Short introduction - it's "scientific research" that portrays the concept of people naturally going from incompetent and unable to judge their own abilities and that of others to getting more humble as they accrue competence and better at judging themselves and others. This is represented with "competence" and "confidence." [Here](https://upload.wikimedia.org/wikipedia/commons/thumb/4/46/Dunning%E2%80%93Kruger_Effect_01.svg/1231px-Dunning%E2%80%93Kruger_Effect_01.svg.png)'s the notorious graph.
Why it's fundamentally flawed:
- It was primarily used on Dunning and Kruger's psychology students. The conclusion could easily be: "Look how stupid our students are" instead of applying it universally.
- The idea of known known, known unknown, unknown unknown, and false knowledge, and the concept of variance in competence are long known to man.
- It is impossible to measure competence, thus also impossible to measure "confidence", the sample size is low and biased (leftists, students, measurable scores, all in big cities in closed environments). For example leftists are fundamentally on the higher end of being delusional.
- Mathematically it's a r=a/b formula, whereas 'a' is competence (x axis) and 'b' is confidence (y axis). The implication is that while 'a' is low, any variance in 'b' causes 'r' to go wildly up or down. To illustrate it with numbers: If competence is scored from 0 to 100 and confidence is the assumed competence, when you are at 10 competence, thinking you are 30 means you overestimated your competence by 200%. If you are at 50 and think you are 70, you overestimated your competence only by 40% - even though in both cases you were off by 20 points. If you are at 70, and you think you are 100, it means you believe you are (one of) the best, yet you are considered only 43% wrong, even though the delusion might be highest in this case. Remember that fancy graph? It is more reflective of bad statistics, because this phenomenon can universally happen. It's simply bad practice.
- If you'd measure the difference between actual score vs expected score in absolute values, you'd rather have a slowly declining line. Wikipedia itself even [shows](https://upload.wikimedia.org/wikipedia/commons/thumb/4/43/Dunning%E2%80%93Kruger_Effect2.svg/512px-Dunning%E2%80%93Kruger_Effect2.svg.png) this.
- If it were done that way, the graph wouldn't be as fancy and popular. Thus various concepts regarding knowledge wouldn't have been arbitrarily attributed to Dunning-Kruger, who felt compelled to explain the very self-evident concept of incompetent people being less able to estimate competence in themselves and others.
- The fact is that most people are well aware of their competence, people are aware of variance in competence, and that there are people who have a confident personality. Combine that type of personality and low competence (think of fat sheboons and pajeets in general), and you have your culprits the Dunning-Kruger effect is all about - in spirit, not numbers. Because there are eternally incompetent entities.
It is said that despite all that it's still true, but the fact is that it's just a fucked up way to visualize and measure it, which merely makes it *look* fancy and scientify, but isn't. There is more information to it like how it cannot be actually reproduced, how it was used by jewish media against right-wingers.
---
In short: It's all a little statistical-mathematical trick. It's vastly overblown by media and doesn't provide any new knowledge. But people are obsessed with it as if it has any value.