1) p value.
p value represents statistical significance. The way that it works is that an experiment is set up with a "null hypothesis". The null hypothesis roughly equates to meaning "no effect". So in a study of whether HCQ improved mortality from covid-19, the null hypothesis is that HCQ makes no difference to covid-19 mortality.
The p value is the likelihood that the null hypothesis is true. Conventionally, p=0.05 (i.e. a 5% chance) is considered the boundary, where if it is equal to or lower, the chance that the null hypothesis is true has become sufficiently low that it is rejected, so an effect was observed. What this means is that every single study that has a "negative" effect (i.e. the null hypothesis is true) will have a p value greater than 0.05. It is inherently what the p value represents, therefore, these are not "shit studies". They are studies that show no significant effect.
That said, there are two issues here: firstly, what this website means by a "negative" effect is that HCQ is harmful when taken for covid-19, not that it is ineffective. Here there may be a complication, because the p value in such cases may actually be a test of whether the null hypothesis is being rejected for the case that it is harmful. You'd need to read the paper to see. There is a second, massive issue, which then relates to how they assess papers as showing benefit ("positive") or harm ("negative"). More on that later.
p value represents statistical significance. The way that it works is that an experiment is set up with a "null hypothesis". The null hypothesis roughly equates to meaning "no effect". So in a study of whether HCQ improved mortality from covid-19, the null hypothesis is that HCQ makes no difference to covid-19 mortality.
The p value is the likelihood that the null hypothesis is true. Conventionally, p=0.05 (i.e. a 5% chance) is considered the boundary, where if it is equal to or lower, the chance that the null hypothesis is true has become sufficiently low that it is rejected, so an effect was observed. What this means is that every single study that has a "negative" effect (i.e. the null hypothesis is true) will have a p value greater than 0.05. It is inherently what the p value represents, therefore, these are not "shit studies". They are studies that show no significant effect.
That said, there are two issues here: firstly, what this website means by a "negative" effect is that HCQ is harmful when taken for covid-19, not that it is ineffective. Here there may be a complication, because the p value in such cases may actually be a test of whether the null hypothesis is being rejected for the case that it is harmful. You'd need to read the paper to see. There is a second, massive issue, which then relates to how they assess papers as showing benefit ("positive") or harm ("negative"). More on that later.