THE EFFECT OF THE TYPE OF SCARCITY MESSAGE ON PURCHASE INTENTIONS MODERATED BY FASHION INVOLVEMENT by MITCHELL MUNGROOP June 18, 2018

THE EFFECT OF THE TYPE OF SCARCITY MESSAGE ON PURCHASE INTENTIONS MODERATED BY FASHION INVOLVEMENT

by

MITCHELL MUNGROOP

THE EFFECT OF THE TYPE OF SCARCITY MESSAGE ON PURCHASE INTENTIONS MODERATED BY FASHION INVOLVEMENT

by

MITCHELL MUNGROOP

University of Groningen

Faculty of Economics and Business

ABSTRACT

The purpose of this study was to help online fashion retailers gain more insights, into the effectiveness of several scarcity appeals on purchase intentions. We argued that different levels of fashion involvement of consumers play a role because of the different ways consumers can respond to advertisements. What is the relationship between types of scarcity messages (limited time, limited quantity and vague scarcity) and purchase intentions and how is this moderated by fashion involvement? An experiment was conducted in a survey for Dutch consumers. The results showed that the effect of the scarcity messages was insignificant compared to the control group and to each other, the fashion involvement did not have a direct effect on purchase intentions. Also, fashion involvement did not positively influence the effect of scarcity messages on purchase intentions when fashion involvement was high. In the future, we recommend research with more product groups, more scarcity appeals and extra measures.

Keywords: Scarcity, Fashion Involvement, Purchase Intentions Research Theme: Persuasion to the limit!

PREFACE

Yours truly has a bachelor’s degree in Business Administration and this thesis in front of you is the product of the final part of my master study Marketing at the University of Groningen. I have been doing research and writing this thesis from February to June 2018.

Since I was little, I have been fascinated by persuasion techniques that influence consumer behavior. Whether it were advertisements on billboards or infomercials on television, I have always wondered why those advertisements worked as well as they did. That is why I was pleased to be selected for the ‘Persuasion to the limit!’ research theme.

I would like to express my gratitude to my thesis supervisor Dr. Keizer for his support, time and valuable input that helped me through the entire thesis process. Another special thanks to my fellow students who I had interesting conversations with about my thesis subject. These conversations made the writing of my thesis very enjoyable.

TABLE OF CONTENTS

INTRODUCTION 6

THEORETICAL BACKGROUND AND HYPOTHESES 8

METHODOLOGY 14

RESULTS 18

DISCUSSION 25

Findings 25

Managerial Implications 27

Limitations and Recommendations for Future Research 27

CONCLUSION 30

REFERENCES 31

APPENDIX A 36

INTRODUCTION

The past years have seen a growth in online fashion purchases in The Netherlands. In 2016, the online fashion expenditures have grown with 15% compared to the previous year (CBS, 2017). Also, since 2013 there has been an enormous growth of 75% in the total number of online pure fashion stores (Emerce, 2017). Due to this tremendous growth, the focus of research on what influences purchase decisions has moved recently from offline to online.

Recent research focused on the online behavior has discovered internal psychological factors of consumers. For instance, the affective states of consumers are a factor (Verhagen & Van Dolen, 2011). Next to that, Zhang, Prybutok, and Strutton (2007) researched the effect of subjective norms on purchase behavior during online transactions. Also, external factors were discovered e.g. that different offers and promotions positively influence the intention to purchase online (Dholakia, 2000; Ramaswamy & Namakumar, 2009), or according to a study in a physical shopping environment of Abratt and Goodey (1990), elevating sales through unplanned buying can be reached through exposure to in-store stimuli.

When we take a look at what predicts purchase intentions in the literature, scarcity appeal messages are fundamental (Inman, Peter & Raghubir, 1997; Aggarwal, Jun & Huh, 2011). For instance, Howard, Shu and Kerin (2007) studied scarcity appeals in advertisements in an offline context in 2004. They analyzed more than thirteen thousand retail display advertisements in four big American newspapers and concluded that the limited time scarcity appeal was the most used scarcity appeal in all these advertisements (84,4%). Considering that scarcity messages are very important indicators of purchase intentions and the huge growth in the online fashion industry, it is interesting to study what the influence is of these scarcity messages on online fashion purchases. While there is a general consensus among researchers that the manipulation of (perceived) scarcity has the potential to gain compliance (Inman et al., 1997), there is not much research about these tactics’ effects in an online environment. The purpose of this study is to help online fashion retailers in gaining more insights, into how effective several scarcity appeals are and thereby knowing what appeal has the potential to lead to higher purchase intentions.

scarcity message (e.g. for a limited time or until sold out, whichever is first) (Tan & Chua, 2014). These restrictions on sales promotions are inexpensive to inflict and can easily be extended when consumers respond well or can be stopped prematurely, by giving the reason (or excuse) that the stock has run out. These types of restrictions are commonly found in sales offers, for instance in the fashion industry.

Fashion retailers like Hennes & Mauritz (H&M) and Zara introduce fashion collections with rapid inventory turnover rate and by that, they create a scarcity atmosphere (Gupta, 2012). The perceived scarcity effects in physical fast fashion retail stores on consumer behavior have been researched in this offline context in a few papers e.g. about instore hoarding (Gupta, 2012), seller induced scarcity (Gupta, 2012) using instore signages with scarcity messages on them (Barnes & Lea-Greenwood, 2010) or reducing price competition by occasionally limiting the availability of products (Balachander & Farquhar, 1994).

This study is interested in the effect of scarcity messages on purchase intentions in an online environment. Since not every consumer is as involved with fashion as others, but everyone has to buy clothing, it is interesting to look at the differences in intentions to purchase when looking at sales promotions containing a scarcity message. Consumer that are highly fashion involved could see the buying of apparel differently than low fashion involved consumers and could therefore be more or less susceptible to advertisements. That is why we are taking the fashion involvement of customers into account when studying the relationship between the type of scarcity message and purchase intentions.

This study investigates the following main research question: “What is the relationship between type of scarcity message and purchase intentions and how is this moderated by fashion involvement?”.

The sub questions that will be answered in this study are the following: “Which scarcity message (quantity limit, time limit or vague scarcity message) has the highest impact on purchase intentions?” and “How does fashion involvement (high fashion involved consumer vs. low fashion involved consumer) moderate the relationship between type of scarcity message and purchase intentions?”

THEORETICAL BACKGROUND AND HYPOTHESES

Scarcity message types

In general, “people want more of what they can have less of” (Cialdini, 2001: p. 78). These scarcity effects are often explained by the commodity theory (Brock, 1968; Lynn, 1991). In essence, the theory explains that a commodity is valued more when it is hard to obtain or not even available (Brock, 1968). The commodity is an object that potentially can be obtained and is desirable, transferable and useful (Brock & Mazzocco, 2004; Eisend, 2008). When a consumer is successful in the purchase of a scarce commodity, he or she can create feelings of being chosen (Brock & Mazzocco, 2004), create a positive perception of the product by satisfying their need to be unique or distinct (Snyder & Fromkin, 1980), or get a feeling caused by the bandwagon reasoning (Worchel, Lee & Adewol, 1975): “The product must be great because everyone wants it”. Next to that, when a person’s freedom of choice is threatened or limited, that person wants to maintain those freedoms (as well the products associated with those freedoms) more than before according to the Psychological Reactance Theory (Brem & Brehm, 1981).

In short, scarcity can be effective because it is a cue to the value of the product (Cialdini, 1993), because products that are difficult to get are more attractive than products that require little to no effort therefore cueing to an item’s desirability (Lynn, 1992) or because of the relentless desire to defend freedom of choice (Brem & Brehm, 1981). It is therefore not a surprise that the scarcity appeal is frequently used in marketing communications.

Scarcity messages are often deliberatively used in marketing and advertising strategies. For example, a large part of the value of luxury items is deducted from their limited availability. The luxury brands like Gucci and Prada are not widely available at local stores, but only at selected flagship stores and dealers. Even in those stores, there are limited edition models. Undeliberate scarcity also exists, in the situation that there is a ban on certain products (e.g. ban on alcohol in the United States in the twenties. The desirability of banned products increases (Mazis, Settle and Leslie, 1973).

Morrison, 1979; Whitlark, Geurts & Swenson, 1993). Purchase intentions are not equal to making actual purchases, but it has been shown that the intention measures do have a predictive utility (Jamieson & Bass, 1989). It is reported in research that consumers who report that they have intentions to purchase have an actual higher purchase rate, than the customers who report that they do not have those intentions to purchase (Berkman & Gilson, 1978).

In this study, different type of scarcity messages will be used. These may lead to different competitive buying scenarios among consumers, because not all restrictions or in this case, scarcity appeals are the same (Inman, et al. 1997). Previous studies have distinguished the scarcity messages into two concepts, namely the limited time scarcity and the limited quantity scarcity (Rice & Keller, 2009). This present study includes both types of scarcity messages. Another researched type of message is the vague scarcity message, that is often used in an online social commerce setting (Tan & Chua, 2004), e.g. Groupon. For this reason, we included this type of message as well.

Limited time scarcity message (LTS). Firms encourage buyers to buy as many products as possible, within a limited time duration. In advertising, this technique is often called the deadline technique (Cialdini, 2008) in an ‘it’s now or never’ type of discount. It is these kinds of offers that result in suggesting exclusiveness and special edition products (Inman, Anil & Raghubir, 1997).

This technique is frequently used by department stores to inform their customers that a special proposition is only available for a limited time or will end on a specific time and date. The given time and date can be completely fictitious. The product can secretly still be sold after the deadline or the discount just stays the same. This could of course affect the trustworthiness of the company, but nonetheless, the scarcity effect could still do the job before the deadline. Following this reasoning, the hypothesis can be formulated:

H1-1 The limited time scarcity message will be more effective on purchase intentions, than the non-scarcity message.

apparel industry (Van Herpen, Pieters & Zeelenberg, 2014), mostly in clothing stores (e.g. Bijenkorf retail store with their “Dolle Dwaze Dagen”).

People are mostly attracted to resources that are scarce, especially if there is competition with others to obtain those resources (Cialdini, 2001). Thus, in this case, consumers have to compete with other consumers to purchase products, due to the fact that the number of products in the store declines as other consumers buy the product (Aggarwal et al., 2011). The fact that quantity is limited because it is popular amongst consumers serves as a social validation (Fennis & Stroebe, 2010).

The demand driven scarcity and the social validation can be assembled and result in a heuristic called ‘double whammy’ (also known as the 'double-whammy-demand-and-supply-based-scarcity-principle) (Fennis & Stroebe, 2010). Clothing store H&M is an example of a company that uses this heuristic. Multiple times a year they have a well-known ‘high fashion’ clothing designer create a collection at discounted prices. These clothing items are available in limitation and these high fashion items are also sold for low prices (which is uncharacteristic for high fashion items). The frenzy amongst the customers heightens the scarcity experience. This leads to the following hypothesis:

H1-2 The limited quantity scarcity message will be more effective on purchase intentions, than the non-scarcity message.

Vague scarcity message type (VS). Combining a time limit with the limited supply, without specifying how many items are left, creates a vague unspecified scarcity message (Tan & Chua, 2014). By combining the persuasive power of both scarcity appeals, as it is often used in social commerce sites as mentioned before, the freedom of choice is even more threatened and therefore may result in a higher purchase intention. For this type of scarcity message, this research chose to not specify the supply, because Aguirre-Rodriguez (2013) discovered that when the appeal is specific (“only 29 items left”) as opposed to formulated in vague terms (“while supplies last”), it lowers the persuasiveness of the ad.

H1-3 The vague scarcity message will be more effective on purchase intentions, than the non-scarcity message.

Since the vague scarcity message combines the persuasion techniques of both the limited time scarcity and the limited quantity scarcity messages, this research hypothesizes the following:

H2-1 The vague scarcity message will be more effective on purchase intentions, than the limited time scarcity message.

The main difference between the limited quantity scarcity and limited time scarcity is the fact that limited quantity scarcity involves both supply and demand scarcity while the limited time scarcity is only due to limited supply. The limited quantity scarcity message has next to the scarcity appeal, also a social validation appeal as aforementioned. The social validation appeal adds to the persuasion power of limited quantity scarcity message, that the limited time scarcity message has less of. Based on this, this research hypothesizes the following:

H2-2 The limited quantity scarcity message will be more effective on purchase intentions, than the limited time scarcity message.

Fashion Involvement

The literature revolving around involvement, argues that customers might be involved with products, purchases and advertising messages (Celsi & Olson, 1988; O’Cass 2000, 2004; Zaichkowsky, 1985). Involvement can be conceptually captured by the interaction between an individual consumer and an object (e.g. a product). Involvement was defined by Schiffman and Kanuk (1983) as the extent of concern for a particular purchase decision by a consumer, that the consumer considers to be important. More specifically in the case of fashion involvement, it can be defined as “the level of customer’s involvement with an object, situation or action, determined by the degree to which (s)he perceives the concept (of fashion) to be personally relevant” (Celsi & Olson, 1988, p. 211). These different involvement levels lead to different consumer responses (O’Cass, 2000, 2004; Zaichkowsky, 1985).

that there is a positive relationship between the level of fashion involvement and apparel purchase (Fairhurst, Good & Gentry, 1989). Therefore, this research assumes that fashion involvement has a positive influence on the intentions to purchase. That is why this research proposes the following hypothesis:

H3 Consumers that are more fashion involved have a higher purchase intention than consumers that are low fashion involved.

Joo Park et al. (2006) discovered that consumers with high fashion involvement are more likely to buy apparel that was recently launched or with a new style. High fashion involved consumers are more involved with the fashion product as it is more personally relevant for them (Celsi & Olson, 1988). When looking at an advertisement, it is probable that the depth of processing of the advertisement is dependent on how personal relevant the advertisement or the advertised product is (Fennis & Stroebe, 2016). So, fashion consumers could think more about the advertisement and the advertised fashion product, than low fashion involvement consumers. The scarcity message can be seen by the fashion involved consumers, as an argument e.g. ‘the fashion product is available in low quantity, so it must be great because it is popular amongst a lot of people and therefore of great quality’. This is not the case when low fashion involved consumers look at an advertisement and see the scarcity message as a heuristic e.g. ‘the product is available for a short time, so I must buy now before it is sold out’. The persuasion by using scarcity messages can lead to an even higher intention to purchase when shown to consumers that have a higher fashion involvement. Therefore, it is proposed that fashion involvement moderates the relationship between type of scarcity message and purchase intentions. This research hypothesizes the following:

H4-1 The effect of the limited time scarcity message on the purchase intentions, will be greater when the fashion involvement is high than when the fashion involvement is low.

H4-2 The effect of the limited quantity scarcity message on the purchase intentions, will be greater when the fashion involvement is high than when the fashion

H4-3 The effect of the vague scarcity message on purchase intentions, will be greater when the fashion involvement is high than when the fashion involvement is low. The conceptual model as depicted in Figure 1, was created to show the causal relationships between the variables that we are going to study.

METHODOLOGY

The data was collected by using a self-administered digital survey with a 4 (scarcity message: time scarcity limit vs. quantity scarcity limit vs. vague scarcity vs. non-scarcity) x 2 (fashion involvement: high vs. low) experiment with a between-subject design. In this research survey, the participants were exposed to one of the four treatment conditions.

Population & Sampling Method

Dutch consumers were the target population of this study. Because this study focused on purchase intentions, the minimum age to participate in this study was twelfth years old because that is the age the Statistics Netherlands (also known as the Centraal Bureau voor de Statistiek) uses for their shopping statistics (CBS, 2014). Also, when children are younger than twelfth, there is no certainty that they are responsible for their own purchase behavior.

We used a convenience sample because the survey respondents were conveniently accessible for this study. Although a convenience sample is not very representative, the primary goal for choosing this sampling method was to get as many respondents as possible in the given time constraint. The survey was promoted multiple times through social media, e.g. in various Facebook groups. Two gift vouchers were raffled among the participants that voluntarily filled in their email to stimulate responses.

Sample

Of the 204 respondents of the survey, 182 respondents finished the survey completely; 67 men (36,8%) and 115 women (63,2%), with an average age of 28 years (M = 28,11, SD = 7,71). 175 respondents (96,2%) have bought products online before and 168 respondents (92,3%) have worn sneakers before.

Operational Definitions and Development of Management Tool

Type of Scarcity Message. The concept fashion involvement was manipulated by showing every respondent the same advertisement for a new sneaker, but every respondent saw a different scarcity message: limited time, limited quantity, vague or non-scarcity message (control condition). The limited time scarcity message “Ends in 7 hours” was selected to resemble consumers looking at an advertisement of a one-day sale during the day. This is similar to the limited time messages on online shops, e.g. onedaysale.nl. The limited quantity message “Only 27 items left” was chosen because Aggarwal et al. (2011) discovered that their respondents indicated that a quantity of hundred units or less is appropriate to experience scarcity, in their study with wrist watches. We chose for an amount lower than hundred, namely twenty-seven units, to intensify the scarcity. And finally, the “Ends in 7 hours or when sold out” message was used as the vague scarcity message. This specific number of hours was chosen to be equal to the limited time scarcity message and the ‘when sold out’ part was chosen because as mentioned before, Aguirre-Rodriguez (2013) discovered that using vague terms is more persuasive than giving the specific number of items left. This wording of the scarcity message is similar to the messages used on social commerce sites, e.g. ibood.com.

In this study, an advertisement of a sneaker was chosen, because the same sneaker can be appealing to both men and women. A non-existing sneaker brand was used to control for existing brand knowledge (Hagtvedt, 2011). The advertisements were made using elements of actual advertisements. Aside from the scarcity messages, all other elements including the size of the text and the picture were exactly the same in all the conditions to control for confounds. The manipulations can be found in Appendix A.

Fashion Involvement. The concept fashion involvement was measured by using the fashion involvement scale from Joo Park, Young Kim & Cardona Forney (2006). The response scale ranged from 1 to 7 (1 = strongly disagree, 7 = strongly agree) and consisted of 4 items (M = 3,70, SD = 1,31, a = 0,788), for example “I usually have one or more outfits of the very latest style” and “I usually dress for fashion, not comfort, if I must choose between two”.

TABLE 1

Translation of scale “Purchase Intentions” by Taylor & Baker (1994)

Original This study

The next time I need the services of a ____, I will choose XYZ.

The next time I need sneakers, I will choose these.

If I had needed the services of a ____ during the past year, I would have selected XYZ.

If I had needed sneakers during the past year, I would have selected these. In the next year, if I need the services of a

____ I will select XYZ.

In the next year, if I need sneakers I will select these.

Next to that, several covariates were measured.

Impulsivity. The concept impulsivity was measured by using the scale from Rook & Fischer (1995). The response scale ranged from 1 to 7 (1 = strongly disagree, 7 = strongly agree). The scale consisted of 9 items (M = 3,52, SD = 0,99, a = 0,817) and one of the items was “I like to buy new clothing that just came out”.

Materialism. The concept materialism was measured by using the scale of Richins (2004) and consisted of 9 items (M = 3,74 SD = 0,95, a = 0,781). A 7-point Likert scale was used again (1 = strongly disagree, 7 = strongly agree). “Buying things gives me a lot of pleasure” was one of the items that measured this concept.

Fashion-Oriented Impulse Buying. The concept fashion-oriented impulse buying was measured by using scale from Joo Park, Young Kim & Cardona Forney (2006) and consisted of 3 items (M = 3,69, SD = 1,35, a = 0,775). An example of an asked item is “I buy clothing with a new style if I see it”.

In Appendix B, all items from the survey can be found. Method of Analysis

RESULTS

The sample consisted of 182 respondents, who were randomly allocated in four conditions. The control condition (M = 2.61, SD = 1.38) consisted of 52 respondents. The scarcity messages limited time scarcity (M = 2.38, SD = 1.58), limited quantity scarcity (M = 2.26, SD = 1.37) and vague scarcity (M = 2.51, SD = 1.24) had respectively 40, 47 and 43 respondents. When looking at the means of the different conditions, it is noticeable that means are on the lower side. Apparently, the intentions to purchase the sneakers were low. In Graph 1, the purchase intentions can be seen for high and low fashion involvement per condition.

Graph 1

TABLE 2

Linear Regression of the model Unstandardized

Coefficients Collinearity

B Standard Error t Sig. Tolerance VIF

Intercept 2.035 .565 3.601 .000 LTS -1.281 .910 -1.408 .161 .071 14.010 LQS 1.336 .832 1.605 .110 .076 13.104 VS -.460 .803 -.573 .567 .087 11.501 Fashion Involvement .155 .144 1.076 .283 .283 3.534 Interaction LTS .257 .225 1.145 .254 .067 15.028 Interaction LQS -.447 .209 -2.132 .034* .072 13.941 Interaction VS .122 .214 .571 .569 .090 11.120 * significant, p < .05

The limited time scarcity message was not significant, (t(1,169) = -1.41, p = .16). The limited quantity scarcity message did also not have a significant effect on purchase intentions (t(1,169) = 1.61, p = .11). Next, the vague scarcity message was insignificant (t(1,169) = -0.57, p = .57). All the scarcity messages did not have a significant effect on purchase intentions. The coefficient of fashion involvement was insignificant, (t(1,169) = 1.08, p = .28). According to this result, fashion involvement can be judged as not important in affecting purchase intentions, thereby rejecting hypothesis 3.

limited quantity scarcity message on the purchase intentions, will be greater when the fashion involvement is high than when the fashion involvement is low. Next, another linear regression was conducted with the entire model, but the covariates were also added (Table 3).

Table 3

Linear Regression of the model with covariates Unstandardized

Coefficients Collinearity

B

Standard

Error t Sig. Tolerance VIF

Intercept 1.454 1.086 1.339 .192 LTS -.984 .919 -1.070 .286 .068 14.631 LQS 1.470 .848 1.735 .085 .072 13.896 VS -.214 .811 -.264 .792 .083 11.981 Fashion Involvement .065 .166 .390 .697 .210 4.763 Interaction LTS .210 .226 .931 .353 .064 15.554 Interaction LQS -.457 .212 -2.150 .033* .068 14.666 Interaction VS .075 .213 .352 .725 .088 11.333 Fashion-Involved Impulsive Buying Behaviour .150 .107 1.399 .164 .477 2.098 Materialism .054 .124 .433 .665 .711 1.406 Impulsiveness .184 .115 1.619 .107 .776 1.288 Age .013 .014 .901 .369 .845 1.184 Sex -.395 .219 -1.803 .073 .889 1.125

Ordered online before -.060 .543 -.110 .913 .910 1.099

Sneaker wear? -.200 .390 -.512 .609 .918 1.089

* significant, p < .05

After looking at the complete model, the hypotheses were tested with ANOVA’s. Hypotheses Testing

To test if there is a difference in purchase intentions between the scarcity conditions, a One-way ANOVA was conducted. This analysis was not significant (F(3,178) = 0.68, p = .64), so a post-hoc test was not necessary. But to give more insights into the differences between the scarcity conditions, we did continue with a post-hoc Tukey HSD Multiple Comparison Test (Table 3).

Table 3

H1-1 To analyze if the limited time scarcity message (M = 2.38, SD = 1.58) was more effective on purchase intentions, than the non-scarcity message (M = 2.61, SD = 1.38) we conducted a One-Way ANOVA. The post-hoc revealed that the effect was not significant (p = .87). The limited time scarcity message and the non-scarcity did not significantly differ in effectiveness on purchase intentions.

Tukey HSD Multiple Comparison Test

H1-2 To analyze if the limited quantity scarcity message (M = 2.26, SD = 1.37) was more effective on purchase intentions, than the non-scarcity message (M = 2.61, SD = 1.38), we conducted a One-way ANOVA. The post-hoc results showed that the effect was insignificant (p = .61). The limited quantity scarcity message and the non-scarcity did not significantly differ in effectiveness on purchase intentions.

H1-3 To analyze if the vague scarcity message (M = 2.51, SD = 1.24) was more effective on purchase intentions, than the non-scarcity message (M = 2.61, SD = 1.38) we conducted a One-way ANOVA. The post-hoc revealed the insignificance (p = .99). The vague scarcity message and the non-scarcity did not significantly differ in effectiveness on purchase intentions. H2-1 To analyze if the vague scarcity message (M = 2.51, SD = 1.24) was more effective on purchase intentions, than the limited time scarcity message (M = 2.38, SD = 1.58), we conducted a One-way ANOVA. When we look at the post-hoc results, it is revealed that the effect was not significant (p = .98). The vague scarcity message and the limited time scarcity message did not significantly differ in effectiveness on purchase intentions.

H2-2 To analyze if the limited quantity scarcity message (M = 2.26, SD = 1.37) was more effective on purchase intentions, than the limited time scarcity message (M = 2.38, SD = 1.58), we conducted a One-way ANOVA. This analysis was not significant (p = .98). The limited quantity message and the limited time scarcity message did not significantly differ in effectiveness on purchase intentions.

H3 To test if fashion involvement has a main effect on the purchase intentions, we conducted a One-way ANOVA. In hypothesis 3, it was argued that consumers that are more fashion involved have a significantly higher purchase intention, than low fashion involved consumers. The main effect was not significant (F(1,180) = 1.51, p = .22). This was also visible in the regression model in Table 2 and 3 as mentioned before. There is no significant difference between higher or lower fashion involvement on purchase intentions.

Next, we performed a Two-way ANOVA per hypothesis.

performed. The Two-way ANOVA revealed that the main effect of the limited time scarcity versus the non-scarcity message did not have a significant effect on purchase intentions (F(1,88) = 0.66, p = .57), nor did it reveal a significant interaction effect with fashion involvement (F(1,88) = 1.676, p = .20). The effect of the limited time scarcity message on the purchase intentions, is not significantly greater when the fashion involvement is high than when the fashion involvement is low.

H4-2 To analyze if the effect of the limited quantity scarcity message (M = 2.26, SD = 1.37) on the purchase intentions, will be greater when the fashion involvement is high (M = 2.31, SD = .24) than when the fashion involvement is low (M = 2.56, SD = .19), a two-way ANOVA was performed. The Two-way ANOVA revealed that the main effect of the limited quantity scarcity message on purchase intentions was not significant (F(1,95) = 1.02, p = .50), nor did it reveal a significant interaction effect with fashion involvement (F(1,95) = 1.036, p = .31). The effect of the limited quantity scarcity message on the purchase intentions, is not significantly greater when the fashion involvement is high than when the fashion involvement is low. This is a different outcome than we found in the regression, where we found a significant result. The difference in the outcome of this particular hypothesis between the regression and ANOVA could be because of amount of data used in the analysis. In the linear regression, the whole model was tested and in the Two-way ANOVA, only the specific data of the limited quantity scarcity condition was used. Another explanation is that in the regression, the continuous results were used where in the Two-way ANOVA, the dummy coded fashion involvement groups were used to compare the groups to each other. The dummy coded variables remove all the nuances in the data; the participants are either in one or the other group. In the continuous data in the regression, there are variances.

TABLE 4

Hypotheses testing with ANOVA’s

Hypothesis Result

H1-1 The limited time scarcity message will be more effective on purchase intentions, than the non-scarcity message.

Rejected H1-2 The limited quantity scarcity message will be more effective on

purchase intentions, than the non-scarcity message.

Rejected H1-3 The vague scarcity message will be more effective on purchase

intentions, than the non-scarcity message.

Rejected H2-1 The vague scarcity will be more effective on purchase intentions, than

the limited time scarcity message.

Rejected H2-2 The limited quantity scarcity will be more effective on purchase

intentions, than the limited time scarcity message.

Rejected H3 Consumers that are more fashion involved have a higher purchase

intention than consumer that are low fashion involved.

Rejected H4-1 The effect of the limited time scarcity message on the purchase

intentions, will be greater when the fashion involvement is high than when the fashion involvement is low.

Rejected

H4-2 The effect of the limited quantity scarcity message on the purchase intentions, will be greater when the fashion involvement is high than when the fashion involvement is low.

Rejected

H4-3 The effect of the vague scarcity message on purchase intentions, will be greater when the fashion involvement is high than when the fashion

involvement is low.

DISCUSSION

The purpose of this study was to take a look at the impact of the type of scarcity message on purchase intentions and how this is moderated by fashion involvement. It was predicted that the scarcity messages (limited time scarcity, limited quantity scarcity and vague scarcity) were more effective on purchase intentions, than the non-scarcity message. However, the results showed us that none of the scarcity messages were more effective on purchase intentions, than the non-scarcity message. The limited time scarcity message and the non-scarcity did not significantly differ in effectiveness on purchase intentions. What could explain this outcome is that the time limit presented in the advertisement (“Ends in 7 hours”) was not scarce enough and the participants did not see the message as a deadline. Also, the limited quantity scarcity message and the non-scarcity did not significantly differ in effectiveness on purchase intentions. An explanation might be that the quantity (“Only 29 items left”) was not limited enough and therefore did not result in the sense of exclusivity, value and quality (Van Herpen, Pieters & Zeelenberg, 2009) or popularity amongst consumers (Fennis & Stroebe, 2010). We chose the “Only 29 items left” message, because Aggarwal et al. (2011) discovered that their participants experienced scarcity in their study when the number of items in the advertisement, was hundred or less. Their study was focused on the wrist watch product category, so that could be an alternative explanation why in this study with sneakers, the number of items did not result in the experience of scarcity. Also, the vague scarcity message and the non-scarcity message did not significantly differ in effectiveness on purchase intentions. This was different from what was expected, because in this case it could be that the “Ends in 7 hours or when sold out” message was not threatening enough to the freedom of choice of the participants.

We will continue with hypotheses that involve the moderator. It was argued that consumers that are more fashion involved have a higher purchase intention than consumers that are low fashion involved. The results showed that there is no significant difference in effect between higher or lower fashion involvement on purchase intentions. Next to that, it was predicted that the effect of the scarcity messages (limited time scarcity, limited quantity scarcity and vague scarcity) on the purchase intentions, will be greater when the fashion involvement is high than when the fashion involvement is low. The results, when testing with ANOVA’s, tell us that the effect of the scarcity messages on purchase intentions were not significantly different when the fashion involvement was high than when the fashion involvement was low. According to our study, fashion involvement does not play a role in the relationship between the different types of scarcity messages and the purchase intentions. When we tested the whole model in a regression analysis, there was one significant outcome. We expected that the effect of the limited quantity scarcity message on the purchase intentions, would be greater when the fashion involvement is high than when the fashion involvement is low. The results showed that effect of the limited quantity scarcity message on the purchase intentions, was not significantly greater when the fashion involvement was high than when the fashion involvement was low. Apparently, this effect is the other way around. It turns out that in this case, when the consumer scores high on fashion involvement, the intent to purchase lowers which is opposed to our prediction. An alternative explanation for this could be that consumers hold a different inference about scarcity than we argued. This other possible inference could be that consumers link scarcity to the fact that a lot of other consumers already have bought this particular item, which could lower their purchase intentions. That other consumers may have bought this product can be less appealing to fashion involved consumers who strive to have the newest fashion themselves. Apparently, it could matter if scarcity is demand or supply driven (Van Herpen, Pieters & Zeelenberg, 2014). This has, to the best of our knowledge not been researched before in an online setting and in relation with fashion involvement. It would be a great starting point for more research, because the difference in effect of demand based or supply based scarcity on purchase intentions in relation to fashion involvement is unclear.

Managerial Implications

The aim of this research was to gain more insights for online retailers in the sale of apparel. More specifically, what the effectiveness was of certain scarcity appeals and with that knowing which of those appeals have the capacity to persuade consumers to buy. Unfortunately, we did not find any evidence for our predictions. When we compared the effectiveness of the scarcity messages with the non-scarcity message, we did not find a significant difference. Another lack of significant results was found when comparing the effectiveness of the limited quantity scarcity and vague scarcity message to the limited time scarcity. Since it has been proven in the literature that scarcity appeals do work, we suggest that more extensive research has to be done to find out more about to effect of scarcity messages to give concrete advice to managers. We do not have substantial evidence for our predictions that included fashion involvement, but we did get insights into the role that fashion involvement plays in the relationship between scarcity messages and purchase intentions in an online environment. On a managerial level, this means that it could be possible that the fashion involvement of the target audience matters, which we can see in testing the hypothesis 4-2 (in the regression). When testing this hypothesis, we found out that when the fashion involvement of consumers is high, that the limited quantity scarcity lowers the intentions to purchase. We suggest that managers take this insight into account when targeting high fashion involved consumers, when considering using a limited quantity scarcity message in their advertisement. But more research has to be done to give more practical insights to managers.

Limitations and Recommendations for Future Research

First of all, the sample used in the research was a convenience sample. As mentioned in the methods section, this is not ideal because questions can be raised about the representability of the sample. This research wanted to give information about Dutch consumers, so it is very likely that the sample was too small to give generalizable information about a group of consumers of this size.

(Grebitus, Lusk & Nayga jr., 2013). Having a high purchase intention does not immediately translate into actual buying behavior. It is recommended for future research to measure the ‘real’ purchases in a more real-life setting, to bolster the confidence in our results. This could be done for instance, by giving participants a real-life budget to spend on a manipulated version of an online retailer’s website with scarcity messages on the different products. We expect that the respondents might be more susceptible to the scarcity when really experiencing the browsing through products part, using their own money and being able to buy the products.

Third, another practical limitation in this research was that only sneakers were tested in this study. To extend this study, it is recommended to use more categories of apparel to see if there are differences between the different apparel product categories, since it can be expected that different product categories can lead to different buying behaviors.

Fourth, the use of the specific scarcity messages was also a limitation. For instance, when trying to measure scarcity with a limited time scarcity message, there were a lot of different ways the scarcity message could be formulated. In this study, the message was “Ends in 7 hours” (as can be seen in Appendix A), to resemble a one-day sale, as mentioned before. The number of hours could be varied in future research and compared to each other. The same goes for the Limited Quantity Scarcity message, where in this study “Only 29 items left” (as can be seen in Appendix A) was used. In future research, the number of products left could be varied.

Fifth, we would recommend that items to measure the perceived scarcity should be added to research on this topic in future research. In this study, we allocated participants to different scarcity groups. After that, we measured the dependent variables. A flaw of this approach was that we do not know anything about how scarcity was perceived. Allocating participants to nominal groups does not give us information about what the participants experience as scarce and the individual differences in their experiences. Therefore, perceived scarcity could possibly be a better predictor for purchase behavior. That is why we recommend including this concept in the future research.

CONCLUSION

REFERENCES

Abratt, R., & Goodey, S. D. (1990). Unplanned buying and in-store stimuli in supermarkets. Managerial and Decision Economics, 11(2), 111-121.

Aggarwal, P., Jun, S. Y., & Huh, J. H. (2011). Scarcity messages: A consumer competition perspective. Journal of Advertising, 40, 19–30.

Aguirre-Rodriguez, A. (2013). The effect of consumer persuasion knowledge on scarcity appeal persuasiveness. Journal of Advertising, 42(4), 371-379.

Balachander, S., & Farquhar, P. H. (1994). Gaining more by stocking less: A competitive analysis of product availability. Marketing Science, 13, 3–22.

Barnes, L., & Lea-Greenwood, G. (2010). Fast fashion in the retail store environment. International Journal of Retail & Distribution Management, 10, 282–300.

Berkman, H. W., & Gilson, C. G. (1978). Consumer behaviour: Concepts and strategies. Encino, CA: Dickenson Publishing.

Brehm, S. S. & Brehm, J. W. (1981). Psychological Reactance: A theory of freedom and control. New York, NY: Academic Press Inc.

Brock, T. C. (1968). Implications of commodity theory for value change. Psychological foundations of attitudes, 243-275.

Brock, T. C., & Mazzocco, P. J. (2004). Responses to scarcity: A commodity theory perspective on reactance and rumination. Motivational analyses of social behavior: Building on Jack Brehm's contributions to psychology, 129-148.

CBS. (2017, September 4). Minder fysieke, meer online kledingwinkels [Press release]. Retrieved from https://www.cbs.nl/nl-nl/nieuws/2017/35/minder-fysieke-meer-online-kledingwinkels

Celsi, R. L., & Olson, J. C. (1988). The role of involvement in attention and comprehension processes. Journal of Consumer Research, 15(2), 210–224.

Cialdini, R. B. (2001). Harnessing the science of persuasion. Harvard Business Review, 79(9), 72-81.

Cialdini, R. B. (2008). Influence: Science and practice (5th ed.). Boston, MA: Pearson Education.

Dholakia, U. M. (2000). Temptation and resistance: An integrated model of

consumption impulse formation and enactment. Psychology & Marketing, 17(11), 955-982. Dittmar, H., Beattie, J., & Friese, S. (1996). Objects, decision considerations and self-image in men's and women's impulse purchases. Acta psychologica, 93(1-3), 187-206.

Eisend, M. (2008). Explaining the impact of scarcity appeals in advertising. Journal of Advertising, 37, 33–40.

Emerce. (2017, August 2). Verkoop online mode met 15 procent gestegen [Press release]. Retrieved from https://www.emerce.nl/nieuws/verkoop-online-mode-15-procent-gestegen

Fairhurst, A. E., Good, L. K. & Gentry, J. W. (1989). Fashion involvement: an instrument validation procedure. Clothing and Textiles Research Journal, 7(3): 10-14.

Grebitus, C., Lusk, J. L., & Nayga jr. R. M. (2013). Explaining differences in real and hypothetical experimental auctions and choice experiments with personality. Journal of Economic Psychology, 36, 11-26.

Gupta, S. (2013). The psychological effects of perceived Scarcity in a retail Setting and its impact on consumer buyer behavior (doctoral dissertation). Retrieved from

https://business.unl.edu/academic-programs/departments/marketing/about/robert-mittelstaedt-doctoral-symposium/docs/2012_SymposiumProceedings.pdf#page=17

Hagtvedt, H. (2011). The impact of incomplete typeface logos on perceptions of the firm. Journal of Marketing, 75(4), 86-93.

Inman, J. J., Peter, A. C., & Raghubir, P. (1997). Framing the deal: The role of restrictions in accentuating deal value. Journal of Consumer Research, 24(1): 68-79.

Howard, D. J., Shu, S. B., & Kerin, R. A. (2007). Reference price and scarcity appeals and the use of multiple influence strategies in retail newspaper advertising. Social Influence, 2, 18–28.

Jamieson, L. F. & Bass, F. M. (1989). Adjusting stated intention measures to predict trial purchase of products: A comparison of models and methods. Journal of Marketing Research, 26(8): 336-45.

Joo Park, E., Young Kim, E., & Cardona Forney, J. (2006). A structural model of fashion-oriented impulse buying behavior. Journal of Fashion Marketing and Management: An International Journal, 10(4), 433-446.

Juster, T.F. (1966). Consumer buying intentions and purchase probability: an

Lynn, M. (1992). Scarcity’s enhancement of desirability: The role of naive economic theories. Basic and Applied Social Psychology, 13(1), 67-78.

Mazis, M. B., Settle, R. B. & Leslie, D. C. (1973). Elimination of phosphate detergents and psychological reactance. Journal of Marketing Research, 10, 390–395.

Morrison, D. G. (1979). Purchase intentions and purchase behavior. Journal of Marketing, 43(2), 65-74.

O’Cass, A. (2000). An assessment of consumers’ product, purchase decision, advertising and consumption involvement in fashion clothing. Journal of Economic Psychology, 21(5): 545–576.

O’Cass, A. (2004), Fashion Clothing Consumption: Antecedents and Consequences of Fashion Clothing Involvement. European Journal of Marketing, 38(7), 869–880.

Patton, J. H., Stanford, M. S., & Barratt, E. S. (1995). Factor structure of the Barratt impulsiveness scale. Journal of Clinical Psychology, 51, 768-774.

Ramaswamy, V. S., & Namakumari, S. (2009). Marketing management: Global perspective, Indian context (4th ed.). New Delhi, India: Macmillan.

Rook, D.W. (1987). The buying impulse. Journal of consumer research, 14(2), 189-199.

Rice, S. & Keller, D. (2009). Automation reliance under time pressure. International Journal of Cognitive Technology, 14(1), 36.

Richins, M. L. (2004). The material values scale: Measurement properties and development of a short form. Journal of Consumer Research, 31(1), 209-219.

Tan, S. J., & Hwang Chua, S. (2004). “While stocks last!” Impact of framing on consumers' perception of sales promotions. Journal of Consumer Marketing, 21(5), 343-355.

Taylor, S. A., & Baker, T. L. (1994). An assessment of the relationship between service quality and customer satisfaction in the formation of consumers' purchase intentions. Journal of Retailing, 70(2), 163-178.

Van Herpen, E., Pieters, R. & Zeelenberg, M. (2009). When demand accelerates demand: Trailing the bandwagon. Journal of Consumer Psychology, 19(3): 302–12.

Van Herpen, E., Pieters, R. & Zeelenberg, M. (2014). When less sells more or less: The scarcity principle in wine choice. Food Quality and Preference, 36, 153–160.

Verhagen, T., & van Dolen, W. (2011). The influence of online store beliefs on consumer online impulse buying: A model and empirical application. Information & Management, 48(8), 320-327.

Whitlark, D. B., Geurts, M. D. & Swenson, M. J. (1993). New product forecasting with a purchase intention survey. Journal of Business Forecasting, 12(3), 18-21.

Worchel, S., Lee, J., & Adewole, A. (1975). Effects of supply and demand on ratings of object value. Journal of Personality and Social Psychology, 32, 906–914.

Zaichkowsky, J. L. (1985). Measuring the involvement construct. Journal of Consumer Research, 12(3): 341–352.

APPENDIX A

The same advertisement for sneakers with different scarcity messages: Advertisement with limited time scarcity message

Advertisement with vague scarcity message

APPENDIX B

Fashion involvement (Joo Park, Young Kim, & Cardona Forney, 2006). 1. I usually have one or more outfits of the very latest style

2. An important part of my life and activities is dressing smartly

3. I am interested in shopping at boutique or fashion specialty stores, rather than at department stores for my fashion needs.

4. I usually dress for fashion, not comfort, if I must choose between two. Impulsiviness (Rook & Fischer, 1995)

1. I often buy things spontaneously.

2. “Just do it” describes the way I buy things. 3. I often buy things without thinking.

4. “I see it, I buy it” describes me.

5. “Buy now, think about it later” describes me.

6. Sometimes I feel like buying things on the spur of the moment. 7. I buy things according to how I feel at the moment.

8. I carefully plan most of my purchases.*

9. Sometimes I am a bit reckless about what I buy.

Fashion-oriented impulse buying (Joo Park, Young Kim, & Cardona Forney, 2006) 1. I buy clothing with a new style if I see it

2. I buy to try out a garment with a new feature 3. I like to buy new clothing that just came out Materialism (Richins, 2004)

1. I admire people who own expensive homes, cars, and clothes. 2. I try to keep my life simple, as far as possessions are concerned. * 3. My life would be better if I owned certain things I don’t have. 4. Buying things gives me a lot of pleasure.

5. I’d be happier if I could afford to buy more things. 6. I like to own things that impress people.

8. It sometimes bothers me quite a bit that I can’t afford to buy all the things I’d like. 9. The things I own, say a lot about how well I’m doing in life.