Analysis of multiple choice questions in the elementary school physics competition

V diplomskem delu je predstavljena analiza nalog izbirnega tipa s tekmovanj za Stefanova priznanja. Predstavljena sta zgodovina in popularizacija tekmovanja. Podrobneje sem pregledala tri šolska leta in tri najbolj obsežne teme: vzgon, silo ter gibanje. Primerjala sem uspešnost reševanja nalog ter analizirala učinkovitost odgovorov. Na podlagi statističnih podatkov sem ugotovila, da ni opaznega napredka pri reševanju nalog od leta do leta in da tekmovalci naloge o vzgonu ter silah bistveno slabše rešujejo kot naloge o gibanju. To pripisujem težavnosti. Za vsako od nalog iz let 2004, 2007 in 2008 s tekmovanj za zlato Stefanovo priznanje sem sestavila graf, iz katerega je razvidno, ali možni odgovori zastavljeni tako, da razlikujejo med uspešnimi in manj uspešnimi tekmovalci. Pri nekaterih nalogah se je razlika jasno pokazala, pri drugih pa ne, kar lahko kaže na to, da je večina nalogo reševala z ugibanjem. Naloge iz leta 2004 so bile glede na učinkovitost dobre, tiste iz leta 2008 pa predvsem lahke.This diploma thesis presents an analysis of multiple choice questions in the elementary school physics competition.. It presents the history and popularisation of the competition. In detail I examined three school years and three rather extensive topics: buoyancy, force and motion. I also compared the successfulness of problem solving and analysed the effectiveness of responses. Based on statistical data I concluded that there was no significant progress in problem solving through the years. I found out that competitors are solving motion related problems fundamentally better than those related to buoyancy and forces which I attribute to different levels of difficulty. I created a graph for every exercise from the elementary school physics competitions from years 2004, 2007 and 2008 from which it is evident if possible solutions are given in a manner to differentiate between successful and less successful competitors. With some exercises the difference was clearly seen, but with others it was not, which can indicate that the majority of competitors were solving that specific problem with speculation. Considering the effectiveness of distractors, the test problems from the 2004 competition were good, while the ones from 2008 were mostly easily solvable