2 ms·

> It is possible to determine whether a skeleton is from a biological male or female using objective observations based on the size and shape of the bones. Tha

> It is possible to determine whether a skeleton is from a biological male or female using objective observations based on the size and shape of the bones. That's not true. There is a probability assignment to pre-established categories. See "A review of sex estimation techniques during examination of skeletal remains in forensic anthropology casework" at https://www.sciencedirect.com/science/article/pii/S0379073816300202 https://www.sciencedirect.com/science/article/pii/S037907381... ] Accuracy of sexing of unknown skeletal remains depends on the element present for analysis and its preservation state ... A minimum threshold of 95% representing the optimum level of sex estimation accuracy and reliability is acceptable in forensic setting ... While the accuracy ranges from 85 to 95% in traditional osteometric methods for sex estimation [54], [61], [73], [74], [75], [76], [77], [78], [79], [80], [81], [82], [83], the accuracy in the tunes of 96% has been reported using virtual morphognostic sexing methods Nor would we expect 100% accurate assignments of people to "biological male or female" because there are many examples which show that a strict binary classification does not have a scientific basis.

FooBarBizBazz 4y ago
> A: It is possible to determine whether a skeleton is from a biological male or female using objective observations based on the size and shape of the bones. > B: That's not true. There is a probability assignment to pre-established categories. This is how literally everything works. "Officer, the 'red-light, green-light' binary is simply false, because estimation errors make the state of the light unknowable!" It's just sophism.
- eesmith 4y ago
  Your "B" summary left out how I then pointed to a paper giving an overview of those accuracy numbers, across a range of methods. What's the error rate in telling if a traffic light is red or green? My own experience says it's very low. Even for people who are color-blind (due to the standardized orientation of the lights, and choice of a blueish-green that's easier for red-green colorblind people to recognize). What's the error rate in assigning gender based on skeletal remains using the approach mentioned in the article? That appears to be 5-15% -- high enough that the article should have used a qualifier like "It is usually possible" or "It is often possible". Even the 5% error rate "acceptable in forensic setting" is far higher than the error rate in your traffic light example.

FooBarBizBazz 4y ago

> A: It is possible to determine whether a skeleton is from a biological male or female using objective observations based on the size and shape of the bones. > B: That's not true. There is a probability assignment to pre-established categories. This is how literally everything works. "Officer, the 'red-light, green-light' binary is simply false, because estimation errors make the state of the light unknowable!" It's just sophism.

eesmith 4y ago
Your "B" summary left out how I then pointed to a paper giving an overview of those accuracy numbers, across a range of methods. What's the error rate in telling if a traffic light is red or green? My own experience says it's very low. Even for people who are color-blind (due to the standardized orientation of the lights, and choice of a blueish-green that's easier for red-green colorblind people to recognize). What's the error rate in assigning gender based on skeletal remains using the approach mentioned in the article? That appears to be 5-15% -- high enough that the article should have used a qualifier like "It is usually possible" or "It is often possible". Even the 5% error rate "acceptable in forensic setting" is far higher than the error rate in your traffic light example.