Can User-Generated Content Change Brand Convictions?

Can User-Generated Content Change Brand Convictions?

Walachnia, Nathalie University of Groningen

S2550466

Author Note

For further correspondence, please feel free to contact n.walachnia@student.rug.nl

January 2018

Master Thesis for MSc Programme Marketing Management Faculty of Economics and Business

University of Groningen

Examiner: dr. J. Hoekstra

Second Evaluator: B. Harms, dr. J. van Doorn

Supervisor: dr. J. Hoekstra, B. Harms

Abstract

With a survey-based research study it was tested whether an experimental manipulation of valence would have an effect on brand conviction. Furthermore, it was hypothesized that focus of user-generated content (UGC) would moderate this effect: Positive UGC content of the focal brand and negative UGC of the competitor brand was assumed to increase brand convictions. Negative UGC of the focal brand was hypothesized to trigger psychological reactance and thereby increase brand convictions. Finally, positive UGC of the competitor brand was hypothesized to create cognitive dissonance and mitigate brand convictions.

Involvement was predicted to moderate the effect of focus on the relationship between valence and brand conviction. One hundred and two participants were randomly assigned to one of the four conditions. Each condition group received an experimental scenario induction, which consisted of a short text that described a coffee machine purchase. Afterwards the experimental manipulation in form of UGC followed. Brand conviction was measured before and after to assess changes that followed the manipulation. Our findings suggest that the effect of user-generated content on brand convictions depends on valence and focus.

Involvement moderates this relationship additionally.

Keywords: valence, user-generated content, involvement, brand convictions

Can User-Generated Content Change Brand Convictions?

Chapter 1. Introduction

What is one of the first things someone does when planning to buy a new TV, coffee machine or restaurant visit? Most likely: go on the internet and look for reviews, media articles and experiences of other people that could give guidance in decision making. In the age of digital it is not only traditional advertising but user-generated content that influences consumer purchase decisions (Cao, Meister & Klante, 2014; Littman, 2008; MacKinnon, 2012).

User-generated content can be defined as content that is produced by the consumer (Tirunillai & Tellis, 2012). It is a form of word of mouth, which is a type of inter-consumer communication. There are various types of user-generated content from a variety of mediums.

Among others these include microblogs (e.g. Twitter), posts and comments (e.g. Facebook), images (e.g. Instagram, Pinterest), blogs (e.g. Wordpress), vlogs (e.g. Youtube), product reviews and recommendations (e.g. Amazon, Yelp) (Liu, Burns & Hou, 2017). Because consumers have more trust in the content that users put out and determine their purchase decisions based on them, user-generated content has become a powerful marketing tool potentially being able to even dictate consumer purchase decisions (Bae & Lee, 2011;

Thoumrungroje, 2014; Huang, Chou & Lan, 2007). One very important subgroup of content- creating users are bloggers, who over the years have transformed from simple internet writers to opinion leaders (Mendoza, 2010). Digital opinions leaders are individuals within a specific reference group who are able to exert influence over others due to their knowledge, personality or abilities (Kotler, Brown, Adam, Burton & Armstrong, 2007; Rogers, 2015; Dippong, Kalkhoff & Johnsen, 2017).

Although user-generated content is certainly powerful, it is not clear yet whether it is

also able to change brand convictions, especially after point of purchase when consumers have

already made their decision. A conviction can be defined as a strongly held belief, opinion or persuasion or as the state of being convinced (Merriam-Webster’s dictionary, 2017; Oxford Dictionary, 2017). Applied to a brand, conviction means the strong belief in or persuasion of a brand. The goal of the paper is to see whether valence, i.e. the goodness (positive valence) or badness (negative valence) of user-generated content has an effect on brand convictions and how this might be moderated by focus of user-generated content: Is one more inclined to believe negative content about a brand to which one is loyal (focal brand) or positive content about a competing brand (competitor brand) (Shuman, Sander & Scherer, 2013)? Consumers who observe that other people have made the same brand purchase choice they made by looking at user-generated content will most likely feel confirmed in their behavior and be more satisfied with it (Cialdini, 1983; Amblee & Bui, 2011; Hennig-Thurau, Walsh & Walsh, 2003; Gupta &

Harris, 2010). The same applies to consumers who are exposed to negative user-generated content regarding the competitor brand since negative experiences made with the alternative brand affirm one’s purchase decision.

The opposite reaction can be expected for consumers who observe other people making

negative experiences with their choice, that is, in form of negative user-generated content for

example. The fact that one is following an action that others followed as well and afterwards

evaluated negatively, might spur a negative reaction, namely psychological reactance, which

consists of feelings of threat and anger (Clee & Wicklund, 1980; Amblee & Bui, 2011; Hennig-

Thurau, Walsh & Walsh, 2003; Rains, 2012). In order to alleviate themselves from those

negative feelings, they might start defending their purchase instead by offering

counterarguments and discard the negative claims (Rains, 2012). Previous research has shown

that highly committed consumers continuously counterargued negative information about their

favorite brand and only supported the positive arguments (Ahluwalia, Burnkrant & Unnava,

2000).

When exposed to positive user-generated content about a competing brand consumers’

choice is not directly being criticized but instead put in question indirectly by praising a different brand. Therefore, consumers might start doubting their purchase and liking it a little less in favor of the competitor brand (Liu, Li, Xu, Kostakos & Heikkilä, 2016). Competitor actions have already been shown to moderate the effect of advertising on sales in the past (Gatignon, 1984). It might as well be that the effect of valence and focus of user-generated content depends on the level of involvement that consumers show. Involvement caused positive reactions even in consumers who show dissatisfaction with a brand (Shiue & Li, 2013). A scenario-based survey experiment will be conducted in order to find answers to all of these questions.

Theoretical implications would include a broadened view on the influence of user- generated content during the post-purchase phase. Due to the fact that user-generated content seems to gain in importance in comparison to traditional marketing tools when it comes to influencing consumer purchase decisions, studying it further in the context of changing brand convictions is important in order to gain more knowledge about this new digital marketing tool (Bae & Lee, 2011; Thoumrungroje, 2014; Huang, Chou & Lan, 2007). Especially the ability of user-generated content to change brand convictions independent of point-of- purchase could additionally spur further research into areas such as consumer brand loyalty and brand commitment. Managerial implications would include: possibilities for companies to target even highly loyal brand consumers of competitor brands and to induce brand switching behavior through the direct use of user-generated content. Finally, user-generated content would gain even more evidence as an effective marketing tool and give companies one more reason to be actively included into firms’ digital marketing strategy plans.

The structure of this report is as follows: After introducing the topic of the underlying

research paper in Chapter 1, Chapter 2 describes the conceptual model and a literature review

that will also include all the hypotheses that will be tested as part of this research. Chapter 3 presents all the important variables and measures while Chapter 4 reports the outcomes of the analyses. Finally, Chapter 5 ends with a discussion and a short review of the research study including a section on limitations, future research as well as practical implications.

Chapter 2. Conceptual Model and Hypotheses

2.1 Conceptual Model

The conceptual model displays all the relevant variables and their hypothesized relation to and effect on each other and is presented in the following:

Figure 1. Conceptual Model.

Valence (positive vs. negative) of user-generated content will influence brand

convictions with the relationship being moderated by focus (focal brand vs. competitor

brand), which in turn will be moderated by involvement (high vs. low). It is expected that

positive user-generated content that focuses on the focal brand as well as negative user-

generated content that focuses on the competitor brand will enhance brand conviction. The

same goes for negative user-generated content that focuses on the focal brand, because it is

assumed to elicit a defensive response resulting in increased brand conviction. Positive user- generated content that focuses on the competitor brand will mitigate brand conviction, because it will increase cognitive dissonance. The effect of focus on the relationship between valence and brand conviction is expected to be moderated by involvement with high

involvement leading to a further increase in the effects predicted in Hypotheses 2-5.

Chapter 2.2 Hypotheses

More and more consumers seek out user-generated content, because it is perceived as more credible than content provided by traditional advertising or by experts. (Cheong &

Morrison, 2008; Ryan & Jonas, 2010; Crespo, Gutiérrez & Mogollón, 2014; Cronin & Taylor, 1992). Although mostly sought out before a purchase decision when consumers feel the most uncertain, user-generated content also affects consumers after a purchase, for example in form of increased brand conviction (Probst, Grosswiele & Pfleger, 2013; Liang, 2016; Merriam- Webster’s dictionary, 2017; Oxford Dictionary, 2017). Because it is likely to increase purchase intention, marketers have discovered this new group of people to become their new brand endorsers (Mendoza, 2010; Littman, 2008; MacKinnon, 2012, Cao, Meister & Klante, 2014).

However, the effect of user-generated content on consumers may be dependent on its valence, that is, the goodness (positive valence) or badness (negative valence) of the user- generated content (Shuman, Sander & Scherer, 2013). Positive experiences of others expressed through positive user-generated content strengthen the belief in our purchase decision and are likely to result in an increase in brand conviction. This is because we feel affirmed in our action when we see that other people have behaved in the same way that we did, which is also known as social proof (Cialdini, 1983; Amblee & Bui, 2011; Hennig-Thurau, Walsh & Walsh, 2003;

Gupta & Harris, 2010). In a purchase context, we are constantly seeking out this form of social

proof by looking at user-generated content in form of reviews, ratings and online recommendations (Amblee & Biu, 2011; Hennig-Thurau, Walsh & Walsh, 2003).

Negative user-generated content expressing negative information may have a detrimental negative effect for both the company and the brand (Tirunillai & Tellis, 2012;

Amblee & Bui, 2011; Bambauer-Sachse & Mangold, 2011; Luo, 2009). Additionally, negative information has a stronger effect on consumer judgement than positive or neutral information and should consequently lead to a significant decrease in consumer’s brand conviction (Herr, Kardes & Kim, 1991; Zou, Yu & Hao, 2011).

Hypothesis 1: Valence of user-generated content influences brand convictions, in that positive user-generated content is expected to lead to an increase in brand conviction and negative user-generated content is expected to mitigate brand conviction.

The effect of valence of user-generated content on brand convictions may depend further on the focus of user-generated content, where positive or negative user-generated content about the focal brand may elicit a different reaction than positive or negative user- generated content about the competitor brand. For example, positive user-generated content about the focal brand confirms the consumer’s purchase choice through social proof and thereby takes away consumer uncertainty by affirming that they have made the right choice (Cialdini, 1983; Amblee & Bui, 2011; Hennig-Thurau, Walsh & Walsh, 2003; Gupta & Harris, 2010). It can be assumed that positive user-generated content about the focal brand will lead to an increase in brand conviction.

Hypothesis 2: User-generated content about the focal brand, when positive, will lead to an increase in brand conviction towards the focal brand.

Consumers may react differently when they are faced with negative user-generated

content of the focal brand. Particularly, the negative experience made by others elicits a reactive

response in them: following Reactance Theory (Clee & Wicklund, 1980), consumers perceive

a differing opinion as an aggressive, freedom-threatening influence attempt towards change that carries and puts pressure on them. In order to re-establish their freedom, consumers move in the direction that lies opposite the influence attempt, which is also termed the “boomerang effect”. Applied to the user-generated content context, it means that when consumers are exposed to negative user-generated content regarding the brand they purchased, it will trigger reactance in them in form of both counterarguing all negative information regarding the focal brand and a maintained belief in all positive information concerning the focal brand (Ahluwalia, Burnkrant & Unnava, 2000; Rains, 2012).

In addition to psychological reactance, cognitive dissonance will be triggered in the consumer: Cognitive Dissonance Theory (Oshikawa, 1969) asserts that individuals who are exposed to an attitude, opinion or behavior that is not consistent with their own will experience dissonance. Dissonance will be triggered in three following situations: after being told to do or say something that is not in line with personal beliefs or attitude (1), when facing discrepant information (2), especially when this information is new and was not available at point of purchase and finally, after making a decision (3). Especially in the latter situation dissonance is most likely to take place, because the individual still has to cope with the attractiveness of the alternative.

Since negative user-generated content regarding a focal brand exposes consumers to information that is not line with their beliefs (1), discrepant (2) and only available after they have made their purchase decision (3), cognitive dissonance will be triggered. In order to reduce dissonance, consumers can try to change their opinion (1), seek more information (2), engage in perceptual distortion by perceiving the discrepant information as untruthful (3) or avoid exposure to a dissonance trigger in the first place (4) (Oshikawa, 1969; Liang, 2016).

Since previous research has already shown that exposure to negative information regarding a

brand elicits a reaction that consists of counterarguing the negative information and

maintaining the belief in positive information, the dissonance triggered by the negative user- generated content of the focal brand is predicted to lead to perceptual distortion by perceiving the negative user-generated content as untruthful (3) (Rains, 2012; Ahluwalia, Burnkrant &

Unnava, 2000). Together, both reactance and cognitive dissonance will lead to an increase in brand conviction towards the focal brand.

Hypothesis 3: User-generated content about the focal brand, when negative, will lead to an increase in brand conviction towards the focal brand.

The opposite effect can be expected for consumers who are exposed to negative user- generated content about the competitor brand. Just like consumers who are exposed to positive user-generated content about the focal brand, negative user-generated content about the competitor brand is likely to confirm consumers purchase decision, since it shows that they have successfully decided against purchasing an alternative brand that caused negative experiences for others, thereby re-affirming their own purchase decision (Cialdini, 1983;

Amblee & Bui, 2011; Hennig-Thurau, Walsh & Walsh, 2003; Gupta & Harris, 2010).

Additionally, negative user-generated content does not only show detrimental effects for a company and its brand but it also impacts consumer judgement a lot more than positive user-generated content (Tirunillai & Tellis, 2012; Amblee & Bui, 2011; Bambauer-Sachse &

Mangold, 2011; Luo, 2009; Herr, Kardes & Kim, 1991; Zou, Yu & Hao, 2011). For this reason, negative user-generated content of the competitor brand can be expected to turn the consumer away from the competitor brand and increasing brand conviction for the focal brand instead.

Hypothesis 4: User-generated content about the competitor brand, when negative, will lead to an increase in brand conviction towards the focal brand.

After having made a purchase, consumers display a selection bias where they expose

themselves to online reviews that support their purchase decision exclusively and avoid those

that do not (Liang, 2016). The purpose behind this is that being exposed to the attractiveness of the alternative product decreases consumer satisfaction with their current product and increases consumers’ switching intention and actual switching behavior (Liu, Li, Xu, Kostakos

& Heikkilä, 2016). Exposure to positive user-generated content of the competitor brand opens up the possibility that consumers may have made the wrong choice by not choosing the alternative brand in favor of the focal brand. Especially shortly after their purchase decision, the attractiveness of the alternative brand may be still very prevalent to consumers, which is also the point when they are most vulnerable to experiencing cognitive dissonance (Oshikawa, 1969).

Experiencing cognitive dissonance is also more likely because through positive user- generated content of the competitor brand consumers are faced with content that is not consistent with their purchase decision and is not in line with their beliefs (1), it represents discrepant information (2) that was not available at the point of purchase and consumers have to deal with the information at point of post-purchase, which is when they are the most vulnerable to information regarding alternative attractiveness (3). It is predicted that consumers will experience a decrease in brand conviction when being exposed to positive user-generated content regarding the competitor brand (Liu et. al., 2016).

Hypothesis 5: User-generated content about the competitor brand, when positive, will lead to a decrease in brand conviction towards the focal brand.

Involvement describes the extent of the perceived personal relevance that a product has

depending on the needs, interests and values of the consumer (Shiue & Li, 2013). The level of

involvement decides about consumers’ decision-making processes and purchases. Brand

involvement in particular is defined as the core equity of a brand that is expressed by the

consumer. Increased liking and brand involvement results from the voluntary acquisition, direct

experience or physical possession of a brand or branded product. Continuous brand

involvement leads to protection against brand dissidents and to repatronage intentions in spite of dissatisfaction with the brand (Shiue & Li, 2013). Involvement showed to moderate the effect eWOM messages on consumers where consumers with higher involvement showed a greater attitude toward the product, website and judged the credibility of eWOM to be larger (Doh & Hwang, 2009). Consequently, when exposed to positive user-generated content about the focal brand or negative user-generated content about the competitor brand, involvement should increase both predicted effects of increased brand conviction.

When exposed to negative information highly brand committed consumers display a bias towards their brand and attest more importance to the positive rather than the negative information (Ahluwalia, Burnkrant & Unnava, 2000). They also avoid attitude derogation by defending against all negative information and delivering counterarguments, while lower committed consumers experience a decline in brand attitude (Ahluwalia, Burnkrant & Unnava, 2000, Rains, 2012). Based on this, involvement should increase the predicted effect of higher conviction following negative user-generated content of the focal brand and increase the hypothesized effect of lower conviction following positive user-generated content of the competitor brand.

Hypothesis 6: Involvement will enhance the effects of Hypotheses 2 - Hypotheses 5.

Chapter 3. Methods

Data Collection

One hundred and eighty-nine participants participated in the survey, that was held in

English. Eighty of these participants did not complete the survey until the end, so that these

responses had to be discarded. Due to a mistake with the images that occurred in the beginning

of data collection in one of the experimental conditions, responses from five participants had

to be additionally discarded, whereupon the images were fixed. Seventeen participants indicated to have suspicions regarding the real study purpose or estimated the purpose of the study correctly. However, after a boxplot showed a normal distribution and indicated no outliers, it was decided to leave these participants in the data set.

Content-related questions following the blog-post were used to assess attentiveness of participant’s reading. Of the 189 participants, 50 participants did not reach the part that contained the control-questions because they stopped the survey beforehand. Regarding the remainder of participants: 23 indicated an incorrect answer to the first control question, 16 for the second control question and four gave a wrong answer to the third control question. All of the participants who answered the control questions incorrectly were excluded from the study, because experimental validity could not be assured for them.

In the end, responses from 102 participants were used for the final data analysis (M

= 27.3, SD = 7.172, Min: 18, Max: 58) of which 37 were male and 65 were female. Participants had various Nationalities such as North-American (2.9%), South-American (2%), Australian (2%), Asian (N = 14.7%) and European (N = 78.4%). Dutch (N = 20.6%) and German (N = 22.5%) participants were in the majority of the sample.

Research Design

In this study, a 2 x 2 (focus: focal brand versus competitor brand x valence: positive or negative) between-subjects factorial experiment was conducted. Through random sampling participants were allocated to the focal brand condition (Saeco) with positive (a; n=24) or negative (b; n=25) valence or to the competitor brand condition (Delonghi) with positive (c;

n=27) or negative (d; n=26) valence.

Before being exposed to the experimental manipulation, participants read a text which

aimed at inducing a post-purchase context. It was created by the experimenter before the study

(see appendix 1) and described a recent coffee machine purchase of the brand Saeco (focal brand). Adapted from Liang (2016) coffee makers were selected as a product category due to the fact that previous research showed low variance for product quality ratings and low reported difficulty for ranking the products. Due to the well-known popularity of the brands, we chose Saeco as the focal brand and Delonghi as the competitor brand.

User-generated content was manipulated using user-generated blog posts that were also created by the experimenter (see appendix 2). It was decided to use blogs as the communication format since user-generated content in form of blogs is generally regarded as higher in credibility than other marketing communication tools used by a brand (Akehurst, 2008). Each blog post consisted of both textual and visual content and either reviewed the coffee maker made by Saeco (focal brand) positively (a) or negatively (b), or Delonghi (competitor brand) positively (c) or negatively (d). Positive valence consisted of three described positive characteristics of the coffee maker and negative valence consisted of three described negative characteristics of the coffee maker. It was agreed upon to limit the amount of arguments to three because three is usually associated with impressions of completeness and sufficient amount of proof of evidence (Antonakis, Fenley & Liechti, 2012).

Content-related questions (see appendix 3) that followed the blog post, controlled for attentive reading of the blog-post. In order to increase the effects of the experimental induction, we aimed to increase personal relevance by telling participants that they may receive the coffee maker they were presented with during the experiment (Lee & Schwartz, 2010; Losciuto &

Perloff, 1967; Liang, 2016).

Data was collected using a web-based survey, which was built in Qualtrics. This

anonymous and uniform method of collecting data has the benefit of reducing social

desirability bias and controlling for response styles (de Leeuw, 2008). The survey link was

distributed within the experimenter’s own network in order to find respondents. Participants

received the survey link and completed the entire survey online and in their own time independently.

Procedure

Firstly, participants read the informed consent. In the beginning, all participants were induced with a written experimental scenario (see appendix 1) about a recent purchase of a coffee maker by Saeco. Afterwards brand conviction was assessed (Kim, 2003). Then, participants were exposed to the experimental manipulation in form of user-generated content (see appendix 2). Subsequently, they answered three content-related control questions regarding the blog post in order to test whether they have read the blog post attentively (see appendix 3). Additionally, brand knowledge was assessed. Afterwards a second assessment of brand conviction followed to test for changes in level of brand conviction. At the end, participants completed the awareness check to ensure that they were still blind to the actual study purpose (see Table 1) (Liang, 2016). At the end, participants reported demographics (sex, age, nationality) and received the written debriefing which included the notification that they would not win a coffee machine.

Measurement of Variables

Involvement. Brand involvement was assessed using the 3-item “Involvement Scale”

by Kim (2003) (9-point-scale ranging from 1 = Do not agree at all to 9 = Absolutely agree), which can be found in Table 1.

Brand Conviction. Following Kim (2003), brand conviction was measured using the

5-item “Attitude Strength Scale” before the experimental manipulation as well as afterwards

to control for changes in level of brand conviction. Table 1 displays the scale.

Brand Knowledge. Brand knowledge was assessed previous to the experimental manipulation using the 5-item “Brand Awareness Scale” by Schivinski and Dabrowski (2014) with a 7-point Likert scale ranging from 1= Strongly disagree to 7=Strongly agree. The scale can be found in Table 1.

Table 1

Summary of Scales: Reliability Analysis and Factor Loadings.

Item Cronbach’s alpha if item deleted Component Loading

Brand Awareness Scale (Schivinski & Dabrowski, 2014) .95 .95

1. I know Saeco. .93 .93

2. I know at least one Saeco product. .93 .94

3. I easily recognize Saeco among other brands. .94 .92

4. I recognize the logo of Saeco. .94 .90

5. I know that there is a brand Saeco .95 .89

Attitude Strength Scale (Kim, 2003) (Pre-measure) .88

1. My attitude towards Saeco is… .85 .83

2. How strong or intense is your feeling towards Saeco in this category? .83 .88 3. How certain do you feel about your attitude towards Saeco in this

product category? .85 .82

4. How important would you say Saeco is to you? .85 .82

5. How knowledgeable do you feel you are about Saeco? .87 .76

Attitude Strength Scale (Kim, 2003) (Post-measure) .89

1. My attitude towards Saeco is… .87 .81

2. How strong or intense is your feeling towards Saeco in this category? .84 .89 3. How certain do you feel about your attitude towards Saeco in this

product category? .87 .81

4. How important would you say Saeco is to you? .86 .82

5. How knowledgeable do you feel you are about Saeco? .86 .82

Brand Involvement Scale (Kim, 2003) .76

1. I care a great deal in selecting this coffee machine from many other

choices available in the market. .68 .83

2. It is important to me to make the right choice of coffee machines. .65 .85 3. I am concerned about the outcome of my choice in making my selection

of this coffee machine. .72 .80

Control Questions .74

1. The blogger is writing during summertime and how she enjoys the sun. .67 .80

2. The blogger was drinking an espresso. .64 .82

3. The blogger does not usually like coffee and prefers tea. .64 .82

Awareness Check

1. What do you think the current study is about?

2. What do you think the purpose of the study is?

Method of Analysis.

Valence. A One-way Analysis of Covariance (ANCOVA) was performed in order to test whether the independent categorical variable valence with the two levels positive valence (+1) and negative valence (-1) has an effect on the continuous dependent variable brand conviction as indicated in Hypothesis 1. Brand knowledge as a continuous variable and pre- measures of the continuous variable brand conviction were used as covariates in this test.

Focus. A Two-way ANCOVA was used to test how the categorical independent variable focus with the levels focal brand (+1) and competitor brand (-1) of user-generated content moderates the relationship between valence and brand conviction. Brand knowledge and the pre-measure of brand conviction were used as a covariate in this test. An interaction term between the independent variable valence and the moderator variable focus was added to test hypotheses 2-5.

Involvement. A multiple regression analysis was used to test for the three-order interaction analysis between valence as the independent variable, focus as the first moderator variable, involvement as the second order continuous moderator variable and conviction as the dependent variable. Brand knowledge and the brand-conviction pre-measure were the control variables in this analysis to test for Hypothesis 6.

Chapter 4. Results

Descriptives and Correlations

Table 2 presents the means and standard deviations as well as the correlations

between the independent, dependent and covariates that were used. No significant correlation existed between the dependent post-measure of brand conviction and valence (r = 0.02, p

= .28) or brand conviction and focus (r = -0.07, p = .51). However, involvement was

significantly correlated with brand conviction (post-measure) (r = 0.48, p > .001). Brand knowledge showed a significant correlation with both the pre-measure of brand conviction (r

= 0.52, p < .001) as well as the post-measure (r = 0.51, p < .001). Interestingly, gender, which was coded -1 for female and +1 for male, was also significantly correlated with age (r

= 0.20, p = .04).

Table 2

Means, Standard Deviations and Correlations of Measures.

M SD 1 2 3 4 5 6 7 8

1 Age 27.25 7.17 –

2 Gender 1.64 0.48 0.20* –

3 Valence 0.00 1.00 -0.01 -0.06 –

4 Focus -0.04 1.00 -0.04 -0.15 -0.02 –

5 Involvement 6.26 1.83 0.02 0.02 0.05 -0.15 –

6 Brand

Conviction (Pre)

5.36 1.81 -0.02 0.48 0.02 -0.05 0.50** –

7 Brand Conviction (Post)

5.29 1.90 -0.16 0.08 0.11 -0.07 0.48** 0.87** –

8 Brand Knowledge

4.75 2.02 0.01 0.20* 0.13 -0.17 0.30** 0.51** 0.51** –

* Correlation is significant at the 0.05 level (2-tailed).

** Correlation is significant at the 0.01 level (2-tailed).

Hypotheses Testing.

Valence. In order to test the effect of valence on brand conviction (Hypothesis 1), a One-way ANCOVA was conducted to determine a statistically significant difference between positive and negative valence on brand conviction controlling for brand knowledge and the pre-levels of brand conviction. Levene’s test and normality checks were carried out and the assumptions met. There was no significant effect of valence on brand conviction after

controlling for brand knowledge and pre-levels of brand conviction, F(1, 101) = 2.61, p = .09,

η p

= .030, so Hypothesis 1 was not supported. The ANCOVA model is summarized in Table 3.

Table 3

Results (Hypothesis 1).

Variable F p ηp

M

M

Valence 3.03 .085 .030 5.50 5.09

Brand

Knowledge 1.50 .224 .015

Brand Conviction

(pre-measure) 211.29 .000 .683

Note. Factor type of valence: group 1 = positive valence, and group 2 = negative valence

Focus. To test the effect of focus of user-generated content on the relationship between valence and brand conviction (Hypothesis 2-5), a Two-way ANCOVA was conducted to determine a statistically significant difference between focal brand focused user-generated content and competitor brand focused user-generated content on the relationship between valence and brand conviction controlling for brand knowledge and pre-levels of brand conviction. Levene’s test and normality checks were carried out and the assumptions met.

There was a significant interaction effect between valence and focus of user-generated content on brand conviction, F(1, 101) = 14.307, p < .001, ηp

= .130. Neither of the main effects was statistically significant, valence: F(1, 101) = 3.787, p = .06, ηp

= .038: focus:

F(1, 101) = .040, p = .84, ηp

< .001. The ANCOVA model can be found in Table 4 and a

plot of the interaction effect can be found in Figure 2. When inspecting the means (Table 5),

we can see that Hypotheses 2, 4 and 5 were supported, but Hypotheses 3 was not.

Table 4

Results (Hypothesis 2-5).

Variable F Sig. (p) ηp

M

M

Valence 3.79 .055 .038 5.50 5.09

Focus 0.04 .841 .000 5.16 5.41

Valence*Focus 14.31 .000 .130

Brand

Knowledge 2.80 .098 .028

Brand Conviction

(pre-measure) 204.11 .000 .680

Note. Factor type of valence: group 1 = positive valence, and group 2 = negative valence;

factor type of focus: group 1 = focal brand, and group 2 = competitor brand

Table 5 Descriptives.

Valence Focus Mean

(Pre)

SD (Pre)

Mean (Post)

SD (Post)

N Positive Focal Brand

Competitor Brand Total

5.63 5.18 5.33

1.59 1.69 1.65

6.00 5.05 5.50

1.64 1.60 1.67

24 27 51

Negative Focal Brand Competitor Brand Total

4.92 5.72 5.33

2.31 1.56 1.98

4.36 5.78 5.09

2.27 1.66 2.09

25 26 51

Total Focal Brand Competitor Brand Total

5.27 5.45 5.36

5.27 1.63 1.81

5.19 5.41 5.29

2.13 1.66 1.90

49

53

102

Note. Dependent Variable: Brand Conviction.

Figure 2. Interaction between valence and focus (‘type’).

Involvement. In order to test whether involvement moderates the effect of focus of user-

generated content on the relationship between valence and brand conviction (Hypothesis 6), a

multiple regression analysis was conducted. A summary of the regression model can be found

in Table 3. There was a main effect for valence (B = 0.21, t(95) = 2.43, p = .017) as well as a

significant three-order interaction between valence, focus and involvement (B = 0.14, t(95) =

2.94, p = .004), so that Hypothesis 6 was supported. The pre-measure of brand conviction

was also significant (B = 0.82, t(95) = 13.26, p <.001) which is explained by the high

correlation between the two. Finally, high multicollinearity existed between the two

interaction terms only, so that results can be interpreted without needing to exclude

predictors.

Table 6

Regression Analysis of Valence, Focus and Involvement on Brand Conviction.

Variable B SE B β t Sig. (p) VIF

Valence 0.21 0.09 .11 2.43 .017 1.05

Focus -0.02 0.09 -.01 -0.27 .785 1.06

Involvement 0.04 0.05 .04 0.69 .493 1.40

Valence*Focus -0.53 0.31 -.28 -1.72 .089 13.81

Valence*Focus*

Involvement 0.14 0.05 .48 2.94 .004 13.70

Brand Conviction

(Pre-measure) 0.82 0.06 .78 13.26 .000 1.78

Brand Knowledge 0.08 0.05 .09 1.62 .109 1.45

R2 0.82 0.84

F 60.17 .00

Chapter 5. Discussion

In this research, we investigated whether valence of user-generated content (UGC) would have an effect on brand conviction. Furthermore, it was predicted that focus would moderate the relationship: positive and negative user-generated content of the focal brand as well as negative user-generated content of the competitor brand would increase brand conviction whereas positive user-generated content about the competitor brand would mitigate brand conviction. Involvement was hypothesized to moderate the effect of focus of user-generated content on the relation between valence and brand conviction.

In our research, Hypotheses 2, 4, 5 and 6 were supported, whereas Hypotheses

1 and 3 were not. The reason why we did not manage to find a main effect for valence, could

lie in the fact that the content of user-generated content is important when it comes to

changing brand convictions. Evidence for this lies in the fact that we managed to find an interaction effect between valence and focus of user-generated content: whether positive or negative user-generated content would have an effect on brand conviction depended on whether the user-generated content focused on the focal brand or the competitor brand.

Especially, highly involved consumers attain greater importance to the focus of user- generated content in form of higher quality for example than lower involved consumers do (Lee, Park & Han, 2008). Previous research has already shown that the effect of user- generated content may be dependent on what the content is focusing on: sales of books by well-known authors were hurt by negative online reviews whereas authors that were not familiar to readers benefited from negative reviews (Berger, Sorensen & Rasmussen, 2009).

Therefore, the focus of user-generated content can be considered an important moderator when it comes to the effect of user-generated content on consumer response variables such as brand conviction.

The highest level of brand conviction was found for positive user-generated content of the focal brand and negative user-generated content of the competitor brand. This was

followed by positive user-generated content of the competitor brand and lastly, negative user- generated content of the focal brand. Although, hypotheses 2, 4 and 5 were supported, hypothesis 3 was not. Since the difference in means between the pre-measure of brand conviction and the post-measure of brand conviction is minimal when it comes to negative user-generated about the focal brand, only tentative conclusions should be drawn here. One possibility for a decrease in brand convictions here could be that participants did not really own a Saeco machine and therefore the effects of cognitive dissonance and psychological reactance were smaller than they would be if participants actually owned the machine

themselves (Quick, Scott & Ledbetter, 2011). Since our sample size was limited, it could also

be that a lack of findings might stem from here.

Finally, we found a third-order interaction effect between valence, focus and involvement: the effect of focus of user-generated content on the relationship between valence of user-generated content and brand conviction was moderated by level of

involvement where high involvement increased the effects that were predicted in hypotheses 2-5 giving support for Hypothesis 6. Highly involved participants seem to gain in brand conviction following positive user-generated content about the focal brand and negative user- generated content about the competitor brand but they also seem to defend against negative claims. When confronted with negative information, they disclaim the negative claims, maintain their positive beliefs and show greater reactance (Quick, Scott & Ledbetter, 2011;

Ahluwalia, Burnkrant & Unnava, 2000, Rains, 2012).

Nonetheless, since we conducted a multiple regression analysis, the interpretation of predictors remains difficult here. However, two factors should lend additional certainty to our results: firstly, despite the risk that young people may have not yet developed a preference within coffee machines because of their so far limited exposure to coffee machines and the potential lack of involvement, this showed not to be case in our study. Instead, there was sufficient variation within the population concerning involvement. Secondly, our final sample size of 102 participants was very limited, but nevertheless we were still able to find

significant results, which with a larger sample size nay even be larger.

Limitations

It is possible that factors other than the valence of user-generated content caused the effect on brand conviction. For example, participants might have disliked the design of the Delonghi coffee machine they saw in the pictures and therefore preferred the Saeco coffee machine without user-generated content being a direct influencing factor in that relationship.

For future research, it is important to control for the design or the external appeal factors of

the coffee machine. Some participants indicated in the final comment section of the study that they were not familiar with the brand. The responses from these participants could bias the results. The same goes for participants who usually do not drink coffee at all, who might be even less familiar with both coffee machine brands, Saeco and Delonghi. Future research should control for coffee drinking in the study and make sure that the brands mentioned in the study are well familiar to the participants.

Finally, the author of the blog post was a girl and the writing style together with the pictures that were chosen were slightly more female-oriented (e.g. less technical, more emotional content; pink flower vases displayed in pictures), which might appeal more to women and have a greater effect on them in comparison to participants who are male.

Therefore, in the next experiment the blog post should be written in a more gender-neutral way.

Future Research

The current research confirms previous research that attested user-generated content a positive brand effect (Vermeulen & Seegers, 2009). However, the effect of user-generated content was mostly studied before purchase (Probst, Grosswiele & Pfleger, 2013; Liang, 2016). This study presented a positive effect of user-generated content after purchase. Future investigations should therefore focus on variables such as brand loyalty and brand

commitment and investigate the relevance of these in the context of user-generated content.

Practical Implications

Before drawing practical inferences for organizational practice, the research should be

replicated first. Overall, our results suggest that positive and negative user-generated content

of the focal and competitor brand impacts brand convictions and that the effect is also

moderated by involvement. If organisations aim to incorporate user-generated content into their overall marketing strategy, it seems important to expose consumers to user-generated content throughout the customer decision journey since brand convictions seem to be malleable independent of point of purchase.

Conclusion

To conclude, our results show that both positive and negative user-generated content

of the focal and the competitor brand is able to change brand convictions during post-

purchase. This effect is also moderated by level of involvement. Due to the major growth

user-generated content has seen within marketing, future investigations into this topic can

assumed to be highly worthwhile.

References

Ahluwalia, R., Burnkrant, R. E., & Uninava, H. R. (2000). Consumer responses to negative publicity: the moderating role of commitment. Journal of Marketing Research, 37, 203-214.

Akehurst, G. (2008). User generated content: the use of blogs for tourism organisations and tourism consumers. Service Businesses, 3,51-61.

Amblee, N., & Bui, T. (2011). Harnessing the influence of social proof in online shopping:

the effect of electronic word of mouth on sales of digital microproducts. International Journal of Electronic Commerce, 16, 91-114.

Antonakis, J., Fenley, M., & Liechti, S. (2012). Learning charisma. Harvard Business Review, 90, 127-130.

Bae, S., & Lee, T. (2011). Product type and consumers’ perception of online consumer reviews. Electronic markets, 21, 255-266.

Bambauer- Sachse, S., & Mangold, S. (2011). Brand equity through negative online word-of- mouth communication. Journal of Retailing and Consumer Services, 18, 38-45.

Berger, J., Sorensen, A. T., & Rasmussen, S. J. (2009). Positive effects of negative publicity:

When negative reviews increase sales. Marketing Science, 29, 815-827.

Conviction. (2017). In Merriam-Webster’s dictionary. Retrieved from https://www.merriam- webster.com/dictionary/conviction

Conviction. (2017). In Oxford dictionary. Retrieved from https://en.oxforddictionaries.com/definition/conviction

Cao, P., Meister, S., & Klante, O. (2014). How social media influence apparel purchasing behavior. Spektrum- Marketingmanagement, 6, 77-85.

Cheong, H. J., & Morrison, M. A. (2008). Consumers’ reliance on product information and

recommendations found in UGC. Journal of Interactive Advertising, 8, 1-29.

Cialdini, R. (1983). Influence. The Psychology of Persuasion. New York, NY: Harper Collins.

Clee, M. A., & Wicklund, R. A. (1980). Consumer behavior and psychological reactance.

Journal of Consumer Research, 6, 389-405.

Crespo, A. H., Gutiérrez, H. S. M., & Mogollón, J. M. H. (2014). Perceived influence on behaviour of user-generated content on social network sites: An empirical application in the hotel sector. Revista Espanola de Investigación de Marketing ESIC, 19, 12-23.

Cronin, J. & Taylor, S. (1992). Measuring service quality: A re-examination and extension.

Journal of Marketing, 56, 55-68.

de Leeuw, E. D. (2008). Choosing the method of data collection. In International Handbook of Survey Methodology, 113-135, edited by E. D. de Leeuw, E. D., Hox, J. J., &

Dillman, D. A. New York: Lawrence Erlbaum.

Dippong, J., Kalkhoff, W., & Johnsen, E. C. (2017). Status, networks, and opinion change:

An experimental investigation. Social Psychology Quarterly, 80, 153-173.

Doh, S.-J., & Hwang, J.-S. (2009). How consumers evaluate eWOM (electronic word-of- mouth) messages. CyberPsychology & Behavior, 12, 193-197.

Gatignon, H. (1983). Competition as a moderator of the effect of the effect of advertising on sales. Journal of Marketing Research, 21, 387-398.

Gupta, P., & Harris, J. (2010). How e-WOM recommendations influence product

consideration and quality of choice: A motivation to process information perspective.

Journal of Business Research, 63, 1041-1049.

Hennig-Thurau, T., Walsh, G., & Walsh, G. (2003). Electronic word-of-mouth: Motives for and consequences of reading customer articulations on the internet. International Journal of Electronic Commerce, 8, 51-74.

Herr, P. M., Kardes, F. R., & Kim, J. (1991). Effects of word-of-mouth and product-attribute

information on persuasion: an accessibility-diagnosticity perspective. Journal of Consumers Research, 17, 454-462.

Huang, I., Chou, Y., & Lan, I. (2007). Effects of perceived risk, message types, and reading motives on the acceptance and transmission of electronic word-of-mouth

communication. Contemporary Management Research, 3, 299-312.

Kotler, P., Brown, L., Adam, S., Burton, S. & Armstrong, G. (2007). Marketing. New South Wales: Pearson.

Kim, J. (2003). Affective and cognitive brand conviction: making loyal relationships (Doctoral dissertation). Retrieved from:

https://www.researchgate.net/publication/266471878_AFFECTIVE_AND_COGNITI

VE_BRAND_CONVICTION_MAKING_LOYAL_RELATIONSHIPS

Lee, J., Park, D.-H., & Han, I. (2008). The effect of negative online consumer reviews on product attitude: An information processing view. Electronic Commerce Research and Applications, 7, 341-352.

Lee, S. W., & Schwarz, N. (2010). Washing away postdecisional dissonance. Science, 328, 709.

Liang, Y. (2016). Reading to make a decision or to reduce cognitive dissonance? The effect of selecting and reading online reviews from a post-decision context. Computers in Human Behavior, 64, 463-471.

Littman, S. (2008). Welcome to the new millennials. Response magazine, May, 74-80.

Liu, Y., Li, H., Xu, X., Kostakos, V., & Heikkilä, J. (2016). Modeling consumer switching behavior in social network games by exploring consumer cognitive dissonance and change experience. Industrial Management & Data Systems, 116, 801-820.

Liu, X., Burns, A. C., & Hou, Y. (2017). An investigation of brand-related user generated

content on Twitter. Journal of Advertising, 46, 236-247.

Losciuto, L. A., & Perloff, R. (1967). Influence of product preference on dissonance reduction. Journal of Marketing Research, 4, 286-290.

Luo, X. (2009). Quantifying the long-term impact of negative word of mouth on cash flows and stock prices. Marketing Science, 28, 148-165.

MacKinnon, K. (2012). User generated content vs. Advertising: Do consumers trust the word of others over advertisers? The Elon Journal of Undergraduate Research in Communications, 3, 14-21.

Mendoza, M. (2010). The Journal of Digital Research & Publishing, Semester One, Seven PM Edition, 114-122.

Oshikawa, S. (1969). Can cognitive dissonance theory explain consumer behaviour? Journal of Marketing, 33, 44-49.

Probst, F., Grosswiele, L., & Pfleger, R. (2013). Who will lead and who will follow: Identifying influential users in online social networks. Business and Information Systems Engineering, 179-193.

Quick, B. L., Scott, A. M., & Ledbetter, A. M. (2011). A close examination of trait reactance and issue involvement as moderators of psychological reactance theory. Journal of Health Communication, 16, 660-679.

Rains, S. A. (2012). The nature of psychological reactance revisited: a meta-analytic review.

Human Communication Research, 39, 47-73.

Rogers, K. B. (2015). Expectation states, social influence and affect control: opinion and sentiment change through social interaction. Advances in Group Processes, 32, 65-98.

Ryan, J. & Jonas, O. (2010). Source credibility of company – Product and user-generated content on the internet: An exploratory study on the Filipino Youth. Philippine Management Review, 17, 121-132.

Schivinski, B., & Dabrowski, D. (2014). The consumer-based brand equity inventory: Scale

construct and validation. GUT – Faculty of Management and Economics, Working Paper Series A, 4, 1-25.

Shiue, Y.-C., & Li, L. S. H. (2013). Brand involvement in retaining customers despite dissatisfaction. Social Behavior and Personality, 41, 643-650.

Shuman, V., Sander, D., & Scherer, K. R. (2013). Levels of valence. Frontiers in Psychology, 261.

Tirunillai, S. & Tellis, G. J. (2012). Does chatter really matter? Dynamics of user-generated content and stock performance. Marketing Science, 31, 198-215.

Thoumrungroje, A. (2014). The influence of social media intensity and EWOM on conspicuous consumption. Procedia - Social and Behavioural Sciences, 148, 7-15.

Vermeulen, I. E. & Seegers, D. (2009). Tried and tested: the impact of online hotel reviews on consumer consideration. Tourism Management, 30, 123-127.

Zou, P., Yu, B., & Hao, Y. (2011). Does the valence of online consumer reviews matter for

consumer decision making? The moderating role of consumer expertise. Journal of

Computers, 6, 484-488.

Appendices Appendix 1 Experimental Scenario Description

You have recently purchased a new coffee machine. After doing some research online and going to the store to look at different coffee machines you have decided to purchase a Saeco coffee machine. This is a full automatic coffee machine that is supposed to fulfill every coffee lover's need.

After your purchase at the local electronics store, you arrived home and unpacked the Saeco machine. You were worried that setting up the machine might take hours, but after reading the manual and going through the installation guide, you managed to set up the Saeco machine within half an hour only. You feel happily surprised to see that the Saeco

machine is so easy to use.

You decide to try out a few different coffee drinks starting with a classic cappuccino.

After putting some fresh milk in the milk carafe and pouring some coffee beans into the bean container you select "Cappuccino". Within one minute, the Saeco machine prepares a freshly hot cappuccino for you. The temperature is hot, the milk froth is thick and the coffee aroma and strength are on point. You feel happy to sit down on the couch and enjoy a delicious cappuccino now.

After you finished your cappuccino, you get up in order to clean the Saeco machine, but you quickly realize that there is no need for this. The Saeco machine cleans itself

automatically each time you prepare a drink. You feel super happy about this high level

of convenience and very satisfied you return to the couch to enjoy the rest of your day.

In the evening you invite your friends to come over for dinner. When they enter the kitchen, they see the new Saeco machine. They are all very impressed by the high-quality design and give you a lot of compliments for purchasing such a nice looking device. You tell them about all the positive experiences you have made with the Saeco machine so far and recommend the purchase to them.

Overall, you feel very satisfied with the purchase of the Saeco machine. From the

taste of the coffee to the ease of cleaning to the nice design it has. You feel like you truly

made a successful purchase here. From now on you plan to stick to Saeco only when it

comes to coffee machines.

Appendix 2 Experimental Conditions.

Positive Valence Negative Valence

Focal Brand

Pleasant Coffee Moments with the SAECO Coffee Machine

Since it’s November again and winter is approaching, nothing is better than staying in bed, reading a book and enjoying a hot cup of coffee next to it. Therefore, I decided to treat myself to a new coffee machine!

After a looot of research, I decided to go for the authentic Italian Saeco coffee machine and I made the right decision.

I have never tasted a better cappuccino ever before and you guys know I have tasted a lot of them!

Such strong coffee aroma together with lots of thick foam, that’s how a cappuccino should look like.

One important thing when it comes to coffee machines is the cleaning. This Saeco machine manages to clean itself each time a drink is prepared. After pouring the beans and making the coffee, I do not have to clean the entire Saeco machine, which is fun :D

Another advantage is the great design: the Saeco machine simply looks good in every kitchen.

Every time I look at it, it makes me feel happy, because it’s just so beautiful!

Better turn the Saeco machine on again and enjoy another delicious cup of coffee. I hope that your coffee moments are as cozy as mine and that you are enjoying this lovely fall. J Yours,

Claire

Unpleasant Coffee Moments with the SAECO Coffee Machine

Since it’s November again and winter is approaching, nothing is better than staying in bed, reading a book and enjoying a hot cup of coffee next to it. Therefore, I decided to treat myself to a new coffee machine!

After a looot of research, I decided to go for the authentic Italian Saeco coffee machine but I made the wrong decision.

I have never tasted a worse cappuccino ever before and you guys know I have tasted a lot of them!

Such weak coffee aroma together with lots of thin foam, that’s how a cappuccino should not look like.

One important thing when it comes to coffee machines is the cleaning. This Saeco machine does not manage to clean itself each time a drink is prepared. After pouring the beans and making the coffee, I have to clean the entire Saeco machine, which is not fun L Another disadvantage is the horrible design:

the Saeco machine simply looks bad in every kitchen. Every time I look at it, it makes me feel unhappy, because it’s just so ugly!

Better turn the Saeco machine off again and enjoy no other horrible cup of coffee. I hope that your coffee moments are cozier than mine and that you are enjoying this lovely fall. J

Yours,

Claire

Competitor Brand

Pleasant Coffee Moments with the DELONGHI Coffee Machine

Since it’s November again and winter is approaching, nothing is better than staying in bed, reading a book and enjoying a hot cup of coffee next to it. Therefore, I decided to treat myself to a new coffee machine!

After a looot of research, I decided to go for the authentic Italian Delonghi coffee machine and I made the right decision.

I have never tasted a better cappuccino ever before and you guys know I have tasted a lot of them!

Such strong coffee aroma together with lots of thick foam, that’s how a cappuccino should look like.

One important thing when it comes to coffee machines is the cleaning. This Delonghi machine manages to clean itself each time a drink is prepared. After pouring the beans and making the coffee, I do not have to clean the entire Delonghi machine, which is fun :D

Another advantage is the great design: the Delonghi machine simply looks good in every kitchen. Every time I look at it, it makes me feel happy, because it’s just so beautiful!

Better turn the Delonghi machine on again and enjoy another delicious cup of coffee. I hope that your coffee moments are as cozy as mine and that you are enjoying this lovely fall. J

Yours, Claire

Unpleasant Coffee Moments with the DELONGHI Coffee Machine

Since it’s November again and winter is approaching, nothing is better than staying in bed, reading a book and enjoying a hot cup of coffee next to it. Therefore, I decided to treat myself to a new coffee machine!

After a looot of research, I decided to go for the authentic Italian Delonghi coffee machine but I made the wrong decision.

I have never tasted a worse cappuccino ever before and you guys know I have tasted a lot of them!

Such weak coffee aroma together with lots of thin foam, that’s how a cappuccino should not look like.

One important thing when it comes to coffee machines is the cleaning. This Delonghi machine does not manage to clean itself each time a drink is prepared. After pouring the beans and making the coffee, I have to clean the entire Delonghi machine, which is not fun L

Another disadvantage is the horrible design:

the Delonghi machine simply looks bad in every kitchen. Every time I look at it, it makes me feel unhappy, because it’s just so ugly!

Better turn the Delonghi machine off again and enjoy no other horrible cup of coffee. I hope that your coffee moments are cozier than mine and that you are enjoying this lovely fall. J

Yours,

Claire

Appendix 3 Control Questions.

Now after reading the blog article we are going to ask you a few questions about it. Please indicate your answer to each question by ticking the appropriate box.

1 = Strongly agree

2 =

Disagree 3 =

Somewhat disagree

4 = Neither agree nor disagree

5 =

Somewhat agree

6 =

Agree 7 = Strongly agree

The blogger is writing during summertime and how she enjoys the sun.

The blogger was drinking an espresso.

The blogger

does not

usually like

coffee and

prefers tea.