The original report discusses real-estate pressure, online connection and mainstreaming. Interactive factor selection requires the page script.
Data story
Fading Spaces.
What happens when gathering places disappear?
Interactive controls require the page scripts. If they do not appear, reload this page. Static reference notes remain available below.
A changing map
of community.
The original report explores changes in U.S. gay-bar listings from 2002 to 2019. Its headline finding is a 37% decline. Figures below reproduce the existing project and require primary-source verification before republication.
−37%
2002—2019
Fewer places.
Uneven losses.
A national trend can conceal very different experiences across communities.
Look beyond the average.
Original reported declines: 37% overall, 52% for women’s bars, and 60% for bars serving people of color. Interactive comparisons require the page script.
Explore the contributing factors.
Geographic unevenness
The original report also compares state-level listing counts. Counts describe the source dataset, not a census of every operating venue.
Reference notes & original examples
The existing project reports a 37% decline in U.S. gay-bar listings from 2002 to 2019 and attributes its framing to sociologist Greggor Mattson. The primary study has not been independently reconciled in this redesign. Listings are not a census of every operating venue.
Read the measures with their limits.
Decline by demographic
Compare the demographic views. The original reported declines are 37% for general male-oriented bars, 52% for women’s bars and 60% for bars serving people of color. Selecting a group reorders the comparison; it does not change the source values.
The original trend illustration
The visible line chart uses the old code’s approximate 2002–2019 trajectory. Its exact values remain in an expandable table. Those illustrative counts have not been validated as a measured time series.
Contributing factors
Explore the three factor explanations: real-estate pressure, changing online connection and mainstreaming. They are narrative contributors, not quantified causal estimates.
Geography
Inspect the state bars and exact table. These seven counts were explicitly representative values in the old source, not verified state-level venue counts. No missing map is implied.