Branded Journalism in the Health Care Industry How journalistic format affects behavioural intention, and the role of involvement

(1)

Branded Journalism in the Health Care Industry

How journalistic format affects behavioural intention, and the role of involvement

Master’s Thesis

Supervisor: Margot van der Goot Date: January 30, 2020

Rebecca Irene España Berganciano (12412791)

Master Communication Science – Persuasive Communication Graduate School of Communication

(2)

Abstract

Branded Journalism is an emerging communication tool that uses the format of an article to positively impact the credibility of the content promoted and the attitudes towards the content and source to promote a specified behavioural intention. In the health industry, this can be translated into promoting a health behaviour through the narrative engagement of an informational blog post, which companies can use to influence attitudes towards their brand and purchase intention. However, health behaviours are often affected by the audience’s level of engagement, as people’s attitudes towards a behaviour are targeted better when the person considers themselves at risk or is personally involved with the related health issue.

The experiment provides support to the claim that branded journalism has a positive effect on content and source credibility, as well as purchase intention. However, a moderating effect of level of involvement was not found.

(3)

Introduction

Lehto and Moisala (2014) define Branded Journalism as “story-based publications blurring the distinction between journalism and advertising”. It contains “journalistic techniques found in both native advertising and content marketing practices” (Serazio, 2019), making it a tool used to promote a company or a product by using a format typically found in journalism and news reporting. It is normally encountered in corporate webpages as part of their blog section; these are “a type of blog used by an organisation to further its goals, they perform as a communications and marketing channel” (Malvey, Alderman & Todd, 2009). It can therefore be seen that the intention behind the use of branded journalism in corporate blogs is to influence its target audience’s behavioural intention to become, or continue to be, a customer or consumer. In the case of companies in the health care industry, branded journalism is used as a means to share relevant health related information for patients or a common audience, boosting their profile as a source of information (Kerr, 2014).

Since branded journalism’s main goal is to influence a behaviour by increasing the intention to perform it, it is important how it can be affected. The Theory of Planned Behaviour indicates that a person’s intention is determined by three factors. These are the attitude towards the recommended behaviour, the subjective norms tied to it, and the perceived easiness of performing the behaviour (Swar, Hameed & Reychav, 2017). Branded journalism can target these aspects as a group or individually, while intending to influence the audience’s perception surrounding the behaviour the company wants them to engage in. Therefore, the current study focuses on the effect of branded journalism on the credibility and attitudes towards both content and source, as well as behavioural and purchase intention.

(4)

However, an important factor that might affect the influence of branded journalism on behavioural intention is the individual’s level of involvement. Since health information is often quite personal, previous knowledge on the issue might determine how this information is received. People who are involved often like to make an informed decision when it concerns health issues, and tend to look for clear and credible information (Ford, Vernon, Havstad, Thomas & Davis, 2006). This drives them to feel like they need to gather more information. As a result of this search for targeted information, they would come across branded journalism. With the surge of the internet and the availability and scope of information readily accessible, one of the most influential mediums by which people consume health information is through online sources. When people are in search of health information, they typically surf the internet looking for more information regarding the issue before they consider whether or not contacting a health professional face-to-face (Hu & Sundar, 2010).

Therefore, the aim of this study is to determine the effect that exposure to branded journalism has on the reader’s perception of content and source credibility, their attitudes towards both content and source, as well as the behavioural and purchase intention. This is within a health-related context, in order to determine the effectiveness that this communication tool can have on the audience’s decision-making process. Additionally, it aims to discern whether or not this effect is moderated by personal level of involvement on the issue.

Academic Relevance

This study builds on previous studies on the effect of branded journalism on audience behaviour by studying branded journalism in a specific health context as well

(5)

as adding the role of involvement as a potential moderating variable. It specifically elaborates further on Cole and Greer’s (2019) findings on audience response to branded journalism and the role of involvement, by identifying involvement as a potentially moderating variable and not a dependent one. It also builds on studies based on the use of blogs to promote a behaviour (Visentin, Pizzi & Pichierri, 2019; Bouhlel, Mzoughi, Ghachem & Negra, 2010), that specify that blogs have a positive effect on credibility and attitudes that are translated into behavioural intention, as well as Yoo and MacInnis’ (2005) research on the use of a journalistic format to affect behavioural intention. The current study takes these findings and applies them to a health context.

Practical Relevance

Branded journalism is a time-consuming tool, as there must be an effort to craft a journalistic-style article and commit to populating a blog with appropriate content. The outcome of this study should guide companies in the health industry into assessing whether or not branded journalism is the right investment by determining if it has the desired effect, which is to influence the target audience’s perception of the company’s credibility and positive brand attitudes. More specifically, this study should guide communication experts and health authorities in assessing the impact of branded journalism and whether or not appropriate regulations should be implemented; should branded journalism have the company’s desired effect, the ethical implications that it carries should be considered and appropriate guidelines developed. This is specifically relevant for companies in the health industry, as the targeted behaviours can have a direct impact on their health.

(6)

Theoretical Framework

In Branded Journalism, a company uses a narrative with a journalistic format, in order to benefit from the credibility that a journalistic article evokes contrary to advertisements (Cole & Greer, 2013). Cole and Greer (2013) go further in their interpretation of branded journalism by declaring that the goal behind this is to influence purchase decisions through promoting trust with customers, and brand-building. Therefore, in essence, branded journalism entails the use of a journalistic format in order to promote a corporate narrative.

This is similar to, but must not be confused with, advertorials which combine editorial content with an advertisement. The main difference lies within the actual branded content presented; while an advertorial will also borrow the editorial style of in this case a magazine, it explicitly mentions the product in order to promote it (van Reijmersdal, Neijens & Smit, 2005). Branded Journalism goes further than advertorials in how closely the editorial format is followed. An advertorial may explicitly mention the product it promotes, but branded journalism refrains from this practice and adheres to a stricter editorial format. Additionally, Branded Journalism is typically shared through owned media, that is, through the company’s magazine or website (Iliff, 2014), while advertorials appear in paid media.

Branded Journalism can be used to influence credibility and attitudes, but more importantly, behavioural intention. This would then indicate the degree to which the reader would agree to partake in the behaviour specified and take future actions in that direction.

But how does this apply to brands in the health care industry? Ford et al (2006) proposed in their research of prostate cancer screening intentions that making health

(7)

information accessible and understandable influences the audience into making the decision to engage in the desired behaviour; branded journalism prevails online in corporate blogs, which increases its accessibility. Therefore, due to the use of a journalistic format which makes receiving and understanding the information more enjoyable (Lehto & Moisala, 2014), branded articles can influence behavioural and purchase intention because the information is accessible for those looking for it.

Credibility

A factor that might explain the success of branded journalism is the journalistic format used, and how it affects not only the content’s credibility, but the source’s as well. Credibility is built due to the sharing of attributes between branded journalism seen in corporate websites and blogs, and online news (Meitz, Ort, Kalch, Zipfel & Zurstiege, 2016). This is due to the fact that these types of blogs borrow stylistic resources from online news articles, which are typically seen as reliable sources of information. Information that is presented in this style will therefore not arise distrust immediately, as the audience is used to seeing the same stylistic attributes from reputable sources.

According to Cole and Greer’s (2013) study on the use of journalistic frames, this type of frame is often deemed more trustworthy by the reader because the format is like that of a news article; therefore, the format here evokes credibility, and builds trust in not only the content but also the source. This is important because reader’s attitudes and behavioural intentions increase according to the perceived expertise of the source (Uribe, Buzeta & Velásquez, 2016), even regarding more health-related issues such as nutraceuticals (Tsai, Chin & Chen, 2010). Thus, by using a journalistic format, brands

(8)

can acquire credibility and this results in a perceived level of expertise of the brand. This trustworthiness is then reflected on the perceived credibility of source and content. It is important to note that the audience can be aware of the intentions of the branded article, which are to influence their perception of the company and thus increase behavioural intention. Disclosures of sponsored content activate people’s persuasion knowledge according to van Reijmersdal, Fransen, van Noort, Opree, Vandeberg et al (2016), which in turn leads to the audience distrusting both the information presented and the source. Branded journalism intends to avoid persuasion knowledge activation by not including disclosures, which can lead to less initial resistance if the audience is not intentionally aware of the source’s bias. Since Branded Journalism uses an owned medium, they are not required to include advertisement disclosures, as it is understood that they are a biased source of information regarding the subject. As Uribe et al’s (2016) research points out, non-sponsored messages increase blog credibility, and branded journalism intends to borrow this credibility by presenting the article in their own owned media. It can take the shape of corporate blogs, in order to avoid sponsor or advertisement disclosures, and avoiding as well the activation of the reader’s perception knowledge as there are no reminders other than the original source that the information is biased.

H1a: Branded Journalism will result in higher source credibility than advertising.

H1b: Branded Journalism will result in higher content credibility than advertising.

(9)

The intention behind Branded Journalism is not to explicitly sell a product like more traditional advertising, but to “position the overall brand in the minds of the target audience” (Cole & Greer, 2013). This means that brands use Branded Journalism to elicit positive attitudes towards the brand, which can be achieved through the narrative engagement that is part of a journalistic format (Murphy, Frank, Chatterjeel & Baezconde-Garbanati, 2013). In this way, persuasion knowledge, which is the awareness of the audience that the source wants to influence them (Tutaj & van Reijmersdal, 2012), is not activated, and thus the negative attitudes elicited by more traditional advertising are not present.

Credibility, although important as explained in the previous section, is not the only aspect that might explain branded journalism’s success over advertising. It was also found as a determining factor that influences brand attitudes when dealing with more informational ads (Yoo & MacInnis, 2005). This means that using a journalistic format not only positively influences credibility, it also positively affects brand attitude because of the informational nature of the content. In this case, credibility is translated into positive brand attitudes. Because readers have more positive attitudes towards a brand when they access this type of content directly from a company source (Cole & Greer, 2013), Branded Journalism can lead to better positive brand attitudes than traditional advertising. This is because the reader is exposed to the message through an owned channel where this type of content is expected, and not a paid one like advertising channels where it may be thrust upon the reader.

It is important to note that while a mediating relationship could be expected between credibility and attitudes when it comes to the effect of branded journalism on behavioural and purchase intention, the current study does not test the underlying mechanisms but focuses on examining whether branded journalism has the expected

(10)

effect on credibility and attitudes separately. Therefore, this study does not test a mediation effect of credibility and attitudes, as it focuses on testing whether an initial effect of branded journalism on these aspects can be tested.

H2a: Branded Journalism will result in better attitudes towards the content than advertising.

H2b: Branded Journalism will result in better attitudes towards the brand than advertising.

Intention

As explained before, behavioural intention refers to the individual’s likelihood to engage in a behaviour (Armitage & Conner, 2010). In the case of the health industry, behavioural intention would then refer to a behaviour within a health context. Purchase intention, following the same line of thought, would then refer to the likelihood to acquire a certain product (Morwitz, 2014).

Perceived credibility and attitudes can be used to target a desired behavioural intention, as explained by Visentin, Pizzi and Pichierri (2019). Their research on the impact of ‘fake news’ on behavioural intentions concluded that people’s perceived credibility of the content and source influences trust, which results in positive brand attitudes. In essence, they identified content and source credibility and brand attitudes as having a potential influencing relationship with behavioural intention. The ‘objective’ aspect took a back seat to the ‘subjective’ one when it came to these factors, meaning that it is the audience’s perception of these that will influence behavioural intention. Again, the current study does not focus on a mediating relationship of credibility nor

(11)

attitudes, but focuses on examining the effect of branded journalism on these aspects independently.

It is important to note that branded journalism targets a behavioural intention as well as a purchase intention, aiming to increase the audience’s proclivity of engaging in the desired behaviour in hopes that this will translate into a purchase intention. Bouhlel, Mzoughi, Ghachem and Negra (2010) found that blogs can be used as tools to increase credibility and generate a positive attitude towards the source, which in turn affect purchase intention. This is why a relationship between branded journalism, which targets credibility and attitudes, and behavioural and purchase intention is expected.

Therefore, the aspects that make behavioural intention successful also positively affect purchase intention. And since these are targeted in branded journalism, the use of this communication tool will result in a positive behavioural and purchase intention.

H3a: Branded Journalism will result in a higher behavioural intention than an advertisement.

H3b: Branded Journalism will result in a higher purchase intention than an advertisement.

Level of Involvement

The previously mentioned factors that might explain why branded journalism can have an effect on behavioural intention. But it is important to notice that the reader needs to engage with this information in the first place, in order to be affected by it. And how actively they engage with it might explain differences in the scope of the effect, thus moderate the effect of branded journalism on credibility, attitudes, and intention.

(12)

In their research, Cole and Greer (2013) found that high-involvement consumers had stronger brand attitudes and purchase intent after reading advice from a peer source, whereas low-involvement consumers had similar responses to corporate sources. Thus, a corporate blog which blends both the authority of a corporate source with the familiarity a blog should result in an effect on behavioural intention for both highly and lowly involved participants. Although, the magnitude of this effect would vary depending on how involved an individual is. While branded journalism might affect lowly involved individuals, it will have a greater effect on highly involved ones. This is because, as Ford et al (2006) explained, highly involved individuals have a greater need to make an informed decision, meaning that they would seek more information. Since they are more involved in the issue, high involvement in an individual would magnify the potential effect of branded journalism on credibility, attitudes, and intention.

As an example, the audience might be overwhelmed with too much information from different sources and thus, not engage in the desired behaviour. However, highly-involved audience members avoid this information overload because they have prior knowledge of the issue (Swar et al, 2017), which makes the amount of new information less, and more digestible. Therefore, a higher level of involvement results in less information overload, and higher engagement.

H4: Level of Involvement will moderate the effect between branded journalism and (4a) source credibility, (4b) content credibility, (4c) content attitude, (4d) brand attitude, (4e) behavioural intention, and (4f) purchase intention.

(13)

Figure 1: Conceptual model of this study

Method

Design

A 2 (Exposure to Branded Journalism vs Exposure to Advertisement) x 2 (Low-Level of Involvement with Issue vs High-Level of Involvement with Issue) between subjects experiment was conducted, in order to evaluate if exposure to branded journalism has an effect on content and source credibility, content and source attitudes, and behavioural and purchase intention, as well as whether or not this effect is

Content Credibility Source Credibility Content Attitude Brand Attitude Behavioural Intention Purchase Intention Branded Journalism Level of Involvement

(14)

Participants

A total of 100 participants completed the experiment. Of these, 40 identified as male, 59 as women, and 1 as other. Participants were between 18 and 78 years old (M = 39.9, SD = 1.15). They were proficient enough in English in order to properly understand the content of the article they were exposed to, and had enough technological literacy to be able to use search engines and surf through web sources. They also had diverse nationalities, with a total of 22, although most of the participants were either Dutch (25 participants) or Spanish (40 participants).

Participants were recruited through snowball sampling online, using survey-sharing Facebook groups where volunteers can be found, and they were asked to share with other potential participants that might be interested in partaking in the experiment.

Stimuli

The product chosen was sunscreen, and the brand Neutrogena. “Sunscreen” was chosen as the health-related topic due to the high occurrence of skin cancer in the general population, as well as the high general concern surrounding it in the recent years as negative effects of prolonged sun exposure have come to light (World Health Organization, n.d.). In the current study, it is important that participants recognize the source of information as a brand, as the attitude towards the source is measured as well, thus Neutrogena was chosen.

In the first condition, they were presented with the branded journalism article, which consisted of a blog post titled “The Dish on Wearing Sunscreen Daily” from Neutrogena’s website. This was presented as a full webpage screenshot where the menu and source could be seen as well as visual cues, in order to elicit a more organic

(15)

reaction. The content of the article consisted of a short narrative introduction into a typical morning routine and continued by listing reasons why wearing sunscreen daily should be important for the reader. The complete full stimulus is shown in Appendix 1.

In the second condition, participants were presented with a two-page magazine spread where Neutrogena’s sunscreen was advertised. This was done to present the advertisement in a manner in which they would encounter it in a non-experiment setting. Appendix 2 shows this spread, consisting of images and information about Neutrogena’s moisturizer with SPF while listing reasons why using UVA and UVB protection daily is important, and how their product targets it.

Procedure

The data was gathered through an anonymous survey, in which participants were randomly assigned to one of the two conditions. After giving consent for their participation, they were presented the article in the assigned condition and asked to confirm they had read it. Afterwards, they completed a survey in which they self-reported the content and source’s credibility, their attitude towards the content and brand, and their intention to partake in the desired behaviour (using sunscreen) and purchase intention (buying the brand’s sunscreen). They self-reported their impressions using a seven-point scale of opposite adjectives.

Measurements

Credibility was based on the participants’ perceived credibility of the source and of the content. Participants were asked to evaluate both factors (Source Credibility, Content Credibility) by answering the statement “I find this content…” and “I find the source of this content…” using a 7-point continuum with 3 pairs of adjectives

(16)

(“unbelievable/believable,” “inaccurate/accurate,” “not trustworthy/trustworthy”). This was based on Cole & Greer’s (2013) measurement of credibility. In the current study, this scale proved a good internal consistency for measuring Content Credibility (M = 4.73, SD = 1.22, 𝛼 = .89) and Source Credibility (M = 4.58, SD = 1.15, 𝛼 = .91).

Attitude was based on the attitudes towards the content presented and the brand. Participants were asked to share their feelings towards first the content, and then the brand, by answering the statement “I find this content…” and “I find this brand (Neutrogena)…” using a 7-point continuum with 5 pairs of adjectives (“unappealing/appealing”, “bad/good”, “unpleasant/pleasant”, “unfavourable/favourable”, “unlikable/likable”). This scale was used according to Spears & Singh’s (2004) scale on measuring attitude towards brand. In the current study, this scale proved a good internal consistency for Content Attitude (M = 4.80, SD = .97, 𝛼 = .89) and Brand Attitude (M = 5.05, SD = .92, 𝛼 = .93).

Intention was based on the participant’s intention to buy the product and to use the product. Using an adapted scale from Spears & Singh’s (2004) scale on measuring behavioural intention, participants were asked to share their intentions on using sunscreen by answering the statement “After reading this content, rate how you would use sunscreen in general” in a 7-point continuum with 5 opposite statements (“never/definitely”, “definitely do not intend to use/definitely intend to use”, “very low interest to use it/very high interest to use it”, “definitely use/definitely not use”, “probably not use it/probably use it”). In the current study, this scale proved a good internal consistency (M = 5.23, SD = 1.24, 𝛼 = .95) Afterwards, participants were asked to share their intentions on specifically purchasing the brand’s sunscreen by answering the statement “After reading this content, rate how you would use Neutrogena’s sunscreen” in a 7-point continuum with 5 opposite statements

(17)

(“never/definitely”, “definitely do not intend to buy/definitely intend to buy”, “very low purchase interest /very high purchase interest”, “definitely buy/definitely not buy”, “probably not buy it/probably buy it”). In the current study, this scale proved a good internal consistency (M = 4.30, SD = 1.31, 𝛼 = .97).

The level of Involvement with the health issue targeted in the content, in this case sun damage, was measured using an adaptation of Flora and Maibach’s (1990) seven-point scale on measuring level of involvement regarding a sexually transmitted disease (“I think about sun damage a great deal”, “I consider myself at risk of sun damage”, “sun damage is a personally relevant topic for me”, “I actively seek the most recent information about sun damage”). In the current study, this scale proved a good internal consistency (M = 4.53, SD = 1.68, 𝛼 = .89). A median split was conducted to divide the participants into two groups, low involvement and high involvement, with a median value of 4.75. 51 participants were in the Low-Level group, and 49 in the High-Level group.

Results

Randomization Checks

Initial randomization checks were conducted for age, gender, and nationality. An independent samples t-test was conducted to compare the distribution of age across both conditions. There was a significant difference in distribution of age for the Branded Journalism condition (M = 36.65, SD = 11.52) and the Advertisement condition (M = 43.87, SD = 14.61; t (98) = -2.76, p = .007, two-tailed). The magnitude of the difference in the means (mean difference = -7.21, 95% CI: -12.40 to -2.03) was

(18)

moderate (eta squared = .072). This indicated that “age” had to be taken into account as a control variable in further tests.

A chi-square test for independence indicated no significant association between gender and the condition assigned, 𝑥2_{(2, n = 100) = 1.77, p = .41. phi = .13.}

Due to the uneven distribution of participant’s nationalities, three groups were formed; “Spanish” with 40 participants, “Dutch” with 25 participants, and the remaining 20 nationalities condensed in “Other” with 35 participants. A chi-square test for independence indicated no significant association between nationality and the condition assigned, 𝑥2_{(2, n = 100) = 1.37, p = .50. phi = .12. Therefore, neither gender nor}

nationality was taken into account as a control variable.

Credibility of Source

A 2 by 2 between-groups analysis of covariance (ANCOVA) was conducted to assess the effect of branded journalism and level of involvement on the perceived credibility of the source, with age as a control variable (Appendix 3, Tables 1 and 2).

There was a statistically significant main effect for exposure to branded journalism, F (1, 95) = 8.26, p = .005, with a moderate effect size (partial eta squared = .08). Participants exposed to branded journalism reported a higher credibility of source (M = 4.82, SD = 1.09) than participants exposed to the advertisement (M = 4.28, SD = 1.17). This confirms H1a.

However, there was no statistically significant main effect for level of involvement, F (1, 95) = 3.90, p = .051, nor a statistically significant interaction effect between exposure to branded journalism and level of involvement F (1, 95) = .08, p = .778. This rejects H4a.

(19)

Credibility of Content

A 2 by 2 between-groups analysis of covariance (ANCOVA) was conducted to assess the effect of branded journalism and level of involvement on the perceived credibility of the content, with age as a control variable. (Appendix 3, Tables 3 and 4).

There was a statistically significant main effect for exposure to branded journalism, F (1, 95) = 4.78, p = .031, with a small effect size (partial eta squared = .05). Participants exposed to branded journalism reported a higher credibility of content (M = 4.94, SD = 1.17) than participants exposed to the advertisement (M = 4.47, SD = 1.25). This confirms H1b.

However, there was no statistically significant main effect for level of involvement, F (1, 95) = 1.16, p = .284, nor a statistically significant interaction effect between exposure to branded journalism and level of involvement, F (1, 95) = .23, p = .635. This rejects H4b.

Attitude towards Content

A 2 by 2 between-groups analysis of covariance (ANCOVA) was conducted to assess the effect of branded journalism and level of involvement on the attitude towards the content, with age as a control variable. (Appendix 3, Tables 5 and 6).

There was no statistically significant main effect for exposure to branded journalism, F (1, 95) = .73, p = .394, which rejects H2a.

Additionally, there was no statistically significant main effect for level of involvement, F (1, 95) = 1.57, p = .213, nor a statistically significant interaction effect

(20)

between exposure to branded journalism and level of involvement, F (1, 95) = .27, p = .606. This rejects H4c.

Attitude towards Brand

A 2 by 2 between-groups analysis of covariance (ANCOVA) was conducted to assess the effect of branded journalism and level of involvement on the attitude towards the brand, with age as a control variable. (Appendix 3, Tables 7 and 8).

There was no statistically significant main effect for exposure to branded journalism, F (1, 95) = 1.70, p = .196, which rejects H2b.

Additionally, there was no statistically significant main effect for level of involvement, F (1, 95) = 3.82, p = .054, nor a statistically significant interaction effect between exposure to branded journalism and level of involvement, F (1, 95) = .06, p = .814. This rejects H4d.

Behavioural Intention

A 2 by 2 between-groups analysis of covariance (ANCOVA) was conducted to assess the effect of branded journalism and level of involvement on behavioural intention, with age as a control variable. (Appendix 3, Tables 9 and 10).

There was no statistically significant main effect for exposure to branded journalism, F (1, 95) = .001, p = .982, which rejects H3a.

However, there was a statistically significant main effect for level of involvement, F (1, 95) = 12.85, p = .001, with a moderate effect size (partial eta squared = .12). Participants with a high level of involvement reported a higher behavioural intention (M = 5.70, SD = 1.24) than participants with a low level of involvement (M = 4.77, SD = 1.15). But there was no statistically significant interaction effect between

(21)

exposure to branded journalism and level of involvement, F (1, 95) = .17, p = .691. This rejects H4e.

Purchase Intention

A 2 by 2 between-groups analysis of covariance (ANCOVA) was conducted to assess the effect of branded journalism and level of involvement on purchase intention, with age as a control variable. (Appendix 3, Tables 11 and 12).

There was a statistically significant main effect for exposure to branded journalism, F (1, 95) = 4.66, p = .033), with a small effect size (partial eta squared = .05). Participants exposed to branded journalism reported a higher purchase intention (M = 4.44, SD = 1.19) than participants exposed to the advertisement (M = 4.12, SD = 1.43). This confirms H3b.

Additionally, there was a statistically significant main effect for level of involvement, F (1, 95) = 10.90, p = .001, with a moderate effect size (partial eta squared = .10). Participants with a high level of involvement reported a higher purchase intention (M = 4.72, SD = 1.39) than participants with a low level of involvement (M = 3.89, SD = 1.09). However, there was no statistically significant interaction effect between exposure to branded journalism and level of involvement, F (1, 95) = .01, p = .947. This rejects H4f.

Table 1 – Summary of the findings. Branded Journalism Level of Involvement Moderation of Branded Journalism x Level of Involvement Source Credibility ✓ Content Credibility ✓ Content Attitude

(22)

Source Attitude Content Attitude

Behavioural Intention _✓

Purchase Intention _✓ _✓

Discussion

The aim of the study was to determinate the effect of branded journalism on source and content credibility, content and brand attitudes, and behavioural and purchase intention, as well as the potential moderating effect of level of involvement. While there was a positive effect of exposure to branded journalism on source and content credibility as well as purchase intention, this effect was not found to be moderated by the participant’s level of involvement.

Exposure to branded journalism resulted in higher source and content credibility, as well as higher purchasing intention, when compared to an advertisement. This experiment supports Cole and Greer’s (2013) finding that a journalistic frame reportedly builds trust on both content and source, making them more credible; meaning that brands looking to increase their credibility as a source, as well as their content’s, would benefit more from using branded journalism compared to an advertisement. It also support’s Uribe et al’s (2016) finding that non-sponsored messages increase credibility, as the branded journalism article reported higher credibility, which could be due to blurring the line between an advertisement and an informational article. Yoo and MacInnis’ (2005) statement that a journalistic format positively affects credibility as well as the use of a corporate blog (Bouhlel et al, 2010) was also supported in this

(23)

experiment, as the stimuli used in the branded journalism condition was an article from a corporate blog.

In this experiment, a positive relationship between exposure to branded journalism and purchase intention was found, when compared to an advertisement. It confirms that branded journalism results in a higher intention as well (Bouhlel et al, 2010); although, contrastingly, the intention affected is purchase intention and not behavioural intention as previously thought (Cole & Greer, 2013). The fact that this experiment found a relationship between branded journalism and purchase intention, but not behavioural intention, is an interesting one. This means that an individual exposed to a company’s branded journalism will consider acquiring the product, but not necessarily engaging in the general behaviour which is promoted.

In addition, involvement was found to also positively affect behavioural and purchasing intention, supporting Cole and Greer’s (2019) finding that a higher level of involvement results in a stronger behavioural intent. This means that, while level of involvement was not found to be a moderating variable between branded journalism and behavioural intention, it did have a separate effect and should therefore also be taken into consideration when brands look to influence purchase intention.

This research, however, could not provide evidence of a relationship between a journalistic format and attitudes towards the source of the content (Yoo & MacInnis’, 2005; Bouhlel et al, 2010). A possible explanation of this finding is presented in the following section.

Limitations and Future Research

Firstly, it is important to note that these results might be influenced by the limited sample size, as this research would benefit from being replicated with a greater

(24)

sample size. Interestingly, of the 100 participants, only 45 that were randomly assigned to the ‘advertisement’ condition completed the survey, contrasting with the 55 in the ‘branded journalism’ condition, resulting in an unequal balance between conditions. This limits the results of this research, and thus, a recommendation for future research would be to recruit more participants that would complete the ‘advertisement’ condition.

The randomization check showed that age was found to be a control variable, which might have limited the results of this experiment. Nationality, on the other hand, was very unevenly distributed, with 20 nationalities added together in order to differentiate between groups. Although a randomization check did not find nationality to be a control variable, future studies focusing on specific nationalities or geographical regions could further elaborate on their potential role, as the nationality group labeled as ‘other’ consisted of nationalities with cultures that are too diverse to be grouped together.

Additionally, brand attitudes prior to exposure to the conditions were not accounted for. This might explain the lack of results demonstrating a relationship between exposure to branded journalism and brand attitudes, as there was no initial brand attitude measured in order to assess if there was a significant change in attitudes after being exposed to branded journalism. A pre- and post-test should be able to provide more insight into the effect of branded journalism on brand attitudes.

Regarding the role of level of involvement, a median split was conducted in order to divide participants in more equally-sized groups. The potential issue with this is, while it does evenly distribute participants across groups, each value above the median is considered equal, as well as each value below the median (Aiken & West, 1991). This means that there might be a greater difference in involvement within groups

(25)

than thought at first. In addition, categorizing a continuous predictor results in a loss of power, which might account as to why the effect of level of involvement on the variables was hard to find (Aiken & West, 1991).

While this research focused on the individual effects of branded journalism on content and source credibility, brand and content attitudes, and behavioral and purchase intention, a mediating relationship of credibility and attitudes on intention was suggested in previous research (Cole & Greer, 2013; Yoo & MacInnis’, 2005; Bouhlel et al, 2010). Therefore, since an effect of branded journalism was found on content and source credibility in this experiment, future research should focus on the potential mediating effect of credibility between branded journalism and purchase intention. This would determine whether the success of branded journalism over purchase intention can be attributed to the credibility it elicits on the individual exposed.

As this experiment found an effect of branded journalism on purchase intention, companies in the health industry should be mindful to report their branded journalism in a way that adheres to the ethical regulations, providing relevant and appropriate sources to back the claims made throughout their articles. Health authorities should consider this impact as well, and determine the best practices for the use of branded journalism in such a delicate topic as health.

(26)

References

Aiken, L. S., West, S. G., & Reno, R. R. (1991). Multiple regression: Testing and

interpreting interactions. Sage.

Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned behaviour: A meta‐analytic review. British journal of social psychology, 40(4), 471-499. Bouhlel, O., Mzoughi, N., Ghachem, M. S., & Negra, A. (2010). Online purchase

intention: Understanding the blogosphere effect. International journal of

e-business management, 4(2), 37.

Cole, J. T., & Greer, J. D. (2013). Audience response to brand journalism: The effect of frame, source, and involvement. Journalism & Mass Communication

Quarterly, 90(4), 673-690.

Flora, J. A., & Maibach, E. W. (1990). Cognitive responses to AIDS information: The effects of issue involvement and message appeal. Communication

research, 17(6), 759-774.

Ford, M. E., Vernon, S. W., Havstad, S. L., Thomas, S. A., & Davis, S. D. (2006). Factors influencing behavioral intention regarding prostate cancer screening among older African-American men. Journal of the National Medical

Association, 98(4), 505.

Hu, Y., & Shyam Sundar, S. (2010). Effects of online health sources on credibility and behavioral intentions. Communication research, 37(1), 105-132.

Iliff, R. (2014, July). Why Your Business Needs an Editorial Strategy. Mashable. Retrieved December 15, 2019, from https://mashable.com/2014/07/10/brand-journalism-owned-media/?europe=true

(27)

Kerr, L. (2014, April). Brand Journalism: Health Providers Embracing New Storytelling Strategy. The Buzz Bin. Retrieved December 12, 2019, from

https://buzzbinpadillaco.com/brand-journalism-strategy/

Lehto, M., & Moisala, V. (2014). Defining Branded Journalism. Academia. Retrieved November, 28, from

https://www.academia.edu/7331254/Defining_Branded_Journalism Malvey, D., Alderman, B., & Todd, A. D. (2009). Blogging and the health care

manager. The health care manager, 28(2), 159-164.

Meitz, T. G., Ort, A., Kalch, A., Zipfel, S., & Zurstiege, G. (2016). Source does matter: Contextual effects on online media-embedded health campaigns against childhood obesity. Computers in Human Behavior, 60, 565-574.

Morwitz, V. (2014). Consumers' purchase intentions and their behavior. Foundations

and Trends in Marketing, 7(3), 181-230.

Murphy, S. T., Frank, L. B., Chatterjee, J. S., & Baezconde-Garbanati, L. (2013). Narrative versus nonnarrative: The role of identification, transportation, and emotion in reducing health disparities. Journal of Communication, 63(1), 116-137.

Serazio, M. (2019). Making (Branded) News: The Corporate Co-optation of Online Journalism Production. Journalism Practice, 1-18.

Spears, N., & Singh, S. N. (2004). Measuring attitude toward the brand and purchase intentions. Journal of current issues & research in advertising, 26(2), 53-66. Swar, B., Hameed, T., & Reychav, I. (2017). Information overload, psychological

ill-being, and behavioral intention to continue online healthcare information search. Computers in Human Behavior, 70, 416-425.

(28)

Tsai, M. T., Chin, C. W., & Chen, C. C. (2010). The effect of trust belief and salesperson's expertise on consumer's intention to purchase nutraceuticals: Applying the theory of reasoned action. Social Behavior and Personality: an

international journal, 38(2), 273-287.

Tutaj, K., & Van Reijmersdal, E. A. (2012). Effects of online advertising format and persuasion knowledge on audience reactions. Journal of Marketing

Communications, 18(1), 5-18.

Uribe, R., Buzeta, C., & Velásquez, M. (2016). Sidedness, commercial intent and expertise in blog advertising. Journal of Business Research, 69(10), 4403-4410.

Van Reijmersdal, E. A., Fransen, M. L., van Noort, G., Opree, S. J., Vandeberg, L., Reusch, S., Van Lieshout, F., & Boerman, S. C. (2016). Effects of disclosing sponsored content in blogs: How the use of resistance strategies mediates effects on persuasion. American Behavioral Scientist, 60(12), 1458-1474. Van Reijmersdal, E., Neijens, P., & Smit, E. (2005). Readers' reactions to mixtures of

advertising and editorial content in magazines. Journal of Current Issues &

Research in Advertising, 27(2), 39-53.

Visentin, M., Pizzi, G., & Pichierri, M. (2019). Fake News, Real Problems for Brands: The Impact of Content Truthfulness and Source Credibility on consumers' Behavioral Intentions toward the Advertised Brands. Journal of Interactive

Marketing, 45, 99-112.

World Health Organization. (n.d.) Sun Protection. World Health Organization. Retrieved December 18, 2019, from

(29)

Yoo, C., & MacInnis, D. (2005). The brand attitude formation process of emotional and informational ads. Journal of Business Research, 58(10), 1397-1406.

(30)

Appendix 1: Branded Journalism Condition

Participants randomly assigned to the ‘branded journalism’ condition were shown the following image.

(31)

Appendix 2: Branded Journalism Condition

Participants randomly assigned to the ‘advertisement’ condition were shown the following image.

(32)

Appendix 3: Results Tables

Table 1 – Effect of Branded Journalism on Credibility of Source

Mean Standard Deviation N Branded Journalism:

- low level of involvement 4.65 1.11 33

- high level of involvement 5.08 1.05 22

- total 4.82 1.09 55

- total 4.28 1.17 45

Level of Involvement

(33)

Table 2 – Results of a two-factor analysis of co-variance for Credibility of Source (N = 100). Sum of Squares df Mean Square F p η2 Condition 10.25 1 10.25 8.26 0.005 0.08 Level of Involvement 4.84 1 4.84 3.90 0.051 0.04 Condition * Level of Involvement 0.10 1 0.10 0.08 0.778 0.00 Error 117.84 95 1.24 Total 2226.33 100

Table 3 – Effect of Branded Journalism on Credibility of Content

Mean Standard

Deviation

N

Branded Journalism:

- total 4.94 1.17 55

- total 4.47 1.25 45

(34)

Table 4 – Results of a two-factor analysis of co-variance for Credibility of Content (N = 100). Sum of Squares df Mean Square F p η2 Condition 7.04 1 7.04 4.78 0.031 0.05 Level of Involvement 1.71 1 1.71 1.16 0.284 0.01 Condition * Level of Involvement 0.33 1 0.33 0.23 0.635 0.00 Error 139.80 95 1.47 Total 2382.22 100

Table 5 – Effect of Branded Journalism on Attitude towards Content Mean Standard Deviation N Branded Journalism:

- total 4.83 0.97 55

(35)

- total 4.76 0.96 45 Level of Involvement

Table 6 – Results of a two-factor analysis of co-variance for Attitude towards Content (N = 100). Sum of Squares df Mean Square F p η2 Condition 0.69 1 0.69 0.73 0.394 0.01 Level of Involvement 1.47 1 1.47 1.57 0.213 0.02 Condition * Level of Involvement 0.25 1 0.25 0.27 0.606 0.00 Error 88.92 95 0.94 Total 2392.48 100

Table 7 – Effect of Branded Journalism on Attitude towards Brand

Mean Standard

Deviation

N

Branded Journalism:

- total 5.11 0.94 55

(36)

- total 4.98 0.90 45

Table 8 – Results of a two-factor analysis of co-variance for Attitude towards Brand (N = 100). Sum of Squares df Mean Square F p η2 Condition 1.41 1 1.41 1.70 0.196 0.02 Level of Involvement 3.18 1 3.18 3.82 0.054 0.04 Condition * Level of Involvement 0.05 1 0.05 0.06 0.814 0.00 Error 79.07 95 0.83 Total 2636.72 100

Table 9 – Effect of Branded Journalism on Behavioural Intention

Mean Standard

Deviation

N

Branded Journalism:

(37)

- total 5.13 1.31 55

- total 5.35 1.16 45

Table 10 – Results of a two-factor analysis of co-variance for Behavioural Intention (N = 100). Sum of Squares df Mean Square F p η2 Condition 0.001 1 0.001 0.001 0.982 0.00 Level of Involvement 17.64 1 17.64 12.85 0.001 0.12 Condition * Level of Involvement 0.23 1 0.23 0.17 0.681 0.00 Error 130.41 95 1.37 Total 2886.00 100

Table 11 – Effect of Branded Journalism on Purchase Intention

(38)

Deviation Branded Journalism:

- total 4.44 1.19 55

- total 4.12 1.43 45

Table 12 – Results of a two-factor analysis of co-variance for Purchase Intention (N = 100). Sum of Squares df Mean Square F p η2 Condition 7.07 1 7.07 4.66 0.033 0.05 Level of Involvement 16.55 1 16.55 10.90 0.001 0.10 Condition * Level of Involvement 0.01 1 0.01 0.01 0.947 0.00 Error 144.26 95 1.52 Total 2014.88 100