rationalskepticism.org

Made of Stars wrote:

slottedbox wrote:I've been told that a big problem with the MB is that subjects can fake it, that is with a bit of nouse one can answer the questions in such a way as to produce the outcome which suits one best.

The question is, why would you?

The problem is that on self-reported measures there is always going to be a bias in how you see yourself and how you'd describe yourself to others, whether intentional or not. For example, people tend to see themselves in a more positive light than is true, they remember positive aspects easier than negative ones, their answers change depending on specifically how the question is asked (e.g. "Do you enjoy your own company?" vs "Are you awkward, shy, and incapable of interacting with other people in social settings?"), etc etc.

The MBTI makes absolutely no attempt to control for any of the biases that we're aware of. The test simply asks you what you think of yourself and then repeats back what you reported thinking about yourself. It's not surprising that many people find that it accurately describes them as it essentially asks: "Do you think you're introverted?" and, if yes, then reports at the end that it thinks you're introverted.

The more damning fact is that even though it is only telling you what you believe about yourself, for many people they find it is an inaccurate reflection of their own personality types. That's an absurd failure.