma3route_crashes_algorithmcode.dta 31064 10 .dta Includes crashes from August 2012 through July 2023; whether a tweet reports a crash and the geolocation of the crash are determined by the described in Milusheva et al. (2021) ma3route_crashes_manualcode.dta 2595 10 .dta Includes crashes from July 2017 through July 2018; whether a tweet reports a crash and the geolocation of the crash are manually coded. Unique crash ID 31064 1 31064 Date/time of the crash (using date/time of the first tweet that reported the crash 31064 Date of the crash (using date of the first tweet that reported the crash) 31064 Latitude 31064 -3.10000000200807 -0.565402313796744 Longitude 31064 36.2839500025994 37.8794900005407 Number of tweets that reported crash 31064 1 66 1 25252 2 3551 3 1047 4 445 5 241 6 163 7 89 8 86 9 40 10 23 11 24 12 20 13 17 14 10 15 4 16 13 17 5 18 3 19 5 20 3 21 3 22 1 23 2 25 2 26 2 27 1 28 1 30 1 31 1 32 1 34 1 35 1 38 1 40 1 41 1 46 1 51 1 66 1 Whether the tweet contains one of the words: 'dead', 'died', 'body', 'killed', or 'fatal' 31064 0 1 0 No 28780 1 Yes 2284 Whether the tweet contains the word: 'pedestrian' 31064 0 1 0 No 30120 1 Yes 944 Whether the tweet contains the word: 'matatu' 31064 0 1 0 No 28523 1 Yes 2541 Whether the tweet contains one of the words: 'boda', 'motorycle', or 'motor cycle' 31064 0 1 0 No 29922 1 Yes 1142 Unique crash ID 2595 1 3115 Date/time of the crash (using date/time of the first tweet that reported the crash 2595 Date of the crash (using date of the first tweet that reported the crash) 2595 Latitude 2595 -4.059868 1.257331 Longitude 2595 34.145808 40.171389 Number of tweets that reported crash 2595 1 67 1 2014 2 321 3 112 4 54 5 29 6 16 7 8 8 8 9 9 10 5 11 2 12 2 13 1 14 4 15 2 16 1 18 2 20 1 35 1 42 1 49 1 67 1 Whether the tweet contains one of the words: 'dead', 'died', 'body', 'killed', or 'fatal' 2595 0 1 0 No 2429 1 Yes 166 Whether the tweet contains the word: 'pedestrian' 2595 0 1 0 No 2534 1 Yes 61 Whether the tweet contains the word: 'matatu' 2595 0 1 0 No 2386 1 Yes 209 Whether the tweet contains one of the words: 'boda', 'motorycle', or 'motor cycle' 2595 0 1 0 No 2500 1 Yes 95