McConney, A., Oliver, M. C., Woods-McConney, A., Schibeci, R., & Maor, D. (2014). Inquiry, engagement, and literacy in science: A retrospective, cross-national analysis using PISA 2006. Science Education, 98(6), 963–980. doi:10.1002/sce.21135
Current science education policy posits that scientific practices of inquiry are the best context for student science learning. McConney and colleagues analyzed data from the Programme for International Student Assessment (PISA), an international assessment of high school science students, to test the relationship between inquiry-based instruction and student outcomes in (a) science literacy, (b) interest in science, and (c) engagement in science. Their analysis showed that while students in high-inquiry classrooms reported higher than average interest and engagement in science, they had lower than average scientific literacy scores. The opposite was also true: students with lower levels of scientific inquiry instruction scored higher than average on science literacy measure and lower on measures of interest and engagement.
The paper provides a succinct and comprehensive overview of different approaches to inquiry-based instruction, ranging from confirmation to structured, guided, and open. The authors cite the literature demonstrating positive correlations between inquiry-based instruction and science literacy, noting that the strongest positive effects have often been associated with guided inquiry, with highest effects when inquiry was structured so that students were engaged in examining and evaluating the quality of evidence and then developing explanations for phenomena. They define confirmation inquiry as investigations that reinforce previously introduced ideas. In structured inquiry, students are required to generate explanations. Guided inquiry has students developing their own research methods to address a research question posed by the teacher. Open inquiry is designed to have students replicate practices of scientists in which they develop their own questions, design and carry out investigations and communicate their results. The authors also cite recent studies that have produced contrary results, where student scientific performance was negatively correlated with student-centered instruction.
In their analysis of PISA data, the researchers use a definition of inquiry that correlates most closely with open inquiry: “practices in which students may be responsible for naming the scientific questions under investigation, designing investigations to research their questions and interpreting findings.”
The researchers investigated the following questions using PISA data from 15-year-old students in Australia, Canada, and New Zealand:
- To what extent do students report high levels of inquiry-oriented learning activities in their science classes?
- What levels of science literacy, interest in science, and engagement in science correlate with students who report high levels of classroom inquiry versus those who report low levels?
New Zealand, Canada, and Australia were chosen for the analysis because, the authors state, they have consistently performed well on the PISA and are similar in their sociocultural histories and traditions. The data set included 1,252 schools and 41,685 students across the three countries.
Six PISA items related to the frequency of instructional strategies were identified as collectively representing evidence of whether students had experienced high or low inquiry-based instruction. These indicators included how frequently students were required to: (1) explain ideas, (2) design how a science question could be investigated in a laboratory, (3) choose an investigation topic, (4) design their own experiment, (5) conduct an investigation to test out their own ideas, and (6) draw conclusions. The researchers used responses to these six items to form a low-inquiry group and high-inquiry group for each country. In all three countries, the low-inquiry and high-inquiry groups were about the same size. The authors note that they could not assess the quality of these instructional strategies, just their existence.
For student engagement in science, researchers used six PISA variables including general science interest, enjoyment of science, personal and general valuing of science, self-efficacy in science, and science self-concept. Science literacy and interest in science were represented by students’ scores on the PISA science items.
In all three countries, students who reported having experienced high levels of inquiry-oriented instruction scored, on average, below their country’s average on the PISA test of scientific literacy. However, they had above-average levels of interest in science and of engagement in science. The contrary was also true: students who had experienced low levels of inquiry-oriented instruction had higher-than-average scientific literacy scores, but lower than average interest and engagement in science.
Implications for Practice
The results of this study not only run counter to common understandings of the power of inquiry-based instruction within the informal science education (ISE) field, but also challenge current science education policies including the recommendations of PISA itself, which advocates for inquiry-based instruction. Rather than suggesting that we discount or uncritically embrace the study results, the authors of the paper suggest a need for closer consideration of why such unexpected results were obtained. They posit two possibilities: The first is that teachers are not adequately supported to provide high quality inquiry-based instruction in ways that would produce improved science literacy. If this is correct, the study suggests a need for further focus on teacher development. But it still begs the question as to why previous meta-studies did not produce the same result. A second possible reason, they suggest, could be that the PISA questions are not well aligned (not sensitive) to positive effects of inquiry-based instruction. The study thus is valuable at least as much for the questions it provokes as it is for the findings it describes. Questions may be one of the more productive ways in which research and practice can relate, with each exploring, in different ways, possible answers to these questions.
This paper is extremely interesting for practitioners. In addition to the above implications for practice, it provides a good review of the literature of inquiry-based science studies, as well as a good reminder to ISE practitioners about the ongoing need to attend to the quality of inquiry-based instruction.
McConney and colleagues’ research also sheds light on the complexity of the relationship research and practice. Research-based knowledge is not produced by one study, but by an accumulating body of evidence that, over the long run and in different settings, produces consensus, if tentative, understandings.