Data Is Plural

The Affordable Care Act, quantified. Last week, the U.S. Department of Health and Human Services released a dataset of state-level Obamacare metrics. The dataset is divided into five main categories: coverage gains, employer coverage, individual market coverage, Medicaid, and Medicare. Between 2010 and 2015, the proportion of Nevadans without health insurance dropped from 22.6% to 12.3% — the largest percentage-point decrease of any state. (In 2015, an estimated 17.1% of Texans still didn’t have health insurance, the highest rate of any state that year.) The metrics come from various sources, including the Census, academic studies, and the department’s own estimates. [h/t Nadja Popovich]

Petroleum rig counts. Since the 1940s, oilfield services corporation Baker Hughes and its predecessor companies have been publishing “rig counts” — the number of rigs actively drilling for oil and/or gas in various parts of the world. These days, the company updates its North America numbers every week and its international counts every month. As of December 16, they counted 637 rigs in — and offshore of — the United States, nearly half of them in Texas. [h/t Jordan Wirfs-Brock]

The birds and the bees (and more). The U.S. Geological Survey’s BISON service brings together “species occurrence” data from hundreds of sources. The service, whose name stands for ”Biodiversity Information Serving our Nation,” currently contains 262 million records, each of which refers to the observation of “an organism at a particular time in a particular place.” Most of the observations are based on direct sightings; others use fossils, written records, or other sources. The data aren’t available for bulk download, but can be accessed via BISON’s free API. [h/t Clare Malone]

Two planes too close. The FAA’s Near Midair Collision System keeps track of incidents where two planes flew uncomfortably close to each other. The system, which is based on reports from pilots and flight crew members, contains more than 7,500 incidents dating back to 1987. The FAA received 305 of these reports for the first 10 months of 2016, including 35 classified as “critical.”

The geography of language on Twitter. Last week, Quartz published an addictive tool that lets you map word usage on Twitter, by U.S. county. It’s based on an academic analysis of 890 million geocoded tweets uttered between October 2013 and November 2014. Data and details available here.