Fact or Fiction:
Narrative versus Factual Text Formats and the Role of Consumer Expertise in Fair Trade Marketing
Maria Tenningås 10630937
Graduate School of Communication University of Amsterdam Thesis supervisor: Dr. Marijn Meijers
Abstract
The goal of this study was to compare the effectiveness of narrative versus factual text format, moderated by level of domain-specific consumer expertise in the context Fair Trade fashion brand. The aim of this research was to influence ad and brand attitudes of readers by communicating the Fair Trade policies employed by an ethical fashion brand in order to allow consumers to understand the external consequences of their purchase. A between-subjects experimental design was employed with two conditions, (text format: narrative vs. factual) and a continuous predictor variable (consumer expertise: low to high). This online experiment was conducted among 294 participants from 53 different nationalities, with attitude towards the ad and brand attitude employed as the dependent variables. In contrast to prior research conducted that compared narrative versus factual advertising, the findings of this study revealed no significant main effect of text format on ad attitude or brand attitude, and no interaction effect between text format and level of consumer expertise on ad attitude and brand attitudes. Further research should be conducted, where additional variables are taken into account to enhance the effectiveness of narratives, such as identification with the protagonist, self-referencing, vividness, and entertainment value of the narrative text.
Keywords: narrative ad copy, factual ad copy, consumer expertise, Fair Trade marketing, corporate social responsibility, ethical consumerism
Introduction
Ethical consumption allows individuals to use their buying decisions to demonstrate their beliefs and opinions, which likens their purchase to a ‘vote’ to support those brands whose ethical policies they support (Anon, 2001). The tragic aftermath of events like the Rana Plaza factory collapse that occurred in Bangladesh in 2012, revealed horrific details about the reliance of major global fashion brands on sweatshop production in developing nations, which has led to global campaigns and boycotts of these brands that are known as ethical offenders (Shaw, Hogg, Wilson, Shiu & Hassan, 2006). However, while consumers are increasingly faced with an overload of negative information about these ethically offensive brands, they lack a supply of positive information on ethical alternative choices (Shaw et al., 2006). While consumers would prefer to support more socially responsible companies, the ethical clothing market lacks positive product information at point of purchase (Shaw et al., 2006). Fair Trade fashion marketing should therefore focus on providing consumers with positive information about a brand’s ethical practices and policies in order to transform ethical intentions into informed, ethical purchase decisions.
Whereas clothing is a functional product that is purchased to fulfill physical needs, fashion is also about expressing one’s identity, satisfying certain emotional needs, and expressing one’s individuality through symbolic production (Easey, 2002). This ‘sign-value’ of fashion also means that consumers manage a part of their ethical identity by choosing to purchase ethical fashion over unethically produced fashion (Shaw, Hogg, Wilson, Shiu & Hassan, 2006). As ethical consumerism is concerned with the effect a purchase has not only on themselves, but on the external world, consumers of Fair Trade fashion need to be given the opportunity to understand the external effects of their purchase that constitute the more intangible benefits of Fair Trade products (Chowdhury & Fernando, 2013). Research has revealed that moral motives are a stronger determinant of purchase behavior of Fair Trade
products than economic factors (Sunderer & Rössel, 2012). Therefore, Fair Trade marketing should strengthen moral motives by communicating the effect a purchase has on the external world, which include those who are involved with the production of these products.
One message strategy that has been largely ignored in Fair Trade marketing research is narrative advertising. Online narrative advertising is an important and rapidly growing venue to engage consumers with target markets (Mooradian et al, 2008). Whereas factual ad copy provides information about a product or brand by presenting facts and descriptions of product features and benefits, narrative advertising tells a story related to the product, which involves a character and is structured by thematic and temporally related sequences of events (Bruner, 1986). Narratives present the intangible benefits of a product by inducing mental simulations of hypothetical scenarios in readers, which elicit transportation into the narrative (Ching, Tong, Chen & Chen, 2013).
The persuasive power of narratives is attributed firstly, to the structural similarity of information acquired in daily life; and secondly, to consumers’ matching of incoming
information with prior knowledge structures (Adaval & Wyer, 1998). People’s interpretation of incoming information relies on previously acquired knowledge and currently activated knowledge structures (Mattila, 2000). Thus, as prior knowledge is a crucial element of information processing, and information processing differs between novice and expert consumers, domain-specific consumer expertise will be explored as a moderating variable in the relationship between the persuasive power of narratives and the brand attitudes and attitude towards the ad that it affects. No such research in the context of Fair Trade fashion marketing has been published to this date, making this study the first of its kind. This leads to the following research question:
Research Question: Is a narrative text ad or a factual text ad more effective in inducing positive ad attitudes and brand attitude towards a Fair Trade fashion brand and how is this effect moderated by domain-specific consumer expertise?
Theoretical Background Narrative vs. Factual Ad Copy
A narrative is defined as a story that consists of a sequence of events that are thematically and sequentially related and involve one or more characters (Adaval & Wyer, 1998). First of all, narratives are characterized by a content component, including actors, actions and motives (Wentzel, Tomczak & Herrmann, 2010). Secondly, narratives contain a structural component that is causal and temporal (Wentzel et al., 2010). In other words, it is a message strategy that aims to influence attitudes, beliefs, and behaviors through a text presented in a story format.
Although empirical research in this field has been scarce, marketers and researchers seem to be intuitively aware of the persuasive potential of narratives, as revealed by a content analysis conducted by Edson Escalas (1998), which found that over 20% of advertising depicts narrative story formats. Narrative advertising is also growing particularly fast in online advertising, in order to engage with consumers and connect with target markets (Mooradian et al., 2008). Narrative advertising presents the intangible benefits of a product by unveiling episodes in which characters are engaged in actions that help them to achieve their goals – these can take the form of drama, transformation, or a slice-of-life (Ching et al., 2013).
In contrast, factual ad copy, also referred to as “list” (Adaval & Wyer, 1998), “expository” (Smith, 1995), or “argumentative” (Padgett & Allen, 1997) ad copy, conveys information about a product or service in a logical, and fact-based manner without a causal or
temporal connection or characters (Padgett & Allen, 1997). In contrast to narrative advertisements, its components can therefore be moved around the text without interfering with the overall meaning of the text, whereas the meaning of a narrative advertisement is a embedded in the structure of the content itself (Smith, 1995). For example, a factual advertisement about Fair Trade policies would list the different policies, in any order, without additional contextual information. A narrative, on the other hand, would embed the same factual statements within the structure of a story of a particular character who experiences the effects of these same policies. Research has found that narrative ads, as opposed to factual ads, generate higher rates of ad recall (Smith, 1995; Tun, 1989), more intense affective reactions (Mattila, 2000), more favorable ad attitudes (Mattila, 2000), more favorable service evaluations (Adaval & Wyer, 1998), as well as stronger service purchase intentions (Mattila, 2000).
Transportation Theory
One key characteristic of narratives is that they foster the potential for readers to become transported into the narrative world. Transportation is a convergent mental process that may occur in response to the immersion into a fictional or non-fictional narrative, where all mental systems and capacities become focused on events occurring in the narrative (Green & Brock, 2000). Transportation is defined as the immersion into a text and the extent to which an individual becomes ‘lost’ in a story (Green & Brock, 2000). In other words, the individual loses track of the real world when being transported in the story and as such experiencing the events in the story being told (Gerrig, 1993; Green & Brock, 2000). Narrative advertising captivates the receiver of the message through the dramatic unfolding of causally related events, which then transports the viewer to the narrative world (Edson Escalas, 2004). The result of effective transportation through narrative advertising is that the
reader may gain a more realistic, vicarious idea of the emotional consequences that are associated with consumption of the brand (Padgett & Allen, 1997). If these are presented as positive emotions in the narrative, transportation is likely to elicit positive affect that may spill over to other attitudes (Wentzel et al., 2010). Research has indeed revealed that highly transported viewers report more positive overall responses, positive attitudes towards the advertisement and brand and positive emotions (Chang, 2009). Additionally, they give less critical evaluations of the text’s arguments and factual information and produce fewer counter-arguments (Chang, 2009).
Further research has found that the greater the level of transportation into a story, the greater the belief that these stories are true because there is a less critical analysis of the strengths of the arguments presented. In contrast, factual or list-form ads lead to more skeptical and critical evaluation of facts (Green & Brock, 2000).
Narrative versus factual information processing
Narrative and factual advertising can be distinguished from one another by the type of information processing they trigger. As discussed previously, both temporal and causal dimensions characterize narratives (Adaval & Wyer, 1998, p. 208). This narrative structure draws causal inferences between events and relates separate elements of the story together (Wentzel et al., 2010) This allows the reader to create a meaningful link between different elements of the content and connect these together into one story that is similar in structure to the way real-life experiences and personal narratives are stored in memory (Wentzel et al., 2010).
Most of the social information and personal experiences one gathers from daily life experiences are transmitted and stored in memory in a narrative form with structures similar to those of narratives. These stories, in turn, provide the basis for understanding and storing new experiences in memory; making judgments and decisions about the objects it refers to;
and further to generate attitudes towards these objects (Adaval & Wyer 1998). When new narrative information is processed, individuals try to match the incoming information with their existing stories in memory, thereby correlating the incoming information with previous memories (Wentzel et al., 2010. When reading a narrative text, therefore, readers are able to connect the brand with their own personal experiences, enhancing the likelihood of a meaningful connection between the brand and the self (Escalas, 2004). This leads to more favorable attitudes towards the brand in narratives, compared to in factual conditions (Adaval & Wyer, 1998; Escalas, 2004; Polyorat, Alden, & Kim, 2007).
Furthermore, a study by Adaval and Wyer (1998) found that narrative forms of information in travel advertising elicited more favorable evaluations of a vacation than did list-forms. They further explain that the effectiveness of narratives is not due to the increased affective reactions that the narratives elicited, but rather to narratives’ “structural similarity to information acquired through daily life experiences and the use of a holistic – as opposed to a piecemeal – strategy for computing judgments” (Adaval & Wyer, 1998). This conclusion further highlights the central role of the different information processing strategies of narratives that have superior potential to those of factual advertisements.
A systematic review of seventeen studies testing narrative versus non-narrative advertisements in various contexts, revealed that narrative information had a stronger influence on decision making compared to non-narratives (Winterbottom, Bekker, Conner & Mooney, 2008). The authors also found evidence that narratives may influence affective mechanisms rather than cognitive processes, suggesting heuristic rather than systematic processing of information, whereas factual ads employed more systematic processing techniques (Winterbottom et al., 2008). This persuasive effect of narratives can be explained by stating that as product benefits in narratives are placed within the larger context of a
realistic life experience, consumers relate to the information presented in a more involved and coherent way (Polyorat et al., 2007).
In contrast to narratives, factual ads typically elicit a more analytical form of information processing with the aim of deriving an overall conclusion and evaluation of the ad, rather than creating a meaningful story (Bruner, 1986). This form of processing leads to more counterarguments due to the evaluation and scrutiny of the facts or arguments presented and will form attitudes based on the strength of the factual ad’s arguments (Padgett & Allen, 1997). This suggests that information presented in a factual way will lead to a less positive attitude towards the ad and the brand compared to a narrative text format, as a more critical processing approach is employed. More positive ad and brand attitudes are the result of less counterarguments and scrutiny of the content.
Reducing Resistance
In order to induce positive attitudes and reduce counterarguments, advertising skepticism must be avoided in order to reduce resistance to the ad and enhance persuasive effects (Friestad & Wright, 1999). When an individual is aware of the persuasive or manipulative intent of an advertisement, persuasion knowledge is activated and a change of meaning may occur in the receiver, which triggers the resistance towards the ad and may result in negative attitudes toward the advertisement and brand (Friestad & Wright, 1999). Narrative ad formats reduce resistance by creating an expectation of entertainment rather than one of persuasion, thereby not activating ad skepticism and by avoiding the activation of persuasion knowledge (Stern, 1994). A study by Wentzel et al. (2010) revealed that when a narrative ad contains no cues that make manipulative intent salient, narratives lead to greater levels of positive affect, stronger self-brand connections, and more positive brand attitudes than factual ads.
In summary, narratives are expected to lead to higher brand and ad attitudes by transporting the viewer into the narrative, which allows readers to gain a more vicarious understanding of the consequences of purchasing a product. Secondly, narratives make it easier to gain meaning from a narrative text, compared to a factual text, due to the similarity of narrative structure and the structure of how real-life experiences are stored in memory. Thirdly, narratives reduce resistance by avoiding ad skepticism and not activating persuasion knowledge. These processes lead to a more positive attitude towards the ad and brand. Based on this information, the following hypotheses are presented:
H1: The narrative ad will lead to more favorable ad attitudes compared to the factual ad.
H2: The narrative ad will lead to more favorable brand attitudes compared to the factual ad.
Moderating Role of Consumer Experience
The effectiveness of narrative advertising, as discussed above, is likely to be limited to certain target markets. The level of expertise a consumer has about a certain product category is crucial in determining the effectiveness of narratives because individuals interpret incoming information depending on their currently activated knowledge structures (Higgins & King, 1981). How familiar an individual is with the information that is being processed is therefore a crucial element in information processing of narratives and non-narrative forms (e.g., factual ad copy) of advertisements (Brucks 1985; Sujan 1985). Consumers process information differently based on the existing knowledge structures, which are then activated when perceiving an advertising message (Bettman & Sujan 1987; Johnson 1988). Consumer expertise is therefore of great importance in the relationship between ad copy type and subsequent ad and brand evaluations.
Novice consumers are those who know nothing or very little about a product type and its background and therefore have very limited already present knowledge structures (Mattila, 2000). Experts, on the other hand, have had sufficient experience with a product category and thus their memory structures are more elaborate, allowing for complex inferential processing (Perachhio & Tybout, 1996). It makes sense that consumers with high levels of product expertise and familiarity use their prior knowledge structures to gain meaning from an advertisement. As a result, they are able to understand new incoming information faster than those with no product expertise (Chi, Feltovich & Glaser, 1981). Due to the superior processing strategies of expert consumers, ad recalls are higher than for novice consumers (Fiske, Kinder, & Larter 1983). Based on this, it is expected that for experienced consumers both narrative and factual ad copy ads will have comparable positive effects.
When processing information in an advertisement, individuals try to match the incoming information with existing knowledge structures (Mattila, 2000). Therefore, expert consumers are likely to match attributes of the ad information with their existing knowledge structures in order to derive meaning from the ad. Novice consumers are not able to do so due to their lack of existing knowledge structures. For this reason, it is more difficult for novice consumers to make sense of a factual or list-based advertisement, as there is nothing to match the incoming information with (Mattila, 2000). The structure of narrative advertisements facilitates this process due to its structural similarities with memories of real-life experiences (Adaval & Wyer, 1998). Expert consumers, on the other hand, rely mainly on their prior knowledge structures to extract meaning from an ad and are therefore not as influenced by the presentation format (Mattila, 2000). Piecemeal processing refers to the computation of evaluations of each attribute into an overall judgment (Mattila, 2000). Based on these differing information processing techniques of novices and experts, it can be concluded that narrative advertising facilitates information processing and learning of new information in novice consumers. Factual advertisements, in contrast, do not aid novices to comprehend the message due to the absence of structural similarity to previously gathered information (Mattila, 2000). Experienced consumers, on the other hand, rely mainly on their existing knowledge structures to make sense of an ad, rather than the structure of the ad itself (Mattila, 2000).
Indeed, an empirical study found that novices evaluated factual format ads significantly less favorably when compared to narrative ads, whereas no difference was found in expert consumers (Mattila, 2000). The author explains this by the facilitating effect of narrative ads in helping the novice consumer make sense of new, incoming information. In the list-format ads, novice consumers must spend considerable effort on trying to expand on the poorly or non-existent knowledge structures, which could also explain the less positive ad evaluations. The expectation is that those consumers who know little about Fair Trade fashion will benefit more than expert consumers from the inherent structure of a narrative because it promotes holistic processing techniques that assist in understanding the message (Adaval & Weyer, 1998). Expert consumers, on the other hand, will be left unaffected by the way the information is presented.
Based on the theories discussed, the following hypotheses are presented:
H3: The more novice consumers will evaluate the narrative ad more positively than the factual ad. Consumers with a higher level of expertise will show no significant difference in attitude towards the ads in the narrative compared to the factual conditions.
H4: Novice consumers will evaluate the brand more positively in the narrative condition compared to the factual condition. Consumers with a higher level of expertise will show no significant difference in attitude towards the ads in the narrative compared to the factual conditions.
Method Stimuli Development
In order to test the presented hypotheses, two ad copies were written for the
hypothetical, ethical fashion brand ‘Fairmark’. The factual ad copy and the narrative ad copy were both based on a shortened segment of the ‘About Our Suppliers’ section of a British green and ethical fashion brand’s website, Komodo (http://www.komodo.co.uk/about-us/).
For the list-format stimulus, the segment was restructured into five bullet points regarding the minimum age requirements and anti-discrimination laws of employees, followed by the work hour policy, and next by the trainings provided, as well as the policy on advance payments to ensure enough time for production. Finally, the brand’s commitment to placing orders far ahead of time to avoid excessive work hours for employees was stated. In total, this stimulus text amounted to 341 words (see appendix).
The narrative stimulus consisted of the same key points, which were presented in the same order and amounted to 365 words (see appendix). A key difference between the two is that the narrative stimulus evolved around a protagonist, Arifa, who is a fictional garment worker at one of the suppliers. The narrative is written from the point of view of a journalist who follows Arifa around at her workplace for one day in order to observe the working conditions at one of the garment factories. The narrative text is constructed in a sequential series of events that make up an average day in Arifa’s working life at one of the supplying factories in Bangladesh.
For both the narrative and the list-format condition, the same photograph of a smiling, middle-aged Bangladeshi lady sitting at a sewing machine, was presented on top of the text. In addition, both conditions were introduced with the following statement:
“Below is a text that the newly founded ethical fashion brand Fairmark is considering placing on their website to advertise the corporate social responsibility practices of the brand. The article is written by Fiona Miller, a research journalist investigating ethical practices in the fashion industry. Please read the text carefully and answer the following questions.”
Pre-Test
The pilot study was conducted in order to test whether the narrative condition scored higher on the transportation scale than the list-format stimulus. As one key characteristic of a narrative text is that it should induce transportation in the reader, transportation was used as the measure for the pilot study. A total of 31 participants were invited to respond to the online survey of the pilot study via e-mail and were subsequently not invited to participate in the main study (N=31). Out of the total number of participants, 56.3% were females and 43.8% were males. The age range of the sample was between 21 and 62 years, with an average age of 30.77 years (Mage = 30.77; SDage = 11.43). This sample originated from 14
different countries, with all except three originating from Europe.
After basic demographic data was gathered, participants of the pilot study were told to read the following texts carefully and subsequently answer the questions. They were then presented first with the narrative stimulus, containing the text that evolves around a
protagonist and is presented in a sequential order of events.
After reading the narrative text, the transportation scale measured participants’ agreement with statements related to transportation on a 7-point Likert scale from ‘not at all’ to ‘very much’. This scale is derived from Green & Brock’s (2000) transportation scale, but six items were removed from this scale, as they do not relate adequately to the context of the stimuli presented in this study. After being presented with the transportation scale,
participants were presented with the list-format text and were again asked to indicate their level of transportation using the same 5-item transportation scale.
A principal component analysis (PCA) shows that the 5 items form a single uni-dimensional scale: only one component has an Eigenvalue above 1 (Eigenvalue 3.05) and
there is a clear point of inflexion after this component in the scree plot. All items correlate positively with the first component and the variable "activity going on in the room around me was not on my mind" has the strongest association (factor loading is .89). Reliability of the scale is good, Cronbach's alpha = .81. Therefore, it appears the scale measures transportation. Looking at the original variable response options we see that high scores on the items
indicate high level of transportation into the text. The higher the reported values, the more transported the participants were into the text.
Participants’ average level of transportation in the narrative text was significantly higher (M=4.83, SD=1.02) compared to the average score on the transportation score for the list-format text (M=2.74, SD=.87). A Paired Sample T-Test confirmed that the narrative stimulus scored significantly higher on the transportation scale than the list-format condition did, t(30) = 9.56, p = .000, 95% CI [4.45, 5.20]. The list-format condition scored significantly lower on the transportation scale t(30) = 17.61, p = .000, 95% CI [1.64, 2.54]. This means that the narrative stimulus did indeed have the expected influence of transporting the readers more than the list-format stimulus did. Therefore, both stimuli were unedited and employed for the main study to effectively compare a narrative text with a list-format text.
Experiment
To test the hypotheses, a study was conducted to test the four hypotheses stated above. Firstly, the main effect of text condition on ad and brand attitudes were tested, followed by a test of the interaction effect of text format and consumer expertise on ad and brand attitude.
Participants
A between-subjects design with two conditions and one continuous was employed in order to test the hypotheses. The variables included were text format (narrative vs. factual ad), which was recoded into a dummy variable, and consumer expertise, which was measured on a Likert scale. A total of 350 participants were gathered via a convenience sample through private e-mails, and private Facebook messages to members of the researcher’s social
network. Only 294 of these participants fully completed the survey (N=294). The sample consisted of 58.5% females and 41.5% males and the participants were between the ages of 17 and 69, with an average age of 27.1 years (SD = 8.89). Among them, 84.4% of
participants were between the ages of 20 and 30 years. Contacted individuals were invited to share the survey with their social networks, which led to a snowball sample that gathered participants from across the world, resulting in a very international sample, which enables generalizations of this research to be made across different cultural contexts.
Participants originated from 53 different countries, of which 58% were European, 17.1% were North American, 8.3% were Asian, and 8% were of African descent. The remaining 2.6% of participants originated from Australia, whereas 1.1% was South American, and 4.9% of the sample reported having dual nationalities. As the target
population for this study was not defined to be from one specific geographic location, this diverse sample was useful in shedding some light on general global attitudes that may derive from narrative versus list-based ad formats in the context of a fashion brand’s corporate social responsibility practices.
A total of 78.2% of participants have earned University degrees (46.9% Bachelor level, 31.3% Master level). Furthermore, 9.9% of the participants earned some college credit but without obtaining a degree, and 7.8% earned a high school diploma. Overall, this sample is rather well educated and from diverse origins.
Research Design
This study was a between-subjects experimental design, with two conditions (narrative vs. factual ad copy) and one continuous predictor (level of expertise). All participants were randomly assigned to either the narrative or the list-format condition and were then asked to answer the same questions regardless of the conditions they were placed in. Their level of expertise was measured separately and is described in more detail below.
Procedure
Participants were invited to participate in the online experiment via e-mail or a private message via Facebook, inviting them to participate and share the survey with friends. The survey opened with an informed ethical consent form in which the procedure of the study was briefly explained. All participants were required to agree with the conditions before
proceeding to the rest of the survey. They were told that they would be shown a text that is designed to express a fashion brand’s corporate social responsibility, yet the aim of the research was not mentioned. Thereafter, demographic variables were asked, such as gender, age, nationality, and highest educational level attained.
Next, the moderating variable of consumer expertise was measured. Following this, they received the instruction to read the following text carefully and subsequently answer the questions provided. No time limit was set for the reading of the texts so participants could take their time to read the texts without time pressure. At this point, participants received either the narrative or the list-based stimulus, accompanied by a brief introduction and a photo of a Bangladeshi garment worker. Next, attitude toward the ad and brand attitude were measured. Finally, a space was left for participants to leave any comments they might have for the researcher.
Dependent measures
Consumer expertise. The moderating variable of consumer expertise was measured using a four-item, seven-point semantic differential scale created by Mishra, Umesh, and Stem (1993). This scale measured the degree of knowledge and experience a person reports having about a particular product, goods, or service. All participants were asked to rate on a scale from ‘strongly disagree’ to ‘strongly agree’ the extent to which they agree with the following items: “I know very much about Fair Trade fashion”, “I am inexperienced with Fair Trade fashion (R)”, “I am informed about Fair Trade fashion”, “I am an expert consumer of Fair Trade fashion”. The second item was measured in reverse, in order to minimize acquiescence and was reversed back into its original direction prior to conducting the statistical analyses. All items were combined to form one scale of the mean values. A reliability test revealed that the scale is reliable, as Cronbach’s alpha = .86.
Attitude toward the ad. Attitude towards the ad was measured using De Pelsmacker, Gueuns and Anckaert’s (2002) six item, seven-point Likert-type scale items. Participants were asked to indicate to what extent they agree with the following statements: “I got a positive impression”, “I found it really something for me”, “I found it interesting”, “I found it credible”, “I found it exaggerated”, and “I found it attractive”. “I found it exaggerated” was a reversed score item which was later recoded. All items were later combined into one mean score, measuring attitude toward the ad. The reliability of the scale was Cronbach’s alpha = .72. If the item “I found it exaggerated” was removed from the scale, its reliability would increase to Cronbach’s alpha = .76. However, as this difference is relatively small, all six items of the scale were included in the analysis.
Brand attitude. Brand attitude was one dependent measure explored in this study and was measured using a seven-point, semantic differential scale of seven bipolar items, as consistent with Machleit and Wilson (1988). The following items were measured to
determine brand attitude: “dislike very much/like very much”, “useful/useless”,
“valuable/worthless”, “unimportant/important”, “beneficial/not beneficial”, “not fond of/fond of”, “enjoyable/un-enjoyable”. Two of the seven items were presented with reversed
endpoints in order to minimize acquiescence and order effects. All items were later combined into one mean score that indicated the overall attitude towards the brand ‘Fairmark’. This scale was treated as an interval level variable and all items were combined to form one scale. The reliability of the scale was good, with Cronbach’s alpha = .90.
Results Randomization Check
In order to check whether randomization of gender across the two conditions was successful, a crosstabulation was conducted where both gender and text format were included as binary variables. The crosstabulation revealed that randomization of gender across the narrative and list-format conditions was successful (Pearson’s Chi Square = x (1) = .11, p< .738).
To check whether age was equally distributed across the narrative and list-format conditions, a One-Way ANOVA was conducted which revealed that age was indeed equally distributed across the narrative and the list-format conditions, F(1, 292) = 1.34, p =.248. The average age of those in the narrative condition was 27.78 years (M = 27.78, SD= 9.95) whereas the average age of those placed in the list-format condition was 26.58 years (M = 26.58, SD = 7.68). Both gender and age were successfully spread across the narrative and the list-format conditions.
Hypotheses testing
Hypothesis 1: Main effect of text format on ad attitude. An Independent Sample’s T-Test was conducted to test the first hypothesis, which states that narratives will lead to a more positive ad attitude compared to the factual condition. The Independent Sample’s T-Test revealed that text format did not significantly influence the ad attitude, F(2, 288) = .14, p = .429. This indicates that the narrative condition did not lead to a more positive ad attitude compared to the factual condition. This is in contrast to the first hypothesis, which predicted a more positive ad attitude in the narrative condition compared to the factual condition and the first hypothesis is therefore rejected.
Hypothesis 2: Main effect of text format on brand attitude. An Independent Sample’s T-Test was also conducted to test the second hypothesis, which states that the narrative text format will lead to a more positive brand attitude compared to the factual condition. The test revealed insignificant results, F(2, 288) = .30, p = .716. This means that narratives did not induce a more positive attitude towards the brand in comparison to factual ads and the second hypothesis is thereby rejected.
Hypothesis 3: Effect of text format moderated by consumer expertise on attitude toward the ad. The third hypothesis, which stated that novices will have a higher ad attitude in the narrative condition, compared to the factual condition whereas more expert consumers will reveal no significant difference in ad attitudes between the conditions, a multiple
regression was conducted. The analysis employed attitude towards the ad as the dependent variable and consumer expertise (a continuous predictor) and text format (categorical predictor) as the independent variables. The multiple regression analysis revealed that the interaction effect of text format and level of consumer expertise on the attitude towards the ad
did not yield significant results, F(3, 291) = 3.26, p = .923. This means that both the more novice, and the more experienced consumers were unaffected by text format. This further means that the first part of the third hypothesis is rejected, as it states that there ad attitudes will be higher for novices and no significant difference will be found for more expert consumers. The second part of the third hypothesis is accepted, as it states that expert consumers will be unaffected by text format and this effect was indeed found.
Hypothesis 4: Effect of text format moderated by consumer expertise on brand attitude. In order to test the fourth hypothesis, which states that novice consumers in the narrative text ad condition will have a higher brand attitude than those in the list-format condition, but for expert consumers there will not be a difference between the narrative and the factual ad, a multiple regression analysis was conducted. The regression model employed brand attitude as the dependent variable and level of consumer expertise (continuous
predictor) and text format (dummy coded narrative vs. factual) as the independent variables. The interaction effect between text format and expertise on the brand attitude also yielded insignificant results, F(2, 287) = 917, p = .447. This model therefore does not accurately predict the dependent variable. This means that the first part of the fourth hypothesis can not be accepted, as both novice and more expert consumers’ brand attitudes were unaffected by the text format. As expert consumers revealed no significant difference between the
conditions as expected, the second part of the fourth hypothesis is accepted.
Discussion
The aim of this research was to compare the effectiveness of narrative versus list-format texts on attitude towards the ad and on brand attitude of novice to experienced
consumers of the Fair Trade fashion product category. This was explored in the context of a Fair Trade fashion brand’s corporate social responsibility practices that would be placed on the ‘About Us’ section of a Fair Trade fashion brand’s website with the aim of presenting a persuasive image of the company’s ethical production standards. It was hypothesized that the narrative text format would lead to a higher attitude towards the ad and a higher brand
attitude than the factual ad. In addition, it was hypothesized that novice consumers will have a higher ad attitude and brand attitude when in the narrative condition, whereas the more experienced consumers would be unaffected by the presentation format and have similar ad and brand attitudes regardless of the text format they read.
Past research has revealed that narratives lead to more favorable brand and ad attitudes through transportation into the narrative (Chang, 2007). Due to the structural similarity of a narrative text format and the storage of real life experiences in memory, narratives make it easier for novice consumers to derive meaning from a text (Adaval & Wyer, 2010). Further, narratives reduce resistance by creating an expectation of
entertainment rather than persuasion, thereby not activating persuasion knowledge and avoiding ad skepticism (Wentzel et al, 2010). As expert consumers already possess
previously acquired knowledge structures related to the information presented, they are likely to rely on these pre-existing knowledge structures to form an evaluation of a brand and ad, and therefore are not affect by presentation format (Sujan, 1985).
The findings of this study revealed no main effects of text format on attitude towards the ad, in contrast to past research and the hypotheses discussed. In addition, novice
consumers’ attitudes towards the ad, too, were unaffected by presentation format. These findings refute the first and third hypotheses. However, a positive correlation between level of expertise and ad attitude was found. This means that those who had more domain-specific expertise in Fair Trade fashion were more enthusiastic about the topic in general which
reflected on a more positive ad attitude. Hence, these experienced consumers have a more favorable opinion of the product category, which spills over to the ad itself. In addition, expert consumers tend to relate new incoming information with pre-existing knowledge structures easily, whereas novice consumers require more effort to process novel information (Mattila, 2000). In this case, the more experienced participants related the information in the text to what they already know about the product category, which must have been mainly positive evaluations that were stable across either text format. The finding reveals that the less experienced a participants was with Fair Trade fashion, the less he or she evaluated the ad as positive. This points to the effects of pre-determined personal experiences that appear to be a stronger predictor of attitude towards the ad than the presentation format. For Fair Trade marketing, this means that providing consumers with possibilities to become familiar and experienced with Fair Trade fashion will lead to a more positive interpretation of a Fair Trade ad. It also means that one challenge Fair trade marketers may face, is inducing positive ad attitudes in novice consumers.
Furthermore, expert consumers reported no significant differences in attitudes
towards the ad, regardless of the condition they were randomly assigned to. This confirms the second part of the third hypothesis and suggests that expert consumers were unaffected by the text format. This result is in line with Mattila’s (2000) finding, which was explained to be a result of expert consumers’ reliance on previous knowledge structures, rather than text format, to make sense of new incoming information (Mattila, 2000). This suggests that the more experienced consumers matched the incoming information from the stimulus with they already knew about Fair Trade fashion policies, and relied mainly on prior knowledge to form an evaluation of the ad, rather than being influenced by the format of the ad itself.
The findings of this study also revealed that there was no main effect of text format on brand attitude, in contrast to the second and fourth hypotheses. Novice consumers of Fair
Trade fashion did not evaluate the brand more positively in the narrative condition compared to the factual condition. This finding is not in line with prior research, which found that novice consumers had significantly more favorable attitudes towards a brand when in the narrative condition compared to the factual condition because the narrative structure (Mattila, 2000). Consumers with no domain-specific knowledge must spend considerable effort when making sense of a factual ad, whereas the information processing is facilitated by the
narrative structure, yet this was not reflected in the results of this study (Adaval & Wyer, 1998).
As the overall brand attitude reported was generally quite positive, the unexpected finding of this study can be attributed to the content of both ads, which contained only positive information about implemented Fair Trade policies. Therefore, even novice participants were able to formulate a judgment regardless of their lack of prior knowledge and easily formed a positive evaluation of the brand based on all the positively ethical statements stated in both the list-format and the narrative condition and the lack of negative information. Another explanation for this finding is that the brand was perceived as positive in comparison to regular fashion brands that do not focus on ethical standards at all. It should also be noted that neither the novice nor the expert consumers had any information about the brand besides what they read in their condition, which contained only positive information.
In this sample, the reported brand attitude for expert consumers of Fair Trade fashion products were unaffected by the presentation format, as stated in the second part of the fourth hypothesis. This result is in accordance with prior research, where no significant differences of brand evaluations were found under narrative compared to factual conditions for expert consumers for experiential services (Mattila, 2000). This finding therefore confirms prior research and suggests that narrative text format is not more effective in inducing positive brand attitudes of expert consumers than factual text format. This is also in line with the
theory that expert consumers are more likely to rely on prior knowledge structures to make sense of incoming information, regardless of the format in which the information is presented (Mattila, 2000).
Interestingly, female participants evaluated the brand more positively compared to males in both text format conditions. A possible explanation for this effect could be that the females in this sample had more strong positive feelings towards Fair Trade fashion than males to begin with, which was then reflected in their reported brand attitudes. Empirical research supports this claim in that females were found to be more likely to purchase, recommend via word of mouth, and advocate Fair Trade to friends and family (Morrell & Jayawardhena, 2010). Another possible explanation for this finding is that females reported higher brand attitudes because they were more affected by the need for social desirability across both conditions than males, which is supported by a study that found that social desirability confounds the relationship between gender and ethical responses (Dalton & Ortegren, 2011). In addition, it should be noted that both the protagonist of the narrative, and the research journalist who wrote the text and was mentioned in the introduction of the stimulus, was female. This could have lead to a higher association between female
participants and the text, as females were able to relate more easily to the female protagonist and author, compared to males.
Limitations
Although the presentation format did not affect the ad attitudes and brand attitudes of expert consumers as expected, it remains unclear exactly why there was no main effect of text format on the dependent variables and why novice consumers were not influenced by the presentation format as was expected. Some possible reasons for this could be related to the limitations of this experiment. For one, there are several limitations in conducting an online
experiment. First of all, some participants may have been distracted, tired, unmotivated, under time pressure, or suffering from a high cognitive load while participating in this study. This could have influenced how much time participants were willing to spend reading the stimulus, which could lead to responses that are not entirely based on the stimulus presented. Studies have found that the transportation effect decreases when one is depleted of cognitive resources (Wentzel et al., 2013). Furthermore, transportation decreases when an interruption of an immersion takes place (Wentzel et al., 2013). Given the crucial role of transportation in narratives and its effects on attitudes, it is plausible that some participants were not fully transported into the narrative due to cognitive depletion or distractions. This a limitation of the experimental setup of this experiment.
Another limitation of this experiment is that of social desirability effects were not controlled for in this study. First of all, having a positive attitude towards Fair Trade products is generally something that is socially desirable, meaning that participants may have
generally answered according to what they thought would be most socially acceptable, regardless of the presentation format. In addition, most participants were in the researcher’s social network and may have decided to answer according to what would be considered most ethical and socially desirable because they were doubtful of their anonymity in this survey. Attitudes may therefore have been reported as quite positive towards Fair Trade as a concept, rather than with regards to the specific textual formats presented. As this product category is quite niche and generally not very accessible, it is also likely that some participants rated themselves higher on the expertise scale than they actually are. This could again be attributed to social desirability effects and could severely influence the data.
Furthermore, the narrative stimulus itself may have been a limitation due to a potential lack several factors that increase the effectiveness of narrative ads. First of all, identification with the protagonist of the narrative may have been difficult for some
participants as the narrative text evolved around a garment worker, rather than a consumer. Research has found that protagonist-reader similarity is an important determinant in effective narrative persuasion (van den Hende, Dahl, Schoormans & Snelders, 2012). This would be an interesting element to explore in further research. Perhaps the identification of the reader with the protagonist was not strong, as this sample consisted of quite well educated and relatively privileged individuals, who may struggle to relate to a Bangladeshi garment worker’s daily life. Future research should consider creating a narrative condition in which the participant is transported into a story of someone who is enjoying the intangible, social, and consciousness-related benefits of a purchased Fair Trade fashion item, rather than the external consequences of a Fair Trade fashion purchase. The research by van den Hende et al. (2012) further reveals that the protagonist-reader dissimilarity can be mitigated by instructing readers to actively imagine themselves in the role of the protagonist persuasion. This method was not employed in this experiment but should be implemented in further research in order to overcome potential reader-protagonist dissimilarities.
Furthermore, narrative online ads that are interactive, vivid, and entertaining, and contain self-referencing lead to more positive product evaluations (Ching, Tong, Chen & Chen, 2013). Perhaps there would have been a more significant difference between narrative and factual ads for novice consumer, had these elements been implemented. For example, the narrative text could be rephrased somewhat to contain self-referencing, and could be
presented in the form of a virtual tour of a garment worker’s daily life in a more vivid and entertaining manner. According to the findings of Ching et al. (2013), these elements increase the effectiveness of narrative advertising. Implementing these different elements is
recommended for further research.
It should also be noted that the results of this study can be generalized across different social and cultural contexts, due to the diverse range of nationalities that participated.
However, these findings cannot be generalized across different product categories besides Fair Trade fashion due to the domain-specific narrative stimulus employed, which relates specifically to labor conditions of a garment worker.
Conclusion
Product categories that contain features and benefits that are intangible to the eye require optimized advertising strategies in order to effectively and persuasively communicate these benefits to the consumer. Narrative advertising has the potential to allow its readers to
vicariously experience the long-term, external consequences of their Fair Trade purchase and their support for an ethical brand. Research revealed that consumers are overwhelmed with negative information about non-ethical brands, yet lack positive product information about ethical companies that they want to support (Shaw et al., 2006). This means that marketers of ethical companies should ensure that their customers and potential customers are provided with enough positive information about the Fair Trade policies employed in production of garments, both prior to purchase and at the point of purchase. In addition, the marketing should be targeted differently at consumers with different levels of domain-specific expertise, in order to ensure that the information is easily processed and understood.
Personalized targeting of consumers with various levels of domain-specific expertise can be achieved through personalized retargeting on websites, for instance. In addition, further experimentation with interactive, vivid, and self-referencing narratives are highly recommended for Fair Trade marketers, as the potential of narrative text formats on novice consumers especially, are extensive, according to previous research. In order for ethical consumers to walk their talk, marketers must tell convincing stories of the consequences of a Fair Trade purchase to allow consumers to vicariously experience the consequences of their purchase.
References
Ethicalconsumer.org (2015). Ethical Consumer: the alternative consumer organisation. Retrieved 28 December 2014, from http://www.ethicalconsumer.org.
Adaval, R., & Wyer Jr, R. (1998). The role of narratives in consumer information processing. Journal Of Consumer Psychology, 7(3), 207-245.
Bettman, J. and W. Park (1980), Effects of Prior Knowledge and Experience on Consumer Decision Processes: A Protocol Analysis, Journal of Consumer Research, 7, 234-48. Brucks, M. (1985). The effects of product class knowledge on information search behavior.
Journal Of Consumer Research, 12(1), 1-16. doi:10.1086/209031
Chi, M., Feltovich, P., & Glaser, R. (1981). Categorization and representation of physics problems by experts and novices. Cognitive Science, 5(2), 121-152.
Ching, R., Tong, P., Chen, J., & Chen, H. (2013). Narrative online advertising: identification and its effects on attitude toward a product. Internet Research, 23(4), 414-438. Chowdhury, R., & Fernando, M. (2013). The Relationships of Empathy, Moral Identity and
Cynicism with Consumers’ Ethical Beliefs: The Mediating Role of Moral
Disengagement. Journal Of Business Ethics, 124(4), 677-694. doi:10.1007/s10551-013-1896-7
Dalton, D., & Ortegren, M. (2011). Gender Differences in Ethics Research: The Importance of Controlling for the Social Desirability Response Bias. Journal Of Business Ethics, 103(1), 73-93. doi:10.1007/s10551-011-0843-8
De Pelsmacker, Patrick, Ben Decock, and Maggie Geuens (1998), “Advertising
Characteristics and the Attitude Toward the Ad – A Study of 100 Likeable TV Commercials,” Marketing and Research Today, 27 (4), 166-179.
Journal Of Consumer Psychology, 14(1-2), 168-180. doi:10.1207/s15327663jcp1401&2_19
Fiske, S., Kinder, D., & Larter, W. (1983). The novice and the expert: Knowledge-based strategies in political cognition. Journal Of Experimental Social Psychology, 19(4), 381-400.
Easey, M. (2000) Fashion Marketing, Oxford: Blackwell Science
Gerrig, R.J. (1993). Experiencing narrative worlds. New Haven, CT: Yale University Press. Komodo | The Uk's number 1 organic and fair trade fashion brand (n.d.). About Us. Retrieved
from http://www.komodo.co.uk/about-us/
Green, M. C., & Brock, T. C. (2000). The role of transportation in the persuasiveness of public narratives. Journal of Personality and Social Psychology, 79, 701–721. Green, M. C., Brock, T. C., & Kaufman, G. F. (2004). Understanding media enjoyment: The
role of transportation into narrative worlds. Communication Theory, 14, 311–327. Mishra, S., Umesh, U., & Stem, D. (1993). Antecedents of the Attraction Effect: An
Information-Processing Approach. Journal Of Marketing Research, 30(3), 331. doi:10.2307/3172885
Morrell, K., & Jayawardhena, C. (2010). Fair trade, ethical decision making and the narrative of gender difference. Business Ethics: A European Review, 19(4), 393-407.
doi:10.1111/j.1467-8608.2010.01598
Moyer-Guse, E., & Nabi, R. (2010). Explaining the effects of narrative in an entertainment television program: Overcoming resistance to persuasion. Human Communication Research, 36(1), 26-52.
Polyorat, K., Alden, D., & Kim, E. (2007). Impact of narrative versus factual print ad copy on product evaluation: The mediating role of ad message involvement. Psychology & Marketing, 24(6), 539-554.
Peracchio, L., & Tybout, A. (1996). The moderating role of prior knowledge in schema-based product evaluation. Journal Of Consumer Research, 177-192.
Shaw, D.S. and Clarke, I. (1999) Belief formation in ethical consumer groups: an exploratory study. Marketing Intelligence and Planning 17(2), 109–120.
Shaw, D., Hogg, G., Wilson, E., Shiu, E., & Hassan, L. (2006). Fashion victim: the impact of fair trade concerns on clothing choice. Journal Of Strategic Marketing, 14(4), 427-440.
Sujan, M. (1985). “Consumer Knowledge: Effects on Evaluation Strategies Mediating Consumer Judgments,” Journal of Consumer Research, 12, 16-31.
Sunderer, G. & Rössel, J. (2012). Morality or economic interest? The impact of moral motives and economic factors on the purchase of fair trade groceries. International Journal Of Consumer Studies, 36(2), 244-250. doi:10.1111/j.1470-6431.2011.01087
Tun, P. (1989). Age Differences in Processing Expository and Narrative Text. Journal Of Gerontology, 44(1), 9-15. doi:10.1093/geronj/44.1.
van den Hende, E., Dahl, D., Schoormans, J., & Snelders, D. (2012). Narrative Transportation in Concept Tests for Really New Products: The Moderating Effect of Reader-Protagonist Similarity. Journal Of Product Innovation Management, 29, 157-170.
doi:10.1111/j.1540-5885.2012.00961
Wentzel, D., Tomczak, T., & Herrmann, A. (2010). The moderating effect of manipulative intent and cognitive resources on the evaluation of narrative ads. Psychology And Marketing, 27(5), 510-530. doi:10.1002/mar.20341
Winterbottom, A., Bekker, H., Conner, M., & Mooney, A. (2008). Does narrative
information bias individual's decision making? A systematic review. Social Science & Medicine, 67(12), 2079-2088.
Appendix
I. Narrative Stimulus
Below is a text that the newly founded ethical fashion brand Fairmark is considering placing on their website to advertise the corporate social responsibility practices of the brand. The article is written by Fiona Miller, a research journalist investigating ethical practices in the fashion industry. Please read the text carefully and answer the following questions.
Arifa is a 33-year old Bangladeshi lady who has worked for one of Fairmark’s supplying factories for ten years. I spent one whole day with her in order to find out what a day really looks like for someone who makes clothes for Fairmark, an ethical clothing line that produces fashion under fair conditions for employees.
As Arifa and I waited for the bus to the factory that morning, she told me that she had applied to work there when she was 17 but was rejected because she was too young. Shortly after Ariva turned 18, she was employed at the factory and has worked there ever since. Arifa’s family comes from a small minority of Bangladeshi society but this did not prevent her from receiving the job, as Fairmark promises no discrimination against employees based on race, color, sex, religion, political affiliation or social origin.
As we arrived at her workplace together a few minutes later, Arifa explained that she works around 40 hours a week but has the opportunity to work for extra money after hours, although she is never forced to do so.
Around 1pm, Arifa and I went to enjoy her one-hour lunch break at a nearby restaurant. She never works on Sundays and is also excused on all religious and cultural holidays. Her wage allows her to support herself and her two children, while saving some money for her
children’s education.
After her monthly skills training session, we walked back towards the factory and she told me about her membership in the democratically elected trade union, which negotiates fair wages every year. Arifa explained that her managers always received a 50% advance payment from
Fairmark so that the workers’ salaries are paid reliably on the same day every month, which is not always the case at other garment factories.
As the sun slowly set on Dhaka a few hours later, I asked her whether she felt exhausted by time pressures set by her bosses, but she shook her head. She told me that time pressure wasn’t an issue for her because orders from Fairmark are always made far ahead of time, which allows her to work at a comfortable pace.
II. Factual Stimulus
Below is a text that the newly founded ethical fashion brand Fairmark is considering placing on their website to advertise the corporate social responsibility practices of the brand. The article is written by Fiona Miller, a research journalist investigating ethical practices in the fashion industry. Please read the text carefully and answer the following questions.
At Fairmark, our mission is to create fashion that is economically, socially, and
environmentally sustainable in order to meet the needs of our present and future generations. In doing so, we guarantee the highest standards of ethical conduct when it comes to all our business practices, with special focus on fair treatment of those who produce the Fairmark fashion garments.
In accordance with the International Labour Conventions (ILO), no child labor is used in the production of our garments. Factories are required to keep an employee register and age certification is mandatory in order to sign a contract of employment, as all employees must be 18 or above. Our factories do no discriminate against employees based on race, color, sex, religion, political affiliation or social origin.
Typically employees work a 40-hour week, their wages being paid regularly with overtime being voluntary and paid. Workers are entitled to a 1-hour break in every 8-hour shift. They also have one day a week off and are allowed to observe religious and cultural holidays of which there are many.
Furthermore, to give all Fairmark employees the opportunity to improve and expand their skills, trainings are provided to all our artisans numerous times a year. All garment workers are paid fair wages, which are reviewed and negotiated annually through democratically elected trade unions.
We are also dedicated to making 50% advance payments on each order. This gives the managers of factories the possibility of paying garment workers on time at the end of each month, no matter when each order is finished with production.
To make sure that workers at garment factories do not face excessive time pressures, each order is made to the factory far ahead of time. This allows for appropriate management of time by managers and allows enough time for hand production at regular speed and without mandatory overtime.
With these strict guidelines in place, Fairmark is sure to provide ethical employment to all workers and therefore allows our valued customers to proudly wear fashion that looks good and feels good.
