Trilogica IQ Test
© Mislav Predavec, November 2007.
Trilogica is a synthesis of 3 IQ tests - Algebrica (32 numerically-algebric-pattern recognition items), Esoterica (32 logically-spatial abstract drawings) and Hieroglyphica (32 verbal associations). Algebrica has been created mostly in the spirit of Strict Logic Sequence Examination I and II. Esoterica has been inspired by fantastic tests Logima Strictica 36 and Logicaus Strictimanus 24 created by Robert Lato yet Hieroglyphica is a try to make one verbal test as much culture fair as possible through the universality of its clues. All parts are high range IQ tests and, as very difficult, time unlimited.
Every correctly solved item will bring a certain amount of points according to the frequency of its solubility by formula
P = 1 + Ö-2´ln(f)
where P is amount of points and f frequency of solubility. Thus, below are examples of solubility frequencies and belonging amounts of points
Test security consistence
Balanced score calculated by scoring system mentioned above is one of warrants of test security. Since the raw score won ´t be revealed to testers it will be much more difficult to figure out which items are correctly solved. That is a good prevention against items correctness analysis and test solutions engineering. Scoring system in itself is very good prevention against solution sharing and team work since with the raising of solubility points lower down. Making the hardest items worthy this system is also a very good stimulant for testers to be persistent with them.
Norms and reports
Testers won ´t receive classic score reports immediately after their submissions because every new submission is an input which influences other scores (points). Scoring system will work pretty similar as a stock exchange. Every tester will receive his temporarily score and percentage of a sum of points of correctly solved items entirely by all testers first time after 7, second time after 20 and third time after 50 submissions by e-mail. Once the statistics stabilize and new submissions would have very little influence on other scores all testers will receive officially score reports and IQ calculated from those statistics (that is between 50 and 100 submissions expected). The sum of points of correctly solved items entirely by all testers will be considered as a synergy of entire population intelligence and theoretically sublimed in one person with IQ 200 (sd=16). In the same way the medium score will correspond with IQ 146 (sd=16), medium IQ on most wide previous high range IQ tests in the world. All scores will be interpolated between those 2 values, IQs will be calculated and one day issued to their testers.