Using Assessment to Evaluate and Improve Inquiry-Based Geoenvironmental Science Activities: Case Study of a Middle School Watershed E. coli Investigation

Journal of Geoscience Education, Mar 2005 by Patterson, Lauren A, Harbor, Jon

Students in school 8A and school 8B, with two additional years of mathematics, were generally able to construct the graph during both the pretest / posttest. Interestingly, me majority of eighth grade students chose to display the data using bar instead of line graphs. It should be noted that while the bar graph is correct, it is best suited to categorical data and is not the best choice to illustrate trends in continuous data, and thus, bar graphs were given a score of 3 out of 4 in the rubric. Furthermore, the hands-on activity encouraged and demonstrated the use of line graphs to display data trends, and a change could not have been detected from pretest to posttest unless the scoring system differentiated oetween line graphs (rated as 4) and bar graphs (rated as 3) (Table 3). The overall level of change from pretest to posttest was not statistically significant (P0 = 0.4484 and 0.2110 for school 8A and school 8B respectively (Table 8)); therefore, this question could be altered in the future to appropriately challenge the knowledge level of eighth grade students.

Question 4: What trends are seen in the above graph, and what do they mean? - The final question in the pretest / posttest was an open-ended question asking students to explain the trends observed in the graph created by question 3 (Table 7). Many students were able to create a graph; however, in the pretest the majority of the students did not know what the graph represented. Most students chose not to respond to the question or to simply state "the upstream line increased, then decreased through time while the downstream line increased".

School 6 students, with less experience in graphs (only 38.5% of the sixth graders had a correct graph in the pretest), were not able to appropriately address this question since their response is contingent on the graph drawn in the previous question. Those students that did draw a correct graph had difficulty grasping the concept of the upstream line always being less than the downstream line. However, after the presentation and the activity, there was a statistically significant improvement in both graphing ability and understanding the meaning behind the graphs (Table 3, Table 8).

School 8A and school 8B students were generally able to state that the upstream line was always less than the downstream line in both the pretest and posttest. The students were able to describe the trend of the lines, but not the meaning of the trends (Table 3). Research indicates that students need to learn how to analyze the significance of data and to practice gathering evidence supporting their findings (Von Seeker and Lissitz, 1999). For example, most students were unable to state that the difference between upstream and downstream E. coli counts indicates the presence of a source of contamination between the two sampling sites. The general trends observed from the responses in the posttest displays a significant improvement in student understanding of the significance of the graph (Table 3). There was also a large improvement from pretest to posttest (P =