There really isn't a simple way to find out if a study should be taken seriously or not.
What you need to do is read the original literature and pay attention to the details. Bad studies are usually bad for a few reasons:
1. Bad experimental design. If you are testing hypothesis X, but your study design doesn't control for all the variables, then the data won't really tell you anything. For example, if you want to see if a high-fat diet increases the risk of cancer and you don't control for smoking or environmental exposure, your data is crap whether it supports or rejects your hypothesis.
2. Bad data. You may have a great experimental design, but if your data isn't strongly supportive of a conclusion, then it's a weak study. If we take the above example and compare a high-fat diet to a low-fat diet across 5 different group and 2 groups show no effect, 2 groups are right on the border of being statistically significant and one group is strongly positive, it doesn't really tell you anything.
3. Bad conclusions. The data says one thing, but the author goes ahead and concludes it means something else. You see this a LOT in bad science. Again if we take the above example, the data may strongly support that a high-fat diet is bad, but then the author goes ahead and says that all fats are bad. The data doesn't support this. This is sort of what the sugar-is-toxic studies do. Are diets that are hyper-caloric and have a high percentage of carbohydrates derived from simple sugars bad for you? Most likely. Should you actively try and exclude them from your diet? The answer to that is no (and the data doesn't support that conclusion).
I was formally trained as a scientist, but no longer work in the lab, but the training on I got in designing experiments (be they in the lab or in marketing) has served me well in filtering out "crap" of all sorts.
What you need to do is read the original literature and pay attention to the details. Bad studies are usually bad for a few reasons:
1. Bad experimental design. If you are testing hypothesis X, but your study design doesn't control for all the variables, then the data won't really tell you anything. For example, if you want to see if a high-fat diet increases the risk of cancer and you don't control for smoking or environmental exposure, your data is crap whether it supports or rejects your hypothesis.
2. Bad data. You may have a great experimental design, but if your data isn't strongly supportive of a conclusion, then it's a weak study. If we take the above example and compare a high-fat diet to a low-fat diet across 5 different group and 2 groups show no effect, 2 groups are right on the border of being statistically significant and one group is strongly positive, it doesn't really tell you anything.
3. Bad conclusions. The data says one thing, but the author goes ahead and concludes it means something else. You see this a LOT in bad science. Again if we take the above example, the data may strongly support that a high-fat diet is bad, but then the author goes ahead and says that all fats are bad. The data doesn't support this. This is sort of what the sugar-is-toxic studies do. Are diets that are hyper-caloric and have a high percentage of carbohydrates derived from simple sugars bad for you? Most likely. Should you actively try and exclude them from your diet? The answer to that is no (and the data doesn't support that conclusion).
I was formally trained as a scientist, but no longer work in the lab, but the training on I got in designing experiments (be they in the lab or in marketing) has served me well in filtering out "crap" of all sorts.