I had a college professor that used to tell us in class: "there are three kinds of liars; liars, damned liars and statisticians". This has proven to be one of the most accurate proverbs of the secular lexicon.
A statistician can make a number "say" just about anything they want it to. It's not exactly lying they will claim, after all they will remind you, they clearly published the data and perimeters of the study. How the data was presented and interpreted may not have been, get ready, "entirely consistent with the observable phenomenon".
Take for instance this little 3P Libertarian gem: Smoking Marijuana is 114 Times Safer Than Drinking Alcohol – Study
My first impression based solely on the headline is that someone has produced a scientific study that indicates that smoking marijuana is a reasonably safe activity. After reading the story that is in fact the conclusion the author reaches after reviewing a sampling of material on the subject.
What I don't like about this type of reporting is that it involves what I see as a lie by implication. The implication is that pot smoking is a relatively safe activity. The article doesn't come right out and state that conclusion as a fact. What they eventually do state is that "the Post noted that Wayne Hall of the World Health Organization said it’s nearly impossible for even those who smoke large amounts of cannabis to overdose on the drug."
This is a very important distinction to make. "Safer" in this article is eventually defined as 'less likely to die from an overdose of the drug'. Smoking pot then isn't really 114 times safer in every sense of the word, its just 114times less likely that a pot head will immediately OD from consuming the drug. That very well may be an important bit of information, but its a long way from the blanket statement of being 114 times safer.
We don't have easy access to the actual study that they base this on, so I'll only interject one further point. We aren't told anything about deaths that occur as a result of using a mixture of marijuana and other substances. It seems to me that this could be very important.
One of the medical benefits of marijuana is that it helps combat nausea. This is an important feature we are told for people who are undergoing chemotherapy as it helps them keep food and drink in their stomach. One other related medical benefit is what is sometimes referred to as "the munchies". Apparently in some cases the use of marijuana stimulates the desire to eat and drink more than the person would normally.
I agree with the medical marijuana enthusiasts that these properties are present in the drug.
The principle way the human body deals with over consumption of alcohol is by vomiting excess liquor out of the stomach. Puking ones guts out is the bodies method of protecting itself after drinking too much. If a person smokes pot and drinks too much alcohol their body has ingested one drug that makes it impossible to expel a poisonous substance.
In cases where alcohol is the cause of death but THC is present in the blood stream, the cause of death is normally recorded as acute alcohol poisoning. It would be interesting to know how many of the alcohol related deaths in the study had a THC component as a contributing factor. My guess is if that data is added in marijuana smoking might not be found to be 114 times safer than booze.