BJANALYTICS

Data Science × Public Health Clear, beginner-friendly content on data science, public health, epidemiology, & biostatistics. 📺 YouTube | 🎧 Apple Podcast New content weekly. 🔗 Links below ⬇️: ➡️ YouTube: @bjanalytics?si=H1M38cWbbA1o_W5d" target="_blank" rel="nofollow noopener">youtube.com/@bjanalytics?s… ➡️ Apple Podcast: podcasts.apple.com/us/podcast/dat…

Averaging across multiple models can sound like humility while leaving the deepest uncertainties almost completely untouched.👇 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #biostatistics #modelaveraging #publichealth #modelselection #healthresearch #prediction

Loss to follow-up can look like a routine data problem when it is actually part of the same risk process the study is trying to measure.👇 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #epidemiology #censoring #publichealth #survivalanalysis #cohortstudies #bias

This Is Why Resource Allocation Models Don’t Work (And Nobody Talks Abou... youtu.be/t9T_vewkDW4?si… via @YouTube 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #publichealth #datascience #resourceallocation #analytics #decisionmaking #operationsresearch #cdc #who

A model can be externally validated and still be far more fragile than the label makes it sound.👇 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #biostatistics #externalvalidation #predictionmodels #publichealth #healthresearch #machinelearning #cdc #who

A competing risks analysis can sound technically careful while still leaving the core question—what the estimate actually means—dangerously unclear.👇 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #epidemiology #competingrisks #publichealth #survivalanalysis

In Theory, Real-Time Health Alerts Work. In Reality… They Don’t youtu.be/_dG7Jtnga7k?si… via @YouTube 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #publichealth #datascience #healthalerts #surveillance #digitalhealth #decisionmaking #healthsystems #analytics

A study can look “adjusted for baseline differences” and still be nowhere close to making the groups truly comparable. 👇 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #biostatistics #baselineadjustment #publichealth #studydesign #confounding #regression #cdc

An attack rate can look sharp and decisive while hiding the fact that the “exposed” group was never exposed in the same way. 👇 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #epidemiology #attackrate #publichealth #outbreakinvestigation #surveillance #cdc #who

Everyone Uses Public Health Scorecards… But They Fail When the Incentive Is the Metric youtu.be/vJg_NciJQUE 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #publichealth #datascience #scorecards #healthmetrics #analytics #healthsystems #digitalhealth #cdc #who

A result can survive every sensitivity analysis in the paper and still be fragile in the ways that matter most.👇 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #biostatistics #sensitivityanalysis #publichealth #causalinference #healthresearch #studydesign #cdc

A secondary attack rate can look like a transmission fact when it is really being shaped by how contacts were found and followed.👇 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #epidemiology #secondaryattackrate #infectiousdisease #outbreak #surveillance #cdc

Everyone Uses AI Triage Tools… But They Fail When the Health System Is t... youtu.be/nlmkV_s6Eew?si… via @YouTube 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #ai #publichealth #datascience #healthcare #triage #machinelearning #predictivemodeling #cdc #who

This change allows for deeper insights while also giving you more time to catch up on past episodes and get the most value from the content. Appreciate you all for listening and supporting our channels.

Starting next week, all three weekly episodes—covering data science in public health, epidemiology, and biostatistics—will be released together every Wednesday.

New Schedule Update: Introducing Triple-Drop Wednesdays We’re making a small but important update to the podcast schedule. To focus on delivering higher-quality, more in-depth content, we’re moving to a new release format: Triple-Drop Wednesdays.

Calling a study “underpowered” or “well-powered” often hides the real issue: what the estimate actually means.👇 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #biostatistics #statisticalpower #publichealth #healthresearch #uncertainty #inference #effectsize #cdc

High prevalence does not always mean high risk. That misunderstanding quietly breaks a lot of public health interpretation.👇 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #epidemiology #prevalence #publichealth #populationhealth #surveillance #healthdata #cdc

This Is Why Health Equity Dashboards Don’t Work (And Nobody Talks About It) youtu.be/6ZTCFEaiMcY?si… via @YouTube 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #publichealth #datascience #healthequity #dashboards #healthdata #ai #analytics #digitalhealth #cdc

A result can be statistically significant and still tell you almost nothing useful. That is the flaw nobody likes to admit.👇 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #biostatistics #statisticalsignificance #pvalues #healthresearch #effectsize #inference

An outbreak curve can look like pure transmission data when it is actually telling a story shaped by delay, revision, and incomplete reporting.👇 🎧Listen to the audio version here: podcasts.apple.com/us/podcast/dat… #epidemiology #outbreaks #epicurves #publichealth #surveillance #cdc