Ambivalence about Social Welfare: An Evaluation of Measurement Approaches

Abstract

<jats:p>Research across disciplines, including political science, has embraced the idea that individuals often possess ambivalent attitudes, but there is considerable disagreement about how to measure this phenomenon. Determining an effective way of capturing such phenomena is important to our under-standing of politics and public opinion. The literature offers several meta-attitudinal and operative measures of ambivalence. I discuss strengths and weaknesses of each of these approaches and con-duct a test of the relative construct validity of two meta-attitudinal and two operative measures of social welfare ambivalence using data from a statewide survey of Florida residents in 2004. The findings suggest that one of the operative measures that forces respondents to rate their positive and negative feelings separately performs better than any of the other approaches currently offered.</jats:p>

Intergenerational ambivalence: new perspectives on intergenerational relationships in the German welfare state

“…The items measuring positive and negative attitudes are contradictory and refer to the same attitude object, namely the elder generation in society ( see also Gainous 2008 a ; Kaplan 1972; Suitor, Gilligan and Pillemer 2011). Thus, the conditions for ambivalence are theoretically given.…”

Digital political information consumption and ambivalent political attitudes.

“…In addition to studying ambivalence as a psychological phenomenon, many researchers have been interested in the study of its measurement (Albertson et al 2005;Alvarez and Brehm 1995;Bassili 1996;Gainous 2008;Holbrook and Krosnick 2005;Kaplan 1972;Thornton 2011). The measurement of ambivalence can be separated into two distinct categories: meta-attitudinal measurements and operative measurements (Bassili 1996;Gainous 2008;Holbrook and Krosnick 2005). In the most basic sense, metaattitudinal measures ask people to evaluate their ambivalence while operative measures attempt to gauge it without subjects' knowledge that ambivalence is being assessed.…”

“…Again, this measurement strategy requires respondents to consider and evaluate their ambivalence. One way this is accomplished is by using a Likert scale with simple statements about the presence of conflicting attitudes (Gainous 2008). Another standard method for collecting meta-attitudinal data on ambivalence is using less direct questions like the one used by Mulligan (2002):…”

“…Generally, this data is collected using traditional survey methods, some studies, however, include measures of response latency (Bassili 1996;Gainous 2008)a more nontraditional survey approachas well as measures that rely on statistical modeling (assessing heteroscedasticity). For those relying on more traditional survey indicators, there are three original formulae worth noting: the Kaplan measure, the Katz and Hass measure, and the Griffin measure.…”

“…Several studies have been conducted measuring the comparative validity of ambivalence measurement strategies (Bassili 1996;Gainous 2008;Holbrook and Krosnick 2005;Thompson et al 1995). In addition to comparing the two different categories of measurement strategies, meta-attitudinal and operative, Gainous (2008), Thompson (1995, and to a lesser degree Bassili (1996) Based on the factor analyses, Bassili continues by creating separate indices for the two measurement classes and running logistical regression models for a dichotomous variable of attitude pliability for all issues. In the model comprised of both indices, the metaattitudinal index failed to reach statistical significance, whereas the operative index was statistically significant in all three cases.…”