Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Using experimental methods to assess the persuasiveness of corporations' trust-repair discourse strategies

Fuoli, Matteo LU (2016) Sixth International Conference of Critical Approaches to Discourse Analysis across Disciplines (CADAAD)
Abstract
Persuasion and manipulation have been central concerns of Critical Discourse Analysis since its beginnings (e.g. Hart 2014; Van Dijk 1996). Traditionally, the persuasive and manipulative potential of texts has been explored by means of interpretive analysis. However, the actual perlocutionary effects of the strategies identified through this process are rarely (if ever) evaluated empirically. This paper aims to show that experimental techniques can usefully complement traditional discourse analysis methods by allowing researchers to test hypotheses about the persuasiveness of discursive strategies, and thus gain new and empirically-grounded insights into the dynamics of social influence, power, and ideology.
The results of two... (More)
Persuasion and manipulation have been central concerns of Critical Discourse Analysis since its beginnings (e.g. Hart 2014; Van Dijk 1996). Traditionally, the persuasive and manipulative potential of texts has been explored by means of interpretive analysis. However, the actual perlocutionary effects of the strategies identified through this process are rarely (if ever) evaluated empirically. This paper aims to show that experimental techniques can usefully complement traditional discourse analysis methods by allowing researchers to test hypotheses about the persuasiveness of discursive strategies, and thus gain new and empirically-grounded insights into the dynamics of social influence, power, and ideology.
The results of two experiments are presented. Using Fuoli and Paradis’ (2014) model of trust-repair discourse as a point of departure, the studies aim to assess the persuasiveness of different trust-repair strategies deployed by corporations after being publicly accused of incompetence or wrongdoing. In experiment 1, apology and denial strategies are compared by means of a scenario-based questionnaire. 448 subjects took part in the experiment. Linear regression analyses reveal that denial is more effective than apology in repairing trust after integrity-based violations, i.e. when a company is accused of having intentionally deceived the public, and that individual assumptions about corporations in general significantly affect context-specific assessments.
Experiment 2 uses a forced-choice, within-subjects design to investigate whether simple declarative assertions (e.g. ‘this will not happen again’) are perceived as more credible than epistemically modalized statements (e.g. ‘we believe that this will not happen again’), and in what circumstances. 29 participants took part in the experiment. The results, which were analyzed using mixed-effects logistic regression analysis, indicate that the type of trust violation as well as the content of the statement significantly affect participants’ choices. Unmodalized statements are preferred when the company’s ability is questioned and the statement is about the future (cf. example above). Conversely, when the company’s integrity is at stake, assertive statements are preferred when the statement concerns the company’s past behavior (e.g. ‘we have always been honest to our customers’). The results of these experiments shed new light on the discursive dynamics of trust and demonstrate the usefulness of experimental techniques to CDA. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to conference
publication status
published
subject
conference name
Sixth International Conference of Critical Approaches to Discourse Analysis across Disciplines (CADAAD)
conference location
Catania, Italy
conference dates
2016-09-05 - 2016-09-07
language
English
LU publication?
yes
id
5611909e-e31d-4f4b-8773-6aab6d365b2e
date added to LUP
2016-09-16 09:30:07
date last changed
2018-11-21 21:25:52
@misc{5611909e-e31d-4f4b-8773-6aab6d365b2e,
  abstract     = {{Persuasion and manipulation have been central concerns of Critical Discourse Analysis since its beginnings (e.g. Hart 2014; Van Dijk 1996). Traditionally, the persuasive and manipulative potential of texts has been explored by means of interpretive analysis. However, the actual perlocutionary effects of the strategies identified through this process are rarely (if ever) evaluated empirically. This paper aims to show that experimental techniques can usefully complement traditional discourse analysis methods by allowing researchers to test hypotheses about the persuasiveness of discursive strategies, and thus gain new and empirically-grounded insights into the dynamics of social influence, power, and ideology.<br/>The results of two experiments are presented. Using Fuoli and Paradis’ (2014) model of trust-repair discourse as a point of departure, the studies aim to assess the persuasiveness of different trust-repair strategies deployed by corporations after being publicly accused of incompetence or wrongdoing. In experiment 1, apology and denial strategies are compared by means of a scenario-based questionnaire. 448 subjects took part in the experiment. Linear regression analyses reveal that denial is more effective than apology in repairing trust after integrity-based violations, i.e. when a company is accused of having intentionally deceived the public, and that individual assumptions about corporations in general significantly affect context-specific assessments.<br/>Experiment 2 uses a forced-choice, within-subjects design to investigate whether simple declarative assertions (e.g. ‘this will not happen again’) are perceived as more credible than epistemically modalized statements (e.g. ‘we believe that this will not happen again’), and in what circumstances. 29 participants took part in the experiment. The results, which were analyzed using mixed-effects logistic regression analysis, indicate that the type of trust violation as well as the content of the statement significantly affect participants’ choices. Unmodalized statements are preferred when the company’s ability is questioned and the statement is about the future (cf. example above). Conversely, when the company’s integrity is at stake, assertive statements are preferred when the statement concerns the company’s past behavior (e.g. ‘we have always been honest to our customers’). The results of these experiments shed new light on the discursive dynamics of trust and demonstrate the usefulness of experimental techniques to CDA.}},
  author       = {{Fuoli, Matteo}},
  language     = {{eng}},
  title        = {{Using experimental methods to assess the persuasiveness of corporations' trust-repair discourse strategies}},
  year         = {{2016}},
}