,
25 tweets,
9 min read
Read on Twitter
1/ Let's talk about how misinfo can spread from failure to fact-check representations of academic research in journalism. This thread is about false claims made about the number of women who seek "late-term" abortions due to maternal health and/or fetal anomaly.
2/ A few days ago, @DouthatNYT tweeted that most 3rd trimester abortions were not sought due to maternal health/fetal anomaly & cited a 2013 paper to support this claim. 
3/ This claim was problematic for a # of reasons, the most important one being that women who sought abortions for health and/or fetal abnormality were EXCLUDED from the study.
Secondarily, the mean time of abortion was 22 weeks, which is in the second trimester, not the third.
Secondarily, the mean time of abortion was 22 weeks, which is in the second trimester, not the third.
5/ Then, today, @xan_desanctis, of @TheAtlantic, repeated Douthat’s false claim, linking to the same 2013 research paper. Notably, she lifts a quote from the paper's lit review that is *not* a description of the paper's findings. theatlantic.com/ideas/archive/…
6/ After I pointed out that the 2013 paper couldn't draw these conclusions bc abortions due 2 maternal health/fetal anomaly were EXCLUDED from the sample, the dregs of twitter put on their genius caps & said, “Well the study says in the lit review that other research found that!”
7/ So, a few things: you should not copy and paste claims from a study’s lit review & cite them as the source for those research findings b/c
1st: that’s not how citations work.
2nd: Lit reviews often contain errors & misrepresentations of other research.
3rd: 🙄🙄🙄🙄🙄🙄🙄.
1st: that’s not how citations work.
2nd: Lit reviews often contain errors & misrepresentations of other research.
3rd: 🙄🙄🙄🙄🙄🙄🙄.
8/ But given this misrepresentation is especially pernicious, I decided to do a public service & fact-check the citations. Plz note I was only able to do this b/c I have a subscription to academic data bases. The research is not available w/out one
So, what did I find? . . . .
So, what did I find? . . . .
9/ The 2013 lit review’s representation of the previous research WAS flawed. Specifically, the claim that “data suggest that most women seeking later terminations are not doing so for reasons of fetal anomaly or life endangerment” is not supported by citations #5 OR #21.
10/ Citation #5: This study examined abortion incidence, as well as ability to access abortions by region. The researchers also looked at prevalence of harassment at clinics.
11/ The researchers did NOT collect data on gestational age!
Nor did they report data on *reasons* for seeking an abortion, such as fetal anomaly.
This means that one cannot draw ANY conclusions about EITHER of these factors from this paper.
Nor did they report data on *reasons* for seeking an abortion, such as fetal anomaly.
This means that one cannot draw ANY conclusions about EITHER of these factors from this paper.
12/ So, again: don’t copy & paste from lit reviews.
13/ Okay, on to the next cited study (ref #21): “Attitudes and Decision Making Among Women Seeking Abortions at One U.S. Clinic.”
Do the results from THIS study support the claims about reasons for seeking "later term" abortions?
Short answer: NO.
Do the results from THIS study support the claims about reasons for seeking "later term" abortions?
Short answer: NO.
14/ This study looked at whether or not women had high/low confidence in their decision to have an abortion. This was the dependent variable.
The independent variables included demographics (age/race, etc), as well as spiritual beliefs, having a supportive partner, etc.
The independent variables included demographics (age/race, etc), as well as spiritual beliefs, having a supportive partner, etc.
15/ The presence of fetal anomaly WAS included as a variable.
However, again, this study was not investigating REASONS for obtaining an abortion. It was looking at the effect of the above variables on *confidence* in the decision.
However, again, this study was not investigating REASONS for obtaining an abortion. It was looking at the effect of the above variables on *confidence* in the decision.
16/ Nor was the effect of fetal anomaly *ever* investigated by *time* of abortion.
There is *no* possible way to examine whether fetal anomaly or maternal health was a reason for seeking a second or third trimester abortion from the data in this study.
There is *no* possible way to examine whether fetal anomaly or maternal health was a reason for seeking a second or third trimester abortion from the data in this study.
17/ So, overall, let's be clear:
Regardless of what these two studies found, both Douthat & DeSanctis misrepresented the original 2013 research they were citing. This was the paper that excluded maternal health/fetal anomaly from the study.
*This* is the primary issue.
Regardless of what these two studies found, both Douthat & DeSanctis misrepresented the original 2013 research they were citing. This was the paper that excluded maternal health/fetal anomaly from the study.
*This* is the primary issue.
18/ Second, neither of the studies cited in the 2013 paper found what the authors claimed they had found. I don’t want to excuse this error, but I will say this is VERY COMMON in academic research.
19/ Which is why your professors try SO HARD to teach you when you are a freshman in college to check PRIMARY SOURCES.
This means: don’t copy & paste claims from lit reviews.
This means: don’t copy & paste claims from lit reviews.
20/ @DouthatNYT should delete his tweets & the @nytimes should issue a statement that one of their columnists was spreading misinformation on the internet
@TheAtlantic should tell @xan_desanctis to correct her article
It shouldn't be on *readers* to fact-check journalistic orgs
@TheAtlantic should tell @xan_desanctis to correct her article
It shouldn't be on *readers* to fact-check journalistic orgs
Update: a "pro-life" organization says that 30 yr old research supports their claim. It does not. The research is sound, for its time, but is being, predictably, misrepresented.
My question to "pro-lifers" is simple: if you have strong moral and/or empirical evidence to support your claims re: "late-term" abortions, why do you keep lying?
This letter, from later-abortion patients, is very powerful: abortionpatients.com
Response from one of the 2013 paper's authors. Please note, this paper's purpose & results have been misrepresented by some opinion columnists:
So much history here. A thread on abortion in the United States:
