Journal Title

Journal ISSN

Volume Title



Global warming is an important scientific issue. As the result of anthropogenic carbon emissions, global warming is already affecting life on this planet and will likely have even more significant impacts on economies, species, and human life in future decades.1 As potential solutions to global warming require economy-wide measures, global warming is also a prominent issue discussed in politics and the media. While there is not much debate regarding the fundamental science of global warming and the greenhouse effect among active, publishing researchers in the field,2 scientific misconceptions abound across broad age ranges and education levels, including politicians and journalists. These misconceptions give rise to a great deal of debate over issues such as the greenhouse effect that have been viewed as settled by the scientific community for decades and sometimes even centuries.3-17 The present science education research was designed with the specific goal of investigating the potential for increasing scientific understanding of a fundamental yet relatively basic physical phenomenon underlying global warming, namely the greenhouse effect. This research lies within the category of science learning in general, and conceptual change research in particular, as this is an area where significant misconceptions are documented to exist.4, 7, 11, 18 For example, a common misconception is that global warming is due to a hole in the ozone layer.4, 19 The Conceptual Change Hypothesis (CCH) holds that misconceptions interfere with the learning of correct concepts and that misconceptions are resistant to the traditional mode of correct instruction alone. According to the CCH, special instruction that effectively addresses learner misconceptions should be added. In the present study, this specialized form of instructional intervention will be called Misconceptions Instruction. The present study examines the relative effects of Traditional Instruction (TI) and Misconceptions Instruction (MI) on the learning of fundamental physics concepts related to the greenhouse effect. In the present study, approximately 200 students in intact course sections of a first-year Atmospheric Sciences course at Texas Tech University were quasi-randomly assigned to one of two instructional treatments: Traditional Instruction or Misconceptions Instruction. Both treatment groups received a pretest, an immediate posttest following treatment, and a delayed posttest approximately 14 days after the treatment. Raw scores were calculated, and mean scores for both groups were subjected to a repeated-measures ANOVA to determine if participants in each group improved over time and if there was a statistically significant difference in their assessment performances. The results of the repeated measures ANOVA reveal no improvement over time and no statistically significant difference between groups for performance on concept questions and on the assessment as a whole, although there was a statistically significant difference between groups for performance on the misconception questions only.
The results carry significant implications for the quantity and type of instruction utilized in this study, as well as implications about the nature of the most common misconceptions about the greenhouse effect and global warming. For example, short reading passages about the greenhouse effect were found inadequate for bringing about conceptual change, in contrast with other studies that had longer exposure times and varied instructional methods.11, 20 Also, the nature of the most common misconceptions concerning the greenhouse effect was found to be categorical, meaning that students used the entirely wrong mental model as opposed to isolated misconceptions about the correct mental model, and thus refutational instructional methods were ineffective. These implications are discussed in light of previous findings related to science education research in general and to research on misconceptions-based instruction in specific. Appendix G: Global circulation models (GCMs) capture the large-scale trends of climate variables, but GCMs operate on a grid size that cannot account for local features which modify the climate and thus the global trend in a region. While the climate is warming globally, many effects are felt locally. Therefore, the GCM output needs to be downscaled to provide regional climate scenarios, and one approach for doing this is empirical statistical downscaling (ESD). There are a few different software packages and methods for performing ESD, one of which is SDSM 4.2, which computes the parameters of multiple linear regression optimized using ordinary least squares or dual simplex algorithms. In this study, the Statistical Downscaling Model software package (SDSM) was evaluated using data from 20 different weather stations and three different atmosphere-ocean general circulation models (AOGCMs) for two variables in order to determine SDSMs ability to meet modern downscaling needs. It was found that as a software package, SDSM was not suitable for projects involving the need to perform many runs of the system. With regard to performance, SDSM was found to perform worse than two of the newer methods against which it was compared.