AAAAaagh. The authors expressed my pent-up discontent buried in my chest for years. I agree that the "reproducibility crisis" isn't just about abuse of statistics. It's symptomatic of perverse...
AAAAaagh.
The authors expressed my pent-up discontent buried in my chest for years. I agree that the "reproducibility crisis" isn't just about abuse of statistics. It's symptomatic of perverse incentives and, in extreme cases, downright corruption.
When understanding, care, and honesty become valued less than novelty, visibility, scale, funding, and salary, science is at risk.
That's it, the one-sentence salient summary, if one is needed.
My only objection is the use of the phrase "cargo cult" as a pejorative. We could simply say "performative", "empty gestures", or similar. I find the generic use of "cargo cult" has a colonial taste and devaluing the real lives and complex histories of the Melanesians. But I'm ready to let this pass and I don't expect this particular matter get a life of its own in this thread.
On the bright side, more people are becoming aware and taking action. Some journals are vetting more rigorously the statistical aspects of the submissions, for example, by appointing dedicated...
On the bright side, more people are becoming aware and taking action. Some journals are vetting more rigorously the statistical aspects of the submissions, for example, by appointing dedicated statistics or data-science editors.
It's not colonial or pejorative to use the term cargo cult; the objective truth of the matter is that the islanders were playing with things they don't understand in the hope of reproducing past...
My only objection is the use of the phrase "cargo cult" as a pejorative
It's not colonial or pejorative to use the term cargo cult; the objective truth of the matter is that the islanders were playing with things they don't understand in the hope of reproducing past prosperity, and thus it is an apt term for uninformed misuse of statistics. It should also be noted the term already sees somewhat widespread use in fields such as software engineering.
Let's agree to disagree, I hope. I apologize for bringing this up in the thread in the first place, which was really a diversion. This is my last word about "cargo-cult" in this thread. How do you...
Let's agree to disagree, I hope. I apologize for bringing this up in the thread in the first place, which was really a diversion. This is my last word about "cargo-cult" in this thread.
How do you think of the current situation of practising statistics in science, industry, or maybe software engineering?
AAAAaagh.
The authors expressed my pent-up discontent buried in my chest for years. I agree that the "reproducibility crisis" isn't just about abuse of statistics. It's symptomatic of perverse incentives and, in extreme cases, downright corruption.
That's it, the one-sentence salient summary, if one is needed.
My only objection is the use of the phrase "cargo cult" as a pejorative. We could simply say "performative", "empty gestures", or similar. I find the generic use of "cargo cult" has a colonial taste and devaluing the real lives and complex histories of the Melanesians. But I'm ready to let this pass and I don't expect this particular matter get a life of its own in this thread.
On the bright side, more people are becoming aware and taking action. Some journals are vetting more rigorously the statistical aspects of the submissions, for example, by appointing dedicated statistics or data-science editors.
That was exactly my experience in my Physics degree too. I'm having to play catchup a bit now I'm doing Data Science post grad :/
It's not colonial or pejorative to use the term cargo cult; the objective truth of the matter is that the islanders were playing with things they don't understand in the hope of reproducing past prosperity, and thus it is an apt term for uninformed misuse of statistics. It should also be noted the term already sees somewhat widespread use in fields such as software engineering.
Let's agree to disagree, I hope. I apologize for bringing this up in the thread in the first place, which was really a diversion. This is my last word about "cargo-cult" in this thread.
How do you think of the current situation of practising statistics in science, industry, or maybe software engineering?