Essay: Sensory marketing

1. Introduction
In today’s stores consumers are overwhelmed by manipulated sensory experiences. For instance, music is being played, scents linger in the air and specific light and colours are used with only one purpose: to trigger purchase behaviour. For example, the well-known brand Abercrombie & Fitch sprays their own fragrances in their stores, which according to the brand is congruent with their brand identity; a confident, bold and masculine attitude. The company also applies loud club music throughout its stores. They argue that younger people can withstand loud music longer, while older ones may run from it, enabling the stores to maintain a more youthful clientele and a ‘fresher’ image. This form of marketing is called sensory marketing and is defined as “marketing that engages the consumers’ senses and affects their behaviors” (Krishna 2012, p. 332).
Sensory marketing has drawn a lot of attention during the last decade. Retailers and manufacturers recognized that multi-sensory brand experiences increase perceived product value, induce shoppers to stay longer at a store and ultimately to spend more money (Krishna, 2011). A specific sub area of sensory marketing that has become increasingly popular nowadays is that of ‘sensory metaphors’. This relatively young research area shows that even subtle, incidental bodily experiences can unconsciously affect thoughts, social perception, attitude, inference and judgement. The power of sensory metaphors is overwhelming: Williams and Bargh (2008) showed that when a subject held a warm drink instead of a colder one, the subject perceived other people as more warm and caring. This example shows that experiencing physical warmth subconsciously promotes interpersonal warmth. Another study has shown that sitting on a soft chair during a sales pitch or negotiation leads to earlier customer agreements. Furthermore, judges sitting on a hard chair will impose higher punishments to criminals in a courtroom (Cherkasskiy et al., 2012).
Since the majority of the sensory metaphor experiments had a social psychology purpose, it would be interesting to examine whether sensory metaphors can also be applied to consumers responses in shops and retail stores for marketing purposes. Therefore, the main research question is: “To what extent do both physical warmth of a drink and the softness of a chair subconsciously influence consumer responses?”
The objective of this study is to investigate the influence of the sensory metaphorical associations of physical warmth (warmth and coldness) and haptic experiences (softness and hardness), and their combination, on consumer responses in a retail environment. The practical relevance of this study is to help retailers to enhance the customers’ attitude, perceived social orientation and purchase intention as parts of consumer responses.
 
2. Theoretical framework
Understanding the process of sensory metaphors on consumer behaviour starts with the underlying processes of sensation, perception and grounded cognition: perception that affects cognition. This paragraph will first clarify the basic concepts of sensory marketing and will then go deeper into sensory metaphors. Thereafter, the independent and dependent variables within this experiment will be clarified and hypotheses will be drawn.
2.1 An introduction to sensory marketing
The two main processes that underlie the way how we interpret the world around us are sensation and perception. Sensation is the stimulation of the senses. Perception is the process of acquiring, recording, interpreting, selecting and organizing sensory information (Matlin, 1988). Thus at sensation, the senses react on the stimulation, whereas the stimulation is interpreted by perception which results in an particular perceptual experience. According to Krishna (2012), the understanding of perception is essential for sensory marketing. She defines sensory marketing as “marketing that engages the consumers’ senses and affects their perception, judgment and behavior” (p. 332).
Sensory marketing is a branch of marketing based on the relationship between the consumers’ senses (vision, audition, smell, touch and taste) and their behaviour. Several marketing channels make use of it, e.g. commercials and printed advertising. As mentioned in the introduction, Abercrombie & Fitch make use of a sensory marketing strategy that combines specific scents and music in their stores in order to sell more. Actually, almost every retail store makes use of sensory marketing: specific music is played almost everywhere. Other types of chains and companies also make use of it; from airlines to hotel chains and from hospitals to restaurants. The Holiday Inn hotels chain for examples uses ‘relaxed’ scents combined with suitable music to keep people in their rooms, lobbies and bars longer.
2.2 Grounded cognition — perception that affects cognition
We know that the mind is connected to the body, but that it also occurs vice versa is relatively new. According to the theory of grounded cognition by Barsalou (2008), all types of memory are directly related to the senses. Thoughts, feelings, and behaviors are grounded in bodily interaction with the environment and cognitive activity is grounded in earlier sensory experience. Barsalou argues that three components constitute grounded cognition, namely: bodily states, situated actions and mental simulations. Various other scientists in marketing and psychology research have supported the theory grounded cognition (Hung & Labroo, 2011; Labroo & Nielsen, 2010; Mazar & Zhong, 2009; Williams & Bargh, 2008).
Bodily state refers to cognition that is affected by an unmoving physical condition that one is in. In all these situations, one holds a particular posture which results in certain behaviors and thought processes. For example, a study by Strack, Martin, & Stepper (1988) showed that facial activity has affected the funniness ratings of cartoons, by having the subjects hold a pen in in their mouths a particular way. The subjects’ smile muscles were stimulated by holding a pen horizontally in the mouth. This example shows that earlier experiences with smiling were grounded in thought, since smiling means liking, fun and pleasure. Another example of the effect of bodily states on cognition is that when subjects wore heavy backpacks, they judged hills as being steeper and distances as longer than when they did not (Proffitt, Stefanucci, Banton, & Epstein, 2003). In these two examples, one holds a particular bodily state which results in a certain perception that affects thoughts and behaviour.
Situated actions refer to cognition impacted by movements when some parts of the body are moved, but not the whole body (as with bodily states). An example of a situated action that affects cognition is provided by a study by Wells and Petty (1980). They showed that vertical and horizontal head movements impacted agreement and disagreement by shaking and nodding heads while listening to the content of a radio broadcast. This example shows that earlier associations with head nodding and shaking has grounded in thought, since nodding means disliking and shaking means liking. Another study that illustrates the impact of situated actions on cognition was conducted by Cacioppo, Priester, and Berntson (1993). The results of their experiment showed that Chinese ideographs were rated more positively when participants flexed their arms during evaluation than when they extended their arms. They argued that when grasping attractive objects, extending the arm is temporally strongly associated with approach by acquiring or consumption of these objects. On the other hand, flexing the arm resulted in avoidance. The reason behind the association between extending the arm and approach on the one hand, and flexing the arm on the other hand is caused by lifelong experiences with such contingencies. These movements and their associations are grounded in our cognition.
The third component of grounded cognition is ‘metal simulations’, which suggest that mental imagery drive cognition. Several studies provided evidence for such mental simulation, whereby imagening sensory perceptions leads to the activation of corresponding regions of the brain. For example, the imagining of hearing a law mower leads to an activation of the brains’ auditory cortex (Zatorre & Halpern, 2005). Aditionally, reading words associated with strong smells like ‘vanilla’ or ‘anise’ activates the brains’ olfactory cortex (González et al., 2006) and viewing images of an apple pie activates the brains’ taste cortices (Simmons, Martin, & Barsalou, 2005). Recently, Elder and Krishna (2012) showed that different visual depictions of a product (e.g. in a webshop) can result in less or more mental simulation of using the product which could affect the purchase intention. For example, a spoon with to the right of the soup bowl (versus left) results in a higher mental simulation and higher purchase intention for right-handed people.
2.3 Sensory metaphors and grounded cognition
The way we think, what we experience and what we do everyday is very much a matter of metaphors. According to Lakoff and Johnson (1980), metaphors are more than mere language and literary devices, they are conceptual in nature and represented physically in the brain. For example, a bad relationship is described as a ‘distant’ one, whereas a good relationship is described as a ‘close’ one, because we are often physically near people we like and physically distant from people we dislike. The combination of bodily states and metaphors figure prominently in poems and songs, for example ‘the heartbroken man’ and ‘her eyes are the sea’. A popular specific sub area of sensory marketing based on grounded cognition is ‘sensory metaphors’. Many studies have provided scientific evidence for the role of metaphors in human thought. For example, the well known idiom ‘something smells fishy’ is metaphorically related to social suspisciousness. People use that metaphor when they conclude that something is wrong. Various studies by Lee and Schwarz’s (2011) showed that fishy smells can arouse social suspiciousness in people in a subconscious way.
Another example of a sensory metaphor is verticality. Ostinelli, Luna, and Ringberg (2014) found that upward movements have a positive effect on one’s self esteem, since upward movements make people feel better about themselves. From earlier experiences, people have learned that upward movements are positively related and downward movements are negatively related, e.g., thumbs up is good and thumbs down is bad. When one is ‘down’, one is more inclined to look down.
Zhong and Liljenquist (2006) found that obscene thoughts increased the urge of cleansing. They also showed that cleaning one’s hands disabling guilt on subsequent behavior. Later studies found that the influence of physical cleansings is not limited to obscene thoughts and behaviors; instead, physical cleansings seem to metaphorically eliminate various traces from the past. An example of a well known idiom is: ‘wash hands of’, e.g. “I washed my hands of it long ago”.
2.4 Consumer responses
The abovementioned examples of sensory metaphors show that even subtle, incidental physical experiences can unconsciously affect thoughts about metaphorically related targets. The majority of the conducted sensory metaphor experiments are related to social psychology, a branch of psychology that deals with social interactions, including their origins and their effects on the individual (person to person). It would be interesting to examine whether these sensory metaphors can also be applied to companies or retail stores. In todays stores many (sensory) marketing activities are already being applied: music is played, specific colors and light and sometimes even scents are used. These sensations will create a certain perception of the environment of the store. However, we believe that the perception of a store can be enhanced by making use of sensory metaphors that affect the consumer responses in an unconscious way.
In order to measure the effect of sensory metaphors on consumer responses, some basic but important aspects of marketing communication research on consumer responses will have to be measured. First, we want to examine whether the general attitude of the respondents will be subconsciously influenced by the metaphorical association of sensory metaphors. The ‘general’ part of this ‘general attitude’ variable refers to the general properties of a company that will be measured. Since ‘attitude’ is one of the most popular constructs in marketing communication research, many measurements and scales are available to examine attitudes (Olson, Zanna, & Mark, 1993). Additionally, we want to examine whether consumers can perceive a company or store as more social after being infleunced by sensory metaphors. Therefore, the second dependent variable in this experiment is ‘the consumers’ perception of the company’s social orientation’, abbreviated to ‘company’s social orientation’. The third and last dependent variable for this experiment is ‘consumer purchase intention’, abbreviated to ‘purchase intention’. Purchase intention is a frequently used construct in both consumer behaviour research and marketing research and can be described as an individual’s conscious plan to make an effort to purchase something (Spears & Singh, 2004).
2.5 Manipulating sensory metaphors
In order to measure the effect of sensory metaphors on the dependent variables: ‘the consumers’ general attitude towards the company’, ‘the consumers’ perception of the company’s social orientation’, and ‘consumer purchase intention’, two sensory metaphors regarding warmth and haptics were selected. They were chosen for further research since their power as sensory metaphors was proven by other studies. This paragraph will further explain the power and associations of the selected sensory metaphors within this experiment.
2.5.1 Warmth metaphors
A common used metaphor in daily life is that of warmth and cold as a personality trait. Perceiving someone as warm or cold entails a broad understanding that focuses on a certain degree of socialilty. We describe people as warm when we perceive them as social, good-natured, friendly, helpful and trustworthy, and on the other hand as cold when we perceive them as unsocial, unfriendly, deceitful and unreliable (Rosenberg, S. et al. (1968), fiske cuddy glick, 2008). The power of this metaphor is also demonstrated as a sensory metaphor by Williams and Bargh (2008). Subjects of their experiment briefly held a cup of hot coffee or a cup of iced coffee after which they had to fill in a personality impression questionnaire. As a result, the subjects who held the warm cup of coffee had a higher perceived social warmth in other people than the subjects who held the iced coffee. Another study by Williams and Bargh (2008) concluded that when holding a warm object instead of a colder one, people will behave in a more socially warm and caring way, for example by rather choosing a gift for their friends than for themselves. The subjects who held a cold object were more greedy and in 75% of the cases they chose a gift for themselves.
The abovementioned studies showed that incidental experiences with physical warmth turned into social warmth in a subconscious manner. How could warm objects produce the same affective states as a ‘warm’ person? According to Asch (1946), most abstract concepts in psychology are metaphorically based on concrete physical experiences and the affective responses are stored together in memory. As a result, the feelings of warmth when one holds a hot cup of coffee or takes a warm bath might activate memories of other feelings associated with warmth (trust and comfort), because of earlier experiences with caretakers who provided warmth, shelter, safety, and nourishment. Because of these frequent early life experiences with the trustworthy caregiver, a close mental association usually develops between the concepts of physical warmth and psychological warmth. This research by Asch has revealed that the insular cortex is implicated in processing both the physical and the psychological versions of warmth information. For these theoretical and empirical reasons, we hypothesize that mere tactile experiences of physical warmth should activate concepts or feelings of interpersonal warmth. Moreover, this temporarily increased activation of interpersonal warmth concepts should then influence, in an unintentional manner, judgments of and behavior toward other people without one being aware of this influence.
Cool temperatures increased the desire for social consumption settings. Their findings highlight the bidirectional relationship between physical and social warmth (Bargh & Shalev, 2012) and converge with another study on social warmth that was conducted by Zhong and Leonardelli (2008). They showed that participants experience a room as physically colder after having been socially rejected. Another study reported that consumers perceived the ambient temperature to be cooler when eating alone than when eating with a partner (Lee, Rotman, & Perkins, 2014). Zwebner, Lee, and Goldenberg (2014) studied whether the association of physical and socio-emotional warmth also extends to products. They found that higher ambient temperatures are associated with more positive affective responses and lower perceived distance to the target product, resulting in increased product valuation.
Based on the abovementioned studies and the positive metaphorical associations of warmth, we expect that consumer responses will also be influenced by experiencing physical warmth. Therefore we expect that:
H1a. Holding a warm drink has a significant positive effect on ‘general attitude’.
H1b. Holding a warm drink has a significant positive effect on ‘company’s social orientation’.
H1c. Holding a warm drink has a significant positive effect on ‘purchase intention’
2.5.2 Haptic metaphors
Common touch-related metaphors are also frequently used in daily life, e.g. “it was a rough day”, “thinking about weighty matters” and “she’s my rock”. According to Ackerman, Nocera and Bargh (2010), these metaphors are also powerful as a sensory metaphors. They found that heavy vs. light clipboards made job candidates appear more important. The metaphorical association of heaviness and lightness is ‘importance’, which implicates the heavier, the more important something is (Jostmann, Lakens, Schubert, 2009). Another study by Ackerman et al (2010) found that that touching a rough vs. a smooth object makes social interactions more difficult. They argued that the experience of roughness and smoothness is metaphorically associated with concepts of difficulty and harshness. They also showed that sensory metaphors regarding to hardness and softness subconsciously affect one’s judgment. In daily life, we describe someone as soft when we perceive him or her as vulnerable, emotional, empathetic and sensitive, and we describe someone as hard when we perceive him or her as invulnerable, unemotional, non-empathetic, rigid and insensitive. In the experiment of Ackerman et al. (2010), participants either touched a soft blanket or a hard block of wood. The participants who touched the hard block of wood judged someone as more rigid and strict than participants who touched the soft blanket. Haptic experiences with respect to hardness and softness are metaphorically associated with (cognitive) flexibility, including stability, rigidity and strictness.
The abovementioned studies showed that experiences that are metaphorically related to haptics subconsciously influence social impression and decision making. The question that arises is how such basic haptic experiences concerning weight, texture and hardness, influence our cognitive processing.
According to several scientists (Barsalou (2003), Mandler (1992), sensorimotor experiences are stored in our mind since infancy, which forms a scaffold for the development of conceptual knowledge. This means that touching something hard activates the grounded conceptual knowledge related to hardness. This conceptual knowledge can also be a metaphor, e.g. feeling a rough piece of wood sensitizes us to rough textures and may also trigger metaphorical roughness. Another study by Ackerman et al. (2010) found that sitting on a hard chair increases rigidity in bargaining, and on the other hand, sitting on a soft chair leads to quicker agreement in bargaining. This experiment moved beyond active touch manipulations to investigate whether passive touch experiences can similarly drive embodied cognitive processing. Instead of having participants touch objects with their hands, they primed participants by the seat of their chair. Eighty-six participants sat in either a hard wooden chair or a soft cushioned chair while completing both an impression formation task and a negotiation task. This latter decision-making task had participants imagine shopping for a new car by making an offer to the dealer, being rejected, and having to make a second offer. Depending on the condition, participants were sitting in a hard wooden chair or a soft cushioned chair. As predicted the participants who sat in hard chairs judged the employee to be both more stable and less emotional. Furthermore, hard chairs indeed produced less change in offer price than soft cushioned chairs did. A study by Cherkasskiy, Song, Malahy, & Bargh (2012) proved that courts impose higher punishments to suspects when sitting on a hard chair. Thus, hardness produces perceptions of strictness, rigidity, and stability, reducing change from one’s initial decisions.
Based on the abovementioned studies and the metaphorical associations of hardness and softness, we expect that consumer responses will also be influenced by haptic experiences regarding to hardness and softness. The metaphorical associations with hardness and softness are less related to social concepts (general attitude and company’s social orientation), but we expect that the soft chair condition will still have a positive influence on the general attitude and company’s social orientation compared to the hard chair condition. Since metaphorical associations of softness are much more related to ‘easiness’ and ‘cognitive flexibility’, we expect that the participants on the soft chair will act more easy and flexible, and will therefore have a higher purchase intention than the participants on the hard chair.
We therefore expect that:
H2a. Sitting on a soft chair has a significant positive effect on ‘general attitude’.
H2b. Sitting in a soft chair has a significant positive effect on ‘company’s social orientation’.
H2c. Sitting on a soft chair has a significant positive effect on ‘purchase intention’.
2.5.3 The connection between the variables; (dis)comfort
Both independent variables within this research are linked to each other, because people use both ‘hard’ and ‘cold’ to explain discomfort and ‘soft’ and ‘warm’ to explain comfort. An example of a frequently used statement is: “It is a cold hard world”. This statement clarifies the negative charge of ‘cold’ and ‘hard’ (Melnick, 1999). The reason why we typically pair ‘cold’ with ‘hard’ and ‘warm’ with ‘soft’, is that associations with ‘cold’ and ‘hard’ are neither pleasant nor positive. A ‘warm soft mommy’ is a frequently used association for a caring mother. This example clarifies the pleasant and positive association with ‘warm’ and ‘soft’.
Apart from the two main effects described above, an interaction effect might occur. When combining both independent variables, ‘physical warmth’ and ‘haptic experience’, with the dependent variables, we expect that the most ‘positive’ condition (warm and soft) will lead to the most positive influence on all these variables. We can maintain that, in general, the positively associated conditions (warm and soft) will automatically lead to a higher mean score on all constructs compared to the negatively associated conditions (cold and hard). We expect a positive interaction between the warm and soft condition on all three dependent variables that is significantly higher than the sum of the two main effects. We therefore expect that:
H3a. Holding a warm drink and sitting on a soft chair has a better postive effect on ‘general attitude’.
H3b. Holding a warm drink and sitting on a soft chair have a postive effect on ‘company’s social orientation’.
H3c. Holding a warm drink and sitting on a soft chair have a postive effect on ‘purchase intention’.
2.5.4 Effect size
Apart from the positive and negative associations of the independent variables, we expect a distinction in the subconscious influence between the constructs regarding to their different metaphorical associations. Since warmth and coldness are mostly metaphorically associated with sociality, trustworthiness, helpfulness and friendliness, we expect that the subconscious influence of the temperature of the drink will be dominant over the type of chair for the ‘company’s social orientation’ and ‘general attitude’ constructs. Conversely, because the metaphorical associations with haptic experience are more related to (cognitive) flexibility, stability, rigidity and strictness, we expect that the subconscious influence of the type of chair will be dominant over the temperature of the drink for the ‘purchase intention’ construct. We therefore expect that:
H4a. The warm drink is a larger predictor of the general attitude than the soft chair
H4b. The warm drink is a larger predictor of the companies’ social orientation than the soft chair
H5. The soft chair is a larger predictor of the general attitude than the warm drink
2.6 Connecting the variables: a conceptualized research model
This study aims to identify the subconscious influence of the metaphorical associations of physical warmth and haptic experience on the participants’ ‘general attitude towards a company’, ‘perception of the company’s social orientation’ and ‘purchase intention’. The focus is on whether the manipulated independent variables will result in a different mean score on the dependent variables. Figure 1 provides an overview of the dependent and independent variables for the experiment.
As mentioned in the theoretical framework, a 2 (cold drink vs. warm drink) x 2 (hard chair vs. soft chair) between subjects design was used for the experiment. The experiment was conducted as a quantitative field experiment at the University of Twente. This section will further explain the sampling procedure and study design, followed by the clarification of the manipulated stimuli. Finally, this section will clarify the validity and reliability of the measures.
3.1 Participants
Since the experiment was conducted at the University of Twente, most of the participants had an academic background. There was however no restriction in educational background or age. The only requirement for the participants was to understand the Dutch language, since the brochure and questionnaire were set up in Dutch. The aim was to gather at least 20 questionnaires for each experimental condition, thus 80 respondents in total, in order to conduct a reliable experiment. Eventually 119 respondents (62% female, 38% male) participated in the experiment. As can be seen in Table 1, the participants were almost equally distributed over the conditions, even as their mean ages. All participants participated voluntarily in the experiment and completed the questionnaire by answering all questions. The respondents’ age (n = 119) varied from 18 to 41 years, with an average age of 22.4 (SD = 2.97).
Table 1. Overview of the experiment dates, including the distribution of participants over the condition
Condition: Cold drink Hard chair Cold drink Soft chair Warm drink Hard chair Warm drink Soft chair
Date of experiment 10-06-15 08-06-15 01-06-15 04-06-15
N 29 31 29 30
Mean age 21.93 21.90 22.41 23.27
Gender % Male
Female 31
69 38.7
61.3 34.5
65.5 46.7
55.3
The subjects of the experiment had to fill in a questionnaire about a brochure of the fictitious green energy company ‘The Green Energy Company’. During the experiment participants were sitting on a hard wooden chair or a soft office chair, and drinking cold or warm tea. Every participant was presented with the same brochure and questionnaire about ‘The Green Energy Company’. In order to prevent that the participants would be aware of the manipulated conditions, four different days were scheduled for conducting the experiment. Every day had another combination of manipulated variables. Table 1 provides an overview of the different experiment dates, including the distribution of participants over the conditions with the mean age and gender percentage.
3.2 Stimuli
In order to conduct a successful, valid and reliable experiment, the purpose of the experiment had to be unclear to the respondents. The manipulated stimuli were well considered before conducting the experiment. Consequently, two small pre-tests were conducted prior to the experiment. This paragraph will further explain why and how the stimulus materials were chosen and manipulated.
3.2.1 Manipulated drink temperature
Warm and cold drinks were used in order to control the manipulated variable of physical warmth. The drink itself (e.g. the taste) does not have anything to do with the purpose of the experiment, since the focus of the experiment is on the warmth or cold that the drink transmits to the participants. Therefore, the temperature of the drink is the only manipulated factor in both the warm and cold condition.
In order to find a drink that has the same properties when being warm and cold, a small pre-test was conducted with 4 people at the researchers home. Both warm and cold drinks were tasted and evaluated on similarities in flavour, colour and substance. After having tasted various drinks with a warm and cold temperature, hot tea and ice tea were chosen as the best drinks for both conditions. The only difference between the drinks was the sweetness between the warm tea and ice tea, since ice tea contains added sugars and regular tea made from a teabag does not. This issue was solved by adding two teaspoons of sugar per can of hot tea. Images of the used drinks can be found in appendix D.
In order to measure the effects of physical warmth, the temperature differences between the drinks were to be as large as possible. It was important that the tea was drinkable from the start of the experiment, to prevent that subjects would not drink the tea at all. The warm tea had a temperature of approximately 55 degrees Celsius and the ice tea had a temperature of approximately 6 degrees Celsius. In order to reach those temperatures, water was boiled for the warm tea and ice cubes were added to the ice tea to cool the drink. To regulate the temperature of the drinks, a thermos flask and thermometer were used for both drinks to keep them at approximately 55 degrees and 6 degrees Celsius. The cups used for the experiment were simple white plastic disposable cups, which are hygienic and easy to use. These cups were good temperature conductors, which is important since physical warmth was the manipulated factor in this experiment.
3.2.2 Manipulated hardness and softness
Within this experiment one hard chair and one soft chair were used to measure the subconscious effects of the metaphorical associations of hardness and softness. The hard chair was to be as hard as possible and the soft chair as soft as possible. In order to find two suitable chairs, a small pre-test was conducted with 3 students of the University of Twente. Both hard chairs and soft chairs were tested and evaluated and the subjects had to keep in mind that apart from the hardness or softness, all other dimensions were to be as much as similar to each other.
The height and depth of the chairs were measured and compared. The hard seats had no ability to adjust the height or seat of the chair, while the softer chairs had. It was therefore possible to adjust the soft seat to the dimensions of the hard seat. After testing the chairs, two chairs were selected, being a soft office chair without arm rests and a hard wooden chair without arm rests. Images of the used chairs can be found in appendix A.
3.2.3 The brochure
A brochure of the fictitious green energy company ‘The Green Energy Company’ was composed in order to measure the subconscious influence of metaphorical associations on the dependent variables. An image of the brochure can be found in appendix B. From this brochure, the participants had to gain all the information about the company. The main reason for using a fictitious company is that the participants would not have an opinion about the company in advance. A benefit for choosing a green energy company is the ability to scale items related to all three constructs.
For the brochure’s lay out, information was retrieved from other Dutch green energy companies such as Windunie, Huismerk Energie, Raedthuys Pure Energie, Qurrent and Greenchoice (Wise, 2014). The brochures of these companies often contained information about the company and how they produce and deliver their green energy.
3.2.4 Introduction
Before the participants started with the actual experiment, a short introduction was attached to the questionnaire. Hereby participants were informed about what was expected from them during the experiment, being providing their attitude to a green drink and a green energy company. An emphasis was made on that the experiment was actually a small pre-test. By framing the experiment as a small pre-test, participants would be more easy going, less critical and purely relying on their sense, without difficulties coming to mind. Since the participants were all recruited from the University of Twente, it is very likely that they would have been extra critical when the actual goal of the experiment would have been mentioned. The introduction that was attached to the questionnaire can be found in appendix C.
3.3 Measurements
After the participants had read the introduction they could start answering the questionnaire, by which the dependent variables were measured. In order to measure the three dependent variables, three different constructs were established with each different items. In this paragraph, the method of measurement for each dependent variable as well as the reliability of the used constructs will be further explained.
3.3.1 Measurement of dependent variables
First the respondents had to supply demographic information, consisting of age and gender. Items on the topic of the drink were included in order to be sure that participants had a physical experience with warmth or coldness before judging ‘The Green Energy Company’. This was important since the purpose of offering a drink was to conduct physical warmth or coldness. The items of the drink, measured on a 7-point-likert-scale, were specifically about taste and purchase intention. However, in the end, the results of the drink do not count for the experiment. They functioned as an unobtrusive manner to let the participants experience physical warmth or coldness before judging ‘The Green Energy Company’. The reason for using multiple items about the drink is to prevent raising suspicion among the subjects regarding the research goals. If the drink itself would only have been covered by one question, the subjects might have concluded that the drink was not of interest to the researchers, and found out the actual research goals. This could have influenced the subjects’ answers to the items and thereby the results from the field study.
After the items about the drink, the participants were presented with the items about the ‘general attitude’ construct. Since measuring a complete attitude is far too extensive, only eight basic attitude-related items were measured. For the creation of the ‘general attitude’ construct, much inspiration has been gained from earlier studies by Mc. Croskey (1989) and Spears and Singh (2004). The items were measured on an eight-item semantic differential scale (Snider & Osgood, 1969). Example answers for the eight-item semantic differential scale are: bad / good, convincing / not convincing, not credible / credible. Today, semantic differential scales are one of the most widely applied scales for the measurement of attitudes. The bipolar adjective pairs can be used for a wide variety of subjects (Himmelfarb, 1993).
Seven items regarding the social orientation construct were asked on a 7-point-likert-scale. This construct was aimed at the social aspects of the company, including customer service, reliability and commitment. The items used in the questionnaire were based on statements from a previous study by Gummesson (2005), in which a sample of 1.000 consumers had to answer, in five to ten words, what first came to mind when thinking of the most important brand aspects. The study identified the most important consumer-based dimensions of brands for the U.S. population, divided in: perceived quality, perceived value, brand loyalty, and sustainability. The social orientation construct contained items regarding the social aspects of the consumer-based dimensions, hence the name ‘company’s social orientation’. The construct contained five positively worded items and two negatively worded items. It is known that varying the direction of questioning minimizes bias produced by the respondents (Spector, 1992). Examples of used statements are: TGEC seems like a customer-friendly energy company to me and TGEC seems like a compassionate energy company to me. The used 7-point-likert-scale ranged from 1=completely disagree to 7=completely agree. It is known that Likert-scales are especially useful when measuring characteristics of people such as attitudes, feelings and opinions, and therefore as well for the sociality related consumer-based dimensions (Likert, 1932).
Finally, another seven items on a 7-point-likert-scale were asked in order to measure the construct ‘purchase intention’. This experiment is not aimed at measuring a complete purchase intention as this would create the need to include other dependent variables, which would make this research too extensive. The construct is adapted and limited to the circumstances of the experiment and the corresponding opportunities. An example of a consecutive limitation is therefore that the setup of the experiment does not concern the existence of a need for a (new) green energy company among the participants. It also does not provide room for comparing alternatives. The ‘purchase intention’ construct focusses on the amount of flexibility and compliance of the participants as the result of the metaphorical associations of the independent variables. The main question is whether participants in the ‘soft chair’ condition will comply faster than respondents in the ‘hard chair’ condition. Therefore, the items included for this construct are about the willingness to receive additional information about the company and the compliance regarding listening to the company’s call centre staff and door to door sales persons. In other words, it is about the interest in the company and her service(s). A few examples of the used statements are: I would consider to receive additional information about TGEC at home and I would consider to switch to TGEC on a short-term, via the free transfer service where they terminate my current contract. The statements were based on previous research by the world’s leading consumer behaviourists: Dodds, Monroe, and Grewal (1991), Grewal, Monroe and Krishnan (1998) and Hardesty, Carlson, and Bearden (2002). The scale of the measured items was ranging from 1=completely disagree to 7=completely agree (Spector, 1992). The questionnaire can be found in appendix D.
3.3.2 Reliability of constructs
After entering the data in the Statistical Program for Social Sciences (SPSS), the two negatively worded items were recoded to be able to measure the internal consistency or reliability of the constructs. As part of the internal consistency calculations a Cronbach’s Alpha was calculated for the three constructs, being ‘general attitude’ (α= .94), ‘company’s social orientation’ (α= .88) and ‘purchase intention’ (α= .88). Table 2 provides an overview of the reliability of the constructs. According to Nunnally (1978), a constructs’ Cronbach’s Alpha should be at least .70 to achieve a reliable scale. Based on the values of the Cronbach’s Alpha we can therefore conclude that all three constructs have a reliable scale.
3.4 Procedure
The participants were selected by the researcher at the University of Twente. The subjects were asked whether they wanted to participate in a small pre-test experiment about green products which would take only 5 minutes. Subjects who were willing to fill in the questionnaire were guided to Cubicus room C227 at the university of Twente. The participants were instructed to first read the introduction attached to the questionnaire. Thereafter they could start filling in the questionnaire. The first items were about the drink, therefore the participants were instructed to taste the drink before filling in the items. After the participants filled in the questionnaire, the researcher asked them for their opinion about the experiment. If they were interested, the participants were informed about the actual purpose of the experiment; that the experiment was not aimed as a pre-test, but as a main experiment with manipulated drinks and chairs. The majority of the respondents were surprised about the actual purpose and were interested in the results of this study
3.5 Data Analysis
Two-way ANOVAs were performed with physical warmth and haptic experiences as independent variables and the ‘general attitude’, ‘company’s social orientation’, and ‘consumers’ purchase intention’ as dependent variables. Individual characteristics of respondents (age and gender) were considered as possible moderators. Bonferroni post-hoc analyses were performed to test the significance differences between the means.

...(download the rest of the essay above)