I was just reading an interesting thread on ECF about a study being done by My (Crystal) Hua, who is an undergraduate at the University of California Riverside.
Summary of her research from this link:
In Silico Approach Using Health-based Interactomes to Analyze the Symptoms Reported by Electronic Cigarette Users in Online Forums.
A few brief thoughts I had on that. From looking at what In My (Crystal) Hua is doing it sounds more like epidemiology (very, very loosely) than anything in the hard data category. It is quite fundamentally flawed because she doesn't have a set sample size. People come and go from forums, and may post that they feel thirsty from vaping and then never come back to update that since they've been drinking more water they feel better. Same for other initial issues, many of which are due to quitting cigarettes, and not actually from vaping at all.
Studies done here where I am are done with a set sample size of people who all give informed consent. Their individual physiology can be measured over time. I thought about volunteering for one but didn't have the time to do it.
In silico is just a form of computational biology. Useful for pulling out real data, but seems pretty silly to try using internet forums as the data mine.
The thread is here for anyone who wants to read it. Some good comments there.
Summary of her research from this link:
In Silico Approach Using Health-based Interactomes to Analyze the Symptoms Reported by Electronic Cigarette Users in Online Forums.
A few brief thoughts I had on that. From looking at what In My (Crystal) Hua is doing it sounds more like epidemiology (very, very loosely) than anything in the hard data category. It is quite fundamentally flawed because she doesn't have a set sample size. People come and go from forums, and may post that they feel thirsty from vaping and then never come back to update that since they've been drinking more water they feel better. Same for other initial issues, many of which are due to quitting cigarettes, and not actually from vaping at all.
Studies done here where I am are done with a set sample size of people who all give informed consent. Their individual physiology can be measured over time. I thought about volunteering for one but didn't have the time to do it.
In silico is just a form of computational biology. Useful for pulling out real data, but seems pretty silly to try using internet forums as the data mine.
The thread is here for anyone who wants to read it. Some good comments there.
