Free Dissertations - Psychology Dissertations

Therefore It Was Assumed That The Items On The Questionnaire Did Fall Into ...

Therefore it was assumed that the items on the questionnaire did fall into the two categories, easy and hard. Given this, it was possible to test the difference between the confidence measure on the easy section and that on the hard section. This result was found to be significant, thus supporting the hypothesis that there would be a difference between the overconfidence effect for easy and difficult items with the overconfidence effect bigger for difficult items. This was further supported by the calibrated curve used to illustrate these results. It should be noted however, that the overconfidence effect on the hard section had a mean of -0.04.
This result is very close to zero which actually indicates no effect, and not an overconfidence effect as the literature would be predicted. It is possibly that methodological problems may have impacted on the result. For example, as the sample was not randomly chose it may have been biased (e.g. it was mostly female and the students may have heard about the overconfidence phenomenon before as they are psychology students). Similarly, it is unclear whether the questions were randomly chosen from the Who Wants to be a Millionaire Bumper Quiz book and whether the questions selected to be used in this book were randomly chosen. As discussed above, the ecological models stress the importance of using randomly sampled questions, however, they suggest that this explains the overconfidence effect. As the opposite was the problem in this study, it is unclear whether the low result relates to the question selection. Juslin et al (1995) showed that it was only below 65% accuracy that the overconfidence effect was evident. Therefore, it may have also been useful to examine the actual level of accuracy (e.g. look at the overconfidence effect for only the items with accuracy below 65%).
Looking at the estimated and actual frequencies of correct answers, no significant difference was found between the easy and hard sections. This finding does not support the research-based hypothesis that there would be a difference. This result may have be influenced by the methodological issues mentioned above. In addition the low sample size may have been a factor that influenced the results.
As a strong overconfidence effect was not found it is difficult to know which explanation for the overconfidence effect best suits the results. If the participants knew about the overconfidence effect prior to completing the questionnaire (which is possible given they were in a psychology class), they may have been consciously trying to reduce the result. Consequently, we may have seen the low result that we did. This justification would lend itself well to the cognitive or information-processing explanation of the overconfidence effect.This explanation suits the findings more closely than ecological models or regression effects.

Order Now. It takes less than 2 minutes.

  1.  
  2.  
  1.  

Please note: The above dissertation snippet was written by a student and then submitted to us to display and help others. Thanks to all the students who have submitted their work to us.