On recent National Assessment of Education Progress exams, in addition to the traditional pen-and-paper test sheets, students were also asked to complete several hands-on and interactive computer-based tasks. The tasks allowed researchers to judge not just the level of science knowledge attained by students, but how well they were able to absorb and apply this knowledge while solving complex reasoning problems. The activities included tasks that mimicked real science experiments, thus giving scientists insight into how kids tackle similar problems in the real-world environment.

The tasks included 40-minute experiments including materials and laboratory equipment. Students were asked to design, and carry out the experiments, and then record and interpret their results. The computer-based tasks involved solving of 20- to 40-minute complex scientific problems that required reasoning through multiple stages in order to arrive at a solution, which frequently involved multiple scientific branches.

In all, almost all students were equally successful in tasks that required investigating limited data sets and only straight-forward observation. Eighty percent of fourth-graders, 84% of 8th-graders and 75% of 12th graders completed the tasks that provided a simple list of instructions, and asked them to record a small set of observations, without drawing any kind of conclusions. The students didn’t do nearly as well on tasks that required collating knowledge, combing it with observable phenomenon to solve a particular problem. Thirty-five percent of fourth-graders and less than a quarter of 8th- and 12th-graders completed the problem that provided limited guidance and asked them to analyze a larger set of variables.

Key Discovery 3: The percentage of students who could select correct conclusions from an investigation was higher than for those students who could select correct conclusions and also explain their results.

On the portion of the exam that required a student to select the correct answer from multiple options, but then subsequently explain their reasoning, over 70% of students were able to choose the correct answer, but less than a third were then able to justify their selection. The discrepancy was particularly stark among 4th graders, were 71% identified the correct answer, but only 15% were able to explain their reasoning in obtaining it.

Female students in all three grades scored higher than males on the hands-on tasks, though males scored higher on the traditional paper-and-pencil science assessment. There was no gender gap in interactive computer tasks.

There was an achievement gap at grades 4 and 8 between students from higher- and lower-income families in both the hands-on tasks and the interactive computer tasks.