Naomi Harada Thyden, PhD MPH
@naomithyden.bsky.social
Social Epidemiologist
Research: structural racism, policy, life course
WNBA fan. 4th Gen Japanese American and WOC. She/her.
📍Minneapolis
Research: structural racism, policy, life course
WNBA fan. 4th Gen Japanese American and WOC. She/her.
📍Minneapolis
I was in a meeting where the facilitator would stop presenting to say out loud when she noticed someone in the process of typing a comment in the chat 😭
April 30, 2025 at 4:33 PM
I was in a meeting where the facilitator would stop presenting to say out loud when she noticed someone in the process of typing a comment in the chat 😭
Same. It’s mostly worth it but I have had some nasty surprises. 🤣
April 18, 2025 at 6:01 PM
Same. It’s mostly worth it but I have had some nasty surprises. 🤣
They made a FB event!
March 28, 2025 at 3:45 PM
They made a FB event!
Why does he have 7 different reasons and none of them make sense. What is the actual reason. 🤨
March 26, 2025 at 11:07 PM
Why does he have 7 different reasons and none of them make sense. What is the actual reason. 🤨
I got the email. Assumed it was everyone.
March 25, 2025 at 9:43 PM
I got the email. Assumed it was everyone.
😂 you could probably pin it yourself and then just hand it over.
March 25, 2025 at 9:41 PM
😂 you could probably pin it yourself and then just hand it over.
I use a tailor for ~$20 because I hate sewing.
March 25, 2025 at 8:24 PM
I use a tailor for ~$20 because I hate sewing.
Haha yes I see friends down south posting salads made from their gardens while it’s still snowing here.
March 20, 2025 at 4:02 PM
Haha yes I see friends down south posting salads made from their gardens while it’s still snowing here.
I referred to it as a “cervical cancer vaccine” and my friend was like there’s a cervical cancer vaccine?? Just didn’t make the connection even though she received it
February 25, 2025 at 3:17 PM
I referred to it as a “cervical cancer vaccine” and my friend was like there’s a cervical cancer vaccine?? Just didn’t make the connection even though she received it
Thanks for reading, Mary!
February 21, 2025 at 1:53 PM
Thanks for reading, Mary!
They are unionized. Some of the firings target people in their probationary period before full union protections kick in. Some of the firings are illegal.
February 19, 2025 at 1:13 AM
They are unionized. Some of the firings target people in their probationary period before full union protections kick in. Some of the firings are illegal.
your Ward 2018 AEP paper is cited, as always!
February 18, 2025 at 11:08 PM
your Ward 2018 AEP paper is cited, as always!
Thanks for pointing that out! I changed my settings so you should be able to now. Looks like I can't DM you since you don't follow me.
February 18, 2025 at 11:01 PM
Thanks for pointing that out! I changed my settings so you should be able to now. Looks like I can't DM you since you don't follow me.
See full commentary for more details on each of these recommendations, additional ethical and statistical justifications, and a treasure trove of citations.
Resist by continuing to move your work forward!
Resist by continuing to move your work forward!
February 18, 2025 at 6:11 PM
See full commentary for more details on each of these recommendations, additional ethical and statistical justifications, and a treasure trove of citations.
Resist by continuing to move your work forward!
Resist by continuing to move your work forward!
Knee jerk “insufficient sample size” justifications for failing to collect and analyze data on marginalized communities is dismissive and harmful.
If our standard practices systematically disadvantage minoritized groups, then we need to find ways to correct them.
If our standard practices systematically disadvantage minoritized groups, then we need to find ways to correct them.
February 18, 2025 at 6:11 PM
Knee jerk “insufficient sample size” justifications for failing to collect and analyze data on marginalized communities is dismissive and harmful.
If our standard practices systematically disadvantage minoritized groups, then we need to find ways to correct them.
If our standard practices systematically disadvantage minoritized groups, then we need to find ways to correct them.
Data owners Rec 3: Become familiar with data ownership frameworks outside of the mainstream.
Mainstream frameworks for data ownership and ethics emphasize concerns about privacy and statistical rigor (aka prioritize sample size), while complementary frameworks emphasize other value systems.
Mainstream frameworks for data ownership and ethics emphasize concerns about privacy and statistical rigor (aka prioritize sample size), while complementary frameworks emphasize other value systems.
February 18, 2025 at 6:11 PM
Data owners Rec 3: Become familiar with data ownership frameworks outside of the mainstream.
Mainstream frameworks for data ownership and ethics emphasize concerns about privacy and statistical rigor (aka prioritize sample size), while complementary frameworks emphasize other value systems.
Mainstream frameworks for data ownership and ethics emphasize concerns about privacy and statistical rigor (aka prioritize sample size), while complementary frameworks emphasize other value systems.
Data owners Rec 2: Reconsider sample size requirements with the size of marginalized populations in mind.
Some marginalized groups simply do not exist in numbers large enough to satisfy sample size req's.
In these cases, are we willing to exclude entire communities from the scope of statistics?
Some marginalized groups simply do not exist in numbers large enough to satisfy sample size req's.
In these cases, are we willing to exclude entire communities from the scope of statistics?
February 18, 2025 at 6:11 PM
Data owners Rec 2: Reconsider sample size requirements with the size of marginalized populations in mind.
Some marginalized groups simply do not exist in numbers large enough to satisfy sample size req's.
In these cases, are we willing to exclude entire communities from the scope of statistics?
Some marginalized groups simply do not exist in numbers large enough to satisfy sample size req's.
In these cases, are we willing to exclude entire communities from the scope of statistics?
Data owners Rec 1: Facilitate the implementation of the above recommendations.
Structural racism researchers encounter structural racism in their efforts to obtain data.
Data owners should allow researchers to make a case for analyzing smaller samples than are customary to with their dataset.
Structural racism researchers encounter structural racism in their efforts to obtain data.
Data owners should allow researchers to make a case for analyzing smaller samples than are customary to with their dataset.
February 18, 2025 at 6:11 PM
Data owners Rec 1: Facilitate the implementation of the above recommendations.
Structural racism researchers encounter structural racism in their efforts to obtain data.
Data owners should allow researchers to make a case for analyzing smaller samples than are customary to with their dataset.
Structural racism researchers encounter structural racism in their efforts to obtain data.
Data owners should allow researchers to make a case for analyzing smaller samples than are customary to with their dataset.
Rec 5b: Avoid analyzing or reporting “multi-racial” as racial/ethnic category.
It is so vague it is meaningless.
One option is to create categories for combos of racial and ethnic groups which exhibit distinct health patterns.
Another option is categories that are not mutually exclusive.
It is so vague it is meaningless.
One option is to create categories for combos of racial and ethnic groups which exhibit distinct health patterns.
Another option is categories that are not mutually exclusive.
February 18, 2025 at 6:11 PM
Rec 5b: Avoid analyzing or reporting “multi-racial” as racial/ethnic category.
It is so vague it is meaningless.
One option is to create categories for combos of racial and ethnic groups which exhibit distinct health patterns.
Another option is categories that are not mutually exclusive.
It is so vague it is meaningless.
One option is to create categories for combos of racial and ethnic groups which exhibit distinct health patterns.
Another option is categories that are not mutually exclusive.
Rec 5a: Avoid analyzing or reporting “other” as racial/ethnic category.
An “other” estimate is just a weighted average, weighted toward whichever remaining group happens to be the largest in that particular dataset – a group which is rarely identified.
An “other” estimate is just a weighted average, weighted toward whichever remaining group happens to be the largest in that particular dataset – a group which is rarely identified.
February 18, 2025 at 6:11 PM
Rec 5a: Avoid analyzing or reporting “other” as racial/ethnic category.
An “other” estimate is just a weighted average, weighted toward whichever remaining group happens to be the largest in that particular dataset – a group which is rarely identified.
An “other” estimate is just a weighted average, weighted toward whichever remaining group happens to be the largest in that particular dataset – a group which is rarely identified.
Rec 4: Publish estimates produced from small samples of marginalized groups even if they are imprecise.
In isolation, imprecise estimates might not be convincing, but if several publications produce similar estimates, we essentially increase the sample size and make progress toward precision.
In isolation, imprecise estimates might not be convincing, but if several publications produce similar estimates, we essentially increase the sample size and make progress toward precision.
February 18, 2025 at 6:11 PM
Rec 4: Publish estimates produced from small samples of marginalized groups even if they are imprecise.
In isolation, imprecise estimates might not be convincing, but if several publications produce similar estimates, we essentially increase the sample size and make progress toward precision.
In isolation, imprecise estimates might not be convincing, but if several publications produce similar estimates, we essentially increase the sample size and make progress toward precision.
Rec 3: Explore ways to thoughtfully increase the analytic sample of the group you’re interested in.
In situation when sample sizes from marginalized groups are too small to analyze with bivariates or regression models, thoughtfully consider who else is similar enough to be categorized together.
In situation when sample sizes from marginalized groups are too small to analyze with bivariates or regression models, thoughtfully consider who else is similar enough to be categorized together.
February 18, 2025 at 6:11 PM
Rec 3: Explore ways to thoughtfully increase the analytic sample of the group you’re interested in.
In situation when sample sizes from marginalized groups are too small to analyze with bivariates or regression models, thoughtfully consider who else is similar enough to be categorized together.
In situation when sample sizes from marginalized groups are too small to analyze with bivariates or regression models, thoughtfully consider who else is similar enough to be categorized together.