Amanda

Numbers & Narratives. R enthusiast.

#BuisnessIntelligence #EconDev #R #Python #Hokie #Tartan

[Hansi Lo Wang (he/him) @hansilowang]

BREAKING: 14 states had significant net over/undercounts in the 2020 census numbers used to allocate House seats and federal money
HI +6.8%
DE +5.5%
RI +5.1%
MN +3.8%
NY +3.4%
UT +2.6%
MA +2.2%
OH +1.5%
TX -1.9%
IL -2.0%
FL -3.5%
MS -4.1%
TN -4.8%
AR -5.0%
https://t.co/xVZNsakL7X

[vicki @vboykis]

All models are wrong, some are easier to implement in production under a deadline

[Amanda @acr_said]

maps, like any other communication tool, can shape narratives, both purposefully and in unintended ways.... (a🧵 worth a read)

[Amanda @acr_said]

My ears perked listening in because @ShafikIslam and I had *just* used an optical illusion analogy for the sneaky biases that worm their way into data work in our essay: https://t.co/q0YSUx6qHv

(is it a message from the universe or just #recencybias & #confirmationbias?) 🤔

[Exploratorium @exploratorium]

Do you see what I see? Explore blind spots––both physical and social––with Exploratorium Sr. Physics Educator @DarthScience, and learn how #STEM can be used to teach about equity @NPR @ScienceRegina on #shortwaveNPR: https://t.co/paY4ijRcpS

[Shafiqul Islam @ShafikIslam]

What is #CertaintyBias?
How does it affect #DataScience #DataAnalytics?
Here is a story you may enjoy @acr_said
https://t.co/w54fp3kV6Z

[Amanda @acr_said]

reading @_cingraham on @substack
is a delight, not just because I now have nuanced info about wolves/deer/hunters/roadkill, but bc the reality of performing thoughtful work while a kid is treating you like a food-dispensing jungle gym is... relatable. https://t.co/5I7Dfb7zXq

[Amanda @acr_said]

⬇ A needed perspective ⬇

[Eric Vance @ervance]

How do we develop data acumen within data scientists? Our approach is to teach data science from a humanities perspective: https://t.co/X3T7PI5rNN
@CUBoulder @NSF

[Amanda @acr_said]

if only there was official recognition for papers that just have fantastic titles (entire basis for the evaluation)... I'd nominate this one: "p < 0.05, < 0.01, < 0.001, < 0.0001, < 0.00001, < 0.000001, or < 0.0000001 …"
https://t.co/RZqV7ExyUY #statstwitter #academictwitter

[César A. Hidalgo @cesifoti]

What are complex systems? And what does it mean to study complexity? The recent Physics Nobel Prize has brought complexity back into the limelight, but also, it is pushing those of us who have dedicated our lives to the of study complex systems to reflect on its history. 🧵 /1

[Amanda @acr_said]

At work, I use #Python #Rstudio and #Tableau to tell stories from data. But I know it’s not a linear path. The imagined line often connecting data to stories concerns me & I had to write about it. Read along via @substack

https://t.co/X3L6gqniya

[Amanda @acr_said]

This is not a cactus. It’s a web-optimized png of a jpg from a digital scan of a color slide that sat in a box for 68 years. What does that even mean?

https://t.co/udIOlr3TPW

#datascience #bigdata #AI #datastorytelling #dataforgood #dataethics

[César A. Hidalgo @cesifoti]

After advising PhD & Master students for over a decade, there is one thing I find most students need to unlearn: the half-ass work mentality acquired during years of tests and homework. Let me explain (thread 🧵). 1/N #AcademicTwitter

[Andrew D. Huberman, Ph.D. @hubermanlab]

Some data too beautiful not to share. Right eye (blue) and left eye (magenta) regions of visual thalamus. No protocol here; perhaps just revel in the fact that you have a map of the visual world in your brain & that all you see & think is a best guess about what is “out there”

[Naeem @Lahorite]

We are introducing PlasticNet (IBM Open Source Project) as part of "IBM Tech for Good" to build a repo of AI models to classify types, brands of plastics, trash on beaches in the ocean, etc.
@PacificWhale @IBM @RedHat
@Grady_Booch @tctjr @calstad
https://t.co/35sSsvFrxN

[Alex Mas @AMLabEcon]

This thread got me to join twitter. I have been following these issues a long time, and the benefit of keeping the unaltered ACS micro data in the public domain far exceeds the cost (which is infinitesimal). Data access=>more research=>improved public policy=>large social benefit

[Amanda @acr_said]

It isn't identical issue to how Census is proposing to alter ACS data. ACS microdata is something of a niche product (but really shouldn't be). Communities need to understand proposed changes across Census products because the changes impact how the data can be responsibly used.

[Amanda @acr_said]

This microdata is some of the most valuable microdata that communities haven't even been using to the extent that they should. This isn't just impacting researchers, this will limit future data analytics efforts that communities are just beginning to build capacity for...

[@Steven Ruggles @HistDem]

/1. Yesterday at the ACS Data Users Conference, the Census Bureau described its plans to replace the American Community Survey (ACS) microdata with “fully synthetic” data over the next three years.

[Amanda @acr_said]

This presentation from CO's Demography office last year highlights some of the issues in Census' approach to differential privacy and Census 2020. It is non-technical and highlights some of the challenges for producers and users https://t.co/yydVVZSjiU

[@Steven Ruggles @HistDem]

/1. The Census Bureau plans to add intentional errors to the 2020 census to protect the confidentiality of census respondents. The Census Bureau insists that the intentional error is necessary to combat the threat of “database reconstruction.”