The phenomenon of unboxing videos : the answer to sponsored video reviews?

(1)

The Phenomenon of Unboxing videos:

the answer to sponsored video reviews?

Inês Isabel de Freitas Fernandes

Student number: 11085649

Date of submission – version: 18th_{August 2017 – final} MSc. in Business Administration – Marketing Track Amsterdam Business School, University of Amsterdam

(2)

(3)

Statement of Originality

This document is written by Inês Fernandes who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(4)

Acknowledgements

I know I should keep it short, but these are the “thank you”s I really want to make, as this piece of (digital) paper is the culmination of a very long journey, that means a lot to me.

First, I would like to thank my family for the unconditional support. My father for the endless patience and keeping me so much company, and my mother for always giving me confidence. To my brother-in-law Pedro and to my sister Dani, who supported me and were always available to help and find solutions for my despairs. To my sister Ana and to my brother-in-law Pedro, for deconstructing my fears and concerns, turning them into light.

I would like to thank my friends for keeping in touch and being supportive. A special thank you to my dear friend Panita, whose friendship assures me that despite the distance, I will never walk alone.

I want to thank my colleagues at Philips, not only for their support and understanding, but also for tolerating my despair and always having encouraging words. A special thank you to dear Rosi, a recent friend that is truly that – a friend.

I have to thank Catarina Carvalho and Anabela Lobo de Carvalho for the encouraging words and for giving me the opportunity to go on this adventure knowing I was safe. I want to thank Mariana Alpoim, my first work colleague, for everything she taught me – lessons that helped more than I could have ever imagined, and that I will carry with me throughout life.

My biggest “thank you” goes to João Pedro for joining me in this adventure, for helping me believe I can do anything, for always dreaming big, and most importantly, for always taking care of me in every single way.

Lastly, I would also like to thank me – for getting this far, for not giving up, for giving my very best. I did it! 

(5)

Abstract

This thesis explored the phenomenon of unboxing videos, as part of the understudied, but very relevant, field of visual eWoM. First, it is important to understand what characterizes unboxing videos in order to understand their popularity and further explore their effects. Fifty-four unboxing videos were analysed resulting in the identification of twelve potentially defining characteristics. Then, under the light of source credibility theory, through an online survey, this research examined whether unboxing videos produced higher message source credibility compared to other non-unboxing videos. The results showed no differences between the effects of unboxing videos and the effects of non-unboxing videos. Lastly, this research examined whether unboxing videos produced different effects on message source credibility and viewers’ attitude towards the brand according to different sponsorship conditions (sponsored, not sponsored, unclear). Both ground truth and perceptual data were analysed, revealing that message source credibility and brand attitude were not significantly impacted by different sponsorship conditions/perceptions of unboxing videos.

(9)

A brand is no longer what we tell the consumer it is – it is what consumers tell each other it is.

(10)

1 Introduction

“Influencer marketing works because people influence people. Today, consumers are choosing to listen to each other over brands.” (Influencer Marketing Agency)

There is a need to know and a will to share, and allied with technological progress, this dichotomy sustains the era of user-generated content. Internet’s increased accessibility and development turned the seemingly impossible into the status quo – anyone can share whatever they want, with whomever they want, wherever they are. While this could solely reshape interpersonal communications, it has proved to affect the ways of Marketing greatly. As King et al. (2014, p.167) summarised: “The Internet has transformed the way we search for information, how we interact with each other and, more importantly, the way we shop”.

Internet and social media extended traditional of-mouth (WoM) into electronic word-of-mouth (eWoM) (Weiss, 2013) by broadening personal networks beyond family and friends, allowing social interactions to happen online, with people we do not necessarily know (Cheung et al., 2009). More technology also enriched the way we communicate by enabling the use of visual cues, such as pictures or videos, to convey or support our message. Instagram posts, videos, snaps, GIFs, memes, emojis, etc., are all part of visual eWoM, which despite its popularity, has remained understudied (King et al., 2014).

An interesting and increasingly popular form of visual eWoM entails sharing an unboxing video (King et al., 2014), i.e., a video that shows a person opening a product for the very first time and reacting to it, for the very first time. A simple search for “unboxing” on YouTube gives an astonishing 60.9 million results (as of May 21st, 2017). Compared to the same period in the previous year, the same search yielded 41.8 million results (as of July 1st, 2016), which means since then, 59,000 unboxing videos on average have been posted each day.

(11)

Despite the evident popularity as a form of visual eWoM, little is known about the defining characteristics of unboxing videos. Assessing these characteristics is important to further understand this phenomenon, especially considering unboxing videos are often shared by digital influencers, who consumers have been progressively preferring to listen to instead of branded communications (The Rise of Influencers Report, 2016; Lockwood, 2016).

As a pivotal piece connecting brands and (potential) consumers, digital influencers can contribute to a product’s success, as much as to its failure. On a study conducted by Twitter and Annalect, 40% of respondents affirmed they bought an item because they saw an influencer talk about it on social media (The value of influencers on Twitter, 2016). However, if their opinion is not favourable, digital influencers might represent a risk for brands, with serious potential to negatively impact both their reputation and sales (Cheung, Lee & Rabjohn, 2008). The fact that digital influencers represent both the risk of losing control over what is shared about a product/brand and the opportunity to approach consumers in a new way (Salyer, 2012), has been determining brands’ interest in working with them, and increasing their influencer marketing budgets (The Rise of Influencers Report, 2016). Thus, sponsored reviews have become more common on social media.

Nonetheless, sponsorship has always been a sensitive topic under consumers’ critical eyes, who are often suspicious of brands’ intentions (Chu & Kamal, 2008). For successful use of influencer marketing, it is imperative that brands carefully assess the way they exploit the digital influencer domain (The Rise of Influencers Report, 2016). Unboxing videos, as a part of visual eWoM – a discipline in need of more research (King et al., 2014) – are then an interesting context to further explore the effects of sponsorship of reviews.

Firstly, a key concern of this thesis is to understand what defines these videos besides the fact that they all show someone opening a product for the first time, on camera. Secondly, this research aims at understanding how unboxing videos compare to other non-unboxing video

(12)

reviews in terms of their effects on message source credibility. The source credibility theory is indispensable to argue on this potential difference. It states that the higher the credibility of the source of the message, the more likely the message is to be accepted and believed (Hovland & Weiss, 1951; Ohanian, 1990; Wu & Wang, 2011). Because unboxing videos show authentic first reactions to a product, they can potentially generate higher message source credibility, even when sponsored, compared to other non-unboxing videos. If a brand sends digital influencers (who in this case are the message source) products for them to review publicly, offers discount codes for them to share with their followers or offers monetary compensation in exchange for mentions and content about the brand, consumers may indeed question their credibility as reviewers (Chu & Kamal, 2008). However, in an unboxing video, the digital influencer supposedly has never opened or experimented the product before. Viewers may then believe that their opinion reflects what they really think despite the existence of a compensation from the brand, regarding the reviewer (message source) as more credible than they would if the review was made in the form of other, non-unboxing video.

Hence, the first questions to be addressed by the present research are:

RQ1: What characteristics are often present in unboxing videos, having the potential to be defining of them?

RQ2: To what extent does the type of video review (unboxing/non-unboxing) influence message source credibility?

Another concern of this research involves the extent to which sponsorship of unboxing videos affects, not only message source credibility, but also viewers’ attitudes towards the brand.

(13)

brand (Cowley & Barron, 2008). It may also signal to viewers that the relationship between the reviewer (message source) and the brand is based on commercial interests, and that the reviewer may not truly like the brand or its products (Colliander & Erlandsson, 2015). Thus, sponsorship may lower message source credibility as well as attitude towards the brand.

Curiously, a study by Colliander & Erlandsson (2015) found contradicting results – despite sponsorship, viewers may believe that the reviewer does like the brand and associates with it not due to compensation but due to real brand appreciation. Moreover, if the reviewer discloses the sponsorship, viewers may also regard the review as honest (Lu, Chang & Chang, 2014).

What happens when the video is not sponsored or when its sponsorship is not clear, as it may depend on message cues and on viewers’ attention to those cues. However, in the context of unboxing videos, again due to the fact that they show someone opening a product for the first time and having their very first reactions to it, this may not decisively influence viewers.

Besides understanding how sponsorship (sponsored, not sponsored, unclear) affects message source credibility and brand attitude, this research is also interested in understanding where that effect is stronger – on message source credibility or on brand attitudes – so as to understand who bears more risks in the sponsoring relationship. Because message source credibility also affects brand attitude (Colliander & Erlandsson, 2015), it may be expected that consumers perceived reviewer credibility varies more depending on whether the unboxing video is sponsored or not rather than on their attitudes toward the brand.

The final questions this research addresses are as below:

RQ3: To what extent does sponsorship of unboxing video affect message source credibility and viewers’ attitude towards the brand?

All in all, this research aims at diminishing the void in visual eWoM literature by examining the specific case of unboxing videos. First, the phenomenon needs to be defined, so it can then

(14)

be properly analysed. Then, it is important to understand how this type of videos compares to other types of video reviews. Lastly, it is important to understand how consumers react to sponsorship of unboxing videos in particular, in order to comprehend whether there are differences compared to other types of sponsored content. This last point is particularly interesting since studies have found contradicting evidence in terms of the effects of sponsorship in consumers’ attitude towards the brand. By bringing more clarity into the phenomenon of unboxing videos and the effects of its sponsorship, this research takes a step further into understanding whether the way consumers react to sponsorship varies with the type of content sponsored.

More specifically, its key contributions include assessing the defining features of unboxing videos, finding whether these videos are an appropriate format to present sponsored content, namely compared to other forms of video reviews (since they may lead to higher message source credibility than other forms of sponsored video reviews), and understanding to what extent sponsorship of unboxing videos affects message source credibility and viewers’ attitude towards the brand, as well as where that effect lies the most (on message source credibility or viewers’ attitude towards the brand). To address RQ1 and understand what the distinct characteristics of unboxing videos are, 54 unboxing videos were watched and salient features of each were noted down, and all cases were organized in a table. A frequencies analysis then revealed which characteristics are likely to be present in most unboxing videos. To address the remaining research questions, two quantitative experimental studies were conducted via two surveys. These were distributed online employing convenience sampling methods: one via email and Facebook, and other via Amazon’s Mechanical Turk. The chosen industry for this research is colour cosmetics/make-up because not only do these products have a mixed nature (search goods and experience goods) but also the industry is relying more and more on digital campaigns and digital influencers.

(15)

After this extensive Introduction, the Literature Review will describe relevant concepts and explore extant literature to help building the hypotheses to be tested, as well as the conceptual model depicting them. Next follows Methodology, where research design and process are detailed. Results are then presented, outlining the supported and/or rejected hypotheses. This thesis ends with result discussion, followed by final conclusions on the research questions proposed.

1.1 Managerial and Academic Relevance

In terms of managerial relevance of this research, it is important to stress that the digital revolution has impacted brands’ marketing strategies, especially as digital influencers have grown into a privileged source of information over branded communications. Naturally, brands have become interested in exploiting this relationship, but it is important that they do it cautiously, to avoid negative reactions from consumers. In an era where the amount of available information is colossal, consumers are constantly looking for authentic, trustworthy opinions that can guide them in making good purchase decisions (The Rise of Influencers Report, 2016; Lockwood, 2016). Also, considering that digital influencers share information in various different formats, more research is needed to understand how consumers react to each format, so that brands know what formats are most effective and safe. Unboxing videos are already popular, so they are a format worth further exploring. Moreover, their nature may make them less susceptible of negative reactions to sponsorship.

Concerning theoretical relevance, this research aims to extend the literature on visual eWoM, which remains understudied (King et al., 2014). Defining the key characteristics of unboxing videos is also a valuable contribution as it provides basis for further research on this type of visual eWoM, namely on the reasons why it is popular and on the potential effects it may have on viewers. Moreover, as King et al. (2014) confirm, there is little research on

(16)

consequences to senders of visual eWoM. Thus, understanding the effects of sponsorship of unboxing videos on the credibility of the reviewer (sender of visual eWoM) provides another context in which light the phenomenon can be observed and the theory tested.

Lastly, a brief note on Tellis (2017), who urges researchers to understand that research is not a quest for truth. For it to be impactful, instead it needs to refute a belief or assumption of its audience. In that sense, this research aims also at proving that there is something unique or special about unboxing videos, believing that due to their nature they are a phenomenon on their own, being much more than simply “another form of visual eWoM” or yet just another video review.

(17)

2 Literature Review

Even though there is considerable literature on WoM and eWoM, references on visual eWoM are still scarce (King et al., 2014). Considering that consumers are trying to adapt and learn how to process all the information so steadily available in the form of pictures and videos, and if influencers and brands are still inferring how to make the most of the consumer-digital influencer-brand triad, it is easy to understand more research is needed.

An in-depth analysis on unboxing videos is given and WoM, eWoM and visual eWoM are amongst its key parent concepts. This section reviews concepts that are relevant to the development and explanation of the conceptual model and its hypotheses.

2.1 WoM, eWoM and visual eWoM

Before proceeding with further, deeper analyses, it is important to understand the key concepts that host unboxing videos, namely eWoM and WoM.

The concept of WoM regards informal communications directed at others concerning the evaluation of the ownership, usage, or characteristics of particular goods and services or their sellers (Anderson, 1998). It has been labelled as “the most effective, yet least understood marketing strategy” (Misner, 1999), being responsible for 20% to 50% of purchasing decisions (Bughin et al., 2010). Since it entails consumers exchanging their own knowledge without a commercial purpose, WoM is usually regarded as more credible than regular advertisements (Wu & Wang, 2011).

WoM has evolved with the Internet and it expanded to accommodate eWoM – a narrower concept that entails “any positive or negative statement made by potential, actual, or former customers about a product or company, which is made available to a multitude of people and institutions via the Internet” (Hennig-Thurau et al., 2004, p. 39). Despite being essentially the

(18)

same as WoM but within the online context, it represents one of the most significant developments in contemporary consumer behaviour (Rosario et al., 2016).

King et al. (2014) identify six characteristics that depict the nature of eWoM: (i) enhanced volume – referring to its wide reach; (ii) dispersion – because product-related conversations happen across different platforms and communities; (iii) persistence and observability – since eWoM is available on demand and remains public virtually forever, and because due to its textual nature, message content and characteristics such as semantics and type of language become indicative of eWoM credibility and usefulness; (iv) anonymity and deception – because the internet is a relatively anonymous environment and sometimes brands take advantage of that by manipulating online reviews to their favour; (v) salience of valence – regarding the numerical scaling that leaves less room for misinterpretation of eWoM valence; (vi) community engagement – on the grounds that eWoM platforms are essentially specialized and consequently consumer communities are now non-geographically bound.

Factors such as risk reduction in the purchasing decision and reduced search efforts are some of the motivations for people to seek eWoM (Hennig-Thurau et al., 2004). However, it is worth noting that eWoM can be sought without the intention to seek information, which might suggest it has an entertaining dimension (Mangold et al., 1999). It has also been proved that consumers that acquire information through eWoM have higher interest on the product/category than consumers that rely solely on information coming from brands (Wu & Wang, 2011). Compared to brand-generated information, eWoM is more persuasive and more desired by consumers (Goldsmith & Horowitz, 2006).

Although its market relevance is recognized, how to manage eWoM successfully remains unclear (Rosario et al., 2016). eWoM can be found in blogs, online forums, review sites, and social networking sites (Goldsmith, 2006). It not only influences purchase decisions but it is also an outcome of it (Duan et al., 2008) – consumers seek eWoM to decide what to buy and,

(19)

after purchase, they become agents of eWoM themselves, for instance, by writing reviews in online forums. How messages are conveyed and assimilated, particularly in the online context, considerably affects the impact of eWoM (King et al., 2014).

In the past, online recommendations were submitted in text form. However, with ever-evolving technology in digital communication, it has become much easier for consumers to share pictures or videos on the internet, sharing product features and consumption-related experiences. Lin et al. 2012 stated that Internet marketers should understand how to utilize visual information properly to better communicate with consumers. Prior eWOM studies focused primarily on text-based instead of visual elements of eWOM (Alon & Brunel., 2006; Lin et al., 2005). Like WoM, eWoM has ripen into a new concept – visual eWoM, which focuses on the types of eWoM that include visual cues such as pictures or videos. If one consumer records a video explaining her/his opinion and decides to share it on the internet instead of simply writing a review, then it would be considered visual eWoM.

On “What We Know and Don't Know about Online Word-of-Mouth” (2014), King et al. explore the state of the literature on eWoM, alerting that despite the increasing interest on the topic, many questions remain unanswered, both from theoretical and managerial perspectives. Rietveld et al. 2016 answered that previous call and aimed to make path in defining and operationalising the concept of visual eWoM and embedding it in the current eWoM literature. Their work also mentioned that visual eWoM has received very little research attention and that unboxing videos are more and more a popular form of visual eWoM.

2.2 Defining Unboxing Videos

Understanding a phenomenon is the first step in both defining and comprehending what characterizes it. The concept of unboxing videos is easy to explain – videos showing someone opening a product for the very first time – but knowing the core characteristics that make them

(20)

a unique type of video review, with its own marketing potential, is what can drive further research.

Referring back to chapter 2.1, unboxing videos fit the concept of WoM because they are shared with others and include information about a product and/or its sellers. However, one of the basic premises of WoM is that its credibility is strongly based on the absence of an inherent commercial purpose, which may be compromised if unboxing videos are sponsored. Nevertheless, from a practical perspective, the decisive factor may depend on how consumers deal with that sponsorship and how they feel towards the reviewer, the review itself and the cues they grasp. As for the six characteristics that depict the nature of eWoM, unboxing videos include all, with a note for observability, since besides textual elements it contains visual elements that can reinforce or reduce credibility and relevance, and an exception for anonymity, since visual cues usually include the face of the reviewer, revealing his/her identity.

In order to properly frame and analyse the phenomenon, this research analysed 54 unboxing videos to assess its potentially defining characteristics. The industry chosen was that of colour cosmetics (make-up) and the reasons that lead to this choice are detailed on section 2.6. The table with all the data collected and a description of the process can be found in Appendix A. This analysis was key to answer RQ1, which inquires what are the potentially defining characteristics of unboxing videos. Analyses that are part of a study should usually be found under the Methodology chapter. However, due to the relevance of this concept for the present study, the insights derived from the analysis of the unboxing videos are explained here. For the scope of this study, only the features that were present in at least half of the videos watched are being considered. Based on those features, it is possible to say that unboxing videos include “unboxing” on the title and involve more than one product. In terms of flow of content, their structure is also often similar. Reviewers mark the moment when the product is about to be opened (start of the actual unboxing) by saying something like “I will

(21)

open it now”, and they state enthusiasm in doing so and in having the product. Product packaging is then described and the blogger usually “swatches” the product (for make-up this means trying the product in the back of the hand, or arm, just to have an impression of colour, texture, effect, finish, etc.), revealing her/his first impressions of the product. Here, sensorial product features such as texture and smell are also described, and interestingly, often only positive aspects about the product are communicated. Information about the brand in general, about its products or about the range of products it belongs to, is also included, alongside information on the price. Another interesting finding is that bloggers often reinforce their enthusiasm in using the products in the long run. The videos usually end with an interactive goodbye note, where bloggers encourage viewers to share their thoughts on the product or the video itself.

It is important to reinforce that all these features are inserted in the context of unboxing videos, when video viewers know or believe that the person is opening the products for the first time. This analysis showed that unboxing videos go beyond simply showing someone opening a product for the very first time and actually provide viewers with more information on the product, its range and the brand. Their content is actually rich and the information provided is diverse, having the potential to entertain and even convince viewers since they are seeing a first review by someone with some degree of expertise, who is able to provide information not only on the product being reviewed but also on the brand. It is also worth noting that these characteristics give unboxing videos the potential to be perceived as more authentic, especially considering they involve lots of emotional expression, with the reactions from the reviewers being well marked. Allied with the fact that they show someone opening a product for the first time, unboxing videos can make a good case to argue that viewers may regard them more favourably than other non-unboxing videos.

(22)

Knowing the potentially defining characteristics of unboxing videos will help further build their case, and is determinant to properly analyse their potential effects.

2.3 Conceptual Model

For simplicity purposes, to illustrate two of the three main research questions (RQ2, and RQ3), two conceptual models are presented below, each addressing one research question, representing the relevant variables as well as the hypotheses to be tested. Next, the remaining of Chapter 2 (Literature Review) will explain the hypotheses and help understanding the conceptual model.

Figure 1.1 – Conceptual model illustrating RQ2

Figure 1.2 – Conceptual Model illustrating RQ3

Type of video review: - Unboxing

- Non-unboxing

Message Source Credibility: - Trustworthiness - Expertise RQ2 H1, H2 Sponsorship of Unboxing video: - Sponsored - Not Sponsored - Unclear Brand attitude RQ3 H3, H4

Message Source Credibility: - Trustworthiness

- Expertise

H5, H6 H7, H8

(23)

2.4 Source Credibility Theory

The source credibility theory affirms that the more the source of the message appears credible, the more likely message receivers are to accept and believe the message (Hovland & Weiss, 1951; Ohanian, 1990; Wu & Wang, 2011). Message source credibility translates into an attitude towards the message source that influences how much the message receiver believes in the sender (Wu & Wang, 2011).

Hovland & Weiss (1951) conducted an experiment to inquire whether transmitting the same piece of information via a highly credible source or via a less credible source would generate similar responses from message receivers. They presented one sub-group with a piece of written information on a topic signed by a highly credible source (e.g., field magazine), and another sub-group with the same information but signed by less credible sources (e.g., mass circulation magazines). The results indicated that despite the judged communications being identical, subjects responded very differently to the "high credibility" and "low credibility" sources, considering the information signed by the latter less fair or justified than the former. Message source credibility unmistakably affects the way message receivers regard the message itself. For this reason, it is considered a relevant variable to assess in whether unboxing video reviews potentially have a considerable impact on how consumers assimilate their content, compared to other non-unboxing video reviews.

2.4.1 Message Source Credibility

Message source credibility is a multidimensional construct composed by three sub-dimensions: trustworthiness, which represents confidence and acceptance from message receivers towards the message source; expertise, depicting the source’s knowledge about a topic; and attractiveness, describing source’s perceived social value, including physical appearance or social status, or similarity to the receiver (Ohanian, 1990; Chu & Kamal, 2008).

(24)

As mentioned by Ohanian (1990), trust can generate positive outcomes, such as acceptance and favourable disposition. It has been indicated as the most influential of the three message source credibility sub-constructs (McGinnies & Ward, 1980). Because unboxing videos show someone opening a product, seeing it and reacting to it for the first time, viewers may perceive the message source as more trustworthy than other type of videos, even when the video is sponsored. A first reaction is something spontaneous that derives from natural impulses, which are not premeditated (Kipper & Shemer, 2006, p.24), and it is seen as expression of one’s authentic, inner thoughts and feelings (Harter, 2002, p. 382). First reactions reflect an uncalculated personality (Liebes, 2001) and thus may cause viewers of unboxing videos to perceive the source as more trustworthy than if they were watching a non-unboxing video, where the reviewer usually discusses past experiences with the product and her/his opinion tends to be more consolidated and potentially biased by other factors over time.

Concerning expertise, studies have proven that when faced with two sources with different levels of expertise, people tend to comply or follow the advice of the source they perceive to have a higher level of expertise (Crisci & Kassinove, 1973; Woodside & Davenport, 1974). Perceived source expertise in video reviews may depend on different factors, and difficulties in discerning an expectation arise. For example, in an unboxing video, the first reactions may be rather superficial and more about the excitement of having the product or the looks of it. However, unboxing videos often include comparisons of the opened product to other products and include more information whether it be about the brand, its products, or even the specific product range. Because they are opening a product for the first time, being able to immediately compare it to other products and give more information about the brand or the product range may signal that that knowledge is readily accessible. Of course this type of information can also be passed on through non-unboxing videos, however, these do not have the same spontaneity as unboxing videos, giving the reviewer time to prepare the review and thus search

(25)

for information on the brand or the product range. Hence, it is expected that for unboxing videos, perceived expertise of the message source is higher than for non-unboxing videos.

When building the initial judgement or impressions of someone, attractiveness is a relevant cue (Ohanian, 1990). It refers to one’s physical appearance, being measured by elegance, sexiness, classiness and beauty (Ohanian, 1991). Because it is so tied with how one looks, it really depends from reviewer to reviewer, from viewer to viewer. Even when comparing the same reviewer, in some instances he/she may present him/herself in a more casual style and thus less classy or elegant, while in other he/she may be presented in the opposite manner. Besides this variability, generally digital influencers from beauty, fashion or lifestyle contexts (which constitute the scope of this research) are considered attractive to some extent, since the segment they work in requires them to be well presented. The relevance of attractiveness in the analysis is relative and it then makes sense to focus on the variables that can explain more about the potentially different effects of different types of video reviews. Thus, this variable will be kept constant throughout the analysis and no hypotheses will be built based on it.

As the nature of unboxing videos is supposed to positively affect at least two dimensions of message source credibility, it is expected that message source credibility will be higher for unboxing videos than other non-unboxing videos. The hypotheses that resume the rationale above are stated as follows:

H1: There is a relationship between the type of video review and trustworthiness such that if the video review is an unboxing video, viewers will regard the reviewer as more trustworthy than if the video review is a not an unboxing video.

H2: There is a relationship between the type of video review and expertise such that if the video review is an unboxing video, viewers will regard the reviewer as having a higher level of expertise than if the video review is not an unboxing video.

(26)

After understanding the potential effects that unboxing videos might have when compared to other non-unboxing videos, it is important to further understand to what extent they make consumers more “tolerant” to sponsorship of reviews.

2.5 Sponsorship of Unboxing Videos

Some definitions of WoM stress that it entails consumer communication that does not have a commercial intention, making WoM a more trusted source of information than branded marketing communications (Wu & Wang, 2011). Moreover, anonymity is one of the characteristics of eWoM that is highlighted by King et al. (2014), since it is easy and common to remain anonymous in eWoM communities. However, considering the current digital panorama and the rise of digital influencers, these points brought by King et al. (2014) and Wu & Wang (2011) do leave out an important part of our reality, and are not necessarily true anymore.

Given the recurrence of posted content, digital influencers can become relatively well known to the public eye, with a status of digital personality/celebrity, with their followers regarding them more as acquaintances or even friends, than strangers. Reputations are then built and a face and a name are frequently attached to it. Moreover, as discussed in the Introduction of this research, brands started acknowledging the marketing potential of digital influencers, thus approaching them for collaborations or paid reviews, which in turn infused eWoM with commercial purposes, whether disclosed or not. And suddenly, a shift in paradigm happens, challenging the given characteristics of eWoM – there are sponsored (visual) eWoM communications initiated by people whose identity is known.

In the specific context of unboxing videos, sponsorship happens when the reviewer receives a compensation, of any kind, to talk about the brand on their social platforms. These compensations can be monetary or can be physical products sent for the reviewer to test.

(27)

As stressed by King et al. (2014), there is not a lot of research investigating the consequences to the senders of eWoM – who, in the scope of this research, are the digital influencers, as initiators or senders of visual eWoM communications. Analysing the effects of sponsorship on message source credibility is then exploring the consequences of sponsored eWoM for eWoM senders.

2.5.1 Sponsorship & Message Source Credibility

To consider the potential effect of sponsorship on message source credibility, it is important to assess that potential effect on trustworthiness and expertise, as these are the sub-constructs of message source credibility.

Considering sponsored reviews, trustworthiness may be compromised since sponsorship signals subjacent commercial intentions, which may cause consumers to doubt the honesty of the review (Chu & Kamal, 2008). Also, when viewers acknowledge they are being targeted and exposed to content with commercial intentions, they process that sponsored information differently, in a more biased, critical and sceptical way since they know they are trying to be convinced of something (Boerman, Reijmersdal, & Neijens, 2012). However, a study by Lu, Chang & Chang (2014) examined consumers’ attitudes towards sponsored recommendations and found that the type of sponsorship (direct vs. indirect monetary compensation to the blogger) does not affect attitude toward the sponsored review, possibly due to the fact that once sponsorship is revealed, viewers may perceive the review as honest, disregarding then the type of compensation the message source got. In the context of this research, it would then be expected that for sponsored and not sponsored unboxing videos trustworthiness will be high, since disclosing sponsorship or reinforcing the absence of sponsorship is a strong positive signal to viewers about the honesty and sincerity of the reviewer (which are measures of trustworthiness; Ohanian, 1990).

(28)

For the unclear condition, trustworthiness is still expected to be high. The uncertainty viewers face when unsure about sponsorship may become secondary as the absence of sponsorship disclosure may avoid the activation of persuasion knowledge, thus allowing for a more unbiased analysis of the content of the unboxing video (Boerman, Reijmersdal, & Neijens, 2012). Thus, since viewers are not told they are watching something sponsored or not sponsored, they will process the information they are receiving in a more unbiased manner, focusing more on the content of the unboxing video more than on its sponsorship. Moreover, because they are watching a first (and supposedly sincere and honest) reaction to a product, that typically involves emotional expression and other content and information, they can focus on those cues to confidently judge the reviewer’s trustworthiness, regardless of the absence of sponsorship disclosure.

As for expertise, the fact that the review is sponsored may lead consumers to question the extent to which the message source is conveying own opinion or just scripted information provided by the brand. However, as argued for H2, considering that in unboxing videos the reviewer often refers other products as comparison and adds information about the brand and its product ranges, expertise may be actually be perceived as higher. Making such comments and adding such information requires previous knowledge, as unboxing videos entail opening a product for the very first time. Thus, it is expected that expertise will be high for all sponsorship conditions.

Overall, message source credibility is not expected to vary significantly among the three conditions. The resulting hypotheses are stated below:

H3: There is a relationship between the sponsorship of unboxing videos (sponsored,

not sponsored, unclear) and trustworthiness such that viewers will regard the message source as trustworthy as when the video is not sponsored or when sponsorship is unclear.

(29)

H4: There is a relationship between the sponsorship of unboxing videos (sponsored, not sponsored, unclear) and expertise such that viewers will regard the message source as much as an expert as when the video is not sponsored or when sponsorship is unclear.

2.5.2 Sponsorship & Brand Attitude

Another important variable that might be affected by sponsorship is brand attitude. If it is plausible to consider that sponsorship has an impact on the credibility of the visual eWoM sender (message source), then it is also natural to expect that it impacts consumers’ attitude towards the brand.

An attitude can be defined as one’s internal evaluation of the brand (Mitchell & Olson, 1981), or as a learned predisposition to consistently respond favourably or unfavourably to an object (Fishbein & Ajzen, 1975), which can be a brand or even a piece of advertisement, for example. Both the Theory of Reasoned Action (Fishbein & Ajzen, 1975) and Theory of Planned Behaviour (Ajzen, 1985) consider that consumer attitudes (either towards an object, i.e., brand, or towards a behaviour, i.e., purchasing) affect their behavioural intention, ultimately influencing actual purchase behaviour. Thus it is relevant to understand the impact of sponsorship on unboxing video viewers’ attitudes toward the brand, as it may be a good indicator of their real intention to purchase. The better the attitudes toward the brand, the higher the purchase intentions.

A good reason to believe that sponsoring a review can lower attitude towards the brand has to do with persuasion knowledge, i.e. people being aware that they are being targeted by brands and the content they are reading/watching is aimed at persuading them (Friestad & Wright, 1994). Studies have found that activating persuasion knowledge impacts consumers’ attitudes towards the brand negatively and may even mitigate the persuasion efforts of the sponsored content (Wei, Fischer, & Main, 2008; Cowley & Barron, 2008; Boerman,

(30)

Reijmersdal, & Neijens, 2012). One of the studies was conducted to find whether there were circumstances under which product placement can actually have negative consequences on consumers’ brand attitude (Cowley & Barron, 2008). Explicit sponsorship (as a form of prominent product placement) was found to activate persuasion knowledge and lead to lower brand attitude. These lower attitudes stem from the fact that people usually do not want to be persuaded and tend to resist such attempts (Campbell, 1995; Wei, Fischer, & Main, 2008). Thus, it is expected that in the sponsored unboxing video condition viewers’ attitude towards the brand will be lower than in the not sponsored condition.

It is also important to note that for persuasion knowledge to be activated, the persuasion attempt has to be recognized (Friestad & Wright, 1994). As argued in chapter 2.5.1, in an unclear condition, not all people will suspect of a persuasion attempt when sponsorship is not disclosed. Considering again the nature of unboxing videos and the potential excitement of seeing someone opening a product for the first time, they may distract from persuasion attempt recognition, avoiding a more critical and sceptical evaluation of the content and consequently a negative regard of the brand (Boerman, Reijmersdal, & Neijens, 2012). Thus it is expected that when sponsorship of the unboxing video is unclear, its effects on brand attitude will be similar to those on the not sponsored condition.

H5: There is a relationship between the sponsorship of unboxing videos (sponsored, not

sponsored, unclear) and viewers’ attitude towards the brand such that when the unboxing video is sponsored, viewers’ attitude towards the brand will be lower than when the unboxing video is not sponsored or when sponsorship is unclear.

H6: There is a relationship between the sponsorship of unboxing videos (sponsored, not

(31)

is unclear (sponsorship), viewers’ attitude towards the brand will be similar as when the

unboxing video is not sponsored.

2.5.3 Brand Attitude & Message Source Credibility

Considering the relationship hypothesized in H3, H4 and the relationship hypothesized in H5, H6, it is important to understand whether the first is stronger than the second. This would allow to understand which dependent variable is more affected by sponsorship of unboxing videos, being it expertise and trustworthiness (MSC) or viewers’ attitudes towards the brand.

An interesting study by Colliander & Erlandsson (2015) investigated the effects of revealing to blog readers that the content they were reading was sponsored. They compared the responses of two groups of blog readers – one group was told that the blogger was sponsored and the other group was not told that. They found that revealing sponsorship broke the trust blog readers had on the blogger, jeopardizing their attitudes towards the blog and lowering blogger credibility.

However and surprisingly, attitude towards the brand was not affected by sponsorship revelation. Disclosing sponsorship could have signalled that the blogger’s relationship with the brand was not be motivated by liking, but by the sponsorship compensation, which in turn could have lead readers to think the blogger did not like the brand and engaged with it just for a compensation. This would then backfire towards the brand, resulting in less favourable brand attitudes (Colliander & Erlandsson, 2015). However, the conclusions of the study were precisely the opposite – despite acknowledging the sponsorship and the potential relationship based on interests, readers still assumed the blogger praised the brand not because of the compensation received, but because the blogger really liked it. This allowed brands to avoid punishment, which may suggest that sponsoring reviews may be relatively safer for brands than for digital influencers. Hence the interest in exploring the effects of sponsorship on message

(32)

source credibility (trustworthiness and expertise) and viewers’ attitude towards the brand, in the context of unboxings videos.

Also, because sponsorship, as a persuasion technique from brands, may not come as a surprise, the attitudes toward them may not be much affected by its presence. However, viewers may develop a closer relationship with digital influencers, based on a high perceived credibility deriving from trustworthiness and expertise. This may make consumers more suspicious when a digital influencer accepts the sponsorship, which of course weakens trust and potentially expertise regards. As hypothesized in H1 and H2, however, this effect is expected to be lower for unboxing videos, compared to other non-unboxing videos.

Still, in the context of unboxing videos, it is expected that sponsorship affects message source credibility (trustworthiness and expertise) more than it affects the attitudes towards the brand.

H7: The relationship between the sponsorship of unboxing videos (sponsored, not sponsored, unclear) and viewers’ attitudes toward the brand is weaker than the relationship between the sponsorship of unboxing videos and trustworthiness.

H8: The relationship between the sponsorship of unboxing videos (sponsored, not sponsored, unclear) and viewers’ attitudes toward the brand is weaker than the relationship between the sponsorship of unboxing videos and expertise.

2.6 Research background: the colour cosmetics industry

The industry chosen for this research is that of colour cosmetics (make-up) – an industry strongly impacted by the heavy presence of an increasing number of digital influencers. These many strong digital influencers not only collaborate with brands but, some of them, eventually launch their own brands. It is common that the same digital influencer publishes sponsored

(33)

reviews and claims independency of opinion. Due to the relevance of digital influencers in the make-up industry, there is a considerable amount of visual eWoM content available.

Moreover, despite being search goods, since it is possible to get complete information about the product attributes before purchase (Park & Lee, 2009), colour cosmetics also have an experiential nature – even though it is possible to read about the features or the effects of the product, it is through usage that one can understand what the product does for them; thus, it is not possible to understand what the product attributes do before purchasing and using (Park & Lee, 2009). These products are also quite subjective since they are usually targeted specific types of skin, for example, and their effect often depends on skill to apply. When looking to purchase experience goods, eWoM is often sought to reduce uncertainty and risk, since it usually includes expert advice and opinions from more experienced users (Park & Lee, 2007). It is then easy to understand why this industry is one where eWoM is of major importance.

Lastly, since industry key players have strong reputations (e.g., L’Oréal Group, Estée Lauder Companies, etc.), are always setting the newest trends and thrive either through hype products or iconic products, it is an interesting industry to explore to what extent sponsoring can be more of a risk or more of an opportunity for brands, and for digital influencers.

(34)

3 Methodology

This section describes the methodology followed to test the aforementioned hypotheses. In this chapter, research design is explained and measures used for each variable are described.

As detailed on chapter 2.6 of the present thesis, the colour cosmetics/make-up industry was chosen to conduct the studies. This research is of explanatory nature since it analyses the effects of the independent variables (types of video reviews) on the dependent variables (message source credibility, brand attitude) based on theoretical models. Two quantitative studies in the form of experiments helped addressing the research questions. The experiments were conducted via surveys by manipulating the stimuli respondents were exposed to. The surveys were distributed online. Pre-tests ensured the stimuli (video reviews) reflected each condition as accurately as possible.

3.1 Research Design – Study 1

Study 1 addressed the second part of RQ1, which inquires to what extent different types of video reviews potentially affect message source credibility differently. The survey entailed watching a video review and answering some questions about how respondents regard the brand, the video itself, and most importantly, the reviewer/blogger (message source). Questions assessing product category involvement and knowledge about unboxing videos were also included. Respondents were randomly assigned to one of the two conditions: watching an unboxing video or watching another, non-unboxing video. To ensure that respondents actually watched a sufficient portion of the video, they were only able to proceed to the rest of the survey after one and a half minutes, which is enough to watch its introduction and have a substantial idea of it. Nevertheless, it was kindly asked to watch the video until the end, if possible. It was also alerted to disregard the advertisements that often precede YouTube videos

(35)

on the survey so it was not necessary to leave the survey to watch the video. The survey can be found in Appendix B.

The surveys were conducted online and hosted on Qualtrics. The population of this study is virtually everyone with internet access, since anyone that can watch the video is eligible to answer questions about it. However, because the colour cosmetics category may not be widely relevant to most people, involvement with the product category was assessed so that it can be seen whether there are differences. Non-probability convenience sampling was employed by asking co-workers, friends, family, and acquaintances to fill in the survey. It is important to note that some respondents were asked to forward the survey link to someone they know that has affinity with colours cosmetics/make-up, hence non-probability snowball sampling was also employed to some extent, though not primarily. Both sampling methods were chosen due to convenience, since they are free and easily allow to gather numerous respondents. Only one link hosted the survey, to facilitate sharing. When respondents opened the survey they were randomly shown one of the four videos selected for this study. The same questions were asked, regardless of the video shown. Respondents were reached via email and Facebook.

3.1.1 Stimuli choice & Pre-test – Study 1

For this study four videos were used as stimuli to represent two conditions: two to represent unboxing videos and other two to represent non-unboxing videos (any type of video review other than unboxing videos). The decision to have two videos per condition was made to allow for the introduction of some variance since videos have a lot of particularities, and some reviewers may b,

The selected videos were about the same brand (Benefit Cosmetics) to control for effects related to different brands. All videos were sponsored in some way, either because the brand

(36)

sent the product to the reviewer/blogger or had them participating in some exclusive event. Thus, minimum variance in terms of effects related to sponsorship was guaranteed. To ensure that respondents could easily understand the videos were clearly sponsored, a pre-test was conducted with 7 random people found at a cafeteria. All videos were readily considered to be sponsored, based on the reviewers’ introduction speech where they disclose they received the products. In terms of the characteristics, the selected unboxing videos contain “unboxing” in the title and effectively show the reviewer/blogger opening the product and reacting to it. For the non-unboxing videos, the criteria entailed not having “unboxing” on the title, nor first impressions. The links to each video can be found in Appendix C.

3.1.2 Survey Structure – Study 1

The survey opens with an Introduction briefly explaining the scope of the research and referring to guaranteed anonymity. First respondents are asked whether they know the brand, and if they respond “Yes”, they are immediately asked about their opinion/attitude towards the brand. Afterwards, they are shown a video and on that same page, they are asked about how they feel towards make-up (product category involvement), whether they have heard about unboxing videos and how often they recur to this kind of video when searching for product information. On the next section, respondents are asked whether they think the video is sponsored or not, giving them the option to also answer “I’m not sure”. They are also asked about their opinion on the video and on the reviewer/blogger (attitudes towards video, brand, and reviewer). The following section asks how respondents’ perceive the reviewer/blogger in terms of trustworthiness, expertise and attractiveness. On the last section, respondents are asked about demographics (age and gender). The survey ends thanking for the participation.

(37)

3.2 Research Design – Study 2

Study 2 addressed the RQ2, which assesses the impact of unboxing video sponsorship on message source credibility and attitude towards the brand. Again, the survey entailed watching a video review and answering questions about how respondents regard the brand (before and after the video), the video itself, and most importantly, the reviewer/blogger (message source). Similarly to Study 1, questions assessing product category involvement and knowledge about unboxing videos were asked.

Respondents were randomly assigned to one of the three conditions: watching a sponsored unboxing video, a not sponsored unboxing video, or an unboxing video that was neither clearly sponsored nor clearly not sponsored. Once again, respondents were only able to proceed to the rest of the survey after one and a half minutes, to ensure they watched at least a portion of the video. The same warning about the possible advertisement preceding the video was made. The video was embedded on the survey so it was not necessary to leave the survey to watch the video. The survey can be found in Appendix D.

The surveys were conducted online and hosted on Qualtrics. The population of this study is the same as Study 1’s. However, because the colour cosmetics category may not be relevant to most people, and because the scope of this study is focused on the particularities of unboxing video sponsorship in the cosmetics industry, it was relevant to inquire people that somehow may be more involved in the category. To be able to reach a more meaningful sample of this population, this survey was shared via Amazon’s Mechanical Turk, where there is a possibility to select the type of respondent wanted. From the criteria available, the selected was “Online Purchase – Health & Beauty” to ensure that respondents accessing the survey potentially had some affinity towards the category and the online world. Also, it was asked that respondents came from an English- speaking country to ensure they understood the videos and

(38)

the questions as well as possible. Thus, non-probability convenience sampling was employed. Only one link hosted the survey, and respondents were randomly shown one of the 3 videos selected for this study. The same questions were asked, regardless of the video shown.

3.2.1 Stimuli choice & Pre-test – Study 2

For this study three unboxing videos were used as stimuli so that each represents one of three conditions: sponsored, not sponsored, and unclear. The videos were again on the same brand (Benefit Cosmetics) to control for effects related to different brands. To ensure maximum variance in terms of the effects of each video, each video was selected so that it was either markedly sponsored, markedly not sponsored or really unclear. A pre-test with 9 random people found at a cafeteria confirmed the selected videos represented accurately the desired stimuli. The selected unboxing videos contain “unboxing” in the title and show the reviewer/blogger opening the product for the first time, describing it and reacting to it. A link to each video can be found in Appendix E.

3.2.2 Survey Structure – Study 2

The structure of survey for Study 2 is the same as that for Study 1, being the only difference that respondents are also asked about their on the brand after watching the video. This is to be able to assess whether the unboxing video, whether sponsored or not, can change previous attitude towards a brand.

3.3 Measurements

3.3.1 Message source credibility

The independent variable is a multidimensional construct composed by trustworthiness, expertise, and attractiveness (Ohanian, 1990). Even though attractiveness was assessed on the

(39)

surveys, it was kept constant in the analysis since the digital influencers of the industry chosen are usually well presented, and this research is interested on the features that can have a higher explanatory power of the different effects of videos within the industry chosen. Each sub-construct was assessed using a five-item bipolar five-point Likert scale, being operationalized according to Ohanian, 1990.

The Trustworthiness scale items are: undependable/dependable, dishonest/honest, unreliable/reliable, insincere/sincere, and untrustworthy/trustworthy. The Expertise scale items are: inexperienced/experienced, unknowledgeable/knowledgeable, unqualified/qualified, unskilled/skilled, and not an expert/expert. The Attractiveness scale items are: not classy/classy, ugly/beautiful, plain/elegant, not sexy/sexy, and unattractive/attractive.

3.3.2 Attitude towards the brand, Attitude towards the video and Attitude towards the reviewer

To measure each viewers’ attitudes toward the brand, the mean of a four-item (like-dislike, positive-negative, good-bad, favourable-unfavourable) bipolar five-point Likert scale was used (Holbrook & Batra, 1987).

3.3.3 Product-category involvement

Involvement is related to the relevance that one finds on a particular product or matter, based on needs, value and interest, and it entails the extent to which one finds a product relevant (Zaichkowsky, 1985; Wu & Wang, 2011). The reduced Personal Involvement Inventory scale developed by Zaichkowsky (1994) was used for measurement. It includes ten items and was measured using a bipolar five-point Likert scale. The scale was adapted to five points to avoid having a survey that was too demanding on respondents. The average of those 10 items marks high and low involvement.

(40)

3.4 Analysing data

3.4.1 Missing values

Participants were required to answer all questions before proceeding to the next set of questions, thus there are no missing values for respondents who completed the surveys, with exception for a question on both studies that did not force answers. This question asked about respondents’ attitude towards the brand and was only displayed if the answer to the previous question (“Do you know the make-up brand Benefit Cosmetics?”) was “Yes”. In total, there were four cases in Study 1 with some missing values and one case in Study 2. This variable is not central for Study 1 as it does not affect neither the independent variable, nor the dependent variable. Thus, pairwise deletion will be employed so that the cases with missing values can still be used for analysis that entail variables without missing values. For Study 2, since the variable is important to measure changes in attitude towards the brand, this case was deleted from the sample. The missing values were signalled by “-99”.

The answers from respondents who interrupted the survey were not considered for the analysis (20 incomplete surveys for Study 1 and 10 incomplete surveys for Study 2).

3.4.2 Computing variables

The data collected was exported from Qualtrics into SPSS for analysis. On Study 1, even though the same questions were asked in each of the four videos, each video had its own variables assigned so that it was possible to grasp its specific results, if necessary. The SPSS file was exported to Excel so that through an “if” function the video each respondent watched could be easily identified. This Excel file was then exported to a new SPSS file containing a new column indicating which video that respondent had seen (column “TypeVideo”: 1 -

(41)

corresponded to 1 instead of “Like a lot”). The same was done for Study 2 (column “TypeVideo”: 1 - “unboxing sponsored”; 2 – “unboxing not sponsored”; 3 – “unboxing unclear”).

On both studies, some counter-intuitive variables were reversed so that all variables in the study pointed at the same direction. The reversed variables referred to opinion/attitude towards the video, the brand and reviewer (all four-item five-point Likert scale; reversed to “Dislike:Like”; “Positive:Negative”; “Good:Bad”; “Unfavourable:Favourable”).

(42)

4 Results & Discussion

4.1 Study 1 – Unboxing videos vs Non-unboxing videos

4.1.1 Basic data analysis

Sample. A total of 109 respondents participated in the survey for Study 1. 69 respondents belonged to the age group from 25 to 34 years old, which represents a large portion of the sample (63.3 percent). 91 respondents were female, which represents the vast majority of the sample (83.5 percent). Thus, the sample is skewed towards female and young-adults, which considering the scope of this research is not a problem since make-up is usually regarded as something in the feminine realm and young-adults are more tech-savvy and likely to search for information online. All respondents indicated their gender and age.

The manipulated conditions were well distributed: 57 respondents watched an unboxing video (52.3 percent) and 52 watched a non-unboxing (47.7 percent). The tables containing information on basic data can be found in Appendix F.

Only 38 respondents (34.9 percent) said they knew the brand. From those that knew the brand, 21 (55.3 percent) had a positive attitude toward the brand, between 4.0 and 5.0 points, while 4 (10.5 percent) had a negative attitude toward the brand, between 2.25 and 1 points. These values were based on attitude towards the brand computed as a mean of four reversed scaled items (reversed to Dislike-Like, Negative-Positive, Bad-Good, Unfavourable-Favourable).

53 respondents (48.6 percent) had moderate to high involvement (from 3.5 to 5 points) with the make-up product category. 41 respondents (37.6 percent) had moderate to low involvement (from 3.4 to 2.5 points) with the product category. The remaining 15 respondents (13.8 percent) had low involvement with the product category. These values were based on

(43)

involvement with the product category computed as a mean of 10 items as per the scale of Zaichkowsky, 1994.

74 respondents (67.9 percent) confirmed they have heard about unboxing videos. From the ones that know unboxing videos, 18 (24.3 percent) claimed to never watch them to gather product information, while the majority (34 respondents, 45.9 percent) claimed to watch them with that purpose “sometimes”.

78 respondents (71.9 percent) considered the video they saw was sponsored. 20 respondents (18.3 percent) claimed to be unsure while 11 (10.1 percent) considered the video was not sponsored. 13 of the “I’m not sure” responses were related to the unboxing videos, while the other 7 were related to non-unboxing videos. As for the 11 responses considering the video was not sponsored, 7 of them related to unboxing videos while the other 4 related to non-unboxing videos. Across the 2 conditions (sponsored non-unboxing and sponsored non-non-unboxing), most people perceived the video they watched as sponsored. These results can be found on Table 1.

Table 1

Cross-Tabulation analysis – Type of Video x Perceived Sponsorship The video is

The phenomenon of unboxing videos : the answer to sponsored video reviews?