Claim at v1.2.0
mob-speedcam-247
e883694d2551e6f029f961eb4941e935447df84b7b029262a545d5d9d24b8859
Data
{
"has_outcome": true,
"kind": "confirmation",
"lineage": {
"audit": "docs/lineage_audit.md; data/derived/lineage_audit.json",
"class": "none",
"class_meaning": "the audit found no city-data source fetched before registration",
"classes": [],
"early_sources": []
},
"outcome": "mob-speedcam-247",
"prior": 0.3,
"registered_at": "2026-11-12T15:15:19-0500",
"registrant": "mobility agent (Claude Opus 5.5), for J. Vernon",
"registration": {
"alternatives": [
"Differential post-pandemic recovery of night and weekend traffic (nightlife) relative to weekday daytime",
"Changes in NYPD crash reporting or in the share of crashes with location and time recorded",
"E-bike and moped growth, concentrated in evening delivery hours",
"Regression to the mean if 2021-22 night crash levels were unusually high"
],
"graph_evidence": [
{
"findings": [
{
"ci_lower": 0.63,
"ci_upper": 0.85,
"pubmed_id": 17666698,
"statistic_type": "relative_risk_ratio",
"statistic_value": 0.73,
"study_id": "w7LPWoPkj7",
"variable_1_name": "speed cameras installed in an urban setting",
"variable_2_name": "number of road collisions"
},
{
"ci_lower": null,
"ci_upper": null,
"pubmed_id": 21994881,
"statistic_type": "mean_difference_unstandardized",
"statistic_value": 5.2,
"study_id": "D_P0qVt-n4",
"variable_1_name": "Installation of speed control cameras (2000-2006 vs 2007-2010)",
"variable_2_name": "vehicular injury death rate per 100,000"
},
{
"ci_lower": 0.42,
"ci_upper": 0.57,
"pubmed_id": 14693888,
"statistic_type": "relative_risk_ratio",
"statistic_value": 0.49,
"study_id": "qVhRrEdQ5j",
"variable_1_name": "Use of mobile speed cameras",
"variable_2_name": "Rate ratio of injurious crashes at intervention and control sites"
},
{
"ci_lower": 0.9,
"ci_upper": 1.1,
"pubmed_id": 17666698,
"statistic_type": "relative_risk_ratio",
"statistic_value": 0.99,
"study_id": "w7LPWoPkj7",
"variable_1_name": "speed cameras on Barcelona's beltway",
"variable_2_name": "people injured (among the population affected by the speed cameras on Barcelona's beltway)"
},
{
"ci_lower": null,
"ci_upper": null,
"pubmed_id": 34424783,
"statistic_type": "mean_difference_unstandardized",
"statistic_value": -4.4,
"study_id": "Gcjc7NFRGZ",
"variable_1_name": "Camera Zone (300 meters of the speed cameras) [vs Non-Camera Zone]",
"variable_2_name": "Average Speed (among Iranian taxi drivers in southern Tehran)"
},
{
"ci_lower": null,
"ci_upper": null,
"pubmed_id": 34424783,
"statistic_type": "mean_difference_unstandardized",
"statistic_value": -1.8,
"study_id": "Gcjc7NFRGZ",
"variable_1_name": "Camera Zone (300 meters of the speed cameras) [vs Non-Camera Zone]",
"variable_2_name": "Average Speed (among Iranian taxi drivers in southern Tehran)"
},
{
"ci_lower": null,
"ci_upper": null,
"pubmed_id": 26648375,
"statistic_type": "mean_difference_unstandardized",
"statistic_value": 6.4,
"study_id": "c0Ao6MYQL4",
"variable_1_name": "vehicles 200 meters after fixed cameras [vs vehicles on roadways without speed cameras]",
"variable_2_name": "speeding vehicles"
}
],
"note": "No speed-camera concept in the graph; variable search found 7 associations in 6 studies."
},
{
"note": "Abstract check: PMID 17666698 (Barcelona) reports RR 0.73 (0.63-0.85) for collisions on the beltway with cameras; the graph's second association from that paper (RR 0.99, 0.90-1.10, labelled 'people injured among the population affected by the speed cameras') is the comparison arterial roads without cameras, so it is mis-extracted. PMID 14693888 (South Wales, mobile cameras) reports injurious crash rate ratio 0.49 (0.42-0.57) within 500 m routes, with decreases in day and night."
}
],
"id": "mob-speedcam-247",
"kind": "confirmation",
"literature": [
{
"citation": "Wilson C, Willis C, Hendrikz JK, Le Brocque R, Bellamy N (2010). Speed cameras for the prevention of road traffic injuries and deaths. Cochrane Database Syst Rev (10):CD004607.",
"doi": "10.1002/14651858.CD004607.pub4",
"finding": "Reductions in injury crashes near camera sites of roughly 8-49%, fatal and serious 11-44%."
},
{
"citation": "NYC DOT (2023, 2025). Speed camera program reports and press releases.",
"doi": null,
"finding": "About 7.6% fewer injuries in the new overnight and weekend hours at camera locations in the first year (agency figure)."
}
],
"plan": "Data: NYPD Motor Vehicle Collisions - Crashes (data.cityofnewyork.us h9gi-nx95), aggregated by SoQL to counts by month, day of week and hour. Exposure groups: new hours = Mon-Fri 22:00-05:59 and all Sat-Sun; old hours = Mon-Fri 06:00-21:59. Outcomes: crashes with >= 1 person injured or killed (primary), persons injured, pedestrians injured, persons killed. Model: Poisson GLM of monthly counts on group x month fixed effects ... specifically log E[count] = group + month(calendar) + group x post, on a 24-month window Aug 2021-Jul 2023, with Pearson-scaled (quasi-Poisson) standard errors. Event-study figure of the new/old ratio by month, Jan 2020 - Dec 2024, to show pre-trends. Placebo: fake start 1 Aug 2021 on Aug 2020-Jul 2022. Mechanism check: camera tickets ('PHTO SCHOOL ZN SPEED VIOLATION') by issue hour in Open Parking and Camera Violations (nc67-uf89) should appear in the new hours from Aug 2022. Design limits: citywide counts dilute a local effect; night and weekend travel recovered from the pandemic at a different pace from weekday daytime travel, which can bias the DiD in either direction; no exposure (vehicle-miles) by hour.",
"prediction": {
"direction": "-",
"magnitude": "Citywide, because cameras cover a small share of street length, the ratio of ratios (injury crashes in new hours / old hours, Aug 2022-Jul 2023 vs Aug 2021-Jul 2022) between 0.92 and 1.00. Persons injured and pedestrians injured are secondary outcomes.",
"refute_if": "Ratio of ratios at or above 1 with an interval that excludes 0.97.",
"support_if": "Ratio of ratios below 1 with a 95% interval below 1."
},
"prior": 0.3,
"registered_at": "2026-11-12T15:15:19-0500",
"registrant": "mobility agent (Claude Opus 5.5), for J. Vernon",
"sha256": "dceaaebfd4c399d50f80cf40c4ef0435101593ff9e9723f0a57f810a9c422043",
"statement": "Extending New York City's school-zone speed cameras to 24-hour, 7-day operation on 1 August 2022 reduced injury crashes in the newly enforced hours (weekday 10 pm-6 am and weekends) relative to the weekday 6 am-10 pm hours that were already enforced."
},
"registration_commit": {
"commit": "f2d4f724d505e61fea1b78c9d80bdc50799e5722",
"committed_at": "2026-11-12T15:15:27-05:00"
},
"registration_file_sha256": "033fdd7cd638a2d19d63dbb0694c2b51ee32e9bf99bef497605570576d9fff7e",
"registration_path": "claims/registered/mob-speedcam-247.json",
"registration_sha256": "dceaaebfd4c399d50f80cf40c4ef0435101593ff9e9723f0a57f810a9c422043",
"registration_text": "{\n \"id\": \"mob-speedcam-247\",\n \"kind\": \"confirmation\",\n \"statement\": \"Extending New York City's school-zone speed cameras to 24-hour, 7-day operation on 1 August 2022 reduced injury crashes in the newly enforced hours (weekday 10 pm-6 am and weekends) relative to the weekday 6 am-10 pm hours that were already enforced.\",\n \"graph_evidence\": [\n {\n \"note\": \"No speed-camera concept in the graph; variable search found 7 associations in 6 studies.\",\n \"findings\": [\n {\n \"variable_1_name\": \"speed cameras installed in an urban setting\",\n \"variable_2_name\": \"number of road collisions\",\n \"statistic_type\": \"relative_risk_ratio\",\n \"statistic_value\": 0.73,\n \"ci_lower\": 0.63,\n \"ci_upper\": 0.85,\n \"pubmed_id\": 17666698,\n \"study_id\": \"w7LPWoPkj7\"\n },\n {\n \"variable_1_name\": \"Installation of speed control cameras (2000-2006 vs 2007-2010)\",\n \"variable_2_name\": \"vehicular injury death rate per 100,000\",\n \"statistic_type\": \"mean_difference_unstandardized\",\n \"statistic_value\": 5.2,\n \"ci_lower\": null,\n \"ci_upper\": null,\n \"pubmed_id\": 21994881,\n \"study_id\": \"D_P0qVt-n4\"\n },\n {\n \"variable_1_name\": \"Use of mobile speed cameras\",\n \"variable_2_name\": \"Rate ratio of injurious crashes at intervention and control sites\",\n \"statistic_type\": \"relative_risk_ratio\",\n \"statistic_value\": 0.49,\n \"ci_lower\": 0.42,\n \"ci_upper\": 0.57,\n \"pubmed_id\": 14693888,\n \"study_id\": \"qVhRrEdQ5j\"\n },\n {\n \"variable_1_name\": \"speed cameras on Barcelona's beltway\",\n \"variable_2_name\": \"people injured (among the population affected by the speed cameras on Barcelona's beltway)\",\n \"statistic_type\": \"relative_risk_ratio\",\n \"statistic_value\": 0.99,\n \"ci_lower\": 0.9,\n \"ci_upper\": 1.1,\n \"pubmed_id\": 17666698,\n \"study_id\": \"w7LPWoPkj7\"\n },\n {\n \"variable_1_name\": \"Camera Zone (300 meters of the speed cameras) [vs Non-Camera Zone]\",\n \"variable_2_name\": \"Average Speed (among Iranian taxi drivers in southern Tehran)\",\n \"statistic_type\": \"mean_difference_unstandardized\",\n \"statistic_value\": -4.4,\n \"ci_lower\": null,\n \"ci_upper\": null,\n \"pubmed_id\": 34424783,\n \"study_id\": \"Gcjc7NFRGZ\"\n },\n {\n \"variable_1_name\": \"Camera Zone (300 meters of the speed cameras) [vs Non-Camera Zone]\",\n \"variable_2_name\": \"Average Speed (among Iranian taxi drivers in southern Tehran)\",\n \"statistic_type\": \"mean_difference_unstandardized\",\n \"statistic_value\": -1.8,\n \"ci_lower\": null,\n \"ci_upper\": null,\n \"pubmed_id\": 34424783,\n \"study_id\": \"Gcjc7NFRGZ\"\n },\n {\n \"variable_1_name\": \"vehicles 200 meters after fixed cameras [vs vehicles on roadways without speed cameras]\",\n \"variable_2_name\": \"speeding vehicles\",\n \"statistic_type\": \"mean_difference_unstandardized\",\n \"statistic_value\": 6.4,\n \"ci_lower\": null,\n \"ci_upper\": null,\n \"pubmed_id\": 26648375,\n \"study_id\": \"c0Ao6MYQL4\"\n }\n ]\n },\n {\n \"note\": \"Abstract check: PMID 17666698 (Barcelona) reports RR 0.73 (0.63-0.85) for collisions on the beltway with cameras; the graph's second association from that paper (RR 0.99, 0.90-1.10, labelled 'people injured among the population affected by the speed cameras') is the comparison arterial roads without cameras, so it is mis-extracted. PMID 14693888 (South Wales, mobile cameras) reports injurious crash rate ratio 0.49 (0.42-0.57) within 500 m routes, with decreases in day and night.\"\n }\n ],\n \"literature\": [\n {\n \"citation\": \"Wilson C, Willis C, Hendrikz JK, Le Brocque R, Bellamy N (2010). Speed cameras for the prevention of road traffic injuries and deaths. Cochrane Database Syst Rev (10):CD004607.\",\n \"doi\": \"10.1002/14651858.CD004607.pub4\",\n \"finding\": \"Reductions in injury crashes near camera sites of roughly 8-49%, fatal and serious 11-44%.\"\n },\n {\n \"citation\": \"NYC DOT (2023, 2025). Speed camera program reports and press releases.\",\n \"doi\": null,\n \"finding\": \"About 7.6% fewer injuries in the new overnight and weekend hours at camera locations in the first year (agency figure).\"\n }\n ],\n \"prediction\": {\n \"direction\": \"-\",\n \"magnitude\": \"Citywide, because cameras cover a small share of street length, the ratio of ratios (injury crashes in new hours / old hours, Aug 2022-Jul 2023 vs Aug 2021-Jul 2022) between 0.92 and 1.00. Persons injured and pedestrians injured are secondary outcomes.\",\n \"support_if\": \"Ratio of ratios below 1 with a 95% interval below 1.\",\n \"refute_if\": \"Ratio of ratios at or above 1 with an interval that excludes 0.97.\"\n },\n \"plan\": \"Data: NYPD Motor Vehicle Collisions - Crashes (data.cityofnewyork.us h9gi-nx95), aggregated by SoQL to counts by month, day of week and hour. Exposure groups: new hours = Mon-Fri 22:00-05:59 and all Sat-Sun; old hours = Mon-Fri 06:00-21:59. Outcomes: crashes with >= 1 person injured or killed (primary), persons injured, pedestrians injured, persons killed. Model: Poisson GLM of monthly counts on group x month fixed effects ... specifically log E[count] = group + month(calendar) + group x post, on a 24-month window Aug 2021-Jul 2023, with Pearson-scaled (quasi-Poisson) standard errors. Event-study figure of the new/old ratio by month, Jan 2020 - Dec 2024, to show pre-trends. Placebo: fake start 1 Aug 2021 on Aug 2020-Jul 2022. Mechanism check: camera tickets ('PHTO SCHOOL ZN SPEED VIOLATION') by issue hour in Open Parking and Camera Violations (nc67-uf89) should appear in the new hours from Aug 2022. Design limits: citywide counts dilute a local effect; night and weekend travel recovered from the pandemic at a different pace from weekday daytime travel, which can bias the DiD in either direction; no exposure (vehicle-miles) by hour.\",\n \"alternatives\": [\n \"Differential post-pandemic recovery of night and weekend traffic (nightlife) relative to weekday daytime\",\n \"Changes in NYPD crash reporting or in the share of crashes with location and time recorded\",\n \"E-bike and moped growth, concentrated in evening delivery hours\",\n \"Regression to the mean if 2021-22 night crash levels were unusually high\"\n ],\n \"prior\": 0.3,\n \"registrant\": \"mobility agent (Claude Opus 5.5), for J. Vernon\",\n \"registered_at\": \"2026-11-12T15:15:19-0500\",\n \"sha256\": \"dceaaebfd4c399d50f80cf40c4ef0435101593ff9e9723f0a57f810a9c422043\"\n}",
"statement": "Extending New York City's school-zone speed cameras to 24-hour, 7-day operation on 1 August 2022 reduced injury crashes in the newly enforced hours (weekday 10 pm-6 am and weekends) relative to the weekday 6 am-10 pm hours that were already enforced."
}History in this collection
- addedv1.0.0Nov 14, 2026this version