The Experimenter Expectancy Effect: An Inevitable Component of School Science?

Abstract

A medium-scale quantitative study (n = 90) found that 10–11-year-old pupils dealt with theory and evidence in notably different ways, depending on how the same science practical task was delivered. Under the auspices of a 2×2 part-randomised and part-quasi experimental design, pupils were asked to complete a brief, apparently simple task involving scientific measurement. One half of the sample carried out the task in a naturalistic whole class context; the other half worked as lone experimenters in solitary conditions where accuracy of measurement was promoted. In the whole class setting pupils exposed to an illustrative lesson displayed behaviour indicative of experimenter expectancy, tending to differentiate theory and evidence to a lesser degree than pupils who experienced an enquiry lesson. In addition, during the illustrative lesson many of the pupils were biased towards their theories in ways that lay well beyond those intended by the research design. On the other hand, in the solitary setting pupils performed equally well with both illustrative and enquiry treatments. Implications are discussed in the light of the problems of excessive pupil theory/data-ladenness. Finally, the advantages and disadvantages of exposing young learners to more authentic versions of professional science are considered.