Nice read, I took a look at the paper and it looks like a really nice metric. Having people twice as poor counting twice as much makes a lot of sense. Very nice way to handle both relative poverty and absolute poverty while not ignoring how some people are much poorer than others.
Yeah, like natural gas
The study pretty much draws no conclusion other than “more research needed”
Yeah, doesn’t get as many clicks as “medium-small” meta study does it.
No significant effect means no statistically significant effect here. The authors never claim an effect doesn’t exist.
Yeah, even the ones where an effect was found have a very low GRADE score (framework for assessing whether conclusions of studies should be used as guidelines) that is affected by risk of biases, directness of evidence, consistency and publication bias.
The author of the study addresses it. He mentions that a lot of studies fail to disclose funding or have a risk of bias based on funding.
Don’t understand the downvotes, they generally work most of the time but can definitely cause issues for grids and home setup without setups to handle it. Plugin solar is really country dependent because of what the exact electrical setup is varies.
To be fair, the population of this meta analysis is severely underpowered to assess whether there is an effect for rare/semi-rare conditions. (population is couple of hundred in most areas) and it shows in the confidence intervals. Wouldn’t exactly call this a “huge” study.
A better title would be “Study unable to determine if cannabis helps or hurts anxiety, depression, or PTSD because of lack of data” which sounds less glamorous.
No significant effect does not mean no effect, very crucial difference. It means that the effect wasn’t statistically significant, i.e. <5% chance of being random.
Wikipedia covers it nicely:
https://en.wikipedia.org/wiki/Advanced_Video_Coding?wprov=sfla1