Files in this item



application/pdf8600327.pdf (9MB)Restricted to U of Illinois
(no description provided)PDF


Title:The Influence of Risk and Regret on Marketing Exchanges
Author(s):Stone, Robert Neil
Department / Program:Business Administration
Discipline:Business Administration
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Business Administration, Marketing
Abstract:Risk is conceptualized in this dissertation as subjective expectation of loss, the more certain one is of that potential loss, the more risk associated with a contemplated exchange. Defining risk this way leads, logically, to the development of regret, defined as the expectation of loss from not proceeding with an exchange. The two constructs are not viewed as mirror images and this was substantiated by the data.
Establishing construct uniqueness, construct validity, and construct legitimacy were primary concerns of this research. Uniqueness for the advocated risk was demonstrated by showing it to differ from other "risks" that have appeared in the marketing literature. For example, one comparison was with risk as uncertainty times consequences; another as the sum of uncertainty about several issues. Results indicated that the various "risks" did not correlate. Construct Validity was demonstrated as only the advocated risk empirically behaved, as hypothesized, in its comparisons with other constructs such as attitudes and intentions. Construct legitimacy was shown by presenting risk's and regret's contributions to understanding the choice process. Within the context of the Fishbein Extended Model, risk and regret were hypothesized to influence intentions by influencing attitude. While many hypothesized results were obtained, the model with risk and regret influencing attitudes and intentions was superior to one with these constructs shown as just influencing attitude. Risk is suggested as an important construct whose influence is subsumed within the attitude construct.
Unique to this study was the development of multiple questions for the risk construct, for each of risk's dimensions, and for the regret construct as well. Though the coefficient alphas were generally favorable, analysis using Lisrel showed that discriminant validity for questions within some risk dimensions must be improved.
Three dimensions for risk were hypothesized: a "personal resource" dimension, subsuming the previous finance, performance, and time dimensions; a "psychological dimension", subsuming the psychological and social dimensions; and an independent, physical risk dimension. Significant inter-correlations were found within the three personal resource variables and also within the two psychological variables. Otherwise, similar to other research with risk, physical risk did not prove to be an independent dimension of risk. Further efforts at substantiating the dimensions of risk are necessary.
Issue Date:1985
Description:248 p.
Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1985.
Other Identifier(s):(UMI)AAI8600327
Date Available in IDEALS:2014-12-15
Date Deposited:1985

This item appears in the following Collection(s)

Item Statistics