Tiktok is once again confirming my belief that we need to teach people to read academic works critically. They're better than tumblr or twitter, but they also need to be considered and poked for flaws. And people will present them in disingenuous ways, even accidentally. If someone's talking about a "100% increase" that could be from 1% to 2%. A 400% increase could be from 0.1 to 0.4. If you're interested in the data it's always worth finding the study yourself and checking what the raw numbers were.

Is a study older than 10 years? Then you need to consider whether the data and conclusion are still relevant in a contemporary social context, especially if it's a social science study (which are the ones I see most misinterpreted, which hurts my little sociology heart). A study on computer usage from 2010 is no longer relevant. It might draw a conclusion that you agree with and could've been accurate at the time, but the data is gonna be skewed because the way we interact with computers and the internet has changed a lot.

Is experimental design playing a role? An example I saw recently was someone presenting the data of a study on sexual deviance that asked participants to masturbate during the experiment as globally applicable. It's not - by asking subjects to perform a sexually deviant act as part of the study it selected for more deviant individuals. Social science experiments always have some level of self-selection (participants agreed to do our experiment), but consider how much of an impact it might have on the result. It can be so big the whole study is bullshit (like a certain study that asked the parents of trans teens about their transitions and recruited participants from a well-known transphobic forum), or just something to keep in mind (a lot of university studies will have a bias towards participants between 18-30 because they recruit a lot of students).

And as always check citations and check author affiliations. But remember citations vary by subject and are not always positive. Niche subjects will have small numbers of citations, and make sure most of the citations are positive and not "look at this bullshit".

You might also be able to find a review of the work on google scholar. These can help, but remember that the same critical lens needs to be applied to them as well.

Academia works because people pick apart each other's work. This does not need to be limited to people who are "qualified" to do so. You, dear reader, can and should read academic publications critically. No work is above criticism. If someone is acting like a work or a theorist or a researcher is above criticism they are either lying or didn't read carefully enough to spot the issue.