Reward, intrinsic interest, and creativity: New findings.

Abstract

These null results were then presented in Eisenberger and Cameron's (1996) Figure 1, contributing to the illusion of a "normal distribution" of effect sizes, and were later entered into the analyses underlying Figure 2, diluting the mean effect sizes reported there. Note that had the identical six comparisons of reward versus no-reward conditions come from six separate experiments rather than from three factorial experiments, their contributions to Figure 1, and their effects on later analyses, would have been very different.Because the use of such procedures is sometimes justified simply as the result of a particular interest in "main effect" conclusions, it may be informative to consider the application of an analogous procedure to a literature in which one might care deeply about the results. Imagine, for example, a review of the literature on the side effects of a new drug, in which (a) only a small number of studies permitted separate evaluations of the effects of the drug under different conditions (e.g., with men vs. women or with large doses vs. small doses), but in which (b) all of those studies had found significant positive effects under one condition and significant negative effects under another condition (e.g., increases in strokes for men but decreases in strokes for women, or vice versa). Readers can decide for themselves how much theoretical value or practical significance they would attach to the conclusions of a review that reported these studies as having tested and as having shown nothing more than an "overall" lack of difference between drug and placebo conditions.These examples are not an isolated case--far from it. Three quarters of the studies included in Cameron and Pierce's (I 994) review involved effect-size estimates obtained only after averaging across conditions designed to yield differential results. In dozens of additional cases, for example, statistically significant effects within experimental groups were averaged with null effects obtained within relevant control groups.Moreover, even after the inclusion of these many small effect-size estimates, created by averaging across theoretically and empirically distinct conditions, Eisenberger and Cameron's (1996) own analyses revealed that their procedure remained inappropriate. For instance, Q statistics, directly testing whether the effect-size estimates within the different reward categories displayed in Figure 2 came from homogenous populations that would justify statistical analyses, revealed that nearly 90% of reported categories involved significantly heterogeneous populations. Even after a principled procedure for eliminating "outliers" was applied, over 60% of these tests for heterogeneity remained significant. Eventually, Cameron and Pierce

The Impact of Social Interaction and Communications on Innovation in the Architectural Design Studio

“…[18] suggests that 'you can have creativity without innovation, but you cannot have innovation without creativity'. [19] examines the work of a number of researchers such as [20][21][22][23], and points out that there was no definite consensus regarding how creativity is defined. He discovers that the creative process looks different to different researchers.…”

A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation.

“…Thus, both studies yielded strong results that were exactly opposite of what the Eisenberger and Cameron model predicts. Eisenberger and Cameron (1998) stated that they had done another meta-analysis in which they compared performancecontingent rewards that had specific performance standards (e.g., rewards for surpassing the 80th percentile) with comparablepositive-feedback control groups, indicating that there were 4 such studies with the free-choice measure and 7 with the self-report measure and reporting that in both cases, these performancecontingent rewards significantly enhanced intrinsic motivation. They did not provide any methodological details, so it is difficult to know what comparisons they made, but there are two important points to keep in mind.…”

Towards a unified theory of task-specific motivation

“…Numerous studies have been carried out on the effect of extrinsic rewards on intrinsic motivation. And although the issue was sometimes fiercely debated (Eisenberger & Cameron, 1998;Hennessey & Amabile, 1998;Lepper, 1998;Sansone & Harackiewicz, 1998), from the available meta-analyses (Cameron & Pierce, 1994;Deci, Koestner, & Ryan, 1999;Deci et al, 2001;Rummel & Feinberg, 1988;Tang & Hall, 1995;Wiersma, 1992) it can be concluded that extrinsic rewards, when they are perceived to have a controlling character, can diminish (intrinsic) motivation (Ryan & Deci, 2000c). Extrinsic motivation was contrasted with intrinsic motivation, which capitalizes on the natural curiosity people possess.…”