Individual’s attitude toward in-stream video advertising on social media platforms in the digital age : the relationship with human personality and the moderating role of age

(1)

Individual’s attitude toward in-stream video advertising on

social media platforms in the digital age

the relationship with human personality and the moderating role of age

Quantitative research of individual’s attitude

Master thesis

Master of Science in Business Administration. Track: Digital Business Amsterdam Business School

Supervised by : Prof. em. dr. ir. Hans J. Oppelland

Author: Tim Walter Student ID: 11416807

Submission date: 13 June 2018 Version: Final

Pages: 46

(2)

Statement of originality

This document is written by Student Tim Walter 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.

(3)

iii

Abstract

Many studies have explored individuals’ attitudes towards advertisement, and studies on consumer attitudes towards Internet -based advertising methods are emerging. In-stream video advertising is a subset of Internet -based advertising methods, which has not yet been the subject of academic research. This study investigates individuals’ attitudes towards in-stream video advertising and their relationship to individuals’ personality traits. It also examines the moderating role of age in interpreting this relationship. An online survey utilizing a non-probability sampling technique was conducted using 167 respondents living in the Netherlands. It found that individuals’ attitudes towards in-stream video advertising are negative, and Internet users perceive them as an irritating and non-entertaining method. Generally, the more open to experience or extroverted individuals are, the more positive their attitude towards in-stream video advertising is. Despite these results, no significant evidence was found that individuals who are extroverted and open to experience have a more positive attitude towards in-stream video advertising. Furthermore, age did not have a significant moderating role in one of the examined relationships. The results of this study suggest that advertising managers should continue applying in-stream video advertisements but not with the purpose of brand awareness. In addition, advertisers should continue creating tailor-made content, think about methods to decrease the irritation level of these videos and take into account personality traits and age to target customers.

1. Introduction—The story of in-stream video advertising

Consider the scenario in which one browses YouTube, clicks on a video that attracts his/her interest, but first YouTube plays an advertisement before the selected video begins. This is an example of an online advertisement tactic called in-stream video advertising.

In the current digital age, online advertising has become a major market and an

essential advertising strategy for firms. Any time Internet users connect to the Internet creates an opportunity for exposure. Internet usage has grown tremendously, and individuals are spending an increasing amount of time and money than ever before on the Internet. In 2017, nearly 47% of the world’s population used the Internet, and Internet usage is projected to be nearly 54% in 2021 (eMarketer, 2017). The highest Internet penetration among individuals in 2017 was in Europe, where 79.6% of the population used the Internet. See Figure A1 in the appendix for a comprehensive view of the Internet penetration. Firms are leveraging

increasing Internet penetration to expose individuals to their advertisements. This can be traced back to firms shifting their funds to greater allocate to marketing spending. The

allocation of marketing spending to traditional or online advertisement has changed in the last years. In the United States, the spending on online marketing represents more than 40% of their marketing budget (Boston Consultancy Group, 2017). A survey of 2,628 top U.S. marketers from for-profit firms performed by Deloitte, Duke and the American Marketing

Association (2017) showed that marketing spending on traditional1 advertising decreased by a

range of 0.1–3.6% from 2011 to 2017. During the same period, spending on online marketing2

increased by 8.2–14.7% (see appendix,Figure A2). Given the fact that research also shows

1_{Traditional advertising refers to non-digital media advertising.} 2_{Online marketing refers to marketing activities using the Internet .}

(6)

2 that firms spend around 6.9% of their firms’ revenue on marketing, the expectation is that this

trend will continue in 2018 (see appendix, Figure A3). Business-to-consumer (B2C) product firms spent the highest percentage of their revenue on marketing at 8.6%; this indicates that a substantial shift occurred (see appendix, Figure A4) (Deloitte; Duke; American Marketing Association, 2017).

The allocation from online marketing budgets to video advertising by firms is rising (IAB, 2017). Nowadays, the majority of online marketing budgets is allocated to online video advertising. Reports from the Interactive Advertising Bureau (IAB) claimed that in 2017, around 56% of firms’ online marketing budgets were allocated to online video advertising. EMarketer expects U.S. online video advertisement spending will experience annual double-digit growth through 2020 (eMarketer, 2017A). The High Performance Marketing Demand Metric (2017) confirmed the emergence of online videos as advertisements. Results from their study of 159 U.S. and Canadian organizations indicate that in 2017, around 95% of

respondents stated that video advertising has become an important advertising method in the last five years (High Performance Marketing Demand Metric, 2017). In line with this

development, online video advertising is increasingly being embraced as a more personal and engaging manner to connect with consumers in the strong rising digital age (Gao, Zhang, & Li, 2014). A substantial driver of this solution is the rise of social media platforms (IAB, 2017). Viewings of online videos by consumers across the three major social media platforms YouTube, Facebook and Snapchat exceeded 20 billion views per day in 2016 (Vidyard, 2017). A report from September 2016 on 5,500 consumers aged 18 years old and over in five countries—Australia, France, Germany, United Kingdom and United States—showed that consumers spend around six hours per week watching video content on social media networks (Brightcove, 2016). The major social platforms demonstrate behaviour in line with the

increase in online video watching. They attribute more value to the use of videos, which leads to an increase in numbers of video impressions on social platforms. In 2015, Facebook

announced that videos will be ranked higher in newsfeeds3 by changing the ranking of videos

in Facebook’s newsfeed algorithm “EdgeRank” (The New York Times, 2017; The Sydney Morning Herald, 2017). Major social platforms extensively use the in-stream video

advertising format to expose their users to videos. Kozloff (2017) recently showed that major social media players are increasingly using in-stream video advertisements. Another recent report suggests that in-stream video advertising is an emerging advertising method (WARC, 2017B). The case study of WARC indicates that in-stream video advertising accounts for nearly 60% of firms’ video advertising spending. Knowing this, it can be stated that in-stream video advertising is a rising and substantive online advertising method in the current digital age and therefore an interesting subject to investigate.

1.1 Research problem

It is clear that online video4_{advertising has become a growing major advertising method in}

the marketing industry. Measuring the effect of online video advertising and obtaining insights into individual’s attitudes towards online video advertisements pose a challenge for

3_{Facebook newsfeed refers to the centre column of a Facebook user's homepage, which shows updates from}

the people and pages that users follow on Facebook.

4_{Online video advertising is a subset of digital marketing and will refer to an advertisement in a video format}

(7)

3 firms (IAB, 2016B). The IAB (2016B) has been interested in individuals’ behaviours and the

effectivity of the firm’s advertisements, which they partially measure through marketing analytics. A joint report by Deloitte, Duke and American Marketing Association (2017) showed that in 2017, 37.5% of firms’ projects used marketing analytics before a decision was made, especially those concerning online marketing. This has always played a role in the decision-making process but has increased a near 7% since 2013. Top U.S. marketers expect marketing budgets for marketing analytics will continue to rise from 2017 to 2020—5.5% to 18.1%, respectively—this is an increase of 229% in three years (Deloitte; Duke; American Marketing Association, 2017).

The marketing analytics measure method is primarily focussed on the clickstream behaviour of individuals and does not represent the point of view of individuals. The metric definitions vary from one online publisher to the next (WARC, 2017A). Little academic research has been conducted to assess the attitudes of individuals towards different

advertising formats, especially online video advertising. However, attitudes of individuals towards advertising is a regular research subject (Lamberton & Stephen, 2016). Several studies exist in the literature concerning the attitudes of individuals towards advertisement in general or towards other advertising formats. Zanot (1984) studied the public’s attitude towards advertising; Mittal (1994), Jin and Lutz (2013) researched individual attitudes towards television advertising. Mehta and Sivadas (1995) analysed the empirical assessment of consumers’ attitudes towards marketing on the Internet. In 1998, Mehta researched advertising attitudes and advertising effectiveness in general (Mehta A. , 1998). A year later Schlosser, Shavitt and Kanfer (1999) conducted research on Internet users’ attitudes towards Internet advertising. Alongside developments on the deployment of several marketing channels, there were several studies that reported on the effect of the combination of

advertising formats on the attitude of individuals towards the advertisements, such as Dijkstra, Buijtels and van Raaij (2005), Havlena (2007), Wakolbinger (2009) and Voorveld (2011). Dijkstra et al. made the following suggestion for further research in their study:

The effects on consumer responses were studied of single- and multiple-media campaigns consisting of print, television, and the Internet. With the advent of new media, investigation of the contributions of media to advertising effects will become more relevant for media planners and advertisers. Especially, the strengths and weaknesses of media in a campaign, and thus their complementarity, need further investigation. (Dijkstra, Buijtels, & van Raaij, 2005, p. 385)

Lim, Ri, Egan and Biocca’s (2015) study, ‘The cross-platform synergies of online video advertising: Implications for cross-media campaigns in television, Internet and mobile TV’, focussed on online video advertising. The most recent study on individuals’ attitudes towards advertisement is by Souiden et al. They investigated consumer attitudes towards advertising in general by contrasting people’s attitudes towards online advertising, taking into account the moderating role of human personality (Souiden, Chtourou, & Korai, 2017).

The above listed research shows that the attitudes of individuals towards video

advertising is an important subject in academic research. While the existing studies broach the topic of individual attitudes towards video advertising, no literature exists regarding the attitude of individuals towards online video advertising, specifically in-stream video advertising. As discussed, in-stream video advertising is a rising advertising method, and

(8)

firms’ expenditures on in-stream video advertising are increasing. Therefore, this research adds value to the literature by filling this knowledge gap.

1.2 Research objective

The intended result of this research is to gain knowledge on the attitudes of individuals towards in-stream video advertising. The objectives of the present study are threefold: first, it investigates the developments in the literature regarding advertising in general and video advertising; second, it investigates the literature regarding individual attitudes towards (video) advertising and identifies a relationship between personality traits and the dependent variable attitude; third, it examines the identified relationship between personality traits and attitudes towards in-stream video advertising and the role of age as a moderator. This study contributes to the field by providing knowledge on individual’s attitudes towards in-stream video

advertising. It can assist managers by providing insights into individual attitudes towards in-stream video advertising, which can be used in the decision-making process regarding appropriate online advertising methods. Furthermore, the relationship between personality traits, attitudes towards in-stream video advertising and the moderating role of an individual’s age contributes to the segmentation or personalising activities of a management’s marketing approach.

1.3 Research method

An online survey was used to collect data. Wright (2005) suggests that online survey research offers advantages in comparison to traditional surveys. These advantages include access to unique populations and saving time and money on paper surveys. Online surveys have the benefit of a large respondent pool that are easily accessible, without geographical restrictions, and researchers can work simultaneously on other activities (Wright, 2005). Wright also notes that online survey research has some disadvantages, but these are not unique to online survey research.

1.4 Structure of the thesis

This thesis is structured as follows: it begins with literature on individual attitudes towards in-stream video advertising and the formulation of the hypotheses. Second, the research design and method are discussed, which is a survey that is then coded as quantitative data. The results are analysed and discussed followed by the conclusion and the implications and future research suggestions.

2. Literature review

The literature review elaborates on the topics that are relevant to this research. There is an overview of the existing key academic literature on each topic, and this is discussed and reviewed. The literature review clarifies frequently used definitions. Next, background information on non-traditional advertising developments related is presented. Thereafter, the existing academic literature on (video) advertising is discussed. Attitudes towards (online)

(9)

5 advertising are subsequently elaborated on. Lastly, the relationship between personality and

the moderating role of age is discussed along with the developed conceptual model of the present study. The literature review closes with the resulting research question.

2.1 Definitions

Traditional advertising refers to the deployment of traditional media as a medium for

advertisements, like communicating an advertisement through televisions, radio, outdoor media (billboards, bus shelters, benches, transit) or print media (newspapers, flyers)

(Rosengren, 2015; Dahlén, 2009). Non-traditional advertising is the utilisation of the Internet as a medium to disseminate advertisements (Dahlén & Edenius, 2007).

Recent research by Jin and Lutz (2013) showed that consumers tend to have a mental representation of the most typical type of advertising, namely television advertising. If online video advertising is not clearly defined, people will interpret this as television advertising (Jin & Lutz, 2013). To avoid misinterpretation, online video advertising will be defined. No academic definition of video advertising is currently available; therefore, this definition is based on existing academic definitions of the related subject, online marketing. Chaffey and Ellis-Chadwick (2012) define online marketing as achieving marketing objectives through applying online technologies, such as the Internet, social media and applications, to online devices. Kotler (2016) describes online marketing as a form of direct marketing that electronically links consumers with sellers using interactive technology. In this research, online video advertising is defined as a subset of online marketing, a means to advertise by deploying videos via online channels on the Internet.

The definition of in-stream video advertising is based on studies from Nielsen (2013), IAB (2016A), Facebook (2018) and WARC (2017). In-stream video advertising involves an individual clicking on a video with the intention to watch that selected video (user-selected video) but is instead redirected to an integrated advertisement in a video format. In-stream video advertising is divided into pre-, mid- and post-roll video

advertisements; these differ by the moment of the occurrence of the video advertisement. Pre-roll occurs when the video advertisement appears before the user-selected video begins; in mid-roll, the video advertisement appears during the video and in post-roll, the advertisement begins after the user-selected video is completed.

Social media is also an important term to define. Social media conceptualisation and

taxonomy research, which analysed 23 academic definitions of social media, defined social media as follows:

a set of mobile and web-based platforms built on Web 2.0 technologies, and allowing users at the micro-, meso- and macro-levels to share and geo-tag user-generated content (images, text, audio, video and games), to collaborate, and to build networks and

communities, with the possibility of reaching and involving large audiences. (Ouirdi, Ouirdi, Segers, & Henderickx, 2014, p. 119)

In this research, the above definition of social media is applied. Social media platforms referenced in this current research include the three major social media websites Facebook, YouTube and Instagram (Statista, 2018A).

(10)

6

2.2 Background information on non-traditional advertisement developments

As previously mentioned, firms have shifted the allocation of marketing budgets. In addition to market research reports that indicate this shift, there is also empirical evidence that suggests a shift in the investment of advertisement mediums. Hudson (2016) found that firms are shifting their marketing investment from traditional mediums (such as television, radio, print, outdoor) to non-traditional mediums, such as social media (Facebook, YouTube, Instagram). Hudson claims that this shift is due to the technological advantages of non-traditional

mediums (Hudson, 2016). The major advantages of non-traditional media over traditional media include the ability to target local and international audiences, efficiency, the potential for personalised marketing, the possibility of many-to-many communications via the Internet (interaction) and real-time results (Boudet, Gregg, Heller, & Tufft, 2017; Trusov, Bucklin, & Pauwels, 2009).

An emerging subset of non-traditional advertisement is in-stream video advertising. Major players are increasingly using in-stream video advertisements—the social media platform YouTube pioneered this advertisement method (Kozloff, 2017). In an official statement, YouTube announced that beginning in 2018, they will discontinue 30-second, non-skippable advertisements (Campaign, 2017; Business Insider, 2017). The reason for this change is to improve the customer experience and because a 30-second advertisement can negatively affect consumers who have smaller Internet data plans. Since the end of 2017, YouTube has been applying the following in-stream video advertisement formats:

non-skippable in-stream ads,5 TrueView in-stream ads6 and bumper ads7 (Google, 2018A).

Facebook launched 5 to 15 second, non-skippable, in-stream video advertisements in 2017. This has been accompanied by social media platform Instagram, which is owned by Facebook (Facebook, 2017). The length of in-stream video advertisements varies from 5 to 30 seconds, but exact lengths depend on the advertiser and Internet platform standards. In some cases, there is a possibility to skip the advertisement directly or after five seconds; conversely, some are non-skippable, which also depends on the social platform standards (IAB, 2016A; Google, 2018B).

In-stream video advertising on social media gives advertisers advantages and new opportunities to provide information to customers (IAB, 2016A). With the broad possibilities of advertisement formats and mediums, a strong need exists to manage and control the possibilities and use ideal advertisement options (Vakratsas D, 1999). To effectively choose the appropriate advertisement format, advertising agencies should understand the differences in the way consumers react to advertisement formats and the need to understand the strengths and weaknesses of the different advertising formats (Buchholz LM, 1991; Vakratsas D, 1999). Furthermore, in a world of excessive product expansion and (e)retailers, coupled with the rise of non-traditional media such as social media platforms, the amount of advertising has been unprecedented for years now (Vreese & Neijens, 2016). As a result, consumers are

overloaded with advertisements from manufacturers and sellers who attempt to inform and

5_{Non-skippable video advertisement: after the advertisement is finished, the video will begin, with a maximum}

length of 30 seconds (Google, 2018A).

6_{After five seconds, the YouTube viewer has an option to skip the advertisement, with a maximum length of 30}

seconds (Google, 2018A).

(11)

7 convince consumers with the aim to increase their purchase intention (McKinsey, 2010).

Literature conducted by Cook (2010) shows that push marketing in combination with the traditional advertising medium worked in the past but is no longer an effective strategy. In today’s consumer marketplace, pull marketing is a more appropriate and effective marketing strategy (Cook, 2010). Along with this shift in advertising strategies, insights into appropriate methods to deliver advertisements without overwhelming consumers (Elliot & Speck, 1998) will be beneficial to advertising managers. It is reasonable to assume that knowledge about individual’s attitudes towards advertising may be able to provide insights into ideal methods for exposing individuals to advertisements.

The deployment of personalised marketing is a recent advertising strategy that can counter current methods that bombard consumers with advertisements. With the availability of customer data and the tremendous growth in Internet usage, personalised marketing is a new trend to approach customers. Firms are creating tailor-made content based on personal customer data. Technology enables marketers to use real-time data to create algorithms that deliver personalised advertisements and content to consumers. Marketing is becoming data driven and data activated and based on customer behaviours, interests and real-time needs (Boudet, Gregg, Heller, & Tufft, 2017; McKinsey, 2017). According to Tucker (2014), the deployment of personalised marketing is useful and increases the effectiveness of displaying advertisements. Tucker’s research on the effects of personalised marketing showed that consumers are nearly twice as likely to click on a personalised advertisement (Tucker, 2014). These results indicate that individuals react in a more positive way with personalised

advertisements than with non-personalised advertisements. However, this research does not reveal comprehensive insights into the attitudes of individuals towards personalised

advertisements. Scant academic research is currently available on the effects of personalised marketing in general. On the other hand, eMarketer (2017) published a report in April 2017, based on research that studied the practices of 513 North American and United Kingdom-based senior marketers across the retail, travel and hospitality, and financial sectors, on personalisation. Their results showed that 48% of the senior marketers reported that personalisation on their website or applications boosted revenues in excess of 10% (eMarketer, 2017B).

Another well-known development relevant to this research is the rise of mobile devices as a medium for advertising. With the upsurge of mobile device use and rapidly developing mobile technology, mobile devices have become a major gateway to deliver advertisements to customers. An advantage of mobile advertising is the ability to deploy location-based advertising, which makes it possible to expose advertisement based on the current and past locations of the mobile device user (Lamberton & Stephen, 2016). The objective of the present study is to understand the attitudes of individuals towards online in-stream video advertising. Therefore, the device type will not be considered as a moderator in the relationship between independent variables and the attitudes of individuals towards in-stream video advertising. There is no academic study that investigates whether the device type as an advertising medium influences the attitudes of consumers towards online video

advertising. It can be reasoned that the device type has no influence on research results concerning the attitude towards video advertisement before or during watching an online video. For this reason, device type is out of the scope of the present study.

(12)

2.3 Academic literature on online (video) advertising

As the literature indicates, online advertising is rising as well as the power of online advertising. Hutter (2015) found that the power of traditional media advertising to attract customer attention is decreasing. By contrast, the power of online advertising is increasing (Hutter, 2015). Research that explores the effects of online or specific advertising formats on consumers’ attitudes is sparse. There seems to be no academic literature that provides insight into the effectiveness of online video advertising. There are only a few reliable reports concerning the effectiveness of online video advertising, which were conducted by market research firms. WARC found that online video advertising is appropriate for long-term brand building—the emotions displayed in online video advertising are the key contributor to generating an effective, long-term online video advertisement (WARC, 2017C). Other findings recently revealed statistics. Statistics indicated that four times as many customers would rather watch a video about a product than read about it, and shoppers who have viewed a video are 1.81 times more likely to purchase than non-viewers (HubSpot, 2017). Other research from 2015 showed that video advertisements have an average clickthrough rate of 1.84%; this is the highest clickthrough rate of all online advertising formats (Business Insider Intelligence, 2015). These statistics are encouraging for firms to start with online video advertisement. However, advertising comes with obstacles for marketers. Recent research reports from IAB provided insights into the challenges of spending more on online video advertising. The top three challenges are ‘ROI of video advertising versus other media’, ‘quality of content’ and ‘price of video advertising’. Respectively, 41%, 40% and 37% of the respondents claimed these as the greatest challenges. A broad assortment of ROI and prices present obstacles for organisations because they must determine which products/services of their inventory are eligible for online video advertisement. (IAB, 2017)

When a video advertisement is produced and ready for distribution, the appropriate channel must be chosen. A literature review from 53 case studies on the effectiveness of advertising in mobile phones assumed that if the same advertisement is presented on a different channel, its effect may be different (Park, Shenoy, & Salvendy, 2008). Stolyarova and Rialp’s research (2014) suggests that the channel used to expose an advertisement to individuals influences the effectiveness of the advertisement. While the literature discusses the influence of the channel on the effectiveness of the advertisement, the present study’s objective concerns video advertisement via social media platforms as the medium. Despite the existing literature on the channel’s influence on the effectiveness of advertisements in general, there is no distinction between the different social media channels in the present study. The first reason is that there is no scientific research which indicates that using different social media platforms as video advertisement channels has distinguishable influences on the effectiveness of the video advertisement. The second reason is that the present study’s objective is understanding individual attitudes regarding in-stream video advertisement. The objective of the existing literature is investigating the channel’s influence on the effectiveness of the advertisements, not the channel’s influence on the attitudes of individuals. Therefore, there is no scientific evidence that the channel has a significant influence on the attitudes of individuals concerning advertisement.

The existing literature above discusses the effectiveness of advertisements and does not provide insights into the individual’s attitudes. In 2005, a study by Dijkstra, Buijtels and van Raaij (2005) explored the effects of single and multiple advertising formats on consumer

(13)

9 responses. An implication of their empirical research is that with the rise of new and

better-targeted media, advertising agencies should not only consider the efficiency perspective of the advertisements and the effectiveness related to the results. Dijkstra et al. suggest that

individual’s attitudes regarding advertisements should be a substantive determinant of the most appropriate advertisements (Dijkstra, Buijtels, & van Raaij, 2005). One measurement method of advertising is sentiment analysis (Piryani, Madhavi, & Singh, 2017). According to Piryani et al., sentiment analysis is an emerging method of analysis for the segmentation of individuals. ‘Opinion mining and sentiment analysis (OMSA) as a research discipline has emerged during the last 15 years and provides a methodology to computationally process the unstructured data mainly to extract opinions and identify their sentiments’ (Piryani, Madhavi, & Singh, 2017, p. 122). Piryani et al. found that marketers became interested in the sentiments of individuals to adapt their marketing strategy to the segmentation. Another well-established segmentation method that can be utilised for designing promotional strategy is segmentation

based on demographics and psychographics8 of individuals (Bearden, Teel, & Durand, 1978;

Samuel, 2016). These studies suggest that the insights into the attitudes and psychographics of individuals towards advertisements has an additional value to marketers. Therefore, insight into individual’s attitudes and psychographics will be a substantive subject in the present study.

Academic articles and studies exist regarding the psychology of marketing and insights into the individual’s attitudes and psychographics. It is widely known that the psychology of individuals has a correlation with marketing. Psychology holds that an individual’s attitude is a predominant determinant of an individual’s behaviour, and for that reason, attitude is an essential aspect to understanding and predicting an individual’s behaviour and decision making (Allport, 1935). In addition, according to the theory of reasoned action, an individual’s attitude influences his/her purchase behaviour (Ajzen & Fishbein, 1980). Accordingly, psychographics such as beliefs and attitudes influence an individual’s intentions while intentions determine an individual’s purchase behaviour. This relationship has been widely confirmed by empirical studies (Kim & Hunter, 1993; Berger, Ratchford, & Haines, 1994). For a considerable time, the concept of the attitudes of

individuals has played a central theme in the science of psychology (Eagly & Chaiken, 1993). Nowadays, the role of attitude still appears as a moderating role in academic marketing research. Furthermore, this characterisation of attitude with regard to advertising is apparent in the following articles: ‘Public attitudes toward advertising’ (Zanot, 1984), ‘Advertising attitudes and advertising effectiveness’ (Mehta A. , 1998), ‘Survey of Internet users attitudes toward Internet advertising’ (Schlosser, Shavitt, & Kanfer, 1999), examining the role of beliefs and attitudes in online advertising’ (Wang & Sun, 2010), ‘Structural effects of cognitive and affective responses to web advertisements, website and brand attitudes, and purchase intentions’ (Hwang, Yoon, & Park, 2011), ‘The typicality and accessibility of

consumer attitudes toward television advertising’ (Jin & Lutz, 2013) and ‘Consumer Attitudes toward Online Advertising’ (Souiden, Chtourou, & Korai, 2017).

(14)

2.4 The evolution of studies regarding individuals’ attitudes towards (online) advertising

According to Hwang, Yoon and Park (2011), the majority of studies on advertising have focussed on the performance of advertising and the determination of the advertisement’s content. Research on the attitudes of people towards online video advertising is limited (Hwang, Yoon, & Park, 2011). While this is true, there are several studies, including recent ones, on the attitudes of individuals towards advertising. The topic of academic research regarding attitudes of individuals towards advertising reveals a shift in the advertisement type and medium as the research objective, which corresponds with the technological

developments. In July 1941, the first legal commercial television advertisement in television history was broadcasted (Business Insider, 2016). This was the beginning of the emergence of television advertisements. Academics subsequently began conducting research concerning attitudes towards television advertisements. Insight into an individual’s attitudes towards television advertising became part of the academic studies. After the 1970s, attitudes towards advertising became increasingly negative, according to the findings by Zanot (1984). Much research indicated that customers had adverse attitudes towards television advertising (Mittal, 1994; Alwitt & Prabhaker, 1994; Schlosser, Shavitt, & Kanfer, 1999). Until the late 20th century, television advertising was the major advertisement format.

From the 1990s until now, academics have conducted several studies concerning the attitudes of consumers towards advertising. Mehta and Sivadas (1995) found that the general attitude of individuals regarding advertising influenced the amount of attention they paid to advertisements. Another study conducted by Mehta (1998) suggested that if individuals have favourable attitudes towards advertisements, they will be more impressionable after viewing an advertisement, and this will affect their attitudes on an advertisement more positively and are more receptive to them (Mehta A. , 1998). In the late 1990s, consumers’ attitudes towards traditional advertising were already becoming more negative; this is a result of the

overexposure of advertisements (Elliot & Speck, 1998). The overexposure of traditional advertisements in newspapers and television broadcasts had a negative effect on the content of the advertisements (Ha, 1997). Ha found that the higher the levels of advertising in traditional mediums such as magazines is, the greater the reduction in the effectiveness of the

advertisement and the sales volume of the magazines will be. In the end this will results in a decreasing profitability (Ha, 1997). To counteract this development, media owners and advertisers had to evolve their advertisement strategies. Rapid technological developments forced advertising managers to find optimal strategies that took into account the traditional- and non-traditional advertisement capabilities. Dahlén and Edenius’s study (2007) provided insights into the distinction of traditional and non-traditional advertisement. They found that when an advertisement is placed in a traditional advertising medium, individuals identify the advertisement more as an advertisement than when the advertisement is placed in a non-traditional advertising medium. Individuals will perceive the intention of the non-non-traditional advertisement message as less persuasive (Dahlén & Edenius, 2007). Dahlén and Edenius concluded that the credibility and attitudes of individuals towards advertisements are higher and more positive, respectively, for advertisements placed in a non-traditional advertising medium. Other academic studies look into the combination of traditional and non-traditional advertisement mediums. Findings by Voorveld, Neijens and Smit (2011) revealed that the combination of television and website commercials led to a more positive consumer attitude towards the advertisement. An earlier study by Havlena, Cardarelli and Montigny (2007)

(15)

11 showed that the combination of television commercials, print and Internet advertisements led

to synergic effects on the attitudes of consumers (Havlena, Cardarelli, & Montigny, 2007). Moreover, results from a study Lim, Ri, Egan and Biocca (2015) determined that individuals who are repeatedly exposed to an advertisement on multiple mediums generated a more positive attitude towards the brand than the individuals who were repeatedly exposed to a single-media medium (Lim, Ri, Egan, & Biocca, 2015).

With the emergence of online advertising halfway through the 1990s, several academic studies on online advertising surfaced. A study by Elliot and Speck (1998) revealed that in the late 1990s, consumers viewed online advertising as annoying, intrusive, disruptive and

irrelevant (Elliot & Speck, 1998). Schlosser et al. saw similar results and provided advertising managers insights into how they could avoid negative consumer experiences. They found that the level of irritation and intrusiveness is less when consumers are browsing the Internet without a specific purpose but more irritating and intrusive when consumers are goal orientated on the Internet (Edwards, Li, & Lee, 2002). In 2008, the results of a study by An and Kim demonstrated that Americans’ and Koreans’ general attitudes towards online advertising are negative (An & Kim, 2008). Studies by Elliot and Speck (1998), Edwards et al. (2002), and An and Kim (2008) concerned online advertising in general. Regarding specific online advertising methods, an exploratory study in 2010 determined that spam

advertisement, banner advertisement,9 pop-up advertisement and unrequested email

advertisements are the formats that were deemed most negative by consumers (Truong & Simmons, 2010). With respect to online advertisement content, Goldfarb and Tucker (2011) found that context-based advertisements were more tolerated by consumers (Goldfarb & Tucker, 2011). Context-based advertisements, a form of personalised marketing, use

individuals’ online behavioural data—commonly collected by way of cookies.10 An example

of context-based advertisement is a Coca-Cola advertisement that appears while a consumer is browsing a grocery store website.

Cho and Cheon (2004) studied the reasons people avoid online advertisements on the Internet. Their research suggested that people avoid advertising on the Internet because of perceived goal impediment, perceived ad clutter and prior negative experience. Marketers, Internet publishers and Internet advertisers should understand that excessive avoidance of Internet advertisements can reduce the collective effectiveness of Internet advertising (Cho & Cheon, 2004). Kelly, Kerr and Drennan (2010) confirmed this in their model that showed people avoid advertisements on the Internet in the social networking environment if they had a prior negative experience or the advertisement is not relevant to the user. This model also suggests that people are sceptical of advertising promises, which results in a negative attitude towards online advertising (Kelly, Kerr, & Drennan, 2010).

Another recent study related to the present study examined the attitudes of individuals towards online advertising (Souiden, Chtourou, & Korai, 2017). Souiden et al. developed a study that investigated the attitudes of individuals towards advertising in general rather than individual attitudes towards online advertising. They explored the moderating role of the personality trait extraversion to explain the attitudes of individuals towards advertising. The

9_{A form of online advertising in which an advertisement is displayed on a webpage; it can be an image or a}

multimedia object such as a video, which redirects to the advertiser’s website.

(16)

study is based on an online survey with nearly 250 respondents, conducted using Canadian consumers. Their results posited the following conclusion:

Attitude toward advertising in general has a positive and significant impact on attitude

toward online advertising. Introversion11 is found to have no moderating impact on the

relationship between both attitudes. However, extroversion moderates this relationship. (Souiden, Chtourou, & Korai, 2017, pp. 207)

Furthermore, results from Souiden et al.’s study support earlier results from studies by Mehta and Sivadas (1995) and Mehta (1998), which determined that attitudes of individuals towards advertising in general influenced the amount of attention they paid to advertisements.

Attitudes of individuals towards advertisements is a recurring subject in the literature. As the results from the previously mentioned studies indicate, individuals’ attitudes play a determinant roll in understanding advertisement effectiveness. These insights on individuals’ attitudes can be useful to marketers. Hence, the present study examines the relationship between personality traits and the attitudes of individuals towards in-stream video advertising, wherein the attitude is the dependent variable and personality traits is the independent

variable.

3. Research concept

3.1 Conceptual model

Figure 1 illustrates the conceptual model of the present study, which hypothesises that the personality traits are independent variables of an individual’s attitude towards in-stream video advertising, wherein this relationship is moderated by the demographic age of the individuals. The creation of the conceptual model is explained in detail in this section, including the associated assumptions detailed below.

Figure 1

(17)

13

3.2 Research assumptions

Basic assumption

To have an overall view of an individual’s attitude towards in-stream video advertising, a basic assumption must be made about an individual’s attitude towards in-stream video advertising in general. The academic literature regarding the attitudes of individuals towards online video advertising has yet to exist; this is true for on-stream video advertisements as well. However, based on the discussed existing literature about individuals’ attitudes towards advertising in general, we can argue that the expectation would be that an individual’s attitude towards in-stream video advertising will tend to be more negative than positive. Therefore, the following hypothesis is formed:

❖ H1: Individuals’ attitudes towards in-stream video advertising is more negative than

positive.

This research assumption does not concern a relationship between independent and dependent variables as well as a mediating or moderating variable. Therefore, a hypothesis was not formulated for the image of the conceptual model sketched above.

Relationship between human personality and attitude

Recall that Souiden et al. included personality traits as a moderating role for the attitudes of individuals towards advertising. Instead of using a personality trait framework in their

research, they relied on a single trait. The majority of studies on the behaviours and reasoning of online consumers use a personality trait framework to measure personality. The five-factor model, also known as the OCEAN model, is the most widely recognised human personality trait model in these studies (Ross, Orr, Sisic, Arseneault, & Orr, 2009) because this model covers the different dimensions of human personality (Tan & Yang, 2014). The OCEAN model has five dimensions that describe human personality: openness to experience,

conscientiousness, extraversion, agreeableness and neuroticism (Goldberg & Sarason, 1990; Costa & McCrae, 1992; Butt & Phillips, 2008). Accordingly, ‘it is a popular human

personality classification method, and a well-established and unifying framework for measuring human personality’ (Tan & Yang, 2014, p. 28). Souiden et al. also used the OCEAN model in their research but only relied on one of the five factors to serve their objective, namely the extraversion trait. As they argue in their study, they chose this trait because existing literature claims that extraversion explains consumer attitudes and invokes feelings (Digman, 1990; Goldberg, 1993). Furthermore, several studies emphasise the important role of extraversion in online communication (Kraut et al., 2002; Butt & Phillips, 2008). Finally, extraversion is considered a substantial human personality trait in current consumer behaviour literature (Barnett, 2013).

In the present study, extraversion is used as well as the personality trait openness to experience from the OCEAN framework. A study by Tang and Yang (2014) showed that of the five traits from the OCEAN framework, individuals who scored high on openness to experience also used the Internet more often for entertainment (online videos and music) objectives (Tan & Yang, 2014). Other studies revealed that the higher the openness to experience score, the more likely that individuals will try and understand a new experience (McCrae & John, 1992; Butt & Phillips, 2008). These findings compelled me to include the trait openness to experience in the research to determine if it plays a moderating role on an individual’s attitude towards in-stream video advertising. Aside from that, Souiden et al.

(18)

suggested to include openness to experience in further research concerning the attitudes of individuals towards online advertisements (Souiden, Chtourou, & Korai, 2017).

There is scarce literature on extraversion in relation to an individual’s attitude towards advertising. The existing literature on extraversion suggests that individuals who score low on extraversion evaluate advertisements as less positive than individuals who score high on extraversion (Chang, 2001). In 2010, an academic study confirmed these findings that

extroverted individuals tend to experience and perceive events in life in a more positive way, while introverts are more oriented towards negative aspects of life—this trait influences the attitudes of individuals towards advertisements (Stysko-Kunkowska & Borecka, 2010). From these findings, we can argue that individuals who score high on extraversion would have a more positive attitude towards in-stream video advertising than individuals who score low. Therefore, the following hypothesis emerges:

❖ H2A: Individuals who score high on extraversion have a more positive attitude

towards in-stream video advertising than those who score low, independent of the individual’s age.

There is no existing literature on the personality trait openness to experience in relation to an individual’s attitude towards advertising. However, there are findings, mentioned prior in the literature review, that may correlate to the human personality trait openness to experience. Recall the academic findings that individuals characterise online advertising as annoying, intrusive, disruptive and irrelevant. If an individual scores high on openness to experience, it can be inferred that he/she likely perceives an advertisement as less intrusive than individuals who score low on openness to experience. Additionally, individuals with a high score on openness to experience will be less likely to perceive a video advertisement as intrusiveness than those who scored low. Moreover, in-stream video advertising is a reasonable method of video advertising, especially when considering the findings that individuals avoid

advertising on the internet because of perceived goal impediment. It can be assumed that an individual who scores high on openness to experience will be less likely to perceive an advertisement as a goal impediment. Since this advertisement method is a non-traditional manner of advertisement, it can be characterised as a new experience for individuals. The expectation is that the higher individuals score on openness to experience, the more positive attitude they should have towards in-stream video advertising. This leads to the following hypothesis:

❖ H3A: Individuals who score high on openness to experience have a more positive

attitude towards in-stream video advertising than those who score low, independent of an individual’s age.

In addition, to prevent misinterpretation of the two personality traits, which are used as independent variables in this study, both definitions follow Tan and Yang’s articulations (2014), who based their definitions on studies by McCrae and John (1992), Digman (1990), and Costa and McCrae (1992).

➢ (Tan & Yang, 2014) define extraversion as a ‘trait…related to heightened level of sociability. Individuals who are high in extraversion are energetic, bold, warm-hearted, outgoing and enjoy the company of others’ (p. 28). In addition, the

(19)

15 extraversion trait can be divided into two facets, extraversion and introversion

(Goldberg & Sarason, 1990).

➢ (Tan & Yang, 2014) define the openness to experience trait as ‘individuals’ receptivity to learning, novelty and change. Individuals who are high in openness to experience tend to be intelligent, curious and like to try new ideas’ (p. 28).

Moderating role of age

In addition to the presence of the independent variable personality traits, the age of individuals is taken into account as a moderator for the relationship between human personality traits and the attitudes of individuals towards in-stream video advertising. In general, the era in which individuals were raised is related to the level of comfort with a particular lifestyle and that level of technology in that era (Levy & Morgan, 1993). At the dawn of traditional advertising, age distinction was already noticeable (Speck & Elliott, 1997). Speck and Elliott (1997) found that older individuals were less likely to avoid

advertising and have higher attention to advertisements compared to younger individuals. In the era of online advertising, research shows that the older the individuals are, the greater their attention to online advertisements (Goodrich, 2013). Other studies on age distinction and advertising-related attitudes found that website duration time was lower for younger individuals as a result of higher levels of advertising, but this is not applicable to older

individuals (Danaher, Mullarkey, & Essegaier, 2006). Findings from a survey of 3,200 people from eight countries (Australia, France, Germany, Indonesia, Japan, Sweden, United

Kingdom, United States) detailed that there is a distinction in individual’s attitudes towards video advertisement that correlated to their age (UNRULY, 2016). The findings showed that online users aged 18–34 years are 25% more likely than the average viewer to feel inspired by video advertisements; however, they are also the age category which scores the highest on muting video advertisements. In the context of the existing literature on age, customer age may exert a moderating role in the relationship between personality and an individual’s attitude towards in-stream video advertising.

With regard to the literature on the moderating role of age, we can argue the older the individual is the more positive the relationship between the personality traits and the attitude towards in-stream video advertising. Therefore, the following hypotheses evolved:

❖ H2B: The higher age of individuals influences the relationship between individuals’

extraversion and their attitudes towards in-stream video advertising in a more positive way than younger individuals.

❖ H3B: The higher age of individuals influences the relationship between individuals’

openness to experience and their attitudes towards in-stream video advertising in a more positive way than younger individuals.

Relationship between personality traits

The above assumptions in the present study hypothesise two personality traits as a single variable. Both personality traits are closely related; it is therefore worth testing for a potential relationship between the two personality traits and their collaborative effect on the attitudes of individuals towards in-stream video advertising. Although no literature exists that has tested this relationship as an independent variable, the following hypotheses will are drawn:

(20)

❖ H4A: Individuals who score high on extraversion and openness to experience have a

more positive attitude towards in-stream video advertising than those who score low, independent of an individual’s age.

❖ H4B: Higher age individuals who score high on extraversion and openness to

experience have a more positive attitude towards in-stream video advertising than younger individuals who score low.

4. Methods selection

4.1 Methods

Given that the population of this is study is large, and there is no sampling frame available, this study uses a non-probability sampling technique. The sampling technique used is convenience sampling. The survey data are collected from a self-administered survey using online survey software Qualtrics. The respondents were obtained digitally through social media, messaging applications and email. The use of online surveys is believed to be the appropriate approach to reach an audience that has experience with Internet and online activities and are current online website users (Kaye & Johnson, 1999; Souiden, Chtourou, & Korai, 2017). Duarte (2014) suggests that respondents need at least two weeks to reply to a mail survey. Therefore, the survey was conducted during a 15-day period in March 2018 in the Netherlands. The expectation was that the majority of respondents are of Dutch

nationality. The measurement items in the survey are derived from existing studies, with the exception of the socio-demographic measurement items. These studies ensure the validity and reliability of these measurement items. A pre-test was also performed to ensure reliability. The preliminary version of the survey was completed by 10 respondents. This version consisted of all four sections of the final survey, which is comprehensively described in the chapter measures. After the preliminary version was completed, the respondents were asked to evaluate the survey through five factors: understandability, length, answer options, interpretation and possible presence of input errors. In addition, the usefulness for analysing the collected data was checked. Based on the results of the preliminary survey, minor

modifications were made. The collected data were analysed using the statistical program IBM SPSS Statistics 25.

4.2 Sample

The population of this study was comprised of social media platform users 18 years and over who predominantly reside in the Netherlands. Information about the respondents’ age, gender, level of education and country of residence were used to assess the similarity of the sample. The minimum number of respondents for analysable data was estimated at 130 respondents. This was determined by using the statistical power analysis program G*Power 3 (Faul, Erdfelder, & Buchner, 2007). The respondents were approached through Facebook, LinkedIn and by invitations to participate in the study through email and messaging applications. Initial lists of contacts were obtained from the researcher’s network and included in the mailing list. The respondents did not recruit other participants for the online survey; therefore, there was no snowball sampling. The majority of respondents live in the Netherlands. As a reward for

(21)

17

participation, two shopping vouchers from bol.com12 with a value of €50 each were raffled

off. Several studies have shown that lottery (equal chance to receive an incentive) incentives for participating in online surveys increase the response rates and are the most common type of incentive used in studies that apply to online surveys (Musch & Reips, 2000; Bošnjak & Tuten, 2003; Göritz, 2006; Laguilles, Williams, & Saunders, 2011). Despite these findings, there are also studies that found contradicting results and suggest that lottery incentives do not increase response rates (Toepel, 2016). Nevertheless, two shopping vouchers were raffled off with the aim to increase the response rate. Shopping vouchers were chosen instead of cash to maintain the privacy concerns of the respondents. The only personal information needed for the incentives was respondents’ email addresses, with respondents given the option to opt out of this measuring item.

4.3 Measures

During the drafting of the survey, several sources were used as guidelines for writing the survey (Dillman, 2007; Toepel, 2016). For all the measuring items in the survey, excluding the sections ‘social media platform’ and ‘socio-demographic’, a seven-point Likert scale was used. This scale is substantiated by the main conclusion of a highly regarded article by George A. Miller (Miller, 1956). Miller found that the human mind is capable of

distinguishing seven different items—less intelligent people can manage up to five items and intelligent people may be able to distinguish nine items. In the academic literature, this is known as ‘the magic number seven’, which refers to using seven categories in a measurement scale so that respondents are able to reply with accurate responses. Nevertheless, there are many academic articles that discuss the most appropriate number of points in a Likert scale. Several studies have concluded that the reliability depends on the number of response categories, for which the reliability increases over scales with three, five, seven and nine response categories, but the reliability decreases for 11-point scales (Bendig, 1954; Komorita, 1963; Boote, 1981). A 1985 study confirmed these findings but found that beyond seven-point scales, the reliability will decrease (Cicchetti, Showalter, & Tyrer, 1985). Respondents were asked to indicate to what extent they agreed with the statements using the numbers 1 to 7, whereby 1 represented strongly disagree and 7 represented strongly agree. The number 4 is the midpoint, representing a neutral answer. These measurement items represent interval data, whereby respondents are able to tick their choice, which minimises entry errors. Because the survey primarily used one measurement level and one structure, the consistency of the survey is high, which results in an increase in the usability of respondents’ answers (Friborg,

Martinussen, & Rosenvinge, 2006).

To prevent misunderstanding, the definition of in-stream video advertising was clarified for the respondents. This term was defined in the literature review. To avoid

acquiescence bias, which is the tendency of people to agree rather than disagree, some of the survey items were reversed (Winkler, Kanouse, & Ware, 1982). The original survey was developed in English and thereafter translated into Dutch. To ensure consistency between the two language versions, the translation was conducted and tested by an independent certified translator. Furthermore, the survey contained measurement items concerning respondents’ socio-demographic and personality traits, and it was emphasised that the surveys were fully

(22)

anonymous and that individual results would not be disclosed, to improve the comfort of the respondents.

Survey instrument and variables

The survey begins with an introductory text that describes and clarifies this study and the survey (see Introduction text survey in the appendix). The survey was comprised of four sections: use and frequency of social media platforms, attitude towards in-stream video advertising, personality traits and socio-demographic characteristics. Four socio-demographic characteristics were obtained: gender (nominal variable), age (ratio variable and discrete), education level (nominal variable) and country of residence (nominal variable). The second section contained two general measurement items related to social media platform use and the frequency of use. Regarding the sections personality traits and attitude towards in-stream video advertising, a review of academic papers were used to develop these measurement items. Respondents whose answers in the first section indicated that he/she did not use social media were excluded from the study, and the questionnaire was terminated.

The measurement items in this study for attitude were adopted and derived from previous studies concerning attitudes towards advertising. To measure individuals’ attitudes towards online advertising, respondents were asked to specify their perceptions of in-stream video advertising on social media from five attitude factors. These measurement items were derived from two studies on attitudes towards Internet advertising by Ducoffe (1996) and Sun and Wang (2010). Both studies completed a factor analysis on the identical two attitude factors information and entertainment. The Cronbach’s alpha and factor loadings of the factor analysis of both studies correspond. In addition to these two factors, the study by Ducoffe (1996) completed a factor analysis on the factor irritation. Along with the two attitude factors information and entertainment, Sun and Wang (2010) completed a factor analysis on the factors credibility, economy and value. In this study, these six factors form the basis of the questionnaire’s section on personality traits. For a detailed overview of the Cronbach’s alpha and factor loadings per factor, see ‘Factor analysis of the attitudinal measurement items of two existing studies’ in the appendix. The mean score of the factors represent the attitude of individuals towards in-stream video advertising. This variable was computed following a factor and reliability analysis of this study. This overall score is defined as ‘Individual’s total attitude toward in-stream video advertising’, abbreviated as ‘attitude score’. The weighting of the factors were assimilated to each other; no distinction was made during the data analysis. The factors and associated measuring items have a solid foundation, and the subject Internet advertising is closely related to in-stream video advertising.

Therefore, this survey format was used to measure the attitude towards in-stream video advertising. To make this survey format applicable to the present study, the term ‘online advertising’ was adjusted to ‘in-stream video advertising’.

In terms of the last section of the survey, there were several studies regarding the assessment of the five-factor personality scales. Those assessments differ in the number of items per personality scale (Jones, 2014). A widely used survey instrument to assess an individual’s personality on the five-factor model is the International Personality Item Pool (IPIP). The personality dimensions extraversion and openness to experience of the

respondents in this study were measured using the IPIP survey instrument. Respondents were asked to indicate to what extent they agreed with the statements, under the terms of which the number 1 represented strongly disagree and the number 7 represented strongly agree. The

(23)

19 measuring items for the dimension extraversion were derived from the IPIP big-five-factor

markers (Goldberg, 1992). Goldberg’s five-factor markers are related to the NEO PI-R theorised by Costa are McCrae (Goldberg, 1992). Goldberg’s IPIP big-five-factor markers were used for the dimension extraversion because the Cronbach’s alpha of this dimension was higher in the IPIP big-five factor than Costa and McCrae’s IPIP-NEO. Based on the principles for writing a survey, the survey for the IPIP big-five factor of the dimension extraversion was more appropriate because the statements and applied words are simplified (Dillman, 2007; Toepel, 2016). Goldberg’s IPIP big-five-factor markers are available as a 50- or 100-item survey; in this study, the 50 items (10 items per dimension) version was used. In the 50 items version, the items in the extraversion dimension have a high internal consistency (Cronbach’s alpha=.87) (Goldberg, 1992). The measuring items for the dimension openness to experience were derived from the IPIP version of the NEO PI-R (Costa & McCrae, 1992). The IPIP-NEO is available as a 50-, 100- or 240-item survey; for this study, the 50-item (10 items per

dimension) version was used. In the 50-items version, the items in the openness to experience dimension have a high internal consistency (Cronbach’s alpha=.82) (Costa & McCrae, 1992). In total, 20 items were used to measure the personality traits of the respondents—10 items per dimension. In both dimensions, 5 of the 10 items were reversed in the survey. The weighting of the factors were assimilated to each other, and no distinction was made during the data analysis.

4.4 pre-test

Before the distribution of the final survey, a pre-test of the preliminary survey was completed by 10 respondents (Table 1). The respondents were asked to complete the preliminary survey via a personal link of the online survey tool Qualtrics. After respondents finished the preliminary survey, the survey was reviewed through an interview in Dutch, see ‘Questions reviewing the pre-test of the preliminary survey’ in the appendix for the measurement items that were asked during the review. Based on the respondents’ feedback, the measuring item ‘In-stream video advertising is a good source of product

information’ and the whole factor ‘Economy’ were reviewed.

Respondents found it difficult to understand these measurement items and therefore did not know how to reply meaningfully. After

examining and revising these measurement items, noting that the factor loading of these measurement items were also relatively low, these measurement items were discarded.

To verify the translation of the survey, pre-test respondents who indicated that their English and Dutch were at intermediate or advanced levels received personal links three days later for the English preliminary. The answers to the English version where compared with

the answers to the Dutch version. Applying the Pearson r correlation resulted in a correlation coefficient value of .91. The comparison did not reveal alarming deviations, so it was

assumed that the translation is valid.

Measuring subject N (respondents) 10 Gender: Female 60% Male 40% Age distribition: 16 - 39 40% 40 - 54 30% 55 + 30% Highest level of education:

Did not complete highs

school 10% High school 20% Bachelor's degree 40% Master's degree 20% Doctoral degree 10% Country of residence Netherlands 100% Language: English 50% Dutch 100% Demographics respondents pre-test

(24)

The questionnaires for the pre-test were completed in the survey tool Qualtrics. Through Qualtrics, the completeness duration time per respondent of both Dutch and English versions was measured. The average completeness duration time was nine minutes; this falls within the communicated duration time of 10 minutes.

The results of the pre-test helped to increase the validity and reliability of the final survey. This is not to suggest that the pre-test 100% guaranteed the validity and reliability of the survey, but it was beneficial. For the final questionnaire template, see Questionnaire template in the appendix.

5. Data collection

A total of 167 respondents participated in the sample of which three respondents were

disregarded from the sample because they did not use social media platforms within the last three months. All remaining 164 participants completed the questionnaire. The average completeness duration time was 7.2 minutes, and 67.7% of the participants completed the Dutch questionnaire while the other 32.3% completed the English version. Of the 164 participants, 85% of the respondents participated in the lottery. The participants of the study were active social media

platforms users; nearly 75% of participants indicated that they use a social media platform daily or multiple times in a day (figure 2). YouTube and Facebook was predominantly used by the participants, respectively 86% of the participants indicated that they have used YoutTube and 83% of the participants indicated that they have used FaceBook in the last three months. Forty-nine percent of the participants had used Instagram in the last three months, and 40% mentioned that they used another social platform from the given options YouTube, Facebook and Instagram. Other social platforms mentioned by respondents were Snapchat, LinkedIn and Twitter. All three other mentioned social platforms have been using in-stream video advertising for more than three months. Regarding the demographic profile of the participants of this study, 52.4% were female, with a mean age of 42.6 years; most

participants attained a bachelor’s degree, and 89.6% reside in the Netherlands. For a detailed demographic profile, see figures A5, A6, A7, A8 and A9 in the Appendix.

6. Data analysis

Preliminary analysis

The measurement items from the questionnaire on attitudes towards in-stream video advertising were derived from existing studies (Ducoffe, 1996; An & Kim, 2008; Sun & Wang, 2010). As mentioned in the methods section, these studies contain several existing factors which represent attitudes towards Internet advertising, whereas this study contains items representing an individual’s attitude towards in-stream video advertising. First, a factor

Individual’s attitude toward in-stream video advertising on social media platforms in the digital age : the relationship with human personality and the moderating role of age