model” (p. 337). If no reasons to be critical can be identified, the individual may accept the new evidence or theoretical interpretation.

Some studies suggest that the strength of prior beliefs, as well as the personal relevance of those beliefs, may influence the evaluation of the mental model (Chinn and Malhotra, 2002; Klaczynski, 2000; Klaczynski and Narasimham, 1998). For example, when individuals have reason to disbelieve evidence (e.g., because it is inconsistent with prior belief), they will search harder for flaws in the data (Kunda, 1990). As a result, individuals may not find the evidence compelling enough to reassess their cognitive model. In contrast, beliefs about simple empirical regularities may not be held with such conviction (e.g., the falling speed of heavy versus light objects), making it easier to change a belief in response to evidence.

Evaluating Evidence That Contradicts Prior Beliefs

Anomalous data or evidence refers to results that do not fit with one’s current beliefs. Anomalous data are considered very important by scientists because of their role in theory change, and they have been used by science educators to promote conceptual change. The idea that anomalous evidence promotes conceptual change (in the scientist or the student) rests on a number of assumptions, including that individuals have beliefs or theories about natural or social phenomena, that they are capable of noticing that some evidence is inconsistent with those theories, that such evidence calls into question those theories, and, in some cases, that a belief or theory will be altered or changed in response to the new (anomalous) evidence (Chinn and Brewer, 1998). Chinn and Brewer propose that there are eight possible responses to anomalous data. Individuals can (1) ignore the data; (2) reject the data (e.g., because of methodological error, measurement error, bias); (3) acknowledge uncertainty about the validity of the data; (4) exclude the data as being irrelevant to the current theory; (5) hold the data in abeyance (i.e., withhold a judgment about the relation of the data to the initial theory); (6) reinterpret the data as consistent with the initial theory; (7) accept the data and make peripheral change or minor modification to the theory; or (8) accept the data and change the theory. Examples of all of these responses were found in undergraduates’ responses to data that contradicted theories to explain the mass extinction of dinosaurs and theories about whether dinosaurs were warm-blooded or cold-blooded.

In a series of studies, Chinn and Malhotra (2002) examined how fourth, fifth, and sixth graders responded to experimental data that were inconsistent with their existing beliefs. Experiments from physical science domains were selected in which the outcomes produced either ambiguous or unambiguous data, and for which the findings were counterintuitive for most children. For example, most children assume that a heavy object falls faster



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement