Isaac Maddow-Zimet
LightNews LightNews
imaddowzimet.bsky.social
Isaac Maddow-Zimet
@imaddowzimet.bsky.social
1.6K followers 210 following 61 posts
Data Scientist at Guttmacher; I lead the Monthly Abortion Provision Study (https://www.guttmacher.org/monthly-abortion-provision-study) but post here mainly about statistics.
Posts Replies Media Videos
imaddowzimet.bsky.social
This is such a great thoughtful and nuanced post by @rmcelreath.bsky.social on open science/computational reproducibility - especially on the need to make research and code *intelligible*, not just transparent.
elevanth.org/blog/2025/07...
Comprehensibility of research. This is not transparency nor openness. What I mean is research has sufficient documentation and justification to reduce error and empower others to make up their own minds about its value. Research should be intelligible. Access is not sufficient. Research can be replicable without being reasonable or correct. Materials and data can be open without being intelligible, and they can be partly closed while still being comprehensible. The rise of AI reinforces the need for comprehensibility, because “I asked the chatbot” is not comprehensible. But for sure AI will be a part of scientific workflow now and into the future. But the researcher must be responsible, not the bot.
July 9, 2025 at 5:12 PM
imaddowzimet.bsky.social
Florida's ban had ripple effects across the Southeast because in 2023 it was one of the few places where people could access in-clinic care after six weeks. In 2023, more than 1,800 Louisianans had abortions in Florida; in 2024, fewer than 500 did (and most before the ban went in effect in May).
June 24, 2025 at 3:23 PM
imaddowzimet.bsky.social
And we saw similar increases in travel to North Carolina, despite the 12 week ban and 72-hour waiting period in effect in that state.
Bar chart showing number of people traveling to NC from FL increased from 210 to 1320, alongside map
June 24, 2025 at 3:23 PM
imaddowzimet.bsky.social
More than ten times more people traveled from Florida to Virginia to access abortion care in 2024 as compared to 2023, for example - despite it being hundreds of miles away.
Bar chart showing number of people traveling to VA from FL increased from 130 to 1620, alongside map
June 24, 2025 at 3:23 PM
imaddowzimet.bsky.social
But we also captured big changes, particularly in the Southeast, where Florida's six week ban led to many more people traveling out of state for care.
Map showing where Florida residents traveled in 2024; bar chart showing increase in travel out of state for abortion care from 3300 in 2023 to 8400 in 2024.
June 24, 2025 at 3:23 PM
imaddowzimet.bsky.social
Some of what we found echoes what the landscape of travel looked like in 2023: In Texas for example, even though many Texans are accessing care through telehealth (as #WeCount recently documented), tens of thousands are also traveling across state lines for care, across huge distances.
Map showing that Texans traveled all across the country for abortion care- many to neighboring states, but also as far as New York, Illinois and Washington.
June 24, 2025 at 3:23 PM
imaddowzimet.bsky.social
Really useful primer, and also includes this major contribution to the literature
April 23, 2025 at 4:12 PM
imaddowzimet.bsky.social
We found that while there was stability overall, there were major shifts at the state level - in particular, Florida saw a steep decline after its six week ban, and (because what happens in one state has ripple effects in others), Virginia saw an increase.
April 16, 2025 at 6:20 PM
imaddowzimet.bsky.social
Another one of these I really like in Brooklyn
Three signs by the waterfront each with one word: “no”, “parking”, “ever”
December 28, 2024 at 11:58 PM
imaddowzimet.bsky.social
also, the drop in response rates is pretty astonishing (I think it was above 60% in 2017-2019 NSFG; it's ~23% in new wave), though there have been RR declines in other major surveys too.
Response rates in household surveys have been declining, and this trend has been observed on NSFG, even prior to 2022-2023 NSFG. The NSFG response rates for 2022-2023 are markedly lower that the response rates for the last 2-year data collection period of 2017-2019. Some of the sharper decline in participation can be attributed to the effects on society of the COVID-19 pandemic, from respondents’ willingness to participate to the ability to recruit and retain interviewers. Complex and long FTF surveys such as the National Health and Nutrition Examination Survey (NHANES) that did not undergo major design changes from before the pandemic, also experienced drops in participation. For example, the NHANES unweighted interview response rate fell from 51.0% for the 2017-2020 data collection to 34.5% for 2021- 2023. The examination response rates fell from 46.9% to 25.6%. Other similarly long and sensitive household surveys to the NSFG, such as the National Survey of Drug Use and Health (NSDUH) that underwent a redesign from only FTF to a web and FTF design, like NSFG, have also observed substantial drops in survey participation. The weighted overall survey response rate for NSDUH dropped from 45.8% in 2019 (pre-pandemic) to 12.3% in 2023.
December 11, 2024 at 3:55 PM
imaddowzimet.bsky.social
Also, some pretty fascinating notes in the User Guide about comparing data across waves:
Data users are encouraged to exercise caution when making comparisons to prior years of data collection and interpreting any differences. The multimode design is very different from the prior FTF-only design, and the COVID-19 pandemic prevented experimental evaluation of the impact of the multimode design on survey estimates. Like other surveys that had to change to a new design due to the disruption of FTF interviewing, changes in estimates from before and after the pandemic are confounded by real changes in the population, measurement differences, and nonresponse, among other possible sources of error (coverage, sampling, and processing). As a result, some estimates may show larger differences from the 2017-2019 NSFG and earlier data compared to those observed between prior data releases. Users are advised to note the design related changes, particularly with regard to mode, if making statements comparing earlier estimates to those from 2022-2023.
December 11, 2024 at 3:55 PM
imaddowzimet.bsky.social
Here's 2021 and 2020 for comparison.
Map of states that did and did not meet response rate criteria for PRAMS (>50%) in 2021; only 2-3 states in the south didn't meet the criteria. Map of states that did and did not meet response rate criteria for PRAMS (>50%) in 2020; only 2-3 states in the south didn't meet the criteria.
November 17, 2024 at 9:36 PM
imaddowzimet.bsky.social
Did anyone else notice that PRAMS response rates in the south seem to have plummeted in 2022? (even aside from the context of overall declines in response rates?)
Graph of states that met and did not meet PRAMS response rate criteria (>50%) in 2022; many states in the south did not meet the criteria.
November 17, 2024 at 9:36 PM
imaddowzimet.bsky.social
Post a meme made by you
Meme of man holding a book and looking at a butterfly. butterfly is labeled “Almost anything”, man is labelled “economists” and is saying “is this an instrumental variable?”
September 1, 2024 at 8:56 PM