European Journal of Innovation Management
Morel Department of Marketing & International Management, Saxon University of Applied Sciences, Ensconced, The Netherlands Abstract Purpose – The purpose of this paper is to develop a scale for measuring consumer doubt toward new products. Design/methodology/approach – The scale was developed in several steps. A large pool of items to represent consumer doubt was generated. Experts reviewed the scale items for conciseness and clarity. An exploratory factor analysis to examine the unintentionally, convergent validity, and discriminate validity of each construct was conducted.
The model was then validated using partial least squares modeling. Finally, the scale and its form were validated, and potential response biases assessed. Data from three studies were used. Findings – The results show that by focusing on reasons for deference, rather than acceptance, the scale yields new insight into innovation success and failure. The CD scale is a reliable and valid measurement instrument to assess consumer doubt toward new products.
Research limitations/implications – For researchers, the results show that only considering positive aspects on innovation adoption can lead o only a partial understanding of how innovation diffuses in the market. Practical implications – By overcoming consumer doubt at early stages of innovation launch, companies could overcome problems related to
Need essay sample on "European Journal of Innovation Management"? We will write a custom essay sample specifically for you for only $ 13.90/page
Yet, a potentially more serious issue that would need to be tackled is why innovations fail. This paper focuses on consumer doubt toward new products, I. E. A lack of conviction that a new product will fulfill its promises. Three studies show that the scale of consumer doubt is valid, and it provides new insights into innovation adoption. Keywords Consumer research, New products, Innovation research, Consumer behavior, Measuring instruments Paper type Research paper European Journal of Innovation Management Volvo. 13 No. 3, 2010 up. 272-293 q Emerald Group Publishing Limited DOI 10. 108/14601061011060120 The authors thank the Academy of Finland for financial support and Oscar Person for detailed comments on previous versions of the manuscript. Most studies in the innovation adoption literature have focused on identifying early adopters and opinion leaders have been considered to be the mantra of an innovation’s success due to their ability to convince other adopter groups of an innovation’s benefits (Rogers, 2003), speed up the product’s diffusion (McMahon et al. , 2000), and shaping the innovation’s lifestyle (Parker, 1992).
Recent research has also emphasized the important role of early adopters in collaborative product development, and lead user designs (von Hippie, 2001). The positive effects of early adopters are documented in a number of studies and have a strong foothold in the literature. However, some researchers have raised doubts about the overly positive influence often attributed to early adopters. They caution that although innovators tend to like new products, they can also be highly demanding customers and require that products development solutions contain the requirements they would like to see in it (Moore, 1999).
Because innovators and early adopters often tend to be experts on an area, they may also favor products that sustain their current expertise and actually only suggest minor improvements to a new product (Survival and Landmine, 2005). The overlooked negative effects of early adopters have lead searchers to question whether it would be more fruitful to look at other constructs that would better capture the success or failure of an innovation in the market. The potentially damaging effects of only focusing on positive aspects of innovation adoption have also been prominent in the literature on innovation resistance (Ellen et al. 1991; Ram, 1989; Ram and Sheet, 1989). According to this literature, firms ought to understand why consumers reject new products instead of only accept them due to the high number of new products that fail in the market per cent). By only sousing on consumer acceptance, companies bypass the reasons for consumers to reject an innovation (Swimming and Focal, 1998). Considering the substantial failure rates of new products in the market, the focus on only positive aspects of innovation adoption seems surprising. By assessing consumer deference rather than acceptance, product failure rates could be minimized.
In a study regarding the failure rates of new product, Collation and Cooper (1979) were able to document 58 per cent of product failures to consumer deference, showing that if deference could be overcome, failure rates could be improved. Despite its importance, the literature on innovation resistance has remained limited in scope as only a handful of studies address the topic (Ellen et al. , 1991; Ram, 1989; Ram and Sheet, 1989). Scales and evidence regarding consumer deference are lacking that would allow researchers to consider the issue in more detail.
Further, rejection is a strong manifestation of deference and can therefore be seen as the end-point of a continuum in which consumers express doubts about whether a product would be appropriate for them to reject it. Rather than outright rejecting an innovation, consumers often postpone session until they learn more about an innovation (Ram and Sheet, 1989) or learn more about it benefits (Rogers, 2003). A more fruitful way to approach innovation deference, rather than outright rejection is to focus on consumer doubt. Consumer doubt in a new product context is defined here as a lack of conviction that a new product will fulfill its promises.
In this study, we develop a scale for measuring consumer doubt toward new products (CD) in an attempt to contribute to the literature in three ways. First, we provide the first empirical assessment of consumer construct. By focusing on doubt instead A scale to measure consumer doubt 273 JIM 13,3 274 of acceptance, we can find out what area product managers would need to focus on to minimize failure rates. Second, we validate the scales across a number of products and consumer segments to provide consistent evidence for the importance of consumer doubt.
In doing so, we show the CD scale to be a reliable and robust instrument that can be used in a variety of settings. In validating the scale, we also develop a immunological network for consumer doubt toward new products, contributing to an understanding of its antecedents and consequences. Finally, we show that there is both convergence and divergence regarding the factors that contribute to consumer acceptance and to consumer resistance of innovations. By considering both types of factors, firms can both make their new products more acceptable to the market, and avoid potential market failure.
In comparing the three studies conducted on consumer doubt to studies on consumer acceptance, we gain insight into the differential reactions consumers have toward acceptance and doubt. To our knowledge, this study represents a first attempt to measure consumer doubt toward new products. Conceptual definition – Adoption of a new product is a future event with uncertain outcomes (Castanet et al. , 2008). At a high level of product newness, risk and uncertainty in particular have been cited as major barriers to adoption (Hoofer, 2003).
Technological innovations are a set of products that consumers consider to be especially risky, as they can evoke negative emotions (Mimic and Fourier, 1998; Wood and Morale, 2006). As such, technological innovations provide a fertile area in which it would be necessary to reduce doubt to gain favorable adoption decisions (Wood and Morale, 2006). Aside from the technology itself, we know from the literature on social shaping of technology that the social context in which a product is used can raise uncertainties and doubt about its appropriateness (MacKenzie and Washman, 1985).
This is especially the case with really new products or products that provide consumers with new benefits, as they involve significant learning costs (Morale et al. , 2001). The social construction of technology approach argues that different social groups perceive technological products differently (Pinch and Bilker, 1984). A product that may be perceived as a success by one group, could be interpreted differently, by another group. Following this line of thought, we are likely to find consumers with differing degrees of doubt in any given population.
Conceptually, doubt is distinct from uncertainty. Uncertainty in a new product context most often refers to performance uncertainty (Hoofer, 2003), that is, all the reasons why a product may not perform well (Hermiston et al. , 2007). Doubt, in turn, refers to the questioning of a potential result of an uncertain outcome or choice (van Pondered et al. , 1996). As such, it is different from uncertainty in that it al. , 2002; Teaser et al. 1983). In a new product context, consumer doubt toward new products can be defined as a lack of conviction that a new product will fulfill its promises.
Since questioning and the lack of confidence signify doubt (Teaser et al. , 1983), they can be considered to differ from uncertainty in three main ways. First, doubt is related to information search and can be reduced if more evidence regarding objective performance can be found (Olsson et al. , 2000). Uncertainty, in turn, does not need to be reduced although more-objective evidence is found, since it pertains to all the ways in which a product may not perform as promised (Hoofer, 003). Second, doubt and uncertainty have different antecedents, and therefore differ conceptually.
Doubt mainly arises from information provided by low-credibility sources (such as advertising) in which objective evidence is lacking (Oboes and Chandler, 1992). According to Hoc and Ha (1986), most consumers do not believe advertising before they receive objective evidence regarding a product’s performance (such as product trial) and thereby doubt the product’s performance. In contrast, uncertainty is evoked based on a wide variety of sources (not Just advertising) and can be manifested in more varied says that doubt (Hoofer, 2003).
Third, the reference point for doubt and uncertainty is different. For doubt, the reference point is external (evoked by advertising and other low credibility sources), whereas for uncertainty, it is internal (the expectation that a product may not perform as promised). Although doubt can also arise from a person him/herself, it becomes evoked in situations in, which sufficient objective evidence is lacking (Olsson et al. , 2000). As such, a person may have high doubts, but low uncertainty that a product may perform as promised.
Given that most consumers stone innovation adoption due to lack of sufficient information, doubt appears to be a closer approximation of the type of uncertainty that is present in consumer reactions to innovations. Doubt is separated from non-acceptance because it does not constitute a decision whether or not to adopt a new product. Compared to attitudes that are defined as lasting, general evaluations of objects (Solomon et al. , 2002), consumers in doubt still have to make up their minds regarding their attitudes toward an object.
As such, doubt can be seen as a pre-attitudinal state in which consumers have yet to decide whether to accept or reject an innovation. As doubt involves questioning, it is generally seen to contribute to innovation deference rather than adoption. Although diffusion of innovations originates in sociology, it has been most widely utilized in the marketing literature (Rogers, 2003). Within this stream of literature, there are a number of studies addressing uncertainty in relation to innovations.
This literature has tended to be either conceptual in nature or focus on performance uncertainty (Hoofer, 2003; Hermiston et al. , 2007). The notion of risk has also been addressed in models of technology acceptance (e. G. Pavlov, 2003; Www ND Wang, 2005), in which it has been posited as a hindrance for the perceived ease of use or perceived usefulness that lead to the acceptance of new technology. Doubt or the questioning and lack of conviction in a product’s performance have not, to the best of our knowledge, been addressed in these streams of literature.
The diffusion of innovations paradigm has traditionally been based on adopter categories and have been discarded as predictors of innovation adoption as they have become outdated in their characteristics and cannot be generalized across product categories (Goldsmith and Hoverer, 1991; Moore, 1999). Thus, the innovation characteristics have been adopted as full-fledged representatives of the diffusion of innovations paradigm (Muter et al. , 2005) and have been found to explain rate of adoption better than other characteristics (Astound, 1974).
Considering the central role of innovation characteristics in the literature of innovation adoption, a fruitful way to measure doubt would be to measure it in relation to these characteristics. Such as view is supported by Rogers (2003), who point out that uncertainty may be found in all attributes of a new product, and can therefore contribute to a consumer’s doubt toward a new product. According to the diffusion of innovations paradigm, consumers evaluate products based on the four product attributes of relative advantage, perceived risk, complexity, A scale to measure consumer doubt 275 276 and compatibility (e. . Bauer, 1960; Rogers, 1962, 2003; Attractor and Klein, 1982). Originally, absorbability, and durability, was also part of the innovation characteristics. However, a meta-analysis by Attractor and Klein (1982) showed that their effect on adoption was weak. Thus, subsequent studies have disregarded them as innovation characteristics (e. G. Morale et al. , 2001). As consumers may rate their acceptance award innovations on all of these characteristics, they may also report their doubt toward the product on the same characteristics.
Thus, in the context of doubt, consumers can question the new product based on these characteristics. This conceptualization builds on the work of Hoofer (2003), who considered performance uncertainty as the major source of uncertainty arising in new products. We also -o expand on the work of Castanet et al. (2008) who suggested that both costs and benefits are uncertain for new products relative to established products. Relative advantage in CD is the degree to which consumers doubt that a new product is received as being better than the one it supersedes (Rogers, 2003).
Many new products have advantages to their predecessors, by being smaller, faster, and containing more features. However, there are also products that are launched for the purpose of increasing a firm’s market share that may not be directly advantageous for consumers (Mimic and Fourier, 1998). Being faced with new products at an increasingly rapid pace, consumers may doubt that each new product that is launched to the market actually brings a relative advantage to them. In this case, they may start doubting the relative advantage of new products.
Perceived risk is a ultrasonically construct consisting of several different kinds of risk (Riskier and Hilting, 2003). In this study, we focus on performance risk, as it is the most important dimension of risk in relation to new products Jacob and Kaplan, 1972). Performance does what it is supposed to do as well as consistency and reliability in relation to performance (Bricks et al. , 2000; Carving, 1987). As products may break down in an untimely manner or perform less reliably than expected, consumers may have to revise their expectations regarding a product’s performance.
Thus, they may experience doubt about the promised performance of a new product. Complexity is the degree to which consumers doubt that an innovation will be easy to use. Although recent research has shown that consumers tend to underestimate the complexity of a new product before use (Thompson et al. , 2005), consumers also learn from experience and may conclude that they need to be cautious about the product’s complexity before they buy it. Many products launched into the market are not necessarily user friendly or easy to use.
Thus instead of assuming that a product is easy to use, consumers may want to learn more about a product before they decide on its ease of use. In doing so, they doubt that the new product will be easy to use. Compatibility is the degree to which consumers doubt that a new product will be consistent with their existing values, past experiences, and needs. A product that is more compatible fits more closely with the individual’s life situation (Rogers, 2003) and can be used to express a consumer’s personality (Bell, 1988).
Many products these days tend to cater to many market segments by customization and by providing consumers with larger selections than ever before, and consumers have also become more aware of their specific styles and preferences and therefore are more apt at owing what products might fit them. However, as consumers are faced with a myriad of choices, it also becomes more difficult to set boundaries of fit, and to be able to know if new products would be suitable for a consumer.
Thus, consumers may also experience doubt regarding how well a new product would fit them. The literature on innovation adoption has concluded that the more advantages, the less risk, the less complexity, and the more compatibility a new product has, the more likely it will be accepted. In the context of doubt, the influence of the innovation attributes is as follows: the less consumers doubt that a new product: provides elating advantages; involves risk; is compatible; and is simple, the less likely a consumer is to reject the innovation.
Consumer doubt leads to less rejection (rather than more acceptance) and provides a comparative basis toward which consumer acceptance can be measured. Our scale of consumer doubt is based on the dimension of relative advantage, perceived risk, compatibility, and complexity, and will be described next in more detail. Methodological approach Based on the recommendations of Devils (2003), we proceeded to develop our scale in several steps. First, we generated a large pool of items to represent consumer doubt.
Second, we showed the items to a set of experts and asked them to review the scale items for conciseness and clarity. Third, we proceeded with scale testing and conducted an exploratory factor analysis (FEE) to examine the unintentionally, convergent validity, and discriminate validity of each construct. The model was then further validated using partial least squares modeling as recommended by Anderson and Garbing (1988). We chose partial least squares modeling as the statistical tool, since it a previous set model.
Finally, the scale and its form were validated, and potential response biases assessed. Scale development Item development Initially, we generated a pool of 56 items to represent consumer doubt toward new products. The items were based on existing literature in innovation adoption regarding innovation attributes as well as a discussion with 27 consumers. The content validity of the items was assessed in several stages (Bearded et al. , 1989). First, the items were shown to a group of experts who were asked to indicate whether they believed the question was appropriate for measuring consumer doubt.
A panel consisting of six academic experts evaluated the appropriateness of the questions. Once the appropriate questions had been chosen, four other experts evaluated the validity of the items to be used in the scale. Items that were considered unclear, ambiguous, or seemed to be related to another construct were deleted. The item development procedures led to a final scale consisting of 22 items. Using data obtained from the first study, we conducted a factor analysis to refine the items. Items that did not have item-to-total subspace correlations above 0. 50 were deleted. Following Bearded et al. 1989), items that did not have statistically higher correlations with the factor to which they were hypothesized to belong than with other factors were also deleted. These analyses resulted in a reduced index of 13 items, shown in the appendix. Samples and stimuli We conducted three studies to establish the scale of CD. The purpose of the first study was to create the scale of CD, whereas studies 2 and 3 were developed to A scale to measure consumer doubt 277 278 validate the scale created in study 1 . The sample for the first study consisted of 260 respondents randomly drawn from a consumer panel.
The panel, is maintained by the university, at which the study was conducted and represents different age groups ND professions, covering altogether 1,600 persons. Of the consumers that were approached, 191 (99 men and 92 women) agreed to take part in the study, resulting in a response rate of 73 per cent. The study was a within-subjects design, with each person seeing three products presented in random order. The pooled sample size was 573 observations. The participants received a small token of appreciation, a pen, for their participation.
For study 2, we recruited one hundred and 19 respondents (60 men and 59 women) from the same consumer panel. This time, each participant saw only one product, resulting in a sample size of 119. The participants received a small gift equivalent to $3 in monetary value for participating in the study. The third study involved a pencil and paper questionnaire with 151 participants (64 men and 87 women) drawn from the same panel. In this study, participants evaluated products from three different product categories; each participant saw two products from each category.
Within these conditions, the six products were randomly presented, rewarded to participants. To avoid dependence in the data structure, all product descriptions were randomized in all samples. In all three studies, we used a large number of actual products as stimuli. As consumers consider technological innovations to be especially risky (Mimic and Fourier, 1998), we focus on such products in our data collection. The products used in the first two studies were the Aibo-robot, auto mower, Tivoli, flat-screen TV, super audio CD player, and wristwatch camera.
Study 3 used a set of 18 products as stimuli, such as mini cameras, palmtops, portable DVD players, MPH players, and a digital weather station. The products were all new at the time of data collection, and could therefore be expected to elicit a range of differing reactions from consumers. In all studies, the products ere presented to respondents using both verbal and pictorial information. No physical products were shown at the time of data collection. Respondents were allowed to examine the products at their own pace. Assessment of the latent structure Following Titan et al. 2001), we assessed the latent structure of the scale of consumer doubt toward new products before deciding on its final form. To find the appropriate form of CD, we estimated several competing models to examine which would best represent the measurement of CD. Theoretically, consumer doubt could be represented by a reflective or a formative construct (Chin, 1998). If it were reflective, it would be subjected to factor analysis, and take the form proposed by the analysis. That is, it would be modeled as a second-order construct with as many subcomponents as a factor analysis solution would propose.
A factor analysis was plausible based on the data, and is shown in Table l. As a formative construct, the scale would be considered an index (Pollen and Lennox, 1991). A first-order formative construct would demonstrate that consumer doubt could best be viewed as a combination of indicators in which each item has the ability to influence the instruct as an independent factor (Foretell and Bookstore, 1982). In contrast, a second-order construct would reveal that doubt consists of several subcomponents that can be meaningfully separated into different groups.