| dc.description.abstract | he authors used a 59-item survey to probe the understanding of climate change by 89 Ontario preservice teachers. The study investigated the usefulness of comparing real survey data from closed, binary choice items, with randomly generated data. Climate change was chosen to be the topic because it is a new emphasis in K?12 science curricula. If teachers had answered the survey randomly, according to Monte Carlo simulations, a normal distribution would result, with 56 of the 59 items answered correctly by 40%?60% of the respondents. A bimodal distribution resulted, however, with 34 items answered correctly by more than 60% and 18 items by less than 40%. Apparently, the teachers knew a lot about climate change, but also had many misconceptions, some identified here for the first time. Item discrimination indices and correlation coefficients, however, were the same for the real versus Monte Carlo data, suggesting that preservice teachers? knowledge was a ?kaleidoscope of understanding,? rather than a coherent picture. This may be because their understanding of climate change came primarily from unconnected sources in the media, or because climate change science involves many different fields of study including astronomy, biology, chemistry, ecology, oceanography, and physics. In conclusion, the analysis herein demonstrates the benefit of comparing real and random data for binary-choice item surveys in multidiscipline topics such as climate change. For those interested in climate change education, these results suggest the importance of emphasizing the difference between reliable and unreliable sources of information and giving careful attention to how to draw on concepts from different scientific fields. | |
