95% of these tests are going to show very blurry results. I've made posts elsewhere regarding why. You are better off doing a 10 page vomit, or taking an assessment (and more than one time) that is inclusive to scoring on a more garnualar range not just functions, account for test bias, desirablility bias, prototype matching... I wouldn't suspect these tests that are made in 5 minutes to hold any insight, unfortunately.
Not advertising the one I'm building that will have a scoring engine behind it, but find something better and retest 2 or 3 times for variance/invariance.