Content uploaded by Rodoula H. Tsiotsou
Author content
All content in this area was uploaded by Rodoula H. Tsiotsou on Aug 23, 2016
Content may be subject to copyright.
1
Rodoula H. Tsiotsou (2016). Predicting high and low engagement in
social networking sites. Proceedings of the 9th AMA SERVSIG 2016 –
International Service Research Conference. June 17-19, 2016,
Maastricht, The Netherlands (p. 1-5)
PREDICTING HIGH AND LOW ENGAGEMENT
IN SOCIAL NETWORKING SITES
As more and more marketers incorporate social media as an integral
part of the promotional mix, rigorous investigation of the factors related to
consumers’ engagement in social networking sites (SNSs) is becoming
pivotal. Given the social and communal characteristics of SNSs such as
Facebook, Linkedin, Twitter, Flirck, You Tube and Google Plus, this study
examines how personal and social factors relate to consumer engagement
with SNSs groups. Specifically, a conceptual model that identifies trust, group
identification, social motives, professional motives and communication
motives as important factors related to high vs. low engagement behavior in
SNSs was developed and tested.
CONCEPTUAL FRAMEWORK AND PROPOSITIONS
Social networking sites attract a fast-growing number of consumers by
enabling them to visualize and articulate their social network and engage in
2
social interactions in a dynamic, interactive, multi-modal form over the Internet
(Boyd and Ellison 2007).
Online community research supports the link between identification
with the consumption group and consumers’ active engagement with it (Men
and Tsai, 2013). Community identification may lead to group-oriented
attitudes and behaviors expressed as participation in group discussions and
activities and increased engagement with the group (Zeng et al., 2009;
Tsiotsou, 2015). Thus, it is expected that consumers highly identified with the
SNS group will be also highly engaged with it.
Trust is considered as an intrinsic feature of any valuable social
relationship and constitutes an important construct in marketing, because it
affects consumers’ positive and favorable attitudes and results in commitment
(Ballester and Aleman, 2001). It has been reported that trust in related to
consumer engagement in brand communities (Brodie et al. 2013; Tsiotsou,
2016) and in social networking sites (Tsiotsou, 2016). Accordingly, it is
proposed here that consumers highly engaged in their favorite SNS group will
exhibit high trust in the SNSs.
Consumers use SNSs for various reasons such as to connect and
reconnect with friends and family members (Subrahamanyam, 2008) to gather
knowledge on people to whom they have an offline connection (Lampe et al.,
2006), to find people and stay up to date on their lives (Rosen, 2007), and to
maintain friendships (Dwyer, 2007). In addition to communication and social
motives, the literature supports professional motives for using SNSs. For
example, Schaefer (2008) has identified three major motives for using SNSs
for business contacts: Staying in contact, reactivation of contacts and most
3
importantly the management of one’s existing contact network. Moreover,
other consumers want to collect contact information on potential future
business partners (Thew, 2008), which is comparable to the collection of
business cards. Therefore, it is expected that consumers highly engaged with
the SNS group will also score high in social, communication and professional
motives.
METHOD
The survey research method has been used to collect data for the
study offline. The questionnaire consisted of three parts. Part I gathered data
on service and social relationships such as trust on SNSs and SNS group
identification and engagement. Part II gathered data on consumers’ motives
(social, communication and professional), and Part III collected demographic
data.
The items for measuring SNS group engagement (four items) and SNS
group identification (four items) were gathered from the work of Algesheimer
et al. (2005). Four items measuring SNS trust were taken from the work of
Chaudhuri and Holbrook (2001) and the motives for using the SNSs were
taken from the literature (Lampe et al., 2006; Rosen, 2007; Dwyer, 2007;
Schaefer 2008; Thew, 2008). All items in the final instrument were anchored
by Strongly Disagree (1) to Strongly Agree (5).
The data was gathered from a convenience sample of 320 members of
various SNSs. The sample consisted of 51% males and 49% females. From
the respondents, 56% were between 25-44 years, 34% were between 18-24
years and 11% were between 45-64 years old. In relation to the internet use,
4
30% of the sample used it for more than 21 hours per week, 19% used for 16-
20 hours, 19% used for 11-15 hours, 20% used for 6-10 hours and 12% used
for up to 5 hours per week. The majority of the sample (298) used Facebook
(93%), 84 respondents used Google + (26%), 72 used Linkedin (22.5%), and
58 used Twitter (18%). Most of the respondents (56%) had more than 201
friends on SNSs, 13% had 151-200 friends, 10% had 101-150 friends, 9%
had 51-100 friends and 13% had up to 50 friends.
To compare consumers with high and low SNS group engagement and
identify differences, classification with discriminant analysis was used. The
mean score in the SNS group engagement scale has been used in order to
identify the high and the low engagement groups. The independent variables
of the study were SNS group identification, trust to SNSs, social motives,
communication motives and professional motives. In a preliminary analysis of
the data, a case analysis was conducted to identify possible outliers and
violations of the assumptions of independence, multivariate normality and the
homogeneity of variance/covariance matrices. No serious violations of the
assumptions were identified.
The overall multivariate relationship (MANOVA) was statistically
significant at the .05 level (chi square = 115.188; Wilk’s Λ = 0.69; p < 0.00)
indicating that the two groups are statistically significantly different. Thus, the
discriminant function extracted was significant and overall the variables used
in the study were able to discriminate between high and low engagement
groups. The analysis continued by evaluating the contribution of each
independent variable to the discrimination of the two groups. All univariate F-
tests were significant (p < 0.00).
5
The next step of the analysis was the classification of the subjects and
the evaluation of the classification procedure. The classification was based on
the Bayesian probability of group membership, assuming group priors equal
to the relative group sizes. The analysis continued with the evaluation of the
performance of the classification procedure, which is set up to maximize the
number of correct classifications. The results for the sample indicated a
74.7% correct classification rate. The precision of correct classification is
satisfactory and for this reason the use of the procedure for classification of
future subjects is recommended.
DISCUSSION
In conclusion, the present study provided empirical evidence for the
applicability of engagement with SNS group as a classification criterion for
SNSs consumers. Specifically, it aimed to classify SNSs’ consumers
according to their engagement level, and further profile them according to
their identification to the SNS group, their trust to SNSs and their motives
(social, communication and professional). Furthermore, a psychographic
profile of low and highly engaged consumers was developed based on the
use of the above personal and social factors. These results enhance our
theoretical understanding of the use of psychographic segmentation in SNSs
consumers and have marketing implications.
SNS engagement is of particular interest to both academics and
managers since high SNS engagement is related with positive attitudinal and
behavioral outcomes (e.g., SNS loyalty and positive word of mouth). The
6
results showed that SNS group engagement can be used as a segmentation
criterion for identifying different SNS consumers. SNSs managers could
benefit from these findings to attract and maintain their consumer base. This
could be done by communicating to consumers the social, communication
and professional benefits they can gain by becoming members of the SNS
Some clear limitations of the study, should, however, be addressed.
First of all, it should be noted that a convenience sample has been used and
therefore its findings cannot be generalized. Furthermore, collection of data
from SNS consumers in other countries could give the chance for cross-
cultural comparisons.
REFERENCES
Ballester, D.E. and Alemán, M.J.L. (2001), “Brand trust in the context of
consumer loyalty,” European Journal of Marketing, Vol. 35 No. 11/12,
pp. 1238-1258.
Boyd, d. m. & Ellison, N. B.(2008). Social network sites: definition, history, and
scholarship. Journal of Computer-Mediated Communication, 13, 210-
230.
Brodie, R.J., Llic, A., Juric, B., and Hollebeek, L. (2013), “Consumer
engagement in a virtual brand community: An exploratory analysis,”
Journal of Business Research, Vol. 66 No. 1, pp. 104-114.
Lampe, C., Ellison, N. and Steinfield, C. (2006) A face(book) in the crowd:
social searching vs. social browsing, Computer Supported Cooperative
Work ACM, Canada, pp. 167-170.
7
Rosen, C. (2007) Virtual Friendship and the New Narcissism, The New
Atlantis, 17.
Schaefer, C. (2008) Motivations and usage Patterns on Social Network Sites,
ECIS, Galway, Ireland.
Subrahmanyam, K., Reich, S.M., Waechter, N. and Espinoza, G. (2008).
Online and offline social networks: Use of social networking sites by
emerging adults. Journal of Applied Developmental Psychology, 29(6),
420-433.
Thew, D. (2008) LinkedIn--a user's perspective: Using new channels for
effective business networking, Business Information Review, 25 (2),
pp. 87.
Tsiotsou, R. H. (2015). The Role of Social and Parasocial Relationships on
Social Networking Sites Loyalty. Computers in Human Behavior, Vol.
48, p.401-414
Zeng, F., Huang, L., and Dou, W. (2009), “Social factors in user perceptions
and responses to advertising in online social networking communities,”
Journal of Interactive Advertising, Vol. 10, pp. 1-13.
