"I make my living from data, yet I consistently find...I have to remind [people] that data is not a perfect representation of reality: It’s a fundamentally human construct, and therefore subject to...meaningful and consequential imperfections."
6/9. I say all this because I think it is important to remember this broader point: it is not tests per se that are a problem, but how we use those tests. They're like other forms of data; as noted, they don't say anything, WE give them meaning
"Data is an imperfect approximation of some aspect of the world at a certain time and place" -- "we’ve conflated data with truth. And this has dangerous implications for our ability to understand, explain, and improve the things we care about."
'What does the data say?' It doesn’t say anything. Humans say things. They say what they notice or look for in data — data that only exists in the first place because humans chose to collect it, and they collected it using human-made tools.
This from is terrific and you should share it widely. I’m a data scientist who is skeptical about data via
“Data is a necessary ingredient in discovery, but you need a human to select it, shape it, and then turn it into an insight.” Thoughtful summary.
I’m a data scientist who is skeptical about #data #datascience via
Data is not information, information is not knowledge, knowledge is not understanding, understanding is not wisdom. A data scientist explains the vagaries of the first link...
What do the data say? Nothing. Data don't talk. Data can't answer questions. They can only inform questions. Humans collect and interpret data. Before accepting their conclusions, consider their competence and integrity. via
I’m a data scientist who is skeptical about data. Data is not a perfect representation of reality: It’s a fundamentally human construct, and therefore subject to biases, limitations, and other meaningful and consequential imperfections —
Data can't answer questions. They can only inform questions. Humans collect and interpret data. Our choices are subject to errors. The quality of data-driven insights depends on the competence and integrity of the people generating them.
The more experienced a person is with data, the more they appreciate its limits. Fantastic piece here by . HT
“What does the data say?” Data doesn’t say anything. Humans say things. They say what they notice or look for in data—data that only exists in the first place because humans chose to collect it, and they collected it using human-made tools.
"Data is a necessary ingredient in discovery, but you need a human to select it, shape it, and then turn it into an insight. Data is therefore only as useful as its quality and the skills of the person wielding it."
“I believe that the usefulness of data & science comes not from the fact that it’s perfect & complete, but from the fact that we recognize the limitations of our efforts...We are only as strong as our humility & awareness of our limitations” via
. professor discusses four errors we need to be looking out for to avoid biased data: - Random errors - Systematic errors - Errors of choosing what to measure, and - Errors of exclusion
Is data science legit? “Data is a necessary ingredient in discovery, but you need a human to select it, shape it, and then turn it into an insight. Data is therefore only as useful as its quality and the skills of the person wielding it.” via ⁦
Why “what does the data say?” is a naive question
I highly recommend this article about data to all those who use numbers to make decisions.
Is data science legit? — Quartz