Читать книгу Theories in Social Psychology - Группа авторов - Страница 24

From its genesis, psychological reactance was conceptualized as arising from situational factors, and research and measurements were consistent with psychological reactance as a state phenomenon. In attempting to study reactance, researchers created experimental freedom-threat situations to arouse psychological reactance. In many of the studies discussed, reactance was inferred by comparing responses across conditions experimentally created (e.g., reactance vs. non-reactance), whether those measures were on attitudes, evaluation of message, or rating of communicator (Wright et al., 2004).

Sharon Brehm (1976) wrote on the implications of reactance theory to that area of clinical psychology concerned with changing behavior and non-successful treatments.

However, she referred to reactance theory as a motivational state that is aroused from situational factors. The construct of reactance was later modified to accommodate a clinician’s environment, with reactance being conceptualized as a personality trait.

A major concern of the clinician was the interaction of this trait (reactant personality) with treatment (Beutler et al., 1991; Shoham-Salomon & Hannah, 1991; Snow, 1991). The genesis toward a psychometric scale emerged from attempts to measure reactance as a disposition (Buboltz et al., 2002; Hong & Page, 1989; Merz, 1983). Reactance is most likely to occur in the clinical setting to persons with volitional/free problem behaviors (e.g., drinking, smoking) as compared to uncontrollable behaviors (e.g., panic, obsessive-compulsive) (Shoham et al., 2004). Psychological reactance as a disposition has been measured by several scales and techniques.

Merz’s (1983) Questionnaire Measuring Psychological Reactance (QMPR) was the first-ever scale designed to capture a measure of psychological reactance and consisted of 26 items utilizing a four-point Likert scale ranging from “not at all appropriate” to “extremely appropriate.” The original scale attempted to obtain measures on resistance, defiance, boomerang effect and the inclination to do the opposite to what others expect of you (Tucker & Byers, 1987). The scale was later modified by Tucker and Byers (1987) to an 18-item reactance scale utilizing a five-point Likert range from 1 = “not at all appropriate” to 5 = “extremely appropriate.” Two major constructs in the scale were behavioral freedom (encompassing measures of reactions to praise, subservient behavior and expectations, and advice of others) and freedom of choice (encompassing measures of freedom and decision making). The QMPR has been found to have convergence validity with the Personality Research Form (PRF), suggesting that psychologically reactant individuals are more likely to be defensive, easily offended, aggressive, autonomous, and impulsive (Dowd & Wallbrown, 1993). These characteristics are consistent with the QMPR’s reactant individual who is seen as dominant, individualistic, with low social ties, and who creates unfavorable impressions on others (Dowd, 1993).

Hong and Ostini (1989), utilizing a four-point Likert format of the translated QMPR, found factorial instability and low correlations among factors. Hong and Page (1989) argue that Tucker and Byers’s QMPR was considered to be psychometrically unstable. Other criticisms came from Donnell et al. (2001), who also questioned Tucker and Byers’s modification of the original scale, noting that the instability of the scale may have also been a result of both the translation of the original scale from German and the adaption of a five-point scale format.

The stability of the QMPR translated by E. T. Dowd was further tested by Donnell et al. (2001), utilizing a six-point Likert-type scale ranging from 1 = “does not apply at all” to 6 = “always applies.” The results of a three-factor scale structure were found to be inconsistent with both Tucker and Byers (1987) and Hong and Ostini (1989). Though reactance appeared to be a multidimensional construct, the QMPR was not accepted as a stable measure of reactance (Donnell et al., 2001).

Hong and Page (1989) 14-item psychological reactance scale attempted to create a factorial stable measure of psychological reactance and emerged as a critique of the QMPR. Hong’s Psychological Reactance Scale (HPRS) utilizes a five-point Likert scale (1 = “disagree completely” to 5 = “agree completely”) and has four distinctive factors: freedom of choice, conformity reactance, behavioral freedom, reactance to advice, and recommendation. Hong (1992), a replication of Hong and Page (1989), supported the factorial structure of the HPRS. The HPRS was later revised (Hong & Faedda, 1996 ) to an 11-item scale. Three items of the original scale were considered to be vague and removed. Shen and Dillard (2005) argue that a single score on the HPRS to measure trait psychological reactance is theoretically and empirically justifiable. However, Jonason and Knowles (2006) and Jonason (2007) indicate that while the HPRS tends toward unidimensionality, it is a moderately stable and a reliable measure of reactance and improvement of the scale is needed. Jonason, Bryan and Herrera (2010), utilizing confirmatory factor analysis, argue for a one-factor 10-item Hong’s Psychological Reactance Scale. For critiques on the HPRS see Rosenberg and Siegel (2018) and Jonason and Knowles (2006).

Another method utilized in the measuring of reactance is the Therapeutic Reactance Scale (TRS). This scale uses a 28-item self-report questionnaire with two subscales (i.e., verbal reactance and psychological reactance) in attempting to measure reactance as a trait. The TRS has recorded convergence validity with personality characteristics of dominance, locus of control (internal), and aggression (Dowd & Wallbrown, 1993; Shoham et al., 2004). A major problem identified with this measurement of reactance is the lack of construct validity. Shoham et al. (2004) argue that what has been conceptually identified as a reactant disposition may not necessarily be a disposition or trait but rather an oppositional style consistent with particular behavioral styles. Inman et al. (2019) added to the understanding of the factor structure of the TRS. Noting that Dowd et al. (1991) TRS identified two factors (i.e., behavioral and verbal) and Buboltz et al. (2002) identified four factors in the TRS (i.e., resentment of authority, avoidance of conflict, susceptibility to influence, and preservation of freedom), their 2019 findings suggest a four-factor scale structure similar to Buboltz et al. (2002) but with the following structures: resentment of authority, conflict seeking, susceptibility to influence and preservation of freedom.

In another attempt to measure reactance, Beutler et al. (1991) utilized two indices from the Minnesota Multiphasic Personality Inventory (MMPI) in developing what they refer to as a reactant score. The two scales that they adopted were Taylor’s Manifest Anxiety Scale and the Edwards Social Desirability Scale. The authors utilized an arithmetic summation of the scores from both scales to derive a reactance score. Critics of Beutler et al. (1991) noted that there is no rationale articulated as to why high anxiety and social desirability should correlate with reactance (Lukin et al., 1985; Shoham et al., 2004). In fact, Spielberg’s State-Trait Anxiety Inventory and the TRS have no statistical relationship. Shoham et al. (2004) reported that Dowd et al. (1994) showed a weak correlation between the TRS and social interaction. A question that can be asked is: Are Beutler et al.’s (1991) measure and the TRS measuring the same construct? Baker et al. (2003) found that there was no correlation between these two reactance measures.

Shoham et al. (1996) utilize a naturalistic method to measure expressed reactance. This method involves the rating of participants’ content-filtered voice in reply to a question that provokes reactance. High-reactant persons’ voice tones sound inhibited, spiteful, and active.

Dillard and Shen (2005) utilize a composite of self-report indices of anger and negative cognitions to measure psychological reactance. This intertwined model integrates the cognitive and affect in understanding reactance and was one of the first meaningful attempts to measure state reactance. Reactance researchers (e.g., Quick et al., 2015; Clayton et al., 2019; Kim et al., 2013; Rains & Turner, 2007; Scott & Quick, 2012; Xu, 2015) have utilized Dillard and Shen (2005) composite index to measure psychological reactance in varying contexts. Their scale is one of the most used reactance scales. Lindsey (2005) constructed a scale to understand the influence of guilt in bone-marrow donation. However, as Quick (2012) notes, her scale has been utilized by reactance researchers. Lindsey’s scale was informed by Brehm (1966) and Hong and Faedda (1996) and has four items modified from Hong and Faedda (1996) to measure state reactance instead of trait reactance. Her scale has been utilized in the literature (see LaVail et al., 2010; Reinhart & Anker, 2012). Quick’s (2012) article provides a comparison between Lindsey’s (2005) and Dillard and Shen (2005) scales. In an attempt to create a scale to measure state reactance Sittenthaler et al. (2015) undertook three studies to develop the Salzburger State Reactance scale (SSR). They identified three subscales in the SSR, namely experience of reactance, aggressive behavioral intentions, and negative attitudes. The scale has nineteen original items with responses to each item varying from “not at all” to “very much” on a five-point Likert scale.

The Crowell Lowery Multicultural Training Reactance Scale (CL-MTRS) is a relatively new scale attempting to measure students’ psychological reactance in a specific context (see Lowery et al., 2020). The scale utilizes 22-items to measure psychological reactance toward multicultural training experience. Cognitive, behavioral, and affective evaluations of the training experience, including course topics, facilitator, and processes, are measured (see Kelly et al., 2019).