The "Positivity Bias" in Evaluations of Public Figures: Evidence Against Instrument Artifacts

Abstract

the past 40 years of large-scale surveys, Sears (1976) has reached an unequivocal conclusion: despite wars, depressions, and public scandals, despite the growing distrust of and cynicism toward government leaders, the public has remained quite positive in their evaluations of those leaders. Fully 76 percent of all public figures were evaluated positively by respondents to Gallup polls between 1935 and 1975. Even at the height of the Watergate scandals only the disgraced Nixon, Dean, Mitchell, and Ehrlichman of 18 leaders were evaluated negatively by respondents to Michigan's 1974 CPS election study. Similarly, only 2 of 13 elected officials were evaluated negatively by blacks during the Watts riots (Sears, 1969). This pattern of positivity certainly contrasts with the image of public cynicism about "politicians" (Mitchell, 1959), and with the currently low level of public trust in government (Miller, 1974;Miller and Miller, 1977). Likewise the practical implications of a Abstract The "positivity bias" is a term used to describe the consistent favorable evaluation of public figures found in surveys over the past 40 years. This paper explored several possible artifactual explanations for this bias, focusing on the survey instrument itself. Two experiments varied the labeling and ordering of scale endpoints, the affective value of the initial context evaluated, and the presence or absence of a prestigious job title associated with the name of the public figure. None of the variations produced significantly different levels of positivity than the standard control condition used in each experiment.

Citizens' Orientations Toward Legislatures: Congress and the State Legislature

“…Such a bias may be present in responses to questions about the performance of political authorities and institutions. See Lau, Sears, and Centers 1979 for an analysis of the extent to which positivity bias may derive from the research instrument. nation of variations in constituents' support for the legislature.…”

Attitude and Attitude Change

“…The balance effect, Heider (1946) argued, was attributable to gestaltic forces toward cognitive consistency. For instance, the attraction effect might be a special case of the positivity bias, which is the tendency fo r people to remember, prefer, and expect positive rather than negative relations between elements, particularly be tween individuals, as exemplifi ed by the high regard people express for public fi gures (Lau, Sears & Centers 1979;Rook et aI 1978). The attraction and agreement effects, on the other hand, might be explained in terms of alternative biases.…”

Differential use of the CAHPS® 0–10 global rating scale by medicaid and commercial populations

“…As a result of this positivity bias, a onepoint difference at the lower end of the scale may be less meaningful than a one-point difference at the higher end of the scale, as it corresponds to a much smaller percentile difference in experience [4,5,6]. A related consideration is response contraction bias, or the tendency for survey respondents to avoid using extreme ends of a scale [7].…”