6 Testing Empowered Involvement (S. 79-80)

6.1 Introduction and Overview

6.1.1 A Deductive Approach

Our previous considerations have equipped us with a theoretical model that can help explain and predict the behaviour of consumers in an Empowered Involvement setting. First, we analysed the connection between involvement and word of mouth based on existing theoretical and empirical findings. Next, we conceptualised a particular type of involvement which we call Empowered Involvement, and which results from the combination of its four formative dimensions.

We are now in a position that allows us to deduce conclusions concerning the relationship between Empowered Involvement and word of mouth (Chalmers, 2001; Popper, 1966). In doing so, we develop a set of hypotheses, i.e., a number of general statements that describe our assumptions about the relationship between Empowered Involvement, its components, and word of mouth. The deductive approach (Popper, 1966) is based on the assumption that our theoretical considerations can be maintained and considered to be valid if we can observe the theoretically assumed relationships between the different components of our model in an empirical test – albeit, only as long as no new test falsifies the hypotheses, or until a new theory is developed which better explains the phenomena observed (Riesenhuber, 2006, pp. 2–4).

6.1.2 Two Stages of Empirical Analysis

The empirical part of our research consists of two stages. In the first stage, we set out to validate the general direction of our research, and more particularly, of our overall research hypothesis – as mentioned above – that an Empowered Involvement approach is likely to stimulate word-of-mouth communication among consumers. This first stage is more descriptive in nature, and less explanatory. In it, we tested whether a very simple form of Empowered Involvement – voting about ads – already leads to the hypothesised word of mouth effects.

On the following pages, we will describe the experiment design, the sample selection, and the results. Once the appropriateness of our direction was confirmed, we moved on to the second stage, based on a more complex conceptualisation, including a measurement of EmI, its antecedents and consequences. This more elaborate test was organised in co-operation with a German word-of-mouth marketing panel, and conducted within a campaign that introduced a new corporate blog.

The second part of this chapter is dedicated to detailing all aspects of this test. It describes the considerations that led to the experiment design, explains the steps taken for the measurement of the different constructs (with a particular focus on the differentiation between formative and reflective specifications), and, finally, presents and discusses the results obtained. 6.2 First Preliminary Research 6.2.1 Introduction In 2006, we conducted an early quasi-experiment with chewing gum maker Wrigley.

The primary objective was to assess whether or not an approach designed to stimulate the four factors thought to contribute to an experience of empowerment, would indeed stimulate word of mouth activity. Chewing gum was found to be suitable for a number of reasons. We follow Zaichkowsky (1985, p. 346), who, based on a literature review, suggested that low-involvement products are characterised by a lack of information-seeking about brands, little comparison between product attributes, perception of similarity among different brands, and no special preference for a particular brand.

In light of this definition, we consider chewing gum to be a low-involvement product category. (Also, in one of Zaichkowsky’s empirical analyses, instant coffee was shown to be a low-involvement product category, as was yoghurt, in Laurent & Kapferer [1985]. We assume that chewing gum, in terms of the level of involvement that consumers experience with it as a product category, can be considered to be similar to these two product categories.)