Graduate School of Communication Master’s programme Communication Science
Non-Profits and the Quest for Engagement:
The Effect of Narrative Perspective and Situational Similarity in Non-Profit Narratives on Instagram
Nicole Fleischer Student ID: 11692634 Dr. Anke Wonneberger
Despite the growing pool of research regarding organizational social media use, one of the recent goals of organizations in general but especially non-profit organizations, is still to understand how social media users can be activated to engage for the organization’s benefit. Past research has shown that in this regard the use of narratives or stories can an effective way for organizations to persuade their social media following to engage with them. One of the underlying mechanisms in this relationship has been defined as narrative transportation, a process in which story receivers become transported into the realm of a story and
consequently, are more likely to undergo attitude and behavior changes. Building on narrative transportation theory, the goal of this study is to investigate the effect of narrative perspective (first-person vs. third-person perspective) and situational similarity (similar protagonist vs. non-similar protagonist) on a user’s willingness to engage with a non-profit organization on Instagram. The results of an experimental study revealed no significant effects in regard to narrative perspective and situational similarity. They did, however, indicate that character identification served as a direct predictor for narrative transportation and as a predictor for online and offline engagement intentions. In addition, this effect was mediated by narrative transportation and in turn, attitude towards the organization. Finally, this study was able to shed new light onto the debate regarding the contrasting views of slacktivism, showing that online engagement intentions were positively associated with future offline engagement intentions.
Introduction
The proliferation of the Internet, and particularly social media in the last decade, has
irrevocably changed the landscape of organizational communication (Chung & Taneja, 2016). The enhanced potential for two-way communication on social media has turned passive stakeholders and publics into active participants in the organization’s conversation and made social media engagement one of the most desired objectives for many organizations
(Hollebeek, 2011). Research has shown that successful online engagement may lead to outcomes such as better stakeholder relationships (Hudson, Huang, Roth & Madden, 2016), increased organizational loyalty (Dessart, Veloutsou & Morgan-Thomas, 2015), enhanced brand image (Ibrahim, Wang & Bourne, 2017) and even increased brand usage intentions (Hollebeek, 2011). Consequently, one of the leading questions that drives current research in this field is what factors or antecedents lead to an increase in social media engagement.
Non-profit organizations have increasingly found themselves in the focus of the debate about online engagement. Research has shown that non-profit organizations can have an easier time generating online engagement than for-profit brands because they are perceived as “doing good” and users who decide to engage with non-profit organizations can use this as a self-expression tool to others, signaling that they also want to do good and help others (Bernritter, Verlegh & Smit, 2016). However, there have been studies that have found that social media engagement may be detrimental to a non-profit’s success because it low-engagement actions such as liking and commenting (often called slacktivism) may decrease user’s future intentions to engage in more meaningful ways offline (Kristofferson, White & Peloza, 2013). The main objective of this study is therefore to investigate how non-profit organizations can shape their external communication to increase their online engagement, while at the same time also investigating the effect on their willingness to move that engagement offline towards real-life action with the goal of gaining more insights into the unanswered debate on slacktivism.
In order to address these questions, this study will build on the Transportation-Imagery-Model established by Green & Brock (2002). According to their hypothesis, story recipients can be transported into a story, meaning they become detached from the world of their own and fully immerse themselves in the realm of the narrative (Banerjee & Greene, 2013). This process has been found to cause changes in affective and cognitive responses, beliefs as well as attitudes and intentions and is therefore seen as a useful tool in persuasion theory (Van Laer, De Ruyter, Visconti & Wetzels, 2013). There are several antecedents that will influence the degree of narrative transportation, mainly categorized into those concerning the story itself and those concerning the story receiver (Green & Brock,2002). One of the aspects that plays an important part when it comes to the story itself is the identifiability of the story’s characters (Van Laer et al., 2013). Prior research has shown that identification with story characters can be impacted by the perspective of the author (Farace, Van Laer, De Ruyter & Wetzels, 2017) as well as by presenting characters that are similar to the story receiver (Igartua, Wojcieszak, Chachón-Ramón & Guerrero-Martín, 2017; Chen, Bell & Taylor, 2016). This study will build on these findings, investigating how the implementation of both of these factors indirectly influences a user’s willingness to engage with a non-profit organization, online and offline. It aims to answer the following research question:
RQ: How does the use of narrative perspective and reader-character similarity on social media affect user’s willingness to engage with non-profit organization? Further, the study will focus on answering the research question in the context of
organizational engagement on Instagram which has been one of the fastest growing social media networks in recent years. The platform counts a total of 800 million active users worldwide (Instagram, 2017), among them more than 8 million active business accounts (Instagram Business, 2017). It has become one of the main social media platforms
organizations can use to communicate with their stakeholders and is therefore of particular interest in the context of this study.
Theoretical Background
Narrative Persuasion and the Transportation-Imagery-Model
In an attempt to motivate individuals to engage with an organization, one of the techniques that has gathered growing support is the use of stories or narratives. Stories have long been an essential part of human interaction and communication (Van Laer, De Ruyter, Visconti, & Wetzels, 2013). However, it is only in recent years that narratives have emerged as an alternative format for organizational communication as part of narrative persuasion theory (Green, 2006). In general, a narrative can be understood as the telling of any type of experience or situation that is cohesive, compelling and includes an apparent beginning, middle and ending (Hinyard & Kreuter, 2007). Providing enough information about the scene or the characters is another element necessary for the story receiver to be able to have a clear picture of the situation (Banerjee & Greene, 2013). While there is an overarching consensus that stories and narratives can indeed benefit persuasion attempts (Green & Brock, 2002; Slater & Rouner, 2002; Dal Cin, Zanna, & Fong, 2004), there are different lines of argumentation as to what the mechanisms are that underlie this process. One of the main approaches that this study will draw upon in order to see if non-profit organizations can increase their engagement by using stories is the theory of narrative transportation.
Initially established by Green and Brock (2000), the theory of narrative transportation describes a mental process in which story receivers become transported into the realm of the narrative and are therefore more likely to be impacted by it (Banerjee & Greene, 2013). This process is thought to be caused by a strong imagination as well as strong feelings of empathy (Van Laer et al., 2013). When story recipients become absorbed in a narrative, they can forget about the real world for a moment and feel like they are actually experiencing what is
happening in the story (Hinyard & Kreuter, 2007). Prior research has shown that returning from such an experience can have a variety of consequences for recipients (Green & Brock, 2002). These can include changes in beliefs, opinions, attitudes, and in this context especially
relevant, changes in behavior intentions (Green, 2004; Van Laer et al., 2013). Narrative transportation therefore provides interesting potential for non-profit organizations trying to influence the behaviors of their audience.
The extent to which a recipient becomes transported by a narrative is mainly
determined by two categories. On the one hand, there are story-related attributes that can be influenced by the story-teller, such as the narrative style, tone or plot. On the other hand, there are the characteristics of the story receiver that may determine narrative transportation, such as the familiarity with the topic, the amount of attention that is being paid to the story or simple demographic attribute such as age or gender (Green & Brock, 2002; Van Laer et al., 2013). Since non-profit organizations have no power to influence the characteristics of their audience, this study will focus on potential story-related attributes that could increase narrative transportation and thus persuade Instagram users to engage with the organization. One of the main story attributes that has been established as influential is having identifiable characters in the narrative (Van Laer et al., 2013). Identifiable characters are characters of a story that are presented in ways, so the recipients of the story can easily identify with them. This makes it easier for the recipients to imagine and experience the actions of the story through the eyes of the character and therefore, should increase the extent to which recipients become transported (DeGraaf, Hoeken, Sanders, & Beentjes, 2012; Van Laer et al., 2013). In order to ultimately understand how non-profit organizations can increase transportation into their narratives on Instagram, the following section will take a closer look at what aspects may increase identification with the characters.
Narrative Perspective
One of the aspects that has been argued to increase character identification, and therefore play an important role in the narrative transportation process, is the narrative perspective (also called narrative person or narrative voice) (Banerjee & Greene, 2012). This concept describes the point of view from which the story is told and is usually distinguished between
first-person and third-first-person narratives. In first-first-person narratives, the story is narrated by the protagonist who is a character within the story plot itself. For third-person narratives on the other hand, the narrator usually remains unidentified and uninvolved in the story (Chen, Bell, & Taylor, 2017; Banerjee & Greene, 2012). The first-person perspective is said to increase the reader’s potential identification with the character more than the third-person perspective because it allows more direct insight into the main characters personal thoughts, emotions and experiences, which in turn can promote the imagination process (Banerjee & Greene, 2012; DeGraaf et al, 2012). This idea is in line with the results of Chen, Bell, and Taylor (2017) who find that using a first-person narrator in a health narrative about Diabetes lead to higher levels of character identification than using a third-person narrator. Similarly, Segal et al. (1997) found that readers of a first-person narrative had higher chances of identifying with the protagonist and were more likely to understand his or her thought processes when compared to third-person narrators. DeGraaf et al. (2012) add to these findings by showing that the perspective from which the narrative was told significantly influenced how much participants of the experiment identified with the characters.
Furthermore, this effect appears to not only be limited to written narratives. Huur, Lim, & Lyu (2017) have found that using first-person images (such as selfies) can also be more effective at creating mental imagery than visuals from a third-person perspective which thus can promote character identification. Therefore, based on all of the findings presented above, the first hypothesis is proposed:
Hypothesis 1: Character identification will be higher for those experiencing a first-person narrative than for those experiencing a third-first-person narrative.
Situational Similarity
Another important mechanism that research suggests could have an impact on how much readers identify with a narrative’s protagonist is the perceived similarity between them (Green, 2006). This can be seen as being both a receiver characteristic as well as a
story-related characteristic. On the one hand, perceived similarity to the protagonist is generally something that is felt by the reader of the story, making it a receiver characteristic. However, on the other hand, the author is also able to present the protagonists in a way that makes them similar to the potential audience, therefore making it a story-related attribute that can be manipulated. Since in this context, this study is interested in the way stories can be shaped to influence identification and in turn narrative transportation, from now on reader-protagonist similarity will be understood as a story-related characteristic.
As mentioned before, identification with a character is partly based on whether a reader can imagine him- or herself as that character, similarity between the two should therefore facilitate identification (Cohen, 2001). Prior research has indeed shown that participants who were placed in an experimental condition in which characters were similar, reported higher levels of character identifications than those who weren’t similar and in turn higher levels of narrative transportation (Igartua et al., 2017). Chen, Bell, and Taylor (2016) provide further evidence for this, showing that protagonist similarity directly influenced identification with the main character in a cancer-related health narrative designed to persuade readers to take action.
Several authors have pointed out that there are many characteristics that reader-character similarity may be based upon, such as age, gender, nationality or more complex characteristics, such as behavior, attitudes, values, personality or situations (Kreuter et al., 2007; Cohen, 2001; Green, 2006; Igartua et al., 2017). In the context of non-profit
organizations’ attempt to engage Instagram users, the audience that they are trying to engage with is generally quite diverse. Since social media platforms do not allow organizations to individualize their posts for different audience segments, it can be difficult to find similarity characteristics to incorporate into their communication that apply to all users at the same time. An organization’s audience will inherently differ in age, gender, nationality or personality, rendering those type of characteristics inefficient when trying to manipulate similarity.
However, one aspect that the audience does have in common, is that they are all using the specific social media platform to consume the organization’s content. Therefore, showing a narrative in which protagonists are in a similar situation, in this case using that social media platform as the reader, could offer great potential to stimulate feelings of similarity and therefore enhance character identification (Cohen, 2001). Along the same lines, Kreuter et al. (2007) have argued that characters in a similar situation might make it easier for story
receivers to process the content of the story which in turn could facilitate character identification. In order to gain more insight into this idea, based on the evidence provided above, I propose the following:
Hypothesis 2: Character identification will be higher for those experiencing a narrative in which the character appears to be in a similar situation than for those where the character does not seem to be in a similar situation.
Based on Hypothesis 1 and 2, this study also expects the following:
Hypothesis 3: Character identification will be highest for those experiencing a first-person narrative in which the character appears to be in a similar situation and lowest for those experiencing a third-person narrative in which the character does not appear to be in a similar situation.
Character Identification
The two previous sections have provided support for the hypotheses that narrative perspective as well as situational similarity may act as an antecedent to character identification. As
already mentioned, character identification has been viewed by many as one of the important story-related factors that influences narrative transportation (Van Laer et al., 2013; DeGraaf et al., 2012). When recipients can identify with a character, this facilitates the process of
imagining what it must be like to live in their shoes and therefore, allows recipients to feel more empathy for the character (DeGraaf, 2013). Consequently, this makes it more likely for story recipients to become transported into the narrative (Slater & Rouner, 2002; Van Laer et
al., 2013). The results of an experimental study by Van Laer et al. (2013) support this argument, showing that higher levels of character identification were significantly linked to more narrative transportation. Similarly, Escalas, Moore, and Brittion (2004) find characters that a story reader can easily identify with lead to more transportation into the story. In a study on the effects of character identification on narrative persuasion in general, Slater and Rouner (2002) find that there is no direct relationship between the two concepts. In an attempt to make sense of these results, they conclude that character identification may predict
narrative transportation instead which then increases the chance of persuasion. Based on these findings, I expect the relationship between character identification and narrative
transportation to be as follows:
Hypothesis 4: The experience of narrative transportation will be stronger when the level of character identification is higher than when character identification is lower.
Attitude Changes
One of the consequences of narrative transportation that has received a decent amount of attention within the field is attitude change (Mazzocco, Green, Sasota, & Jones, 2010). Green and Brock (2000) were one of the first to argue that due to the affective and cognitive
responses that narrative transportation generates, the opinion formation of recipients can be influenced and therefore, attitudes are more likely to be affected. Since then, many studies have been able to show that narrative transportation has a significant impact on attitude change (Mazzocco et al., 2010; Van Laer et al., 2013; Braddock & Dillard, 2016). The findings of Banerjee and Greene (2012) provide further support for this claim, showing that transportation significantly influenced the attitude towards alcohol and this effect was mediated by cognitive as well as affective responses. In an experimental study on the effects of brand image narratives, Lim and Childs (2017) have provided results that argue brand consumers’ level of transportation into the narrative is significantly linked to an increase in positive attitudes towards the brand. Similarly, in 2016 the authors found that transportation
into a brand narrative on Instagram positively influenced the attitude towards that brand’s Instagram account (Lim & Childs, 2016).
Following the line of reasoning above, I would propose that, in the case of non-profit organizations, being transported into a narrative written by and about the organizations will consequently have a positive impact on the readers’ attitude towards the organizations. Therefore, I hypothesize:
Hypothesis 5: Attitudes towards the non-profit organization will be more favorable for those experiencing higher levels of narrative transportation than for those experiencing lower levels of narrative transportation.
Behavior Intentions
As mentioned at the beginning of this chapter, one of the main prospects of narrative transportation for organizations is that it has been found to lead to changes in behavior intentions. An experimental study by Van Laer et al. (2013) has shown that higher levels of narrative transportation can lead to significantly stronger behavior intentions. Similarly, findings by Braddock & Dillard (2016) provide additional evidence for this relationship, showing a positive link between being exposed to a transporting narrative and narrative-consistent behavior intentions. Building on Hypothesis 4, Slater, Johnson, Cohen, Comello, and Ewoldson (2014) propose the argument that being transported is likely to have an influence on behavior intentions because of results in more narrative-consistent values and attitudes. The findings of an experimental study by Oliver, Dillard, Bae, and Tamul (2012) on the attitude towards stigmatized groups support this argument. The authors are able to show that after being exposed to a narrative, an empathetic attitude towards stigmatized groups directly influenced behavioral intentions.
In the case of non-profits, promoting behavior in form of engagement has become a desirable goal for many organizations (Hollebeek, 2011). Engagement can vary between low, medium and high and can generally take place online as well as offline (Neiger et al., 2012).
Online forms of engagement are most commonly categorized as low to medium engagement and include actions such as liking, sharing or creating social media posts. Offline engagement on the other hand is considered a form of high engagement that consists of actions such as donating, volunteering or participating in other real-life events (Neiger et al., 2012). As the findings presented above have shown, positive attitudes towards the organization due to narrative transportation can increase a story receiver’s behavior intentions. When applied to the pursuit of creating more online and offline engagement, I propose that these results suggest that a positive attitude towards the organization will results in higher engagement intentions. Therefore:
Hypothesis 6: The intention to engage with the non-profit organization a) online and b) offline will be higher for those with a more favorable attitude towards the
organization that for those with a less favorable attitude.
In summary, the previous hypotheses therefore propose that readers experiencing a narrative in the first-person perspective and with characters similar to themselves will show the highest behavior intentions because they will identify more with the characters which in turn will lead to higher levels of narrative transportation and thus a more positive attitude towards the organization. In order to test this research model (see Figure 1) for significance, the following hypotheses are also being proposed:
Hypothesis 7: Those experiencing a first-person narrative will show higher online and offline engagement intentions than those experiencing a third-person narrative. This effect will be mediated by higher character identification which in turn will lead to higher narrative transportation which in turn will lead to a more favorable attitude towards the organization.
Hypothesis 8: Those experiencing a narrative in which the protagonist appears to be similar will show higher online and offline engagement intentions than those who are not. This effect will be mediated by higher character identification which in turn will
lead to higher narrative transportation which in turn will lead to a more favorable attitude towards the organization.
Similarly to Hypothesis 3, the interaction between narrative perspective and situational similarity is expected to have the highest impact on online and offline engagement intentions. Therefore, I finally hypothesize:
Hypothesis 9: Those experiencing a first-person narrative in which the protagonist appears to be similar will show the highest online and offline engagement intentions and those experiencing a third-person narrative in which the protagonist appears to be non-similar the lowest. This effect will also be mediated by higher character
identification which in turn will lead to higher narrative transportation which in turn will lead to a more favorable attitude towards the organization.
Slacktivism
In addition to the previously mentioned hypotheses, the present study also aims to look into the relationship between online and offline engagement intentions. In general, prior studies have shown that it can be fairly easy for non-profit organizations to foster online engagement by users on social media due to their perceived brand warmth (Bernritter, Verlegh, & Smit, 2016). Since non-profit organizations are generally perceived as “doing good”, users who decide to engage with non-profit organizations can use this as a self-expression tool, signaling to peers that they also want to do good and help others (Bernritter, Verlegh, & Smit, 2016). However, there has been a growing consensus among some scholars in the field that argue that increased social media engagement may actually be detrimental to a non-profit’s success because it can negatively affect offline engagement intentions. This concept is generally known as slacktivism or clicktivism and is defined as the willingness to perform acts that require little to no effort, such as liking or sharing online, which then leads to a decrease in willingness to engage in real and more meaningful types of action (Kristofferson, White, & Peloza, 2013; Hu, 2014). This is generally thought to happen because forms of public online
engagement may satisfy a user’s need to “do good” and create an outward illusion of activism which in turn however, reduces the need to engage with the cause in more traditional and effective ways (Christensen, 2011).
While there are several studies that provide evidence for the existence of slacktivism and the negative effects of online engagement (Kristofferson, White, & Peloza, 2013; Schuman & Klein, 2015), there have also been contradicting studies that show that online engagement can increase future engagement intentions (Hogben & Cownie, 2017), money contributions (Mano, 2014; Lee & Hsieh, 2013) and offline forms of participation (Hu, 2014). Therefore, in order to gain a more in-depth, clearer picture of the relationship between online and offline engagement, the following exploratory research question is proposed:
RQ1: How does the intention to engage with a non-profit organization online influence the intention to engage offline?
Figure 1: Proposed Research Model
Method
Experimental Design and Stimulus
In order to test the above-mentioned hypotheses, this study conducted an online experiment, using the survey platform Qualtrics. Participants were randomly assigned to one of four experimental conditions based on a 2 x 2 factorial design. The two experimental factors were narrative perspective (first-person vs. third-person) and situational similarity (similar vs.
non-Narrative Perspective Character Identification Narrative Transporation Attitude Towards Organization Online Engagement Intentions Offline Engagement Intentions Situational Similarity + + + + + + ?
similar situation of the protagonist). All respondents were asked to read a fictional narrative spread out over four individual Instagram posts. The narrative itself told the story of a girl called Nora who donated to an elephant rescue organization by adopting one of the organization’s newly rescued elephants. She then found out that the adopted elephant was pregnant when it was rescued and followed up by deciding to become an active volunteer. The story finished off with a happy ending about the healthy mother and her baby who were soon to be released back into the wild.
To manipulate narrative perspective, the story was either written from a first-person or from a third-person perspective. In addition, based on Farace, Van Laer, de Ruyter and
Wetzels (2017) conceptualization of narrative perspectives in visualized images, two of the four pictures used for the Instagram posts that showed Nora were manipulated by using either selfies (first-person) or elsies (third-person). And lastly, first-person narratives are generally understood to be more persuasive due to the fact that they allow the reader more insight into the protagonist’s thoughts and feelings (Banerjee & Greene, 2012). Therefore, the first-person condition material was written with more personal insight into Nora’s emotions by including phrases such as “So Excited!”, “I’m so grateful” or “I can’t believe…” instead of more neutral phrasing for the third-person condition.
Situational similarity was manipulated by identifying the source that Nora used in order to become engaged with the organization. In the similar condition, the story depicted Nora as a volunteer who was activated after seeing the organization Instagram content and their call to action on an Instagram post. On the contrary, in the non-similar condition, Nora was said to have seen a newspaper article about the organization. The source of activation was mentioned in the first as well as last Instagram posts that participants read (See Appendix A for all pictures and story text).
Data Collection and Sample
Participants for the experiment were recruited with an anonymous link that was distributed online via Facebook over a 5-day period. The sample was collected by means of a
convenience sample. There were no age restrictions or criteria of any other kind when recruiting respondents.
The initial sample included a total of 160 respondents that completed the experiment. Two participants were then removed for straight-lining on four out of the five scales used and two were removed because they completed the survey in under 2.5 minutes which was
deemed as unrealistic (M = 7.73, SD = 3.85). The final sample therefore included a total of 156 respondents that were evenly distributed across all four experimental conditions by being assigned to one of the four experimental groups in Qualtrics (n=44; n=38, n=37, n=37). The average age of participants was 29.96 (SD = 13.16), ranging from a minimum of 15 years to a maximum of 75 years. There were 131 female respondents, 24 male and one who identified as other. In regard to education, 82 participants had completed a Bachelor’s degree, 40 were high school graduates (or equivalent), 28 had completed a Master’s degree, 5 participants had finished a Doctorate degree and one respondent reported having less than a high school degree.
Measures
Following the story, respondents were asked to rate how strongly they agreed with two statements in order to check whether both manipulations were effective. For narrative perspective, respondents were asked to rate on a 7-point Likert scale how much they agreed that “The story I just saw was told directly from the perspective of Nora”. For situational similarity, this statement was “The main character of the story decided to volunteer after seeing an Instagram post”.
Identification with the protagonists was measured with a scale by Chen, Bell and Taylor (2017), asking participants to rate their agreement with five items on a 7-point Likert
scale from “Strongly disagree” to “Strongly agree”. Sample items were “The main character and I are similar kinds of people” or “I like the main character in the story a lot”. A principal component analysis showed that these five items form a single uni-dimensional scale with only one component that had an eigenvalue above 1 (eigenvalue 3.51) and explained a total of 70.27 % of variance. In addition, the scale was found to be highly reliable with a Cronbach's alpha of .89. Therefore, the scale was deemed to successfully measure participants’
identification with the main character Nora and a new variable was computed based on the average of all items (M = 4.49, SD = 1.17).
Narrative transportation was measured on using parts of the Narrative Transportation Scale by Green & Brock (2000). This experiment included eight of the 13 original items that were deemed most appropriate to the context of this narrative. Participants were presented with statements such as “While I was reading the narrative, I could easily picture the events in it taking place” and “While reading the narrative I had a vivid image of the main character”. Again, all items were measured on a 7-point Likert scale, ranging from “Strongly disagree” to “Strongly agree”. The results of a principal component analysis showed that this scale was also found to be uni-dimensional. All items were positively correlated with one factor that had an eigenvalue of 3.47 and accounted for 43.11% of variances. Furthermore, the reliability of the scale was reasonable, Cronbach's alpha = .79, indicating that the scale accurately
measured narrative transportation. The newly-computed variable indicated a mean of 4.60 and a standard deviation of .94.
To measure the attitude towards the organization, the attitude scale by Dunlop, Wakefield and Kashima (2010) was adapted to the context of this study. This scale consisted of a total of six items, asking participants to rate their attitude towards the organization on 7-point scale between bad to good, undesirable to desirable, unfavorable to favorable, not
important to important, not effective to effective and lastly, worthless to valuable. Another
previous scales, this scale was uni-dimensional. Only one component had an eigenvalue above 1 (eigenvalue = 4.77) and this component explained 79.46% of variance. The
Cronbach’s alpha was .95, suggesting the scale used was highly reliable. Therefore, the final variable was constructed based on the average of all items (M = 5.85, SD = 1.11).
Finally, for online and offline engagement, questions were based on the potential actions of engagement by Neiger et al. (2012) that were perceived to be relevant to the context of Instagram and non-profit engagement. For online engagement, participants were asked how likely they would be to like, comment or share the organization’s content as well as follow their account. For offline engagements, respondents were asked to rate the
likelihood of them engaging within the organization by telling a friend about it, donating, attending an event or volunteering. Both scales for online and offline engagement were also found to form uni-dimensional scales, demonstrated by the results of two further principal component analyses. All four online engagement items positively correlated with one component that had an eigenvalue of 2.81 and explained a total of 70.20 % of variance. The items were also found to be highly reliable, Cronbach's alpha = .86 (New variable: M = 4.00,
SD = 1.51). Similarly, for offline engagement, the results indicated that the scale was
uni-dimensional with a one component solution and an eigenvalue of 2.97, accounting for 74,21% of the variance. In addition, a reliability test confirmed that the scale was highly reliable, with a Cronbach’s Alpha of .88 (New variable: M = 3.65, SD = 1.53).
The full wording of all items used in this experiment as well as a list reporting the factor loadings for all items can be found in Appendix B and C.
Results
Randomization
In order to test whether participants in the four experimental conditions varied significantly on any relevant factors, a randomization check was conducted. These factors included age, gender, education level as well as frequency of non-profit engagement. First, an ANOVA was
carried out in order to see if participants’ mean age differed between experimental groups. The results showed that there were no significant differences, F (3, 152) =0 .64, p = .592 Furthermore, the results of independent sample t-tests showed that the distribution among conditions did not vary significantly in regard to gender, t (6) = 4.71, p =.582 or education level, t (12) = 14.67, p =.260. Lastly, participants did not differ between groups when it comes to how frequently they engage with non-profit organizations online, F (3, 152) =0 .32,
p = .811. The same was also found for how frequently participants engaged with non-profits
offline, F (3, 152) =0 .6, p = .619. Thus overall, the randomization between the four experimental conditions was successful.
Manipulation
To check whether the two experimental factors were manipulated successfully, participants were asked to report how much they agreed with the statements “The story I just saw was told from the perspective of Nora” and “Nora decided to become active after seeing an Instagram post”. For narrative perspective, the results of an independent sample t-test showed that the average agreement with the statement was significantly higher for the first-person perspective condition (M = 6.11, SD = 1.07) than for those in the third-person perspective condition (M = 3.08, SD = 1.65), t (156) = -13.83, p < 0.001, indicating that the manipulation of this factor was successful. For situational similarity, an independent sample t-test showed comparable results. Participants in the similar condition reported higher agreement with the statement (M = 5.66, SD = 1.35) than those in the nonsimilar condition (M = 3.49, SD = 2.21), t (156) = -7.53, p < 0.001. Therefore, the manipulation was assumed to be successful for this factor as well.
Model Estimation
First, in order to asses each component of the proposed mediation model, multiple regression analysis was conducted (see Figure 2 for overview of results). The examination of both
experimental factors showed that there was no significant main effect of narrative perspective on character identification, b = .34, SE = .27, t (152) = 1.26, p = .211, 95% CI [-.20, .87] (First-person perspective: M = 4.59, SD = 1.13, Third-person perspective: M = 4.37, SD = 1.20), and no significant main effect of situational similarity on character identification either,
b = .17, SE = .27, t (152) = .65, p = .527, 95% CI [-.37, .71] (Similar: M = 4.51, SD = 1.16,
Non-Similar: M = 4.46, SD = 1.18). In regard to the interaction effect, the results were also found to not be significant, b = -24, SE = .38, t (152) = -.65, p = .517, 95% CI [-.99, .50]. The model revealed that only approximately 1.2% of the variance in character identification was explained by the predictors (R2 = .012). An additional ANOVA confirmed these findings, showing no significant main effects for narrative perspective, F (1, 152) = 1.84, p = .250, situational similarity, F (1, 152) = .01, p = .788, or their interaction, F (1, 152) = .58, p = .517. Thus, Hypothesis 1, 2 and 3 could not be supported.
Further, as expected by Hypothesis 4, the multiple regression analysis revealed a significant association between character identification and narrative transportation, b = .53,
SE = .05, t (153) = 10.62, p < .001, 95% CI [.43, .63] and 42,55% of variance in narrative
transportation was explained (R2 = .425), therefore supporting the hypothesis. Similarly, the results confirmed expectations regarding Hypothesis 5, showing that narrative transportation was a significant predictor of attitude towards the organization, b = .25, SE = .11, t (152) = 2.30, p = .023, 95% CI [.04, .47] with 23,9% of variance explained (R2 = .239).
And lastly, Hypothesis 6 expected that a positive attitude towards the organization would serve as a predictor for a) online engagement intentions and b) offline engagement intentions. The results showed that there was no direct effect of attitude on online engagement intentions, b = .15, SE = .09, t (152) = 1.75, p = .083, 95% CI [-.02, .33] and no direct effect on offline engagement intentions, b = .11, SE = .10, t (152) = 1.07, p = .287,
95% CI [-.09, .31], thus providing no evidence for the support for Hypothesis 6. However, they did reveal that character identification served as a direct predictor for both online
engagement intentions, b = .25, SE = .11, t (151) = 7.45, p < .000, 95% CI [.55, .75] and offline engagement intentions, b = .25, SE = .11, t (152) = 5.87, p < .000, 95% CI [.45, .91]. In regard to online engagement, approximately 51,3% of variance was able to be explained by the predictors (R2 = .513). For offline engagement, it was roughly 37,9% of variance (R2 = .379).
Figure 2: Mediation Model and Coefficients
Note: *p ≤ .05. **p ≤ .01. ***p ≤ .001.
Next, in order to test the indirect effects of narrative perspective and situational similarity on online and offline engagement intentions for the mediational hypothesis 7 and 8, mediation analysis with bootstrapping procedures and bias-corrected confidence estimates was
conducted. In this study, a 95% confidence interval was obtained by using 5,000 bootstrap samples as suggested by Hayes (2018). As expected due to the non-significance of Hypothesis 1 and 2, the results revealed that for narrative perspective, there was no significant direct or indirect effect on online or on offline engagement intentions. Furthermore, the results for situational similarity also indicated no significant direct or indirect effects on both online and offline engagement intentions (see Appendix D for full table of results). Therefore, there was no evidence for the support of the mediation effects proposed in Hypothesis 7 and 8.
Similarly, when testing for a moderated mediation effect between narrative
perspective, situational similarity and online/offline engagement intentions, as suggested in
Narrative Perspective Character Identification Narrative Transporation Attitude Towards Organization Online Engagement Intentions Offline Engagement Intentions Situational Similarity .34 .17 .53*** .26* .15 .11
Hypothesis 9, using the same bootstrapping procedure by Hayes (2018), the results were also found to not be significant (results also reported in Appendix D). Therefore, there is no indication for the support for Hypothesis 9.
Since the experimental manipulations were both not significantly correlated with any of the outcome variables, an additional revised model in which both narrative perspective and situational similarity were dropped was considered. Figure 3 provides an overview of the results for the revised mediation model.
Figure 3: Revised Mediation Model
Note: *p ≤ .05. **p ≤ .01. ***p ≤ .001.
The mediation analysis conducted with 5000 bootstrap samples, obtaining a confidence level of 95% (Hayes, 2018) confirmed a significant partial mediation effect of character
identification on online engagement intentions, b = 0.90, SE = .08, t = 11.90, p < .001, 95% CI [.75, 1.05], while explaining 47.9% of variance, F (1,154) = 141.54, p < .001. In addition, the direct effect of character identification on online engagement intentions was found to be significant, b = 0.73, SE = .10, t = 7.24, p < .001, 95% CI [.00, .53], thus suggesting a partial mediation. Similarly, for offline engagement, the results of the mediation analysis provided further support for the mediating role of narrative transportation and attitude towards the
organization between character identification and offline engagement, b = 0.79, SE = .08, t = 9.25, p < .001, 95% CI [.62, .95], while explaining 35.7% of variance, F (1,154) = 85.54, p < .001. The direct effect of character identification on offline engagement was also found to be significant, b = 0.66, SE = .12, t = 5.72, p < .001, 95% CI [.43, .89], thus suggesting another partial mediation.
Research Question 1
Lastly, in regard to the exploratory research question on the effect of online engagement intentions on offline engagement intentions, the results of a linear regression revealed a positive relationship. Online engagement intentions were found to positively predict offline engagement intentions, b = 0.704, t = 11.98, p < .001, 95% CI [0.59, 0.82], explaining 48.2% of variance (R2= .48). The entire model was also found to be significant, F (1,154) = 143.58,
p < .00, providing strong evidence for a positive association between online and offline
engagement intentions. Discussion
The findings of the analyses presented above provide insight into this study’s leading research question as well as the proposed hypotheses. Overall, the results suggest that narrative
perspective, situational similarity as well as their interaction do not significantly influence a user’s willingness to engage with a non-profit organization on Instagram, directly or by mediation through character identification, narrative transportation and attitude towards the organization (Hypothesis 7-9). With regard to the Hypothesis 1 and 2, this study did not find support for the two main effects claiming that narrative perspective and situational similarity increase character identification. However, the results did show that a story receiver’s identification with the character of the story positively influences online and offline engagement intentions and that this relationship was partially mediated by narrative
the debate about slacktivism, the results suggest that intentions to engage with a non-profit organization online do not lower offline engagement intentions. On the contrary, this study was able to show that increased online engagement intentions positively influence a user’s future offline engagements, thus providing new valuable insights that contradict the idea that online engagement may reduce an individual’s willingness to engage in more effective ways.
Implications
The results of this study have both practical and theoretical implications. Practically, the findings suggest that including narratives in Instagram communication can be an effective way for non-profit organizations to create online and offline engagement intentions among their audience. While the results did not confirm the hypotheses that writing from a first-person perspective and with similar characters would increase engagement intentions, they do imply that higher character identification can increase them. Therefore, if non-profit
organizations are able to find other ways to increase character identification, this might increase narrative transportation into their narratives and thus, improve attitude towards the organization as well as engagement intentions. Both could be vital in helping non-profit organizations further their causes and reach their goals. In addition, the results regarding the notion of slacktivism also have immense implications for non-profit organizations. Recent findings by Kristofferson, White, and Peloza (2013) or Schuman and Klein (2015), in support of the negative effects of slacktivism have created increasing caution and concern among non-profit organizations when it comes to using social media channels. This study has
demonstrated that when it comes to engagement intentions, these concerns could be unsubstantiated and non-profit organizations should be encouraged to promote online
engagement, given that the intentions to engage online might lead to a subsequent increase in intentions to engage offline. Theoretically, these results provide additional evidence in support of the Transportation-Imagery-Model by Green and Brock (2000, 2002). They highlight both the importance of character identification as an antecedent of narrative
transportation as well as the significance of improved attitude and engagement intentions as resulting consequences of it. This study therefore adds new knowledge to the understanding of narrative persuasion theory.
Limitations and future research
While this study provides additional support for the effective use of narratives in non-profit communication, it does not remain without limitations. One of the main faults of this study can be found in both manipulations. First, in regard to narrative perspective, the manipulation might have been less effective due to the attempt to uphold internal validity. In a more natural setting, the first-person perspective perhaps would have given more in-depth and personal insight into the main characters thoughts and emotions. However, for the sake of internal validity, the content in both experimental groups was kept to the same length and contained the same amount of information. Therefore, the difference between first- and third-person in the experimental setting might have been less pronounced than it would be in a real-life setting which consequently could have had an influence on the effect of narrative perspective. Second, in regard to situational similarity, the study perhaps failed to fully assess whether participants felt similar to the protagonist or not. All respondents were told to imagine they were active Instagram users when being shown the manipulation material. However, there was no explicit question asking whether participants actually use Instagram and the open comment section at the end of the survey revealed that many did not use Instagram. While participants were told to imagine being on Instagram, this might have not been strong enough to actually elicit feelings of similarity to the protagonist’s situation and results should
therefore be interpreted with hesitation. The substantial majority of female participants also raises questions in regard to the external validity and whether the results can be generalized to both male and female social media users.
In addition, the conclusions regarding engagement, especially in the context of slacktivism, also have to be interpreted with care. This study was only able to measure
behavior intentions in an experimental setting. Scholars and organizations alike should
therefore be wary of drawing any definite assumptions from this in regard to real-life behavior of social media users. However, studying actual behavior after exposure to narratives on Instagram, perhaps in a longitudinal study, would be a good proposition for future research and provide more insight into the effect of narrative transportation on real-life behavior.
The findings of this study should also encourage further exploration of the antecedents of narrative transportation, particularly in regard to factors affecting character identification. Future research could explore possible other factors aside from narrative perspective and situational similarity in order to understand how non-profit organizations can make full use of the potential of narrative transportation. Another recommendation for future research could be to expand the study’s hypotheses to other social media platforms than just Instagram. In general, Instagram can be seen as a very visual-based platform that allows little room for written word. Future studies could look at how the combination of text/photo/video on other social media platforms might influence character identification as well as narrative
transportation differently. With the growing popularity of Instagram Stories and Snapchat, it could be helpful for non-profit organization to gain more insight into video narratives on social media.
In conclusion, this study makes a promising case for the use of narratives in non-profit communication, however, additional research is needed in order to fully understand how to properly construct and benefit from these narratives in the future.
References
Banerjee, S. C., & Greene, K. (2013). Examining narrative transportation to anti-alcohol narratives. Journal of Substance Use, 18(3), 196-210.
Bernritter, S. F., Verlegh, P. W., & Smit, E. G. (2016). Why nonprofits are easier to endorse on social media: The roles of warmth and brand symbolism. Journal of Interactive
Marketing, 33, 27-42.
Braddock, K., & Dillard, J. P. (2016). Meta-analytic evidence for the persuasive effect of narratives on beliefs, attitudes, intentions, and behaviors. Communication
Monographs, 83(4), 446-467.
Chen, M., Bell, R. A., & Taylor, L. D. (2016). Narrator point of view and persuasion in health narratives: The role of protagonist–reader similarity, identification, and
self-referencing. Journal of Health Communication, 21(8), 908-918.
Christensen, H. S. (2011). Political activities on the Internet: Slacktivism or political participation by other means? First Monday, 16(2).
Chung, S., & Taneja, H. (2016). Reassessment of audience in public relations Industry: How social media reshape public relations measurements. Asia Pacific Public Relations
Journal, 17(1), 60-77.
Cohen, J. (2001). Defining identification: A theoretical look at the identification of audiences with media characters. Mass Communication & Society, 4(3), 245-264.
Dal Cin, S., Zanna, M. P., & Fong, G. T. (2004). Narrative persuasion and overcoming resistance. In E. S. Knowles & J. Linn (Eds.), Resistance and Persuasion (pp. 175– 191). Mahwah, NJ: Erlbaum.
De Graaf, A., Hoeken, H., Sanders, J., & Beentjes, J. W. (2012). Identification as a mechanism of narrative persuasion. Communication Research, 39(6), 802-823. Dessart, L., Veloutsou, C., & Morgan-Thomas, A. (2015). Consumer engagement in online
brand communities: A social media perspective. Journal of Product & Brand
Dunlop, S. M., Wakefield, M., & Kashima, Y. (2010). Pathways to persuasion: Cognitive and experiential responses to health-promoting mass media messages. Communication
Research, 37(1), 133-164.
Escalas, J. E., Moore, M. C., & Britton, J. E. (2004). Fishing for feelings? Hooking viewers helps! Journal of Consumer Psychology, 14(1-2), 105-114.
Farace, S., van Laer, T., de Ruyter, K., & Wetzels, M. (2017). Assessing the effect of narrative transportation, portrayed action, and photographic style on the likelihood to comment on posted selfies. European Journal of Marketing, 51(11/12), 1961-1979. Green, M. C. (2004). Transportation into narrative worlds: The role of prior knowledge and
perceived realism. Discourse Processes, 38(2), 247-266.
—— (2006). Narratives and cancer communication. Journal of Communication, 56(1). 163-183.
Green, M. C., and T. C. Brock (2000). The Role of Transportation in the Persuasiveness of Public Narratives. Journal of Personality and Social Psychology, 79 (5), 701– 21. —— (2002). In the Mind’s Eye: Transportation-Imagery Model of Narrative Persuasion. In
M. C. Green, J. J. Strange, and T. C. Brock (Eds.), Narrative Impact: Social and
Cognitive Foundations (pp. 315–341). Mahwah, NJ: Erlbaum.
Hayes, A. F. (2018). Introduction to mediation, moderation, and conditional process analysis,
second edition : A regression-based approach. New York: The Guilford Press.
Hinyard, L. J., & Kreuter, M. W. (2007). Using narrative communication as a tool for health behavior change: a conceptual, theoretical, and empirical overview. Health Education &
Behavior, 34(5), 777-792.
Hogben, J., & Cownie, F. (2017). Exploring slacktivism: Does the social observability of online charity participation act as a mediator of future behavioural intentions? Journal
Hollebeek, L. (2011). Exploring customer brand engagement: definition and themes. Journal
of Strategic Marketing, 19(7), 555-573.
Hu, C. W. (2014). Health slacktivism on social media: Predictors and effects. In International
Conference on Social Computing and Social Media (pp. 354-364). Springer, Cham.
Hudson, S., Huang, L., Roth, M. S., & Madden, T. J. (2016). The influence of social media interactions on consumer–brand relationships: A three-country study of brand
perceptions and marketing behaviors. International Journal of Research in
Marketing, 33(1), 27-41.
Hur, S., Lim, H., and Lyu, J, (2017). “I” or “She”? The impact of visual perspectives on creation of consumption imagery on Instagram. In International Textile and Apparel
Association (ITAA) Annual Conference Proceedings. 92. Retrieved from:
https://lib.dr.iastate.edu/itaa_proceedings/2017/presentations/92
Ibrahim, N. F., Wang, X., & Bourne, H. (2017). Exploring the effect of user engagement in online brand communities: Evidence from Twitter. Computers in Human Behavior, 72, 321-338.
Igartua, J. J., Wojcieszak, M., Cachón-Ramón, D., & Guerrero-Martín, I. (2017): “If it hooks you, share it on social networks”: Joint effects of character similarity and imagined contact on the intention to share a short narrative in favor of immigration. Revista
Latina de Comunicación Social, 72, 1085 - 1106.
Instagram (2017). Strengthening our commitment to safety and kindness for 800 million. Instagram Blog. Retrieved from: http://blog.instagram.com/post/165759350412/1709 26-news
Instagram Business (2017). Welcoming 1 million advertisers. Instagram Business Blog. Retrieved from: https://business.instagram.com/blog/welcoming-1-million-advertisers/ Kreuter, M. W., Green, M. C., Cappella, J. N., Slater, M. D., Wise, M. E., Storey, D., ... &
framework to guide research and application. Annals of Behavioral Medicine, 33(3), 221-235.
Kristofferson, K., White, K., & Peloza, J. (2013). The nature of slacktivism: How the social observability of an initial act of token support affects subsequent prosocial
action. Journal of Consumer Research, 40(6), 1149-1166.
Lee, Y. H., & Hsieh, G. (2013, April). Does slacktivism hurt activism? The effects of moral balancing and consistency in online activism. In Proceedings of the SIGCHI Conference
on Human Factors in Computing Systems (pp. 811-820). ACM.
Lim, H. and Childs, M. (2016). Brand storytelling on Instagram: How do pictures travel to millennial consumers' minds? In International Textile and Apparel Association (ITAA)
Annual Conference Proceedings. 96. Retrieved from:
http://lib.dr.iastate.edu/itaa_proceedings /2016/presentations/96
Lim, H. and Childs, M. (2017). Why Instagram? The role of telepresence and transportation in visual content marketing. In International Textile and Apparel Association (ITAA)
Annual Conference Proceedings. 28. Retrieved from:
https://lib.dr.iastate.edu/itaa_proceedings/2017/presentations/28
Mano, R. S. (2014). Social media and online health services: A health empowerment
perspective to online health information. Computers in Human Behavior, 39, 404-412. Mazzocco, P. J., Green, M. C., Sasota, J. A., & Jones, N. W. (2010). This story is not for
everyone: Transportability and narrative persuasion. Social Psychological and
Personality Science, 1(4), 361-368.
Neiger, B. L., Thackeray, R., Van Wagenen, S. A., Hanson, C. L., West, J. H., Barnes, M. D., & Fagen, M. C. (2012). Use of social media in health promotion: purposes, key
performance indicators, and evaluation metrics. Health Promotion Practice, 13(2), 159-164.
Oliver, M. B., Dillard, J. P., Bae, K., & Tamul, D. J. (2012). The effect of narrative news format on empathy for stigmatized groups. Journalism & Mass Communication
Quarterly, 89(2), 205-224.
Schumann, S., & Klein, O. (2015). Substitute or stepping stone? Assessing the impact of low-‐‑ threshold online collective actions on offline participation. European Journal of Social
Psychology, 45(3), 308-322.
Segal, E. M., Miller, G., Hosenfeld, C., Mendelsohn, A., Russell, W., Julian, J., Delphonse, J. (1997). Person and tense in narrative interpretation. Discourse Processes, 24, 271–307. Slater, M. D., & Rouner, D. (2002). Entertainment—education and elaboration likelihood:
Understanding the processing of narrative persuasion. Communication Theory, 12(2), 173-191.
Slater, M. D., Johnson, B. K., Cohen, J., Comello, M. L. G., & Ewoldsen, D. R. (2014). Temporarily expanding the boundaries of the self: Motivations for entering the story world and implications for narrative effects. Journal of Communication, 64(3), 439-455. Van Laer, T., De Ruyter, K., Visconti, L. M., & Wetzels, M. (2013). The extended
transportation-imagery model: A meta-analysis of the antecedents and consequences of consumers' narrative transportation. Journal of Consumer Research, 40(5), 797-817.
Appendix A. Manipulation Content – Group 1 (First-Person & Similar Protagonist)
Appendix A. Manipulation Content – Group 2 (First-Person & Non-Similar Protagonist)
Appendix A. Manipulation Content – Group 3 (Third-Person & Similar Protagonist)
Appendix A. Manipulation Content – Group 4 (Third-Person & Non-Similar Protagonist)
Appendix B. Item Measures.
Character Identification (Chen, Bell, & Taylor, 2017)
The main character and I are similar kinds of people. I like the main character in the story a lot.
I feel that I know the main character in the story well. I identified with the main character in the story. I felt a connection to the main character in the story.
Narrative Transportation (Green & Brock, 2000)
While I was reading the narrative, I could easily picture the events in it taking place. While I was reading the narrative, activity going on around me was on my mind. (R) I could picture myself in the scene of the events described in the narrative.
I was mentally involved in the narrative while reading it. The narrative affected me emotionally.
I found my mind wandering while reading the narrative. (R)
While reading the narrative I had a vivid image of the main character. While reading the narrative I had a vivid image of the organization.
Attitude towards the Organization
After hearing about the organization, how would you describe your attitude towards them and their work?
(a) bad or good
(b) undesirable or desirable (c) unfavorable or favorable (d) not important or important (e) not effective or effective (f) worthless or valuable
Online Engagement
How likely would you be to … (a) like the organization’s content?
(b) comment on the organization’s content? (c) share the content with a friend?
(d) follow the organization’s account?
Offline Engagement
How likely would you be to …
(a) tell a friend about this organization? (b) donate to the organization?
(c) attend an event in support of the organization? (d) spend time volunteering for the organization?
Appendix C. Factor Loadings, Eigenvalues and Variance Explained
Measure Factor 1 Factor 2
Character Identification V1 V2 V3 V4 V5 .795 .773 .823 .909 .884 - - - - - Eigenvalue Variance Explained 3.51 70.27% - - Narrative Transportation V1 V2 V3 V4 V5 V6 V7 V8 .689 .260 .792 .760 .739 .346 .732 .717 -.132 .765 -.092 .157 -.038 .733 -.192 -.336 Eigenvalue Variance Explained 3.47 43.11% 1.32 16.54% Attitude towards the
Organization V1 V2 V3 V4 V5 V6 .862 .883 .991 .897 .879 .915 - - - - - - Eigenvalue Variance Explained 4.77 79.46% - - Online Engagement Intentions
V1 V2 V3 V4 .850 .806 .822 .872 - - - - Eigenvalue Variance Explained 2.81 70.20% - - Offline Engagement Intentions
V1 V2 V3 V4 .811 .857 .917 .858 - - - - Eigenvalue Variance Explained 2.97 74.21% - -
Appendix D. Direct and Indirect Effects
Direct Effect Bootstrap Results for Indirect Effect
Effect SE t p LLCI 95% LLCI 95% Effect SE LLCI 95% ULCI 95% Narr. Perspective Narr. Perspective à Online Engagement -.308 .174 -1.766 .079 -.653 .036 .144 .178 -.209 .489 Narr. Perspective à Offline Engagement -.292 .200 -1.460 .146 -.686 .103 .135 .153 -.149 .459 Sit. Similarity Sit. Similarity à Online Engagement .216 .197 1.092 .277 -.175 .606 .055 .149 -.234 .353 Sit. Similarity à Offline Engagement -.058 .174 -.332 .740 -.401 .285 .067 .176 -.274 .424 Interaction Effect
Index of Moderated Mediation (Difference Between Conditional
Indirect Effects) NP * SS à Online Engagement -- -- -- -- -- -- -.005 .011 -.031 .014 NP * SS à Offline Engagement -- -- -- -- -- -- -.004 .008 -.023 .011 n = 156
