The effects of self-regulatory focus and risk reduction strategy on consumers’ willingness to pay for innovative products

(1)

The effects of self-regulatory focus and risk

reduction strategy on consumers’ willingness to pay

for innovative products

Martine Ruigrok

10273441

Amsterdam, June 18, 2014

Thesis seminar Business studies Supervisor: Bram Kuijken Academic year: 2013/2014 Semester 2, Block 2/3

(2)

2

Abstract

The aim of this study was to provide a deeper insight into the effect of chronic- and induced self-regulatory focus and the use of a risk reduction strategy in advertisements on consumers’ willingness to pay for innovative products. By means of an online Vickrey auction the willingness to pay was measured for five different advertisements. In contrast to what was expected, chronic self-regulatory focus did not seem to have an effect on consumers’ willingness to pay. The induced self-regulatory focus did have an effect on the bids the respondents placed, however, in contrast to what was expected, the induced prevention focus advertisement resulted in higher willingness to pay than the induced promotion focus advertisement. Adding the review to the advertisements resulted in a significant higher willingness to pay regardless of chronic self-regulatory focus or the induced self-regulatory focus. For managers, these results suggest that inducing self-regulatory focus could be an effective marketing tool. However, caution should be taken when creating certain cues to improve the consumers’ willingness to pay. Furthermore the addition of a risk reduction, such as a review, in an advertisement should be considered when marketing innovative products.

1 Introduction

New product innovation has received a great amount of attention among academic researchers (Lewis, 2001; Cooper & Kleinschmidt, 1987) as well as in companies searching for competitive advantage. Microsoft and Adobe Systems Inc., for example, spend more than 20 percent of their total revenues on developing new products (Krishnan & Zhu, 2006, p. 813). Even though companies spend a lot on their research and development, the success rate of new products is only 45 to 67 percent (Cierpicki, Wright & Sharp, 2000, p. 777). Cooper and Kleinschmidt (1987) however argue that success or failure is not just a simple result of environmental or situational factors; it should be earned through technological synergies as well as marketing synergies. Whether or not a new product will be successful depends on the consumers’ willingness to pay for such a product.

Previous research has shown that product advantage is a very important factor in determining success (Cooper & Kleinschmidt, 1897). Items comprising product advantage include unique benefits to the customer, higher quality, reduced customers’ costs and the level of innovativeness. When consumers perceive the product as new or innovative they value it more and are more likely to adopt it. When a firm comes up with an intended innovation but the consumer fails to perceive newness, adoption will be lacked (Ram, 1987, p. 208). However, it has also been argued that innovative products are harder to understand and categorize, and therefore result in lower consumers’ preferences of the product (Moreau, Markman & Lehmann, 2001). This leads to an important question for managers about how to market new products in order for the customers to perceive it as new but still understand the benefits. According to Gregan-Paxton and Moreau (2003) consumers can learn about and understand the benefits of really new products better when marketers give them categorization cues rather than analogy cues.

Research in the psychology has shown that there is a difference between people in regulatory focus and that this difference might also influence which dimensions of evaluation people’s preferences and choices in life determine (Higgins, 1998, p. 39). Promotion focused people are associated with an openness to change, whereas people with a high level of prevention focus are associated with a preference for stability (Liberman, Idson, Camacho & Higgins, 1999). These findings may suggest that promotion focused people will have a higher likelihood of purchasing really new products than prevention focused people, because the adoption of really new products does go hand in hand with change. The effects of self-regulation on the adoption of new and really new products have also been studied by

(5)

5 Herzenstein, Posavac & Brakus (2007). These findings indeed state that promotion focused consumers have higher purchase intentions for really new products. This effect is due to the concerns about the performance of the new technology, which are more pronounced among prevention focused consumers.

This study builds on the findings of Herzenstein et al. (2007) by measuring the differences in self-regulatory focus on the consumers’ willingness to pay. Research has shown that consumers follow through less often on positive purchase intentions to buy really new products than on intentions to buy incremental new products (Alexander, Lynch & Wang, 2008). This study therefore goes further than merely examining the purchase intentions by conducting an experiment by means of a real online auction that measures the genuine consumers’ willingness to pay. According to Herzenstein et al. (2007) self-regulatory focus can be induced and therefore may be an effective marketing tool in new product promotion. This study investigates if promotion and prevention cues in advertisements have an impact on consumers’ willingness to pay. Because it has been found that prevention focused consumers have more concerns about the performance of a new product and that this results in a lower purchase intention, this study also examines potential risk reduction strategies in advertisements. Altogether this results in the following research question: What is the effect of self-regulatory focus on consumers’ willingness to pay for innovative products and how can this effect be influenced by certain cues in the advertisements of these products?

This paper continuous with an overview of the literature, which explores the research topic. Based on previous literature, a conceptual framework is developed from which several hypotheses are drawn. Next, the methodology and the research method are described in a more extensive way. This is followed by a presentation and an analysis of the results of the study, which ends in a discussion and conclusion of the research.

(6)

6

2 Literature review

2.1 Innovative products

An innovation is defined as a product that is new to the market and is perceived by the consumers as new (Ram, 1987, p. 208). This perceived newness can be due to radical changes in the product concept or due to changes in just one attribute of the product. Technological changes facilitate the rapid creation of really new products. These are radical innovations that represent entirely new product concepts and create entirely new products categories or at least substantially expand an existing category (Gregan-Paxton & John, 1997, p. 275). In contrast (incremental) new products are new models or brands in existing categories.

Due to the fact that innovative products create new product categories they are harder to understand because consumers face the challenge of constructing new knowledge structures rather than just changing existing ones (Moreau et al., 2001). Research has shown that consumers follow through less often on positive purchase intentions to buy really new products than on intentions to buy incremental new products (Alexander, Lynch & Wang, 2008). It is important for firms that potential consumers fully understand the benefits of their new products as this will influence the performance expectancies and in turn the consumers’ preferences. To make sure consumers understand the new product, learning strategies, such as categories, analogies and mental simulation, are used.

According to Gregan-Paxton and Moreau (2003) consumers can learn about and understand the benefits of really new products better when marketers give them categorization cues rather than analogy cues. A categorization cue is communicating that a new product is a member of an existing category. This signals a similarity wherein both the attributes and the relations associated with the category are transferred to the new product. An analogy is when it is communicated that a new product is like a member of an existing category. This signals a match wherein only the relations associated with the category are transferred to the new product.

Another learning strategy for learning new information is mental simulation. Consumers can make decisions about future consumption through autobiographical anticipations or consumption visions of the future that visualize themselves with the new product or service. Mental simulation make events concrete and therefore can provide a means to deal with uncertain futures and help consumers learn about new benefits (Castaño, Sujan, Kacker & Sujan, 2008).

(7)

7 2.2 Self-regulatory focus

The basic motivational principle throughout the history of the psychology has been the hedonic principle that people approach pleasure and avoid pain. From models in the biology as well as in the social psychology and the empirical support for them it is clear that people are motivated to approach pleasure and avoid pain. According to Higgins (1998), however, the hedonic principle is too basic and is therefore not sufficient to understand human strategic behavior. Higgins proposes that the way in which the hedonic principle works in motivation might be even more important than the fact that it works. He describes two different ways in which the hedonic principle operates; with a promotion focus and with prevention focus. This is called the regulatory focus theory (Higgins, 1998). Prevention and promotion focus are two ways of regulating pleasure and pain. Evidence has shown that this regulatory focus has a great impact on people’s feelings, thoughts and actions and is independent of the hedonic principle (Higgins, 1998). People’s feelings, strategies and choices can vary greatly. The hedonic principle provides only limited understanding of such variability. It simply accounts for the basic differences between positive and negative motivational states. Regulatory focus, in contrast, can account for the variability across people and across situations that occurs within positive and negative motivational states. Emotions, decision making, performance and preferences all vary depending on the self-regulatory focus (Higgins, 1998). The two separate self-regulatory systems, promotion and prevention, are concerned with the pursuit of different type of goals.

The promotion system is concerned with the pursuit of hopes and aspirations, called ideals. Promotion focus generates sensitivity to the presence or absence of positive events. To attain their goals, people in a promotion focus adopt approach strategic means. For example, promotion focused individuals approach matches to attainment such as earning extra money to buy something. Regulatory focus also influences people’s preferences and choices in life. Some dimensions of evaluation are more relevant to promotion focus concerns. For example, when buying a new apartment luxury or promotion attributes are big, capacious rooms.

In contrast prevention focus is concerned with the fulfillment of duties, obligations or even necessities, termed oughts. Prevention focus generates sensitivity to the presence or absence of negative events. To attain their goals, people in a prevention focus use avoidance strategic means. Individuals with a prevention focus, for example, avoid mismatches to attainment such as saving their money to buy something by not spending it on other things. Dimensions of evaluation more relevant to prevention focused individuals are security or prevention attributes such as the presence of reliable smoke detectors.

(8)

8 Self-regulatory focus can be activated chronically and temporarily. Chronic self-regulatory focus varies across individuals. An individual can have a strong promotion focus or a strong prevention focus or both. Differences between individuals in prevention and promotion focus are reflected in the difference between ideal and ought self-regulation (Higgins, 1998). According to Higgins this difference can arise from differences in interactions between a child and its caretaker (1998, pp. 14-16). Chronic promotion focus can be initiated by hugging and kissing the child for behaving in the desired manner and withdrawing love as discipline. When caretakers encourage the child to be alert to potential dangers and punish the child when they behave undesirably this will result in a chronic prevention focus.

Regulatory focus is not plainly an individual variable; it can also vary across momentary situations (Higgins, 1998, pp. 19-20). It is possible to experimentally induce a promotion or a prevention focus through situational activation. This can be done by, for example, task feedback, task instructions or by using slogans and mottos. Herzenstein et al. (2007) manipulated self-regulatory focus with scenarios framed either in promotion or in prevention terms that created the need to purchase a new car/boat. Participants were asked to imagine themselves living next to a lake but working on the other side, and being able to save commute time if they purchased this new car/boat to drive/sail across the lake. The promotion scenario read: “It is very important for you to find ways to shorten this long commute time”. The prevention scenario read: “It is very important for you to avoid not being able to shorten this long commute time”.

Another example of manipulating self-regulatory is by using slogans and mottos (Faddegon, Scheepers & Ellemers, 2008). The slogan “Where there’s a will, there’s a way”, for example, emphasizes the prospect of gain and the pursuit of aspirations and induces a promotion focus. In contrast a slogan such as “An ounce of prevention is worth a pound of cure” induces a prevention focus because it emphasizes the prevention of loss and the fulfillment of duties.

Previous research has shown that induced self-regulatory focus can have an influence on consumer behavior (e.g. Chernev, 2004; Florack & Scarabis, 2006; Lee & Aker, 2004). Chernev (2004) assumed that consumers prefer products when the attributes of the products are superior on a dimension that is relevant to their regulatory focus. In three experiments, Chernev manipulated self-regulatory focused and found that prevention focused consumers are more likely to overweight utilitarian, reliability-related, and unattractive attributes than

(9)

9 promotion focused individuals, who are more likely to relatively overweight hedonic, performance-related, and attractive attributes.

The impact of regulatory focus is not limited to attributes of choice options but can also be effective in advertising campaigns (Florack & Scarabis, 2006, p. 742). Florack & Scarabis (2006) found that the fit between an advertising claim and consumers’ regulatory focus has an impact on product preferences. Participants were more likely to prefer products presented in an advertisement with a claim that corresponded with the experimentally induced focus. This is in line with the findings of Cesario, Grant & Higgins (2004). Their findings suggest that promotion focused individuals are more likely than prevention focused people to be persuaded by a message that emphasizes approach means to reach a goal. In contrast a message that induces the use of avoidance means is more effective when individuals are in a prevention focus than when they are in a promotion focus. Lee and Aker (2004) have also studied the role of regulatory focus on message-framing effects. The results of their experiments showed that brand attitude was rated more favorably when promotion products were grain-framed rather than loss-framed. In contrast, participants rated brand attitude more favorably when prevention products were loss-framed rather than gain-framed.

2.3 Risk reduction strategies

Because previous research (Herzenstein et al., 2007) has shown that prevention focused consumers have more concerns about the performance of a new product and that this results in a lower purchase intention, this study examines potential risk reduction strategies in advertisements. A risk reduction strategy is a strategy devised by a consumer from a set of potential risk relievers that is most likely to diminish the risk level until it reaches the level perceived low enough for the consumer to decide to purchase the product (Cases, 2002). Which risk reduction strategy is best suitable can depend on the perception of risk (Roselius, 1971) or on the type of product (Derbaix, 1983).

The perception of risk causes the buyer to select which device or strategy seems to be best suited for the type of risk involved (Roselius, 1971, p. 57). Financial risk, functional risk, psychological risk, physical risk, social risk and time risk are all dimensions of perceived risk (Cases, 2002, p. 377). For most kinds of products, performance or functional risk is the most predictive dimension of overall perceived risk (Kaplan, Szybillo & Jacoby, 1974). Ram and Sheth (1989) defined functional risk as performance uncertainty due to insufficient testing of a new technology by its manufacturer (in Herzenstein et al., 2004, p. 252).

(10)

10 Examples of risk reduction strategies are: brand loyalty, word of mouth, free sample, money back guarantee, store image and warranties. Given the design of this study, only risk relievers that can be portrayed or mentioned in the advertisements on the auction platform will be discussed.

A warranty can be a suitable risk reduction method to relieve the financial risk occurred in purchasing an innovative product. Consumers appear to think: “I’m not sure whether this innovative product will work very well, but even if it doesn’t I can have the product replaced or get a refund” (Shimp & Bearden, 1982, p. 44). According to Van den Poel & Leunis (1996) a money-back guarantee is one of the best risk relievers for performance risk as well as financial risk. In Shimp & Bearden’s study a warranty however did not significantly reduce the subject’s perceptions of performance risk.

Another type of risk reliever that can be implemented in advertisements is an endorsement. Endorsements or testimonials can be from a person like you, from an expert on the product or from a celebrity (Roselius, 1971, p. 57). The most popular risk reliever type studied by Tan (1999) was reference group appeals. According to Shimp & Beaerden (1982) extrinsic cues appear to be incapable in reducing the uncertainty associated with whether an innovative product will actually perform its designated function as it should. This product performance uncertainty can only be reduced by actual product usage experiences or word-of-mouth assurances (Shimp & Bearden, 1982, p. 44). According to Kalish (1985) experience type information produced by actual adopters can reduce the uncertainty associated with the new product. As experience information becomes available, the uncertainty associated with the new product is reduced and this results in an increase in the customers’ perceived value of the product.

2.4 Consumers’ willingness to pay

Consumers’ willingness to pay is the maximum price a buyer is willing to pay for a given good (Wertenbroch & Skiera, 2002). This maximum price a consumer is willing to pay for a product is also called the consumers’ reservation price and equals the product’s value to the consumer (Kalish & Nelson, 1991). This reflects the perceived value of the product as well as the perceived sacrifice involved in acquiring the product (Simonson & Drolet, 2004, pp. 681-682). The reservation price is thus a monetary ratio scaled measure of preference that does not take the purchase price into account. When shopping for a certain good a consumer compares her reservation price for each product with its purchase price and in order to maximize her

(11)

11 utility the consumer then chooses the product with the largest differential (Kalish & Nelson, 1991, p. 328).

2.5 Literature gap

The effect of self-regulation on the adoption of new and really new products has been studied by Herzenstein et al. (2007). This effect, however, was studied on purchase intentions. Previous research has shown that consumers follow less often on positive purchase intentions to buy really new products than on intentions to buy incremental new products (Alexander, Lynch & Wang, 2008). Therefore, when studying the effect on really new products it is better suited to measure consumers’ actual willingness to pay. Knowledge of consumers’ willingness to pay is crucial in estimating demand and creating optimal pricing schedules. This is done by means of a Vickrey online auction that measures the genuine consumers’ willingness to pay.

For marketers it can also be very useful to know how to influence the effect of self-regulatory focus in a positive manner to increase consumers’ willingness to pay. Since previous research has already shown that inducing self-regulatory focus can have an influence on consumer behavior (e.g. Chernev, 2004; Florack & Scarabis, 2006; Lee & Aker, 2004), this study builds on these findings by exploring other cues in advertisements that can influence consumers’ willingness to pay. Prevention focused consumers have more concerns about the performance of new products which results in lower purchase intentions (Herzenstein et al., 2007); therefore, this study examines the use of a risk reduction strategy in advertisements. Experience type information produced by actual adopters can reduce the uncertainty associated with the new product, which results in an increase in the customers’ perceived value of the product (Kalish, 1985). This research therefore studies the effect of self-regulatory focus and a review in an advertisement on consumers’ willingness to pay.

(12)

12

3 Conceptual framework

Based on the literature review, several hypotheses are drawn and these are visualized in a conceptual framework (Figure 1).

Figure 1 : Conceptual Framework

The influence of self-regulatory focus on willingness to pay for innovative products

Promotion focused people are associated with an openness to change, whereas people with a high level of prevention focus are associated with a preference for stability (Liberman et al., 1999). These findings may suggest that promotion focused consumers will have a higher likelihood in purchasing really new products than prevention focused people, because the adoption of really new products does go hand in hand with change. The effects of self-regulation on the adoption of new and really new products have also been studied by Herzenstein et al. (2007). These findings indeed state that promotion-focused consumers have higher purchase intentions for really new products. In this research the consumers’ willingness to pay is measured instead of the purchase intentions.

H1: Promotion focused people will have a higher willingness to pay for innovative products than prevention focused people.

(13)

13 The influence of induced self-regulatory focus on willingness to pay for innovative

products

Regulatory focus can vary not only across individuals but also across momentary situations (Higgins, 1998, pp. 19-20). It is possible to experimentally induce a promotion or prevention focus through situational activation. Previous research has shown that induced self-regulatory focus can have an effect on consumers’ behavior (e.g. Chernev, 2004; Florack & Scarabis, 2006; Lee & Aker, 2004). According to Herzenstein et al. (2007) self-regulatory focus can be induced and therefore may be an effective marketing tool in new product promotion. This study investigates if promotion and prevention cues in advertisements have an impact on consumers’ willingness to pay. Because previous research has shown that promotion focused consumers have higher purchase intentions than prevention focused consumers the following is also expected concerning induced self-regulatory focus:

H2a: Promotion focused cues will have a positive effect on consumers’ willingness to pay for innovative products

H2b: Prevention focused cues will have a negative effect on consumers’ willingness to pay for innovative products

Furthermore, it is expected that the effect of induced self-regulatory focus on willingness to pay will be moderated by chronic self-regulatory focus. This is in line with previous research such as Cesario, Grant & Higgins (2004). According to their findings promotion focused individuals are more likely than prevention focused people to be persuaded by a message that emphasizes approach means to reach a goal. In contrast a message that induces the use of avoidance means is more effective when individuals are in a prevention focus than when they are in a promotion focus.

H3: The effect of induced self-regulatory focus on consumer’s willingness to pay will be moderated by chronic self-regulatory focus

The mediating role of product performance uncertainty

Consumers discount the value of a product due to uncertainty associated with it (Kalish, 1985, p.1569). It can therefore be suggested that a high level of perceived product performance uncertainty will result in a lower willingness to pay. According to Herzenstein et al. (2000) the difference between promotion focused and prevention focused consumers and their purchase intentions for new products is due to the concerns about the performance of the new technology. The results suggest that prevention focused participants are more likely to generate concerns about the purchase and use of a really new product than promotion focused

(14)

14 participants (Herzenstein et al, 2000). This study tries to verify this finding and also tests if product performance uncertainty mediates the effect of self-regulatory focus on consumers’ willingness to pay.

H4: Prevention focused consumers will perceive more product performance uncertainty than promotion focused consumers.

H5: Product performance uncertainty will have a negative effect on willingness to pay H6: Product performance uncertainty will mediate the effect of self-regulatory focus on consumers’ willingness to pay

The influences of risk reduction strategies

According to Kalish (1985) experience type information produced by actual adopters can reduce the uncertainty associated with the new product. As experience information becomes available, the uncertainty associated with the new product is reduced and this results in an increase in the customers’ perceived value of the product. Therefore a potential risk reduction strategy to increase consumers’ willingness to pay might be adding a positive customer’s review in the advertisement. According to Tan (1999) the most preferred risk reliever in internet shopping is reference group appeal. It is therefore expected that the use of a review in an advertisement results in a decrease in perceived performance uncertainty and an increase in consumers’ willingness to pay. However, risk reduction is not necessary a requirement for innovative behavior. Consumers who have a high stimulation level may actually prefer to take risks (Raju, 1980). Every individual prefers a certain level of stimulation, which is termed optimum stimulation. When the environmental stimulation, determined by ambiguity, complexity and novelty, comes below optimum, an individual will try to increase stimulation (Raju, 1980, p. 272). Younger, more educated and employed people appear to have a higher optimum stimulation level. According to Raju innovativeness might be even primarily motivated by risk taking (1980, p. 280). Therefore, it is expected that the positive effect of risk reduction strategies is stronger on consumers with a relatively high level of prevention focus.

H7: The use of a risk reduction strategy will lead to a decrease in perceived performance uncertainty.

H8: The use of a risk reduction strategy will lead to an increase in consumers’ willingness to pay

H9: The positive effect of risk reduction strategies will be stronger on prevention focused people than on promotion focused people.

(15)

15

4 Methodology

This research is based on an empirical quantitative study, where the data is collected through an experiment by means of a real online auction in combination with a short questionnaire. In this research the level of analysis consists of individuals. This study researches the influences of self-regulatory focus and a risk reduction strategy on the willingness to pay of individual men and women for innovative products.

Because the language of the auction platform is Dutch, the data in this research is only collected from people who understand Dutch. The participants also had to live in the Netherlands because the products that are won in the auction can only be shipped within the Netherlands. The conceptual framework could be applicable for all companies marketing an innovative product. Marketers can use the data collected through this experiment to increase their consumers’ willingness to pay by taking into account the results of this research in designing their own advertisements.

4.1 Overall design

The overall design used in this study is a multimethod quantitative study (Saunders, Lewis & Thornhill, 2012, p. 165). The data is collected using an online experiment in combination with short questionnaire in a single phase design. Therefore both sets of results can be interpreted together to provide a richer and more comprehensive response to the research question in comparison to the use of mono-method design (Saunders et al., 2012, p. 167).

A classical experiment is conducted, whereby the sample of participants is randomly assigned to either one of the experimental conditions or to the control group (Saunders et al., 2012, p. 175). The use of a control group helps remove the possible effects of an alternative explanation to the planned intervention and eliminates threats to internal validity. A limitation of an experiment, however, is the external validity (Saunders et al., 2012, p. 176). The use of the real online auction platform, however, ensures the internal validity as well as improves the external validity because of the realistic consumer context rather than a laboratory context.

4.1.1 Vickrey Auction

In order to determine willingness to pay in market research, several techniques may be used. Market researchers can estimate willingness to pay from survey data that show the stated preferences. The key advantage of survey data is that they can be used in concept testing and new product development. The downside of survey data however is that they provide little

(16)

16 incentive to consumers to truthfully reveal their willingness to pay because the responses are hypothetical (Hoffman, Menkhaus, Chakravarti, Field & Whipple, 1993, p. 320). Hoffman et al. (1993) suggest the use of experimental Vickrey auctions (Vickrey, 1961) for determining consumers’ willingness to pay.

The propositions are tested by using a Vickrey auction, also called a second-price-sealed-bid auction. This means that the participants don’t know what other participants are bidding and that the winner of the auction pays the amount of the second highest bid (Ausubel & Milgrom, 2006, p. 2). This design of the auction, in theory, results in participants bidding their true value because only by bidding their true value, they can be sure to win exactly when they are willing to pay the price (Ausubel & Milgrom, 2006, p. 2). In comparison to using questionnaires or hypothetical experiments, which can only measure the value or purchase intentions by what the participants say, this auction design can be very effective in measuring consumers’ true willingness to pay because the participants can’t just say they will pay a certain amount, because if they win they will really have to buy the product.

According to Noussair, Robin and Ruffieux (2004) the Vickrey auction can indeed be an effective tool to reveal the demand for certain products. However, they do note that sufficient practice and appropriate training in the rules of the auction is important (Noussair et al., 2004, p. 739). In this study the rules are explained before every auction. When panel members decide to participate in an auction, they are first send to a page where they have to agree with the rules. The rules state that the participants are only allowed to bid once, the bid they place is legally binding so they are not able to cancel. Furthermore, the winner pays the second highest bid and the winner has to pay within 48 hours. On the same page it is also explained why it is so important to bid exactly their maximum price they are willing to pay. It is explained that bidding higher than your true willingness to pay is unwisely since there is a chance you will have to pay more than your willingness to pay if the second highest bid is also above your true valuation. Bidding lower than your true valuation is also not recommended since there is a chance you will not win the auction even though you were willing to buy the product for a higher price than the winning bid. The results of Noussair et al. (2004) have, however, shown that explaining and suggesting the optimal strategy to the participants directly is less effective than a technique to learn on their own. Participants don’t seem to learn the process by just reading the rules, but rather by active participation. Also Coppinger, Smith and Titus stated that participants first need a certain amount of practice to learn that the best strategy is indeed bidding what they are truly willing to pay for the auctioned good (1980, pp. 18-21). This is seems to be a limitation of the Vickrey method in

(17)

17 measuring consumers’ willingness to pay. However, since this study is not interested in the precise quantitative amount the consumers’ are willing to pay but rather in the relative differences between different forms of advertisements, the Vickrey auction still is a very useful method. Since the possible bias would be across all groups of the different conditions, it would not affect the results of the differences in regulatory focus and the use of risk reduction strategy on consumers’ willingness to pay. Since the participants are allocated to either the control or experimental condition, the possible bias would affect all conditions equally. Therefore the results can be compared and the relative differences in consumers’ willingness to pay can be explained entirely by the difference in the condition (Saunders et al., 2012, p. 175).

4.1.2 The auctioned good:

The auctioned product should be an innovative product with high perceived newness. The product chosen for this study is a vacuum cleaning robot (Irobot.com, 2014). With just a push of a button the robot gives you clean floors. This type of vacuum cleaning robot is new to the market and is seen as an innovative product due to the new technologies used such as intelligent navigation, responsive cleaning technology and a new brush design, which improves the cleaning results substantially compared to other vacuum cleaning robots.

Due to time constraints, only one product was auctioned. Therefore the results are less generalizable since there might be differences between different product categories. For example, it can be expected that there might be differences between utilitarian and hedonic products. According to Sen and Lerman (2007), for example, the perceived usefulness of a review is moderated by product type. Readers are less likely to find negative reviews of hedonic products useful, compared with utilitarian products.

4.1.3 The advertisements

The bidders in this study are randomly assigned to one of the five advertisements. These advertisements all promote the same product but to test the propositions there are differences between these advertisements. Self-regulatory focus in this study is induced by means of either a promotion focused cue or a prevention focused cue. This is similar to previous research, which has also used slogans (Faddegon et al., 2008) or advertising claims (Florack & Scarabis, 2006) to induce self-regulatory focus. The five different treatments are as follows:

1. Control: no self-regulatory focus cue + no risk reduction strategy 2. Promotion focused cue + no risk reduction strategy

(18)

18 3. Prevention focused cue + no risk reduction strategy

4. Promotion focused cue + risk reduction strategy 5. Prevention focused cue + risk reduction strategy

See appendix A for a copy of the advertisements used. A promotion focused cue is supposed to emphasize the presence or absence of positive events and induce the use of approach means to reach a goal (Higgins, 1998). In contrast, prevention focused cues should stress the presence or absence of negative events and induce the use of avoidance means (Higgins, 1998). In the advertisements the promotion focused cue was gain-framed and consisted the positive event of having a clean house. In contrast, the prevention focused cue in this study was loss-framed and emphasized the avoidance of having a dirty home.

The risk reduction strategy used in condition four and five is the addition of a positive review in the advertisement. For this study, it has been chosen to use a review as risk reduction since experience type information produced by actual adopters can reduce the uncertainty associated with the new product, which can result in an increase in the customers’ perceived value of the product (Kalish, 1985). The reviewer, of the review used in the advertisements, is very positive about the product and confirms the convenience and ease of the use of the product (see appendix A for the review in Dutch).

4.1.4 Questionnaire

In order to be able to test the propositions, the bidders also had to fill in a short questionnaire concerning questions that measure their chronic self-regulatory focus, perceived product uncertainty and perceived newness of the product (see appendix B for the complete questionnaire in Dutch). Only if the bidders answered these questions their bid was submitted. Questionnaires allow the collection of standardized data from a sizable population in a highly economical way, which allows easy comparison (Saunders et al., 2012, p. 177). However, since people generally don’t want to fill in a long questionnaire, the number of questions had to be kept to a minimum in order to maximize the response rate (Saunders et al., 2012, p. 178). Since the number of questions was kept to a minimum, it was not possible to have alternative forms or check questions for measuring perceived newness and perceived performance uncertainty, therefore the reliability of these questions cannot be tested. Also the self-regulatory focus had to be measured with an abbreviated scale, only six of the eleven questions were included. However, for the self-regulatory focus scale it was possible to measure the internal consistency since the scale still concluded multiple scale items.

(19)

19 4.2 Research Sample

The focus of this study is on Dutch consumers. A reliable sample had to be made from this population in order to be able to generalize the results (Saunders et al., 2012, p. 265). The sample size was strived to be as large as possible, because the larger the size, the closer its distribution will be to the normal distribution and thus the more robust it will be and the better the results can be generalized (Saunders et al., 2012, p. 265). The minimum requirements for the sample size was stated of at least 30 people for every category compared since statisticians have shown that a sample size of 30 or more will usually result in a sampling distribution for the means that is very close to a normal distribution (Saunders et al., 2012, pp. 264-265).

In total, 484 individuals participated in the auction. However, not everyone completed the short questionnaire following the auction (n=430). Of the Dutch population, 49,5% is male, and 50,5% is female (CBS, 2013). The men in this sample are, compared to the Dutch population, slightly overrepresented. Of the respondents 55,2% (n=267) are male and 44,8% (n=217) are female. The vast majority of the respondents are between 20 and 65 years old (90%, n = 91), compared to the Dutch population this group is highly overrepresented. The participants of the experiment are relatively high educated. Of the participants, 52,2% are in possession of a Bachelor’s or Master’s degree, of the Dutch population only 28,3% are high educated (CBS, 2012).

4.3 Operationalization of constructs and variables 4.3.1 Willingness to pay

The dependent variable in this study is the consumers’ willingness to pay. This is measured by the bidding prices the participants place in the auction.

4.3.2 Chronic self-regulatory focus

The self-regulatory focus is assessed by means of the Regulatory Focus Pride Questionnaire (Higgins, Friedman, Harlow, Idson, Ayduk & Taylor, 2001). This questionnaire however consists of eleven questions. In order to maximize the response rate the questionnaire was therefore abbreviated. An example question of the Regulatory Focus Pride Questionnaire that measures prevention is “How often do you obey rules and regulations that were established by your parents?”. A question that measures promotion is “How often have you accomplished things that got you ‘psyched’ to work even harder?”. These questions have been converted into statements, such as: I often accomplish things that get me ‘psyched’ to work even harder.

(20)

20 The participants were asked to specify their agreement or disagreement on a 5-point likert scale for the following six statements:

Prevention

1. Growing up, I often “crossed the line” by doing things that my parents would not tolerate. -

2. I got on my parents’ nerves often when I was growing up. - 3. Not being careful enough has gotten me into trouble at times. -

Promotion

1. I have often accomplished things that got me psyched to work even harder? + 2. I feel like I have made progress toward being successful in my life. +

3. Compared to most people I am typically unable to get what I want out of life. -

All items measuring prevention and one question measuring promotion are reverse-phrased. This has been done to reduce response bias (Field, 2009). Furthermore, the questions are disordered for this same reason (See appendix B for the actual sequence of the statements). The difference between the promotion and prevention score was calculated, whereby a low figure indicates a high prevention focus and a higher figure indicates a promotion focus.

4.3.3 Perceived performance uncertainty

New product performance uncertainty includes all the reasons a new product may not perform well (Herzenstein et al., 2000, p. 252). Performance uncertainty for example can be due to insufficient testing of a new technology by its manufacturer or due to consumers’ problems like having difficulties in using a new complex technology. The participants were asked to specify their agreement or disagreement on a 5-point likert scale for a statement concerning their uncertainty about the new product. The measurement used in this study is similar to the measurements used in previous research to measure perceived performance uncertainty. To measure performance risk, Cases, for example, asked her participants to state their (dis)agreement on the following statement: “The quality of the jacket purchases on the internet may not meet expectations” (2002, p. 392). Also Tan used statements such as “The product would fail to perform to my satisfaction”, to measure perceived product risk (1999, p. 178). Of the respondents, 54% are unsure whether the product will meet their expectations. To correctly interpret the uncertainty mean values in the results section, it is important to note that higher mean values correspond to lower uncertainty.

(21)

21 4.3.4 Perceived product newness

The control variable, perceived product newness, needs to be kept constant to avoid it influencing the effect of the independent variables on the dependent variable (Saunders et al., 2012, p. 174). Because it has been shown that only for new products there is an influence of self-regulatory focus on purchase intentions (Herzenstein et al., 2007), there is a question included in the short questionnaire concerning the perceived product newness. The product is considered new by 44,3% of the respondents.

5 Results

In this section the results of the research are presented. First, the descriptive statistics of willingness to pay and chronic self-regulatory focus are provided. Next the hypotheses are tested by using ANOVA’s and regression analyses.

5.1 Descriptive statistics 5.1.1 Auction Bid Data

As mentioned in section 4, a total of 484 panel members participated in the auction and placed a bid. Looking at the histogram of the frequency distribution of the bids, the sample looks highly positively skewed and seems non-normally distributed. Quantifying the skewness and kurtosis confirms this. The skewness value is 2,620 with a standard error of 0,111, which results in a standardized z-score of skewness of 23.60 (p < ,001). This confirms significant positive skewness. The kurtosis is also highly significant with a standardized z-score of 39.80 (p < ,001). Since the Shapiro-Wilk is more appropriate for smaller sample sizes than the Kolmogrov-Smirnov test, the Shapiro-Wilk test is used to test for normality (Ahad, Yin, Othman & Yaacob, 2011). The Shapiro-Wilk test on the auction bid data shows a significant deviation from normality (F(484) = 0,713, p < ,001). Therefore, it cannot be assumed that the data of the auction bids is normally distributed.

Since the data is highly positively skewed, consisting of a cluster of low bids, it can be proposed to exclude these lower bids to improve normality. Therefore, in order to reduce the highly positive skewness of the auction bid data, only the bids in the top 50 percentile of each treatment are selected. In table 1 the descriptive statistics of the adjusted sample (n=238) are provided. Chart 1 visually presents the mean bids for each treatment of the adjusted sample. As expected, treatment 4 (induced promotion + risk reduction) has the highest mean value. Remarkably, in contrast to what was expected, the induced prevention treatment (treatment 3)

(22)

22 has a higher mean bid than the induced promotion treatment (treatment 2). Before extensively testing all the hypotheses, normality for the adjusted sample is tested first.

Table 1: Descriptive statistics bid amount – excluding bids bellow treatment median

Treatment Minimum Maximum Mean Std. Error mean Std. Deviation

1 (n=42) 20,00 225 55,11 7,27 47,12 2 (n=42) 16,00 185 48,55 6,36 41,22 3 (n=45) 20,00 200 59,11 6,11 40,97 4 (n=53) 25,00 300 77,53 8,89 64,75 5 (n=56) 22,50 190 57,46 4,68 34,99 Total (n=238) 16,00 300 60,25 3,10 47,87

In order to improve normality even more, on top of including only the bids in the top 50 percentile of each treatment, the auction bid data is also transformed by ln. Including only the bids above the median of its treatments, resulted in a decrease of the sample (n=238) but the ln bids seem to be distributed closer to a normal distribution than before the transformation. However, it still doesn’t seem normal distributed. Quantifying the skewness and kurtosis partially confirms this. The overall skewness is non-significant with a value of 0,692 and a standard error of 0,158, which results in a standardized z-score of skewness of 4,38. This confirms significant positive skewness. However, the data of treatment 3, 4, and 5 is just a little skewed, and does not differ significantly from normality. The overall kurtosis (as well as the kurtosis of the individual treatments) is not significant either with a standardized z-score

0 10 20 30 40 50 60 70 80

Treatment 1 Treatment 2 Treatment 3 Treatment 4 Treatment 5

€

(23)

23 of -0.34. The Shapiro-Wilk test on the ln auction bid data does show a significant deviation from normality for all treatments.

Based on these tests it cannot be assumed that the data of the auction bids is normally distributed and accordingly the assumption of normality is validated. Therefore, parametric tests, such as an ANOVA or a regression analysis, might give inaccurate results (Field, 2009).

5.1.2 Chronic self-regulatory focus

Chronic self-regulatory focus, in this study, is measured with two scales consisting each of three items. The internal reliability of the scales is checked with Cronbach’s Alpha. The Cronbach’s Alpha’s are commuted using the adjusted sample (n=238) consisting the participants who’s bid was above the median bid of their treatment. In total 218 participants completed the promotion scale and 216 participants completed the prevention scale. The prevention scale, consisting three items, turned out to be sufficiently reliable (Cronbach’s Alpha=0,703). The Cronbach’s Alpha could be marginally improved by excluding the statement: Not being careful enough has gotten me into trouble at times. However, deleting this item would only increase the Cronbach’s Alpha to 0,705, therefore this item was not excluded from the scale. The Cronbach’s Alpha for the promotion scale is 0,587, however by excluding the statement: Compared to most people I am typically unable to get what I want out of life the Cronbach’s alpha is improved to above the cut-off point of 0.6 (r=0,602).

Subsequently, the overall chronic self-regulatory focus is measured by calculating the difference between the promotion and prevention score. A low figure indicates a high prevention focus and a higher figure indicates a promotion focus. The Shapiro-Wilk is non-significant, which means normality of the data can be assumed (F(215) = 0,994, p = ,528).

5.2 Hypothesis testing

Even though, the data of ln bid amount does not seem to be normally distributed, an ANOVA is generally known to be a robust test even when the data is skewed (Field, 2009). Therefore ANOVA’s on the data were carried out to test some of the hypotheses. An assumption of the ANOVA is homogeneity of variance, which means that the variances in the different groups should be identical. To test this, the Levene statistic is used. If the Levene statistic is significant, then the null hypothesis of equal variances is rejected. Levene’s test for ln bid amount indicated equal variances (F = 1,862, p = ,118) and thus satisfies the homogeneity of variance assumption.

(24)

24 5.2.1 Effects of chronic self-regulatory focus on consumers’ willingness to pay (h1)

In order to test the first hypothesis, a one-way independent ANOVA was carried out to compare the mean ln bid values for prevention focused and promotion focused people. A median split was done to determine whether a respondent was more prevention or more promotion focused (Med = 0,333). The results of the ANOVA do not show a significant difference between the mean values (F(1,214) = 0,067, p = ,795). Accordingly, the results do not suggest that there is a difference between promotion and prevention focused people, and therefore also the hypothesis that promotion focused people have a higher willingness to pay than prevention focused people is rejected (H1).

5.2.2 Effects of the different treatments on consumers’ willingness to pay (h2, h8)

As mentioned earlier, the mean value of the auction bids is highest for treatment 4 (table 1). The treatments in descending order of their mean bids are; treatment 4 (M = 77,53), treatment 3 (M = 59,11), treatment 5 (M = 57,46), treatment 1(M = 55,11) and treatment 2 (M = 48,55). Participants who were assigned to the promotion focus, on average, placed the lowest bids. This is in contrast with hypothesis 2. To test whether the differences between the mean bids per treatment are significant, an ANOVA was carried out. The results of the ANOVA show a significant effect of the treatments on ln bid amount at the p<0,01 level for the five conditions (F(4,234) = 3,458, p = ,009). To determine where the significance exists a LSD post-hoc test is conducted. The results show that not all treatments are significantly different from each other (appendix C, table 7).

Effects of induced self-regulatory focus on consumers’ willingness to pay

The results of the LSD post-hoc test show that the ln bid mean value of induced prevention treatment is marginally significantly higher than the mean value of the induced promotion treatment (p = ,051). This result is in contrast with hypothesis 2a and 2b.

However, it was expected that self-regulatory focus only had an influence on willingness to pay for new products. Therefore an ANOVA is also carried out including only the participants who perceived the product as new. The Levene Statistic of the ln bid amount including only these participants indicated equal variances (F = 0,772, p = ,546) and therefore satisfies the homogeneity of variance assumption. The ANOVA results show a significant effect of the treatments on ln bid amount at the p<0,05 level for the five conditions (F(4,99) = 2,912 p = ,025). The post-hoc LSD tests showed that the promotion focused treatment had significantly lower ln bids than the prevention focused treatment at p< 0,01 level (appendix C,

(25)

25 table 8). When a risk reduction strategy was included in the promotion and prevention treatments, the promotion focused treatments did have higher ln bids. However, both ANOVA’s (appendix C, table 7 and 8) showed non-significant results. The planned contrast test (table 2) between the two treatments including induced promotion focus and the two treatments including prevention focus did not show significant results for the sample including participants who didn’t perceive the product as new (t(234) = 0,465, p = ,642). The planned contrast test (table 2), however, did show marginally significant results for the sample including only participants who perceived it as new (t(99) = 1.975, p = ,051). However, the value of the contrast was in the opposite direction, of what was expected. The treatments including a prevention focus (treatment 3&5) had a positive contrast value compared to the treatments including a prevention focus (treatment 2&4). Concluding, it can be assumed that inducing self-regulatory has an effect on willingness to pay for innovative products. However, this effect was in the opposite direction of what was expected. Therefore, hypotheses 2a and 2b, stating that induced promotion focus has a positive effect on willingness to pay and induced prevention has a negative effect on willingness to pay, are rejected.

Table 2: Planned Contrasts ln bid amounts treatment 2&4 vs. 3&5

Value of contrast Std. Error T Df Sig. (2-tailed)

Not controlled for perceived

newness 0,082 0,176 0,465 234 ,642

Controlled for perceived

newness 0,559 0,283 1,975 99 ,051

Effects of risk reduction strategy on consumers’ willingness to pay

At the p<0,05 level, treatment 4 is significantly higher than treatment 1 and significantly higher than treatment 2. This suggests that an induced promotion focus in combination with a risk reduction strategy (treatment 4) results in a higher willingness to pay than an induced promotion focus (treatment 2) or than the control treatment (treatment 1). The mean bid of treatment 5 is slightly higher than treatment 3. The difference however, is not significant (p = ,915). Therefore, it cannot be assumed that the addition of a risk reduction strategy in a prevention focused advertisement will lead to a higher willingness to pay. The planned contrast test (table 3) did show a significant difference at the p<0,01 level between the treatments excluding risk reduction (treatments 1,2&3) and the treatments including risk reduction (treatments 4&5) (t(234) = 2,838, p = ,005). According to these results hypothesis 8, stating that the use of risk reduction strategy will lead to an increase in consumers’ willingness to pay, is accepted.

(26)

26

Table 3: Planned Contrasts ln bid amounts treatment 1,2&3 vs. 4&5

Value of contrast Std. Error T Df Sig. (2-tailed)

0,680 0,240 2,838 234 ,005

5.2.3 Effects of different treatments on perceived uncertainty (h7)

Hypothesis 7 states that the use of risk reduction strategy will lead to a decrease in perceived performance uncertainty. To test this, a one-way independent ANOVA is carried out to compare the perceived uncertainty mean values across the five treatments. Levene’s statistic showed non-significant results for the uncertainty data, satisfying the homogeneity of variances assumption (F = 0,250, p = 0,909). The results of the ANOVA do not show a significant difference between the mean values (F(4, 215) = 0,678, p = ,608). Accordingly, the results do not suggest that there is a difference in perceived uncertainty between the five different treatments. The planned contrast (table 4) between the treatments without a risk reduction strategy (treatment1,2&3) and the treatments including a risk reduction strategy (treatment 4&5) did not show significant results either (t(215) = 0,772, p= ,441).

The perceived uncertainty data for the sample controlled for perceived newness satisfied the assumption of homogeneity of variances (F = 2,434, p = ,052). The ANOVA did not show significant results for the sample including only the respondents who perceived the product as new (F(4,99) = 1,909, p = ,115). The LSD post-hoc test (appendix C, table 9), however, did show a significant mean difference between treatment 5 and treatment 3 (p = ,032). These results suggest that adding a risk reduction to an induced prevention focus results in less perceived uncertainty. The planned contrast test (t(99) = ,407, p = ,685), however, did not show significant results either (table 4). These results suggest that there is no difference in the use of risk reduction strategy compared to no risk reduction strategy in perceived uncertainty. To conclude, hypothesis 7 is accepted only for induced prevention focus advertisements for innovative products, but not for the control- or promotion advertisements. Therefore, hypothesis 7 is partially supported.

Table 4: Planned Contrasts perceived uncertainty 1,2&3 vs. 4&5

Value of contrast Std. Error t Df Sig. (2-tailed) Not controlled for perceived

newness 0,320 0,419 0,772 215 ,441

Controlled for perceived

(27)

27 Controlled for perceived newness, the LSD post-hoc test (appendix C, table 8) showed significantly higher perceived uncertainty for treatment 3 compared to treatment 1 (p = ,044) and compared to treatment 2 (p = ,014). These results suggest that inducing prevention focus in an advertisement for innovative products results in more perceived uncertainty compared to no induced self-regulatory focus or compared to the induced promotion focus. Therefore, it can be concluded that inducing self-regulatory focus in an advertisement has an effect on perceived uncertainty. This is similar to hypothesis 4 that states that prevention focused consumers perceive more product performance uncertainty compared to promotion focused consumers. However this finding is based on induced self-regulatory focus and hypothesis 4 concerns chronic self-regulatory focus. When a risk reduction strategy is included, the induced promotion focus treatment actually had a lower level of uncertainty than the induced prevention focus treatment. This difference, however, was not significant (p = 0,392). Therefore, it can be concluded that when a risk reduction strategy is added, there seems to be no effect of induced self-regulatory focus on consumers’ perceived uncertainty.

5.2.4 Effects of perceived uncertainty on consumers’ willingness to pay (h5)

Hypothesis 5 states that product performance uncertainty will have a negative effect on willingness to pay. In order to test this, a one-way independent ANOVA was carried out to compare the mean ln bid values for people with different levels of perceived uncertainty. The results of the ANOVA do not show a significant difference between the mean values (F(46, 168) = 0,737, p = ,887). The planned contrast test between people with a high level of uncertainty and people with a low level of uncertainty did not reveal a significant difference either (t(215) = 0,403, p = ,687). Accordingly, the results do not suggest that the level of perceived uncertainty has an effect on consumers’ willingness to pay and thus hypothesis 5 is rejected.

5.2.5 Effects of chronic self-regulatory focus on perceived uncertainty (h4)

As stated in hypothesis 4, it is expected that prevention focused consumers will perceive more product performance uncertainty than promotion focused consumers. To test whether chronic self-regulatory focus has an impact on perceived uncertainty a simple linear regression is done. As mentioned before the higher the value of the variable chronic self-regulatory the more promotion focused the respondent is. The results suggest that chronic self-regulatory focus significantly predicts the level of perceived uncertainty, b = ,195, t(214) = 2,916, p = ,004. Chronic self-regulatory focus also explained a significant proportion of variance in the

(28)

28 level of perceived uncertainty, R2 = ,038, F(1,214) = 8,501, p = ,004. Since higher scores on the perceived uncertainty variable indicate lower perceived uncertainty, these results suggest that the more promotion focused a consumer is the less uncertainty it perceives. Based on these results, hypothesis 4 is accepted.

5.2.6 Moderating effect of chronic self-regulatory focus (h3, h9)

Hypothesis 3 states that the effect of induced self-regulatory focus on consumers’ willingness to pay will be moderated by chronic self-regulatory focus. In order to analyze this moderating influence, an interaction variable is created by multiplying the two independent variables together (Baron & Kenny, 1986). The two predictors and the interaction variable were entered into a simultaneous regression model (table 5). Model 1 including the two predictors is significant, F(2,13) = 5,142, p < ,01. Also model 2 with the interaction variable is significant, F(3,12) = 3,656, p < ,05. However, model 2 with the interaction between induced self-regulatory focus and chronic self-self-regulatory focus did not account for significantly more variance than just the induced self-regulatory focus and the chronic self-regulatory focus level by themselves, R2 change = 0,003, p = ,404. This indicates that there is no significant moderation between induced self-regulatory focus and chronic self-regulatory focus on consumers’ willingness to pay. Accordingly, hypothesis 3 is rejected.

Table 5: Regression Model Summary: dependent variable: ln bid amount

Model 1 Model 2

Effect B Std. Error B Std. Error

Constant 8,535** ,046 8,531** ,046

Induced promotion -0.326** ,106 -0,321** ,106

Chronic Self-regulatory focus -0,037 . ,031 -0,025 . ,034

Induced * Chronic -0,067 . ,080 Model 1 Model 2 F 5,142** 3,656* Df1;Df2 2;213 1;212 R2 ,046 ,049 R2 Change ,003 F Change 0,698 * p< ,05 ** p< ,01

Hypothesis 9 tests the moderating effect of chronic self-regulatory focus on the relationship between the use of a risk reduction strategy and consumers’ willingness to pay. The two predictors and the interaction variable were entered into a simultaneous regression model. Model 1including the two predictors is significant, F(2, 213) = 3,070, p < ,05. Model 2, however, including the interaction effect is not significant, F(3, 212) = 2,041, p > ,05.

(29)

29 Model 2 with the interaction between risk reduction and chronic self-regulatory focus therefore did not account for significantly more variance than just risk reduction and chronic self-regulatory focus level by themselves, R2 change < 0,001, p = ,911(table 6). These results indicate that there is no significant moderation between risk reduction strategy and chronic self-regulatory focus on consumers’ willingness to pay. Accordingly, hypothesis 9 is rejected.

Table 6: Regression Model Summary: dependent variable: ln bid amount

Model 1 Model 2

Effect B Std. Error B Std. Error

Constant 8,390** ,055 8,391** ,056

Chronic Self-regulatory focus -0,033 ,031 -0,036 ,040

Risk reduction 0,190* ,083 0,188* ,085

Chronic * Risk reduction 0,007 ,063

Model 1 Model 2 F 3,070* 2,042 Df1;Df2 2;213 1;212 R2 ,028 ,028 R2 Change ,000 F Change 0,012 * p< ,05 ** p< ,01

5.2.7 Mediating effect of product performance uncertainty (h6)

As stated in hypothesis 6, it was expected that product performance uncertainty mediates the effect of chronic self-regulatory focus on consumers’ willingness to pay. However, since hypothesis 1 and 5 were not accepted, mediation cannot exist. That is, according to Baron and Kenny (1986) in order for the possibility of mediation to exist there has to be a significant effect of the predictor on the outcome variable, the predictor has to have a significant relationship with the mediator and the mediator should be significantly related with the outcome. However, this study found no significant relations between chronic self-regulatory focus and willingness to pay (H1) and no significant relation between perceived uncertainty and willingness to pay (H5). Accordingly, hypothesis 6 is not supported.

The effects of self-regulatory focus and risk reduction strategy on consumers’ willingness to pay for innovative products