Pregled bibliografske jedinice broj: 1255255
When the answer is correct, but the test-taker is unsure - on the correlates of confidence level in test item answers
When the answer is correct, but the test-taker is unsure - on the correlates of confidence level in test item answers // 23. Dani Ramira i Zorana Bujasa - knjiga sažetaka / Arambašić, L. ; Erceg, I. ; Kamenov, Ž. (ur.).
Zagreb: Odsjek za psihologiju Filozofskog fakulteta Sveučilišta u Zagrebu, 2017. str. 51-51 (predavanje, domaća recenzija, sažetak, znanstveni)
CROSBI ID: 1255255 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
When the answer is correct, but the test-taker is unsure - on the correlates of confidence level in test item answers
Autori
Ljubotina, Damir
Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni
Izvornik
23. Dani Ramira i Zorana Bujasa - knjiga sažetaka
/ Arambašić, L. ; Erceg, I. ; Kamenov, Ž. - Zagreb : Odsjek za psihologiju Filozofskog fakulteta Sveučilišta u Zagrebu, 2017, 51-51
Skup
23. Dani Ramira i Zorana Bujasa
Mjesto i datum
Zagreb, Hrvatska, 06.04.2017. - 08.04.2017
Vrsta sudjelovanja
Predavanje
Vrsta recenzije
Domaća recenzija
Ključne riječi
computer based testing, confidence level, test-taking strategies
Sažetak
The essential information for interpretation of knowledge and ability test results is the response accuracy. Computer based testing enabled conceptually new information on time spent solving the item and also enabled adding the ratings of the degree of respondents’ confidence in their responses. Although confidence ratings have been used in psychology research, in the field of test theory they have been seldom considered (Jonsson & Allwood, 2003 ; Yen et al., 2010). Some authors reported that alpha coefficient increased with the confidence in the answers, while others discussed the option of weighting responses with confidence ratings. The aim of this study was to investigate the utility of confidence ratings in item analysis, in evaluation of individual results, and to explore some of its correlates. Results were obtained on a sample of 279 university students and 97 high school students. Some personality measures (conscientiousness, impulsivity, neuroticism, self–competence, perfectionism and three dimensions of temperament), test-taking strategies, test-taking behaviour, and metacognition were investigated as possible correlates. Information on accuracy and solving speed, and confidence ratings were collected for each item of the cognitive ability test. On the item level, average confidence rating had relatively strong correlations to difficulty level and average solving speed. Number of correct responses and metacognition were significant predictors of confidence ratings, and to a lesser extent self-competence and test anxiety. The hypothesis that participants with more developed ability will show significantly higher confidence when giving a correct answer than when giving an incorrect one was partly supported. This kind of the difference in confidence ratings is usually smaller for participants with less developed ability. These analyses suggest that correct and incorrect answers should be analysed separately. Confidence rating as a supplemental indicator of test quality demonstrated satisfactory psychometric characteristics.
Izvorni jezik
Engleski
Znanstvena područja
Psihologija