1.

The Effect of User Psychology on the Content of Social Media Posts: Originality and Transitions
Matter Lucia Lushi Chen*, Walid Magdy and Maria K. Wolters School of Informatics, University
of Edinburgh, Edinburgh, United Kingdom

This pilot study was designed to examine the following research questions: 1. Changes in Affect:
To what extent do changes in the affect of social media posts correlate with users’ personality
traits and mental well-being? 2. Originality: To what extent does the use of non-original material
in their posts correlate with users’ personality traits and mental well-being?

Although the psychology literature suggests a strong association between negative mood states
and neuroticism (Rusting and Larsen, 1995), we did not find this in our data. Our results are in
line with previous studies of verbal cues to personality traits in social media (Yarkoni, 2010;
Golbeck et al., 2011; Schwartz et al., 2013; Park et al., 2015). Golbeck et al. (2011) found that
social media users who were more likely to talk about anxiety were on the higher end of the
neuroticism scale. We speculate that self-presentation bias may influence how social media
users regulate their expression of negative emotions in their public posts. The only relevant
association we found was that social media users on the high end of neuroticism are more likely
to switch between posting positive and negative affective content. This finding aligns well with
the fact that high neuroticism is associated with high emotional instability (Costa and McCrae,
1992). The link between posting non-original content and elevated depression symptoms
appears to be moderated by neuroticism. This suggests that high levels of neuroticism
predispose users both to depressive symptoms and to an indirect disclosure of emotions
through quotes and lyrics. In our sample, the prevalence of depressive symptoms is higher than
would be expected in the general population. In the original CES-D paper, Radloff (1977)
proposed three levels of depression severity: low (0–15), mild-to-moderate (16–22), and high
(23–60). They found that only 21% of the general population scored above the low symptom
level. In contrast, in our sample, nearly half of the participants exhibit a high level of symptoms
(>22). Within the context of social media studies of depression, however, our data set is not
exceptional. For many studies in the area, high symptom individuals account for nearly half of
the data set (De Choudhury et al., 2013; Tsugawa et al., 2015; Nadeem, 2016; Reece et al., 2017;
Orabi et al., 2018). Our results support the claim that affect expressed in social media data text
is associated with social media users’ affect patterns in real life. However, the data set used in
this study is from the early 2010’s and only covers the well-established social media platform
Facebook. The associations found in this study are likely to be slightly different from those found
in another social networks (e.g., Instagram) or in a new data set collected 10 years later.

Corpus Annotation 3.3.1. Social Media Affect For the purpose of this study, we refer to the affect
shown in social media posts as social media affect. In this study, following (Mohammad, 2016),
we operationalize valence as the post-author’s attitude toward a primary target of opinion. We
refer to the “dialectic” affective state as mixed valence. If there is no clear trend toward positive
or negative affect, the associated valence is neutral. 3.3.2. Originality We define posts that
consist of quotes from sources, such as song lyrics, books, or movies as non-original content; all
other content was defined as original. Since non-original content might not directly reflect the
 user’s moods or emotions, annotators were instructed to annotate such posts according to the
likely emotions of the author.

In future work, we plan to enrich our data set with more in-depth analyses of original vs. non-
original content, extend coverage by including a larger sample of the myPersonality data set,
and construct statistical models that allow us to observe long-term trends in posting patterns.
Future studies should also examine the extent to which affect expressed in nonoriginal content
is aligned with the users’ affect when they post the material.

2. FACTORS AFFECTING CONSUMER ATTITUDES AND INTENTIONS TOWARD USER-GENERATED

PRODUCT CONTENT ON YOUTUBE Imran Anwar MIR Iqra University 5 Khayaban-e-Johar, H-9
Islamabad, Pakistan e-mail: Mr.Imran.Mir@gmail.com Kashif Ur REHMAN Iqra University 5
Khayaban-e-Johar, H-9 Islamabad, Pakistan e-mail: drkashif@iqraisb.edu.pk Management &
Marketing Challenges for the Knowledge Society (2013) Vol. 8, No. 4, pp. 637-654

The current study aims to examine the influence of quantity of posts, views, and reviews on the
credibility and usefulness of product related content generated by users on YouTube. It also
examines the influence of perceived credibility and usefulness on consumer attitudes toward
product related content generated by users on YouTube (ATU) and their behavioural intentions
(BI). This research is founded on Simonsen’s (2011) methodological suggestion to analyse the
utility of YouTube as a communication channel and browsing system from the users’ side
because most of the content on social media sites (e.g. on YouTube) is created by users and
most of this content is used by end users.

Data was collected from a sample of 231 university students at Islamabad (the capital of
Pakistan). Samples were drawn using convenience sampling procedure and an offline self-
administrative questionnaire was distributed to each respondent. To give a brief description of
the resulting sample we can state that 61% of the respondents were male and 39% female, 7.4%
were under 20 years of age, 63.8% had between 20 and 29 years, 24.8%, were between the age
of 30 and 39, and 4% were 39 or above. 54.3% students were enrolled in undergraduate degree
programs, while 45.7% were enrolled in graduate degree programs. To measure the constructs,
we adapted several items from previous studies. To measure QPVR, two items were adapted
from Bailey (2005) and two items from Jones et al. (1986). To measure the perceived credibility
(PC) nine items were adapted from Chi (2011). To measure perceived usefulness (PU) three
items were adapted from Chi (2011) and three items from Patwardhan and Ramaprasad (2005).
To measure consumer attitudes toward product related UGC on YouTube (ATU) two items were
adapted from Liu et al. (2009) and four items were adapted from Lai and Chang (2011). To
measure the BI, two items were adapted from Liu et al. (2009). Responses were measured on a
seven-point Likert scale ranging from one (“Strongly Disagree”) to seven (“Strongly Agree”).
Principal component analysis (PCA) with Varimax rotation was conducted to screen the items
and check their validity. PCA was run on 4-items of QPVR, 9-items of PC, 6-items of PU and 6-
items of ATU. On the first iteration, PCA extracted only one factor of QPVR validating all the 4-
items. The PCA values of QPVR are eigen value = 2.120, percentage of variance = 53.009, KMO=
0.734, Bartlett’s test of sphericity = 0.000 (p < 0.05) and Cronbach’s alpha = 0.704. Table 1 shows
the factor loadings of QPVR. At the first iteration, PCA produced two components of PC with
some low loading items. The third iteration, after deleting 4 invalid items, produced one
component. This component consisted of 5 items. The PCA values of PC are eigen value = 2.515,
percentage of variance = 50.296, KMO = 0.789, Bartlett’s test of sphericity = 0.000 (p < 0.05) and
Cronbach’s alpha = 0.748. Table 1 shows the factor loadings of PC. PCA extracted only one factor
of PU. PCA validated all the 6-items of PU. The PCA values of PU are eigen value = 3.065,
percentage of variance = 51.075, KMO = 0.762, Bartlett’s test of sphericity = 0.000 (p < 0.05) and
Cronbach’s alpha = 0.806. Table 1 shows the factor loadings of PU. PCA extracted only one
component of ATU. However, only 4-items were chose and 2-items were deleted as their initial
extraction values were very low. The PCA values of ATU are eigen value = 2.184, percentage of
variance = 54.599, KMO = 0.703, Bartlett’s test of sphericity = 0.000 (p < 0.05) and Cronbach’s
alpha = 0.742. Table 1 shows the factor loadings of ATU. Due to the least number of items (i.e. 2)
only reliability of BI was examined. Guttman Split-Half Coefficient was used to test the reliability
of BI. Guttman Split-Half Coefficient of BI was 0.700. A confirmatory factor analysis (CFA) was
performed to assess the goodness of fit of the measurement models of QPVR, PC, PU and ATU.
Amos version 18 was used for the structural equation modelling (SEM) analysis. Traditional cut-
off criteria of model fit (see Bentler, 1983:1990; Browne and Cudeck, 1993; Marsh and Grayson,
1995; McDonald and Ho, 2002; Schumacker and Lomax, 1996) was used to assess the goodness
of fit of measurement models of QPVR, PC, PU and ATU. Measurement models of QPVR, PC, PU
and ATU showed a good fit to data (see Table 2). Minimum standardized path coefficients
should be 0.20 and above 0.30 is considered ideal to accept the relationship between the
variables (Chin, 1998). Figure 2 shows the CFA item loadings of the constructs of this study.

the current study found that the quantity of posts, views, and reviews has a positive effect on
the perceived credibility and usefulness of the product content which users generate on
YouTube. The current study found that perceived credibility positively influences the user
attitudes toward product content which other users generate on YouTube. The message sources
that consumers perceive more credible have strongly positive effect on their attitudes toward
the message (Friedman and Friedman, 1979; Ohanian, 1990). Consumers perceive UGC sources
(e.g. bloggers, video up-loaders etc.) as more credible than social media advertisers and possess
positive attitudes toward UGC (Jonas, 2010). The current study also found that perceived
credibility has a positive effect on the perceived usefulness of the product content which other
users generate on YouTube. Some past studies conducted in web context support this finding.
For example, McKnight and Kacmar (2007) found that information credibility influences the
perceived usefulness of the information on the web positively. Consumers perceive online
comments and product reviews credible as well as useful aids in purchase decisions (Cheung et
al., 2008). The current study found that consumer attitudes toward the product related UGC on
YouTube and their intentions of using it in purchase decisions are associated positively.
Consumers’ positive attitude toward the UGC leads to the consumption of UGC (Daugherty et
al., 2008)
The findings of the current study have some important implications for those businesses which
advertise their products and services using social media (e.g. YouTube). These findings suggest
that advertisers should sponsor social media users to promote their products. Users who
generate product content or information on social media are viewed as opinion leaders by other
 users (Cheong and Morrison, 2008). These findings also suggest that social media advertisers
should embed their advertising messages in user generated YouTube videos with the permission
of the video uploader. This will expose more users to the advertiser’s message. This is justified
by the fact that young consumers perceive UGC publishers (e.g. bloggers, video up-loaders) as
credible and like to watch user generated videos on YouTube (Jonas, 2010). Furthermore,
findings of the current study imply that advertisers should use real product users in their social
media ads instead of celebrities. Consumer endorsements enhance perceived credibility of the
endorsed product. It enhances the audience’s attitudes toward the endorsed product (Wang,
2005).
QPVR:QAUNTITY OF POSTS,VIEWS,REVIEWS
PC:PERCIEVED CREDIBILITY
ATTITUDE TOWARDS YOUTUBE
PU: PERCIEVED USEFULNESS
BI: BEHAVIOURAL INTENTIONS

Future studies should treat quantity of posts, views and review as separate variables. Future
studies should also examine the impact of advertising messages embedded in user generated
YouTube content on consumer attitudes and behavioural intentions.

3. Social Media Brand Community and Consumer Behavior: Quantifying the Relative Impact of
User- and Marketer-Generated Content Khim-Yong Goh, Cheng-Suang Heng School of
Computing, National University of Singapore, Singapore 117417, Republic of Singapore
{gohky@comp.nus.edu.sg, hengcs@comp.nus.edu.sg} Zhijie Lin

4. User Perceptions of Social Media: A Comparative Study of Perceived Characteristics and User
Profiles by Social Media Sylvia M. Chan-Olmsted, University of Florida, USA Moonhee Cho,
University of South Florida, USA Sangwon Lee, Kyunghee University, Korea

social media sites , ugc , Promotional Self-presentation, Brand Centrality, Marketer-directed

Communication, Response to Online Marketer Action, Factually Informative about the Brand, Brand
Sentiment

H1. Brand-related UGC on YouTube is more likely than brand-related UGC on Twitter or Facebook to
feature consumer self-promotion.
H2. Brands are most likely to be central in brand-related UGC on Twitter and least likely to be central in
brand-related UGC on YouTube.

H3. Marketer-directed brand-related UGC is less likely on YouTube than on Twitter or Facebook.

H4. Brand-related UGC is least likely to be posted in response to an online marketer action on YouTube.

H5. Brand-related UGC on Twitter, Facebook or YouTube is equally likely to feature brand-related factual
information.

H6. Sentiment towards brands in brand-related UGC on Twitter, Facebook, and YouTube is similar across
all three social media sites

Openness

Participation

Connectedness

Conversationality

Commonality

Brand relations is also one of the major variable that affect ATUGC. It is classified into three different
categories that are 1.) Promotional Self-Presentation , 2.) Brand Centrality , 3.) Marketer-Directed
Communication.

1.) Promotional Self-Presentation

Self-presentation is a performance (Goffman 1959): an “effort to express a specific image and
identity to others” (Zywica and Danowski 2008, p. 6). Consumers commonly use possessions,
brands, and other symbols to construct their images in both offline and online contexts (Belk
1988; Schau and Gilly 2003). Self Presentation at some attempts can be considered as Self
Promotion.Research suggests that people engage in self-presentation practices on all three
social media sites (e.g., Lange 2008; Marwick and Boyd 2011; Zywica and Danowski 2008).
Explicit self-promotion, as noted by some (e.g., Marwick and Boyd 2011), may create tensions,
depending on the culture of the site; in fact, being labelled self-promotional by others could be
the result of a failed self-presentation strategy. While the mass media sometimes depicts
Twitter as being a site that hosts posts on the mundane details of one's daily activities, blatant
self-promotion on the site is often considered inappropriate (Marwick and Boyd 2011). And
while the site offers more easy options for self-presentation through UGC than does Twitter,
research suggests that one's personal profile (e.g., ‘what I like’) and “display of friends”
(Papacharissi 2009, p. 210) are the chief means by which users self-present. Facebook users
may, of course, use brands (e.g., through listing one's schools, favourite sports teams, etc.) in
 their self-presentation; however, the culture of user self-promotion through brand-referencing
communications appears to be less highly developed on this social networking site. While
YouTube users still need to negotiate a balance between self-promotion and connecting with
others, the site is far more geared toward self-presentation than are Twitter or Facebook.

2.) Brand Centrality

Brand centrality refers to the role of the brand in brand-related UGC. recent empirical studies
suggest that brand centrality may vary across content (c.f. Jansen et al. 2009). With Twitter's
140- character limit, plus hyperlinks, it is difficult to introduce multiple topics in a single tweet.
While the site hosts content that varies in its brand centrality (Jansen et al. 2009), its technical
design and stronger cultural focus on sharing information and facilitating discussion – including
with marketers – comparatively privilege brand centrality. Specifically, compared with Facebook,
which is more oriented toward facilitating social connectedness, Twitter is posited to host a
greater percentage of brand-central posts. YouTube features several video types, such as
reviews, demonstrations, and ‘unboxing’ videos, which could, in principle, feature the brand
more centrally; however, research suggests that consumers find it hard to recall seeing product-
related information on the site, indicating that brands may often play a more peripheral role
(Cheong and Morrison 2008). As the dominant culture of the site is one that highlights the self, it
is hypothesized that brands are more likely to be featured peripherally – as identity-supporting
props (e.g., Schau and Gilly 2003) – on YouTube than on Twitter or Facebook. These Social media
sites not only help the brand to grow but also gives an instant impact on people who surf on
these sites.

3.) Marketer-Directed Communication

Many social media sites allow consumers to communicate with marketers (c.f. Deighton and
Kornfeld 2009). In brandrelated UGC, consumers may pose questions or complaints to
marketers, as well as respond to companies’ questions or comments. Twitter has hosted a
number of high-profile consumermarketer exchanges that have attracted media attention (c.f.
Deighton and Kornfeld 2010). While Twitter is recognized as being a site on which users ask for
information and complain (Naaman, Boase, and Lai 2010) – two reasons consumers might direct
content towards marketers – Facebook allows consumers to pose similar content to marketers
on their brand pages. Facebook, as the world's most popular site in terms of time-spent
(Radwanick 2011), has attracted many marketers. It establishes a trust between the consumer
and the user.
Psychological Response Towards Social Media

One reason people post on social media, according to an article in the Journal of

Experimental Social Psychology, is because social media sharing can link to positive social
media feedback and self-esteem. More directly, the quest for likes or follows on social
media heavily influences why people post. Psychological Response consists of four types of
responses, 1.) Emotional Response, 2.) Cognitive Response, 3.) Immediate behavioral Response, 4.)
Latent Behavioral Response.

1.) Emotional Response

Sometimes surfing on different social media sites for a long time can affect emotions of a person
as a result they can either feel a sense of pleasure which indeed happens when a person goes
through a positive review or any positive news. But on the other hand these sites can also evoke
a feeling envy which simply means that the person may feel exhausted, depressed and also
moody which can ultimately affect the health and overall emotion of that person. Hence,
instead of a unified definition, a model that has garnered much interest, particularly by
introducing a classification scheme rather than a general definition of emotions, is
the Circumplex Model of Affect proposed by Russell. This model, which has also been applied to
the social media context before, uses the two dimensions of valence (positive vs. negative) and
arousal (arousing vs. relaxing) to classify affective states. The basic premise of this model is that
emotions are the outcome of an appraisal process (i.e., a stimulus is processed, which then leads
to an observable affective response). In this paper, we focus on the observable part of
emotional processing, though we retain the general classification using valence and arousal.
Emotions have already been recognized to have an essential influence on human behavior in the
marketing context. For example, Bagozzi et al. stressed emotions as markers, mediators and
moderators of human responses when conducting a literature review. As such, the emotional
context of an ad (e.g., happy or sad TV program) effects the evaluation of these ads and their
recall. 

2.) Cognitive Response

The importance of information quality has received considerable attention over the last decade.
While most of the prior studies focused on defining the dimensions of information quality (Arazy
et al., 2017; Pow and Li, 2015), there is a large body of literature that attempted to develop
various frameworks to measure information quality (Lillrank, 2003; Liu, 2004; Rieh and
Danielson, 2007; Wang and Strong, 1996). In response to the development of the Internet,
additional concerns about online information quality and user-generated content quality are
growing (Eysenbach et al., 2002) due to the lack of standards for evaluating online information
quality (Mai, 2013; Savolainen, 2011). Different dimensions of information quality encounter
varying challenges in terms of specific context (Arazy et al., 2017); therefore, the term
information quality should be inherently situational, which is more complex with respect to the
research. In recent years, studies in various fields have emphasized the user perspective and
 perceived information quality. A study showed that a product quality not only is related to the
‘best’ in every dimension but also meets all the expectations for users (English, 1999). Similarly,
Kahn et al. (2002) further explained that the quality should meet and may even exceed users’
expectations. Many scholars have Zhang and Yuan 3 explored the relationship between the
perceived information quality and other online activities. For example, Wu et al. (2010) found
that perceived information quality positively influenced perceived usefulness; in other research,
Kim et al. (2008) revealed that online perceived information quality significantly affected users’
trust. The process of information quality perception is complex, which includes the fluid
interplay of attributes, purposes, values and prior knowledge. In a review study, Ghasemaghaei
and Hassanein (2016) revealed that the extant online information quality literature largely
focused on the consequences of the information quality evaluation; however, few studies
focused on the antecedents of information quality. This research gap motivated us to explore
the framework that can explain why and how users perceive information quality. In our research
context, perceived information quality is the cognitive judgment made by users on the value of
information, including not only the intrinsic characteristics of the information content, but also
the satisfaction of the information requirements or expectations of the user. The following
section will be combined with our mixed-method analysis of the users’ cognition process.

3.) Immediate behavioral Response

This has two factors 1) Information Pass-along and 2) Impulse buying.

1) Information pass-along
The unique features of Facebook facilitate users sharing information with others by clicking
“like” on posts made by other users, or “share” by posting information on a friend's wall
(Chen et al., 2013). Passing along brand-related content on Facebook is eWOM activity, as
the content conveys opinions, facts, or user experiences with brands or products. Examining
the emotional and the cognitive aspects of WOM, Ladhari (2007) found that pleasure and
arousal (O) had significant influences on WOM intention (R). Kim and Niehm (2009) found
perceived information quality of a website (O) positively influenced recommendation
intention (R). Ha and Im (2012) examined the influence of website design on WOM
activities.
2) Impulse Buying
Impulse buying refers to making a purchase based on a sudden urge to buy something
immediately (Adelaar, Chang, Lancendorfer, Lee, & Morimoto, 2003). Impulse buying is
different from purchase intention because decision-making time is noticeably short and the
purchase is unreflective (Weun, Jones, & Beatty, 1998). Conceptualizing impulsive buying
behavior, Stern (1962) identified suggestion impulse buying as one of four impulse buying
categories (i.e., pure, reminder, suggestion, planned). According to Stern (1962), suggestion
impulse buying occurs when a consumer sees a product for the first time and visualizes a
need for it. Applying this category of impulse buying to an experience of encountering
brand-related UGC, a consumer exposed to such content may also see the product featured,
imagine what it would be like to use the product, and consequently want to purchase it. Due
to the development of e-commerce, consumers can easily act upon their impulses and
 immediately purchase featured products. In fact, features on Facebook fan pages allow
consumers to make instant purchases by providing links to online stores. Thus, UGC could
motivate impulse purchases.

4.) Latent Behavioral Response

This also has two factors 1) Future Purchase Intention and 2) Brand Engagement.

1) Future Purchase Intention

In general, purchase intention refers to a future plan to buy a particular product or service
(Adelaar et al., 2003). Purchase intention represents an intention to act favorably (i.e.,
approach behavior) in response to informational stimuli related to brands or products.
Regarding relationships between emotional response and purchase intention, researchers
have demonstrated that intention to purchase follows positive internal states. Examining
the influence of retail store environment on consumer response, Baker et al. (1992) found
that participants’ willingness to purchase was enhanced as pleasure and arousal increased.
Although previous researchers did not examine the influence of eWOM on purchase
intention via cognitive response, findings on the consequences of eWOM have shown that
eWOM influences purchase intention (Christodoulides et al., 2012; Yu & Natalia, 2013).

2) Brand Engagement
According to Kim and Ko (2012), brand-related content on Facebook influences relationship
equity, that is, the tendency of consumers to stay in a relationship with a brand. Brand
engagement is a key component in building relationships between brands and customers
(Keller, 2001). Brand engagement describes the emotional tie that connects customer to
brands (Goldsmith, 2012). A brand engaged customer shows willingness to be involved with
a brand and gather information about the brand, talk about it, and exhibit its use to others
(Keller, 2001). Brand engagement (R) is a consequence of emotional and cognitive states (O)
evoked by the brand (Allen, Fournier, & Miller, 2008; Goldsmith, 2012) and that could be
evoked by brand-related UGC.
NAMES OF THE VARIABLES AND FACTORS IN THE TABLE

1. ATUGC : Attitude towards User Generated Content

2. NORMATIVE
3. INFORMATIVE
4. SOCIAL FACTORS
5. BRAND RELATION
6. STS : Social Media Interaction Ties
7. SI : Social Media Influence
8. SC : Social Media Commitment
9. PSP : Product Self Presentation
10. MDC : Marketer Directed Communication
11. BC : Brand Centrality
12. P&I : Platforms And Intermediaries
13. PC : Perceived Credibility
14. PR : Perceived Risks
15. PU : Perceived Usefulness
16. PRSM : Psychological Response towards Social Media
17. ER : Emotional Response
18. CR : Cognitive Response
19. IBR : Immediate Behavioral Response
20. LBR : Latent Behavioral Response
21. INFORMATION PASS ALONG
22. IB : Impulse Buying
23. FPI : Future Purchase Intention
24. BE : Brand Engagement