Sign in

Mark Ramos

@mframos.bsky.social
1.1K followers 20 following 225 posts

Teacher, Data Scientist, Gamer, Ph.D. Statistics. My views are my own.

PostsRepliesMedia
Reposted by Mark Ramos
The Conversation U.S. @us.theconversation.com · 27/09/2026
Scientific studies often say researchers “controlled” for certain variables. But what does that actually mean? A statistician explains what “control” means and how it can help you judge reports of new research:
theconversation.com
What does it mean to ‘control’ for variables in a scientific study? A statistician explains
Understanding what a ‘control’ is can help you decide for yourself how insightful new studies are.
01611
Mark Ramos @mframos.bsky.social · 21/08/2026
The way to improve public trust in science lies squarely in helping the public become voraciously informed consumers of science. A public capable of critically evaluating statistical evidence creates strong incentives for higher-quality research. #stat #research #science www.psu.edu/news/researc...
psu.edu
More than half of adults in U.S. say they lack basic statistical understanding | Penn State University
Many people in the U.S. believe that their statistical knowledge is limited, but they would rely on statistics for decision-making more if they understood statistics better, according to a new study l...
130
Mark Ramos @mframos.bsky.social · 18/08/2026
093
Mark Ramos @mframos.bsky.social · 18/08/2026
Art by @bonkerific.bsky.social
000
Mark Ramos @mframos.bsky.social · 18/08/2026
Half of 274 statistically significant claims in social science #research were replicated in the SCORE project. How good/bad that sounds depends on how you phrase the headline. How good it actually is can be considered using some math, #statistics intuition, & common sense. doi.org/10.1093/jrss...
110
Reposted by Mark Ramos
Verasight @verasight.io · 17/08/2026
Congratulations to Samuel Anyaso and @mframos.bsky.social on a new article in PLOS ONE, published with Verasight survey data. "Self-reported perception of statistical literacy: Evidence from a National Survey of U.S. Adults" 1/n
111
Mark Ramos @mframos.bsky.social · 04/08/2026
#JSM2026 Day 2 shirt.
000
Mark Ramos @mframos.bsky.social · 02/08/2026
Day 1 shirt at #JSM2026
000
Mark Ramos @mframos.bsky.social · 09/07/2026
Very excited to attend!!! #statsky #statistics #jsm2026
010
Mark Ramos @mframos.bsky.social · 09/07/2026
@usapolling.bsky.social
000
Mark Ramos @mframos.bsky.social · 08/07/2026
From our latest research published in @plosone.org, it's refreshing to see the U.S. public’s willingness to acknowledge what they don't know and openness to use such knowledge if they understand it better. Wisdom is honest humility paired with the appetite to learn. #science doi.org/10.1371/jour...
doi.org
Self-reported perception of statistical literacy: Evidence from a National Survey of U.S. Adults
Amid growing recognition of the role statistical literacy plays in informed decision-making, we examined self-reported statistical literacy and willingness to apply statistical reasoning in everyday c...
111
Mark Ramos @mframos.bsky.social · 09/03/2026
That’s exactly what was done here. Ai wrote the code. Not everyone who teaches stats in high school knows how to code in R. Now they can code simple things like this using their native language.
010
Mark Ramos @mframos.bsky.social · 22/02/2026
In a coin toss, the probability of heads is taught to be half. Is it really? In this preprint, I argue that helping students understand statistics better entails explicit and sustained inclusion of the observer-relative reality of randomness in your pedagogy. #education #STEM doi.org/10.35542/osf...
doi.org
OSF
000
Mark Ramos @mframos.bsky.social · 04/02/2026
Sharing this essay I wrote that's just been published in @significancemag.bsky.social about loving the things we do, especially when those things involve the people we love. I hope you enjoy it. significancemagazine.com/autism-bayes...
significancemagazine.com
Autism, Bayesian probability, and why we do what we do - Significance magazine
Statisticians can apply their trade anywhere it’s needed, bringing benefits to friends, family and the wider world. Find out how one statistician and father used his training to help improve...
001
Mark Ramos @mframos.bsky.social · 02/12/2025
YES! #stat #statsky #research #researchsky #statistics #pvalues #science #sciencesky
000
Mark Ramos @mframos.bsky.social · 27/11/2025
"Absence of evidence is not evidence of absence." Comprehending this one statement can clear up a host of misunderstandings about how null hypothesis testing works and why failure to reject the null hypothesis does say less than rejecting it. doi.org/10.64628/AAI...
doi.org
Absence of evidence is not evidence of absence – and that affects what scientific journals choose to publish
Researchers design studies that might disprove what’s called their null hypothesis – the opposite of the claim they’re interested in exploring.
000
Mark Ramos @mframos.bsky.social · 10/11/2025
The idea of "Wisdom of the Crowd" may seem strange at first glance, until you realize that the historical foundations of the Gaussian distribution is based on observations of human measurement error. Then, the idea becomes quite... Normal. #Statsky #statistics #edusky
030
Reposted by Mark Ramos
Theresa Kim, PhD, MS @tyk314.net · 08/11/2025
"Association does not imply causation" is the truth and continues to be an important mantra in the practice of statistics. What does it actually mean? Can we prove causal relationships using observational data? If no, does it mean that detected associations are useless? #statssky #causalinference
He-Man gives advice. "Causal relationships CAN'T be proven using observational data. Acting on association as causal is ALWAYS a leave of faith. Be skeptical, but not too skeptical. Until next time!"
032
Mark Ramos @mframos.bsky.social · 08/10/2025
Join me on 10/18/25 as I talk about ubiquitous bias for publishing statistically significant results in peer-reviewed journals and why it's not the boogeyman it is made out to be. The event is virtual, free and hosted by the R Users Group - PH. Register now! #science www.meetup.com/r-user-group...
meetup.com
#141 Balanced Examination of Positive Publication Bias, Sat, Oct 18, 2025, 8:30 AM | Meetup
Dr. Mark Ramos will examine the conventional view that positive publication bias, favoring statistically significant findings, is inherently detrimental to research quality
000
Mark Ramos @mframos.bsky.social · 30/09/2025
Happy to share that one of my most enduring insights from over two decades of teaching statistics and mathematics subjects is now published in Statistics Teacher by @amstatnews.bsky.social. #statsky #STEM #STEAM #education #mathed #iteachmath www.statisticsteacher.org/2025/09/25/p...
statisticsteacher.org
On the Pedagogy of Randomness: Effectively Teaching How Random Is Relative in High School | Statistics Teacher
011
Mark Ramos @mframos.bsky.social · 26/09/2025
All parents want the best for their child. Thus, the issue on Tylenol use & autism risk is so consequential. It's important to communicate the science objectively, thoroughly & compassionately. Put yourself in the shoes of the pregnant mom suffering from a fever. theconversation.com/what-parents...
theconversation.com
What parents need to know about Tylenol, autism and the difference between finding a link and finding a cause in scientific research
Trump’s definitive statements that Tylenol use during pregnancy causes autism suggest certainty that the science does not support.
000
Mark Ramos @mframos.bsky.social · 22/09/2025
Our latest paper on drivers of HPV vaccination rates is now published in Vaccine. #science #sciencesky #research #researchsky #hpv #vaccine doi.org/10.1016/j.va...
doi.org
Redirecting
000
Mark Ramos @mframos.bsky.social · 09/09/2025
That weakness may have been inherited from the training data. What proportion of people that you meet have the ability to admit that they don't know? I don't know.
010
Mark Ramos @mframos.bsky.social · 27/08/2025
For my fellow Dungeons and Dragons fans who may taking a #statistics course this semester. This will come in handy when the time comes to learn hypothesis testing. #DnD #D&D
040
Mark Ramos @mframos.bsky.social · 25/08/2025
They say MTG has been dying since 1995. :D
010
Mark Ramos @mframos.bsky.social · 16/08/2025
Just finished Eyes of Wakanda and can say that it soundly makes up for both What if...? Season 3 and Season 2.
000
Reposted by Mark Ramos
Amstat-American Statistical Association @amstatnews.bsky.social · 15/08/2025
There are many “statistically significant” birthdays in August, including the term “artificial intelligence,” coined in August 1955. Read about more historical happenings in August’s This Month in Statistics History: magazine.amstat.org/...
A screenshot of the August Time Series article
032
Mark Ramos @mframos.bsky.social · 14/08/2025
For more information, read our #openaccess paper in @accountabilityair.bsky.social here: www.tandfonline.com/doi/full/10....
tandfonline.com
Balanced examination of positive publication bias impact
Positive publication bias is the tendency to favor studies that reject null hypotheses for publication and is widely regarded as detrimental to research enterprise quality. However, this view overs...
000
Mark Ramos @mframos.bsky.social · 14/08/2025
Curious how #publicationbias shapes scientific record? This Shiny app shows hypothetical distributions of published papers across levels of bias. Play with assumptions on the research landscape and how researchers behave to see what happens. #science #statsky #bias levelalpha005.shinyapps.io/shiny1/
levelalpha005.shinyapps.io
131
Mark Ramos @mframos.bsky.social · 11/08/2025
For sure. Sample size does get larger the smaller the effect assumed. I think your paper can benefit from including more context on how expensive clinical trials get wrt sample size. Small world, I saw Adrian's graphic on linkedin and discussed suggestions with him. www.linkedin.com/feed/update/...
linkedin.com
Clinical trial biostatisticians do hypothesis testing differently than what you learned in Stats 101 - and you should pay attention! | Justin Bélair posted on the topic | LinkedIn
Clinical trial biostatisticians do hypothesis testing differently than what you learned in Stats 101 - and you should pay attention! You see, there is so much discussion around the idea that statisti...
110
Mark Ramos @mframos.bsky.social · 10/08/2025
Thank you. Yes, it is an easy mistake to make. I like to say the problem with statistics isn’t that it is hard to understand, but rather that it is so easy to misunderstand.
010
Mark Ramos @mframos.bsky.social · 10/08/2025
"On the other hand, if researchers and publishers are competent, good faith actors and the only risk in positive publication bias comes from inflation of type 1 error rate by chance, ....(we) show that the negative impact is outweighed by the much larger improvement in true positive rates."
230
Mark Ramos @mframos.bsky.social · 09/08/2025
Plus, people are free to propose other frameworks. Bayesian inference is one example. It isn't strictly better than a frequentist approach, but neither is it strictly worse. As always, context is key. Whatever you propose, it must be open to scrutiny on such matters as t1 & t2 error rate control.
100
Mark Ramos @mframos.bsky.social · 09/08/2025
Statistical evidence is not the only piece of evidence. Your sample mean, odds ratio, etc. do not change if a hypothesis test fails to reject Ho. We only offer a reasonable framework backed by ways to control type 1 error rate and power in the context of uncertainty and sample variability.
100
Mark Ramos @mframos.bsky.social · 09/08/2025
That's why clinical trials have different stages. Earlier stages may be looking for any effect, but later stages (as mentioned in the paper) set standards for clinical relevance using various, non-statistical bases (i.e. they test that the effect is at least larger than this or it does not pass).
000
Mark Ramos @mframos.bsky.social · 09/08/2025
pvalues and ci's are two parts of the same inference. Both are taught, used, and often misunderstood. The truth value of hypotheses are typically unknown even after one has made a decision. Best we can hope for is to have some control over our error rates, which is why we set alpha for t1 error.
100
Mark Ramos @mframos.bsky.social · 09/08/2025
In this paper, I discuss an often misunderstood aspect of inference that can trip up even seasoned researchers. If you wonder what "practical significance" means in #statistics, I guarantee you that reading this will make you a better scientist. #statsky #StatsSky #science doi.org/10.1093/jrss...
360
Mark Ramos @mframos.bsky.social · 06/08/2025
Had a lot of fun doing this talk at #JSM2025. The transformed and relative Lorenz curves are available for use in the glorenz R package. #statsky #StatsSky #datascience #dataliteracy #jsm cran.r-project.org/web/packages...
033
Mark Ramos @mframos.bsky.social · 05/08/2025
Today I will be presenting our recently published paper on New Graphical Displays and Related Statistical Measures of Health Disparities Among Groups. Hope to see you there! (CC-210, 2pm-3:50pm). #JSM #JSM2025 #statsky #StatsSky #statistics. academic.oup.com/jrsssa/advan...
010
Mark Ramos @mframos.bsky.social · 04/08/2025
#jsm2025 kicks into high gear! So many exciting sessions to watch. What session are you looking forward to? Let’s gooooo! #jsm #statsky #statssky
000
Mark Ramos @mframos.bsky.social · 04/08/2025
Awesome estimathon game by #janestreet at #jsm. #jsm2025 #statssky. We were team two sided XD.
010
Mark Ramos @mframos.bsky.social · 03/08/2025
Behold, hot chicken!
000
Mark Ramos @mframos.bsky.social · 03/08/2025
#jsm2025 is here in Music City! #stats #statsky #jsm #statssky Anyone try the hot chicken yet?
100
Mark Ramos @mframos.bsky.social · 02/08/2025
Off to Nashville for the Joint Statistical Meetings. #jsm #jsm2025 #statsky. Doing a talk on our paper about some new disaggregated displays and measures for comparing continuous quantities between groups. Wish me luck!
031
Mark Ramos @mframos.bsky.social · 31/07/2025
Published studies often show more significant results than unpublished ones. Is this bad practice by journals? Not really. Read why in my latest paper published by @tandfresearch.bsky.social #sciencesky #statsky #researchsky #science #statistics #bias #research www.tandfonline.com/doi/full/10....
tandfonline.com
Balanced examination of positive publication bias impact
Positive publication bias is the tendency to favor studies that reject null hypotheses for publication and is widely regarded as detrimental to research enterprise quality. However, this view overs...
030
Mark Ramos @mframos.bsky.social · 26/07/2025
No method is perfect; no assumptions hold exactly. We use #statistics to make sense of uncertainty as objectively as we can. Most importantly, proper statistical practice keeps us honest, especially when results don't conform to what we hoped for. #datascience #statsky #dataliteracy #science #edusky
020
Mark Ramos @mframos.bsky.social · 26/06/2025
Congratulations to the finalists of Significance Magazine's Statistical Excellence Award for Early-Career Writing! While my essay didn't make the shortlist, I'm honored to be mentioned & happy that the judges enjoyed reading it. #datascience #autismawareness significancemagazine.com/2025-writing...
significancemagazine.com
2025 writing competition shortlist announced - Significance magazine
We are delighted to reveal the three finalists of the 2025 Statistical Excellence Award for Early-Career Writing. Launched earlier this year, the competition attracted entries from across the world, w...
010
Mark Ramos @mframos.bsky.social · 14/06/2025
Statistical power is the chance of detecting an effect given that a specific one exists. Being transparent with how much power we can get for a study before spending resources on it gives confidence to funders that their money is not being spent on a coin toss. #academicsky #research #science #STEM
131
Mark Ramos @mframos.bsky.social · 05/06/2025
Thanks, Dave!
010
Mark Ramos @mframos.bsky.social · 05/06/2025
Our latest paper is now published in @jrssa.bsky.social. We constructed new ways to describe disparity between groups in survey-weighted data. Software to use these methods are now available in the glorenz R-package. #statsky #datascience #dataliteracy #academicsky academic.oup.com/jrsssa/advan...
academic.oup.com
Graphical displays and related statistical measures of health disparities between groups in complex sample surveys
Abstract. Different methods for describing health disparities in the distributions of continuous measured health-related variables among groups provide mor
184