Types of tasks analysis in testing chemistry

Abstract

Testing in chemistry teaching, as a means of estimating students achievement, is viewed through an analysis of results achieved in the tests with alternative or multiple choice. The aim of the work is to find a more appropriate form of such tasks and increase their reliability in testing certain topics at various levels. The research included 687 students of I, II and III forms of grammar school, mathematical and science direction. The percentage of correct answers to the tasks without an explanation (A variety in which the student is required to choose among a number of distracters; B variety in which the student is required to explain his answer) was greater compared to the tasks that needed explanation. Comparison of the results obtained in the alternative choice tasks showed that there is a statistically significant difference (alpha=0,01) between the percentages of correct answers in tasks with and without explanation for the A and B varieties only in the tasks at the level of understanding. For the multiple choice tasks the required explanation of the chosen answer also leads to a statistically significant difference (alpha=0,01) between the percentages of correct answers to tasks with and without explanation, for all levels of knowledge. From the results presented it is clear that both the alternative and multiple choice tasks in chemistry testing may be a valid parameter of achievement in chemistry only if the answer chosen is proved by an explanation.

U ovom radu se testiranje u nastavi hemije, kao način za proveru znanja, posmatra kroz analizu rezultata postignutih u zadacima tipa alternativnog i višestrukog izbora. Cilj rada je pronalaženje što podesnije forme postavljanja ovih zadataka i povećanja njihove pouzdanosti pri proveravanju određenih natavnih sadržaja na različitih nivoima znanja. Uzorkom je bilo obuhvaćeno 687 učenika I, II i III razreda gimnazije prirodnog smera. Procenat tačnih odgovora za zadatke bez obrazloženja je za oba tipa zadataka (varijanta A - u kojoj je potrebno kao odgovor izabrati neki od ponuđenih distraktora; varijanta B - u kojoj je izabrani odgovor trebalo obrazložiti) bio veći u odnosu na zadatke sa obrazloženjem. Poređenje rezultata dobijenih u zadacima alternativnog izbora pokazalo je da postoji za varijante A i B samo u zadacima na nivou razumevanja. Za zadatke višestrukog izbora traženo obrazloženje izabranog odgovora takođe dovodi do statistički značajne razlike (alfa-0,01) između procenata tačnih odgovora u zadacima sa i bez objašnjenja, za sve nivoe znanja. Iz rezultata iznetih u radu se vidi dda zadaci alternativnog i višestrukog izbora, u testiranju iz hemije mogu biti validan parametar usvojenosti znanja iz hemije samo ako je izabrani odgovor potvrđen obrazloženjem.