Influence of social networks on the purchase decisions of university students

Abstract

The emergence of social networks has not just had a great impact in the way companies promote their products and services, but also in the decision-making process of consumers regarding their purchases. Using the application and extension of the models proposed by Okazaki et al. (2012), the present study tries to understand the factors that motivate the use of social networks in the purchase decisions of young university students, for this a self-administered questionnaires were applied to 224 university students. Some limitations were found at the time of evaluating the adjustment of the model through a structural equations method, due to the number of indicators per construct. However, the results show that the proposed model presents a good adjustment. Therefore, it can be concluded that the study develops an appropriate approach to the knowledge of the factors that can influence university students, who intend to use social networks to buy, validating the conclusions drawn by Okazaki et al. (2012) in their study. At last, it is recommended for companies that wish to promote their products through social networks to look out for strategies that combine information transparency and stimulation of word-of-mouth communication among users, generating a larger impact on the purchase decisions of clients.

Full text

ISSN: 1131 - 6837 Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 61
Inuence of social networks on the purchase
decisions of university students
Inuencia de las redes sociales sobre las decisiones
de compra de estudiantes universitarios
Lore na este fani a Gu ti érr ez fLó rez 1
Mar ia isabe L Corr ea esCoba r1
andr és Hen ao restr ep o1
dia na ara nGo -bot ero2
aLeja ndro VaLenC ia-aria s2
Universidad Nacional de Colombia (Colombia)
Instituto Tecnológico Metropolitano (Colombia)
Recibido el 5 de septiembre de 2015, aceptado el 20 de marzo de 2017
Nº de clasicación JEL: M31, L81
DOI: 10.5295/cdg.150577lj
Abstract:
The emergence of social networks has not just had a great impact in the way companies promote their
products and services, but also in the decision-making process of consumers regarding their purchases.
Using the application and extension of the models proposed by Okazaki et al. (2012), the present study
tries to understand the factors that motivate the use of social networks in the purchase decisions of young
university students, for this a self-administered questionnaires were applied to 224 university students.
Some limitations were found at the time of evaluating the adjustment of the model through a structural equa-
tions method, due to the number of indicators per construct. However, the results show that the proposed model
presents a good adjustment. Therefore, it can be concluded that the study develops an appropriate approach to
the knowledge of the factors that can inuence university students, who intend to use social networks to buy,
validating the conclusions drawn by Okazaki et al. (2012) in their study. At last, it is recommended for companies
that wish to promote their products through social networks to look out for strategies that combine information
transparency and stimulation of word-of-mouth communication among users, generating a larger impact on the
purchase decisions of clients.
Keywords:
Trust, purchase decisions, information, marketing, social networks.
1 Department of Organization Engineering. Carrera 80 No 65-223. Postal code: 050034.(Colombia). legutierrezf@
unal.edu.co; maicorreaes@unal.edu.co; anhenaore@unal.edu.co
2 Departament of Management Sciences. Carrera 31 No 54-10. Postal code: 050013 (Colombia). dianaarangob@
itm.edu.co; jhoanyvalencia@itm.edu.co

Inuence of social networks on the purchase decisions of university students
Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 ISSN: 1131 - 6837
62
Resumen:
El auge de las redes sociales no solo ha tenido un alto impacto en la manera como las empresas promocionan
sus productos y servicios, sino también en la toma de decisiones de los consumidores en cuanto a sus compras.
Partiendo de la aplicación y extensión de uno de los modelos propuestos por Okazaki et al. (2012), este estudio
busca entender los factores que motivan el uso de las redes sociales en las decisiones de compra de los jóvenes
universitarios, para lo cual, se llevó a cabo la aplicación de cuestionarios auto-administrados a 224 estudiantes
universitarios. Aunque a la hora evaluar el ajuste del modelo por medio de la metodología de ecuaciones estruc-
turales se encontraron algunas limitaciones por el número de indicadores por constructo, los resultados señalan
que el modelo propuesto presenta un buen ajuste, con lo cual se concluye que este estudio permite obtener una
aproximación del conocimiento de esos factores que pueden inuenciar la intención de uso de las redes sociales
para comprar en los jóvenes universitarios; validando las conclusiones obtenidas por Okazaki, et al. (2012) en
su trabajo. Por último, se recomienda a las empresas que deseen promocionar sus productos a través de las redes
sociales, el buscar estrategias que incorporen transparencia en la información y estimulen el “voz-voz” entre los
usuarios, generando así un mayor impacto en las decisiones de compra de los clientes.
Palabras clave:
Conanza, decisión de compra, información, mercadeo, redes sociales.

L.E. Gutiérrez Flórez / M.I. Correa Escobar / A. Henao Restrepo / D. Arango-Botero / A. Valencia-Arias
ISSN: 1131 - 6837 Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 63
1. INTRODUCTION
The growth of the Web 2.0 along with the advance and popularity of online social
networks, have had a high impact in the way companies promote and commercialize their
products and services (Bolotaeva and Cata 2010). Online communication has turned into a
dominating channel that inuences the purchase decisions of consumers, where social net-
works become a strong source of information to clients (Cheung et al. 2009, cited in Akar
and Topçu 2011). The present study focuses on analyzing the different possibilities that
companies have through the use of social networks to generate new marketing strategies,
and examine the inuence that these strategies have on the purchase intentions of university
students.
Initially, a literature review is made on the emergence of social networks, its inuence
in the market and other determining factors of the social networks use in purchase deci-
sions. Afterwards, a new model is proposed and tries to identify those factors that inu-
ence directly and indirectly the intensions of the social networks use to make a purchase,
applying the model proposed by Okazaki et al. (2012) with their main hypothesis and later
extending it by adding a new cause-effect relation based on exposed by the Theory of Rea-
soned Action (TRA). The method of structural equations is used to evaluate the adjustment
of the model. Finally, conclusions and recommendations are made for companies that wish
to use social networks for increasing their competitiveness in the market.
2. LITERATURE REVIEW
2.1. Contextualization of Web 2.0, social media and the emergence of social networks
Information and Communication Technologies (ICTs) have affected various dimen-
sions of modern societies, modifying even economic, politic, social and cultural structures
(Bran et al. 2017). Internet has changed how people work, communicate and live, and has
become one of the most ideal mediums to generate interaction among users (Akar and
Topçu 2011; Hoyos and Valencia 2012), allowing a faster, a more synchronized and oppor-
tune exchange and availability of information (Hsu et al. 2007), generating an invaluable
management system tool for knowledge creation, communications and marketing research
(Valencia et al. 2013).
The term Web 2.0 is commonly used to dene technologies and applications developed
in the Internet. Also, along with the advance and popularity of online social networks, have
had a high impact in the way companies and consumers commercialize their products and
services (Bolotaeva and Cata 2010).
The emergence of social networks have attracted millions of users that can share pic-
tures, videos, opinions and experiences with other users since their rst launch in 1997
(Maurer and Wiegmann 2011) and given the possibility of acquiring information of prod-
ucts from companies (Mangold and Faulds 2009), a good example is proposed by Quinton
and Harridge (2010, cited in Soares et al. 2012, p.46) of the remarkable increase in the
participation of users in the Web 2.0 emerging content.

Inuence of social networks on the purchase decisions of university students
Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 ISSN: 1131 - 6837
64
In a similar way, Cheung et al. (2009, cited in Akar and Topçu 2011) suggest that online
communication has become a dominant channel that inuences the purchase decisions of
consumers, being a valuable source of information to clients. These networks open the
possibility to establish communities that provide the means to access and share publicity,
adds and novelties that brands promote, with the purpose of exploiting their advertising
capacities and positioning their brands in the target market (Alvarado 2012).
2.2. Importance and inuence of social networks in marketing
According to Pookulangara and Koesler (2011), social networks have become
one of the most important tools to identify and create marketing strategies, given that
their use has generated big possibilities and opportunities to learn about consumers’
needs, and consequently answer opportunely, proactively and creatively to clients.
Besides, each network helps to understand the client, plainly identify their biggest
unsatisfied needs, answer their demands, and give the group value (Cheung and Lee
2010).
Moreover, one of the main challenges of businesses nowadays is to identify the
target audience they must address, and also to determine the proper timing to present
their marketing strategy (Hill et al. 2006). A series of efficient tools arise from this
concern, the first tool is the Word-of-Mouth (WOM) technique. According to an
investigation made by Kalpaklioglu and Toros (2011), around 78% of consumers in
47 countries trust on the recommendations of other consumers. This is why sellers
have increasingly become more interested in new techniques like the Word-of-Mouth
(WOM) than in conventional advertising techniques.
Among other techniques, there are product samples, which are distributed by
companies through social networks to their potential consumers, using digital bo-
nuses for physical products and periods of time for services (Nooy et al. 2011, cited
in Schlereth et al. 2013). Nonetheless, the success of this strategy depends on an
adequate selection of consumers to guarantee a widespread and fast broadcast of the
information.
Taking the above into account, Kunz and Hackworth (2011) found that consumers
today have turned to look for information about new products, sales and discounts by
following brands on social networks. On one hand, companies have the opportunity
to create cheaper and friendlier campaigns for their products. On the other hand,
there are certain weaknesses about social networks and businesses, such as aggres-
sive advertising, invasion of privacy, and legal issues (Bolotaeva and Cata 2010).
Likewise, Ontario (2008, cited in Akar and Topçu 2011) exposes how companies
use social media to increase the corporation’s visibility, presenting their products
and services, in order to interact with members of the community, thus an exchange
of ideas and information is generated. Additionally, according to Tanuri (2010), com-
panies are starting to understand the use of social media as a component to develop
their market research, marketing strategy, the monitoring of their consumers’ feel-
ings, public relations, customer management, and also using it as an advertising
channel to create assertive campaigns to successfully reach their target audience.
(Tanuri 2010, cited in Akar and Topçu 2011).

L.E. Gutiérrez Flórez / M.I. Correa Escobar / A. Henao Restrepo / D. Arango-Botero / A. Valencia-Arias
ISSN: 1131 - 6837 Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 65
Complementing the above, Cocheo (2009, cited in Kunz and Hackworth 2011)
assures that consumers’ response is more favorable because they have control over
their opinions and decisions. The possibility to generate, edit, and share online in-
formation about companies, products and services, makes consumers more confident
towards other users’ opinions by considering them more objective and trustworthy
information than the messages generated by the companies. Therefore, opinions
among peers become an important influence in purchase behavior (Akar and Topçu
2011). Particularly, a study made by DEI Worldwide3 about the consumers’ points of
view toward social media (Akar and Topçu 2011), shows how consumers that inter-
act with social networks engage better purchase decisions than the ones that do not.
Furthermore, Miller and Lammas (2010), based on the same study, suggested that
70% of consumers visit social networks to get information and 49% of these clients
take their purchase decisions based on the information collected there.
All of the above can be summarized in a new trend called Shopping 2.0 or social
shopping, in which social networks try to become a place to consult and purchase,
where users can answer all the questions they have concerning the products they are
interested in, they can read and share comments, and of course, buy the products and
services offered through social networks (Alvarado 2012).
2.3. Determining factors of the use of social networks in purchase decisions
Brymer (2009, cited in Wang and Chang 2013) states that merchandising meth-
ods are changing remarkably nowadays business interaction and the most valuable
information today about the consumers’ decision-making is linked to the trust that
social media members generate, while experts, authorities, mass media and mass
advertising have consequently lost strength. In this way, the consumers’ participation
in social networks is motivated by different factors. Dholakia et al. (2004) propose
in their study the following five factors: i) intentional value, the perceived satisfac-
tion following the success of a predetermined purpose; ii) self-discovery, given that
interactions can help define each person’s preferences, likes, dislikes and values
(McKenna and Bargh 1999, cited in Dholakia et al. 2004); iii) interpersonal connectivity,
because of the social benefits given by establishing relationships with other people
(social support, friendship, intimacy) (Dholakia et al. 2004); iv) social recognition,
that is a strong motivation to certain consumers that consider it important to win the
acceptance and approval of others (Baumeister 1998, cited in Dholakia et al. 2004);
and v) value of entertainment, which comes from the fun and relaxation that arises
from sharing with other members of the Web (McKenna and Bargh 1999, cited in
Dholakia et al. 2004).
Another study shows that virtual communities, despite being flexible and based
upon a wide range of cultural interests and social affiliations (Bermudez et al. 2017),
compared with traditional organizations are fragile because they do not guarantee
what is expected (Hsu et al. 2007). The effects on the decision-making are based
3 DEI Worldwide is a leading social media marketing agency with specialties in conversational outreach, integrated
strategies, market research, viral content creation and application development.

Inuence of social networks on the purchase decisions of university students
Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 ISSN: 1131 - 6837
66
on the strength of the created bonds, homophily, source credibility, and also on the
impact over personalities of virtual community members (self-esteem, leadership,
online behavior) (Acar and Polonsky 2007). Additionally, Soares et al. (2012) in-
vestigate whether social interactions among social network users (trust and social
relationships) have a positive impact in a series of marketing interactions (infor-
mation broadcast, WOM, and attitude toward the advertisement). The present study
was made in 150 university students between 18 and 35 years old, concluding that
trust is an important human factor that stimulates the tendency to give and receive
information within a social network and that reduces uncertainties and simplifies
decision-making. Likewise, trust has a positive impact on WOM, the exchange of
information and its broadcast amongst consumers (Soares et al. 2012).
Furthermore, a study applied to bachelor students of eight different faculties of
the University of Afyon Kocatepe between 18 and 24 years old to determine the
influencing factors in the attitudes of consumers towards social media marketing,
presents an analysis of 7 factors: responsive attitudes toward merchandising through
mass media, use of mass media, knowledge of social media, level of affectation by
social media, participation in any social network, prospective over social networks,
and the fear of marketing related to social media. The study concluded that the de-
gree of information processing, comparisons, and purchase behavior is affected by
the previously acquired knowledge from the opinions of other consumers by the
digital experience. It was also concluded that experience in social media generates
a better attitude while interacting in the Web, and that consumer attitudes positively
affect purchase through the previous gathering of information and WOM. It can be
assumed from the above that people are looking for more products and services on-
line before making a purchase, being social networks a means that allow customers
to reach products, services and brands related to their own experiences (Akar and
Topçu 2011).
3. MODEL OF OKAZAKI ET AL. (2012)
In accordance with the literature review made above, the proposed work regard-
ing social networks have been oriented towards its influence in marketing strategies
and interactions related to advertising and information spreading. The model pro-
posed by Okazaki et al. (2012), which understand the basic mechanism of the social
influence that takes place in the usage of social networks, it will be used to identify
the factors that can affect the intention to use social networks to make a purchase.
The model proposed by Okazaki et al. (2012) derives from the brand communi-
ty model designed by Algesherimer et al. (2005), which is based on the literature
about brand communities (Mcalexander, et al. 2002; Muniz and O’guinn 2001), so-
cial identification (Bhattacharya and Sen 2003), and interactions among consumers
(Dholakia et al. 2004). The definition for each construct is structured in the Table 1.

L.E. Gutiérrez Flórez / M.I. Correa Escobar / A. Henao Restrepo / D. Arango-Botero / A. Valencia-Arias
ISSN: 1131 - 6837 Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 67
Table 1
Constructs of the model of Okazaki et al. (2012)
Construct Denition
Identication with the
social network (IWSN)
Refers to the strength of the relationship between the user and the
social network, as well as the convergence between psychological
attributes of the users and the website’s content.
Commitment with the
social network (CWSN)
It refers to the positive inuences of the personal identication with a
social network, which are dened as the intrinsic motivations of inte-
racting and cooperating with members of the social network.
Attachment degree
(AD)
It is the intensity of the interactive and personalized relationship bet-
ween an individual and a website.
Perceived usefulness
(PU)
It is the value obtained by achieving a predetermined instrumental pur-
pose and by the exchange of information through participation in the
virtual community. This construct combines the value of the informa-
tion and the instrumental value of the means of communication.
Social recognition (SR)
Obtained social benets, such as social support, friendship and
intimacy that arise by establishing and maintaining contact with other
people.
Entertainment (E) It is the motivation for online media use.
Social pressure (SP)
It is an important element of the attitude in the theoretical formula-
tions of behavior that can consist in public conformity or in the private
acceptance of rules. When public conformity is not accompanied by
a complete acceptance at a private level, the person suffers from peer
pressure, which plays a signicant inuence in his or her behavior.
Electronic Word-of-
Mouth (e-WOM)
Refers to the exchange of information and interaction among users
where debates about personal opinions on certain subjects occur.
Intention to search
information (ISI)
Refers to the behavior towards a brand and the search of information
about a brand.
Source: Okazaki et al. (2012).
The model is made up of three blocks. The rst block is based on the main constructs
of the model proposed by Algesherimer et al. (2005). The two remaining blocks constitute
precisely the extension suggested by Okazaki et al., where antecedents are derationed on
the topic of brand identication (degree of identication, entertainment in the Web, social
recognition and Web’s perceived usefulness) and the commitment with the network conse-
quences (Word-of-Mouth and behavior toward the brand) illustrated in Figure 1.

Inuence of social networks on the purchase decisions of university students
Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 ISSN: 1131 - 6837
68
Figure 1
Model by Okazaki et al. (2012)
Source: Okazaki et al. (2012).
The constructs and the relations among them compose the rst block. These constructs
are considered in the brand community model of Algesherimer et al. (2005), and are adapt-
ed by the model of Okazaki et al. (2012) with the purpose of identifying the way users react
to brand advertising through social networks, carried out by companies. The hypotheses
that constitute the rst block are:
H1a: Identication with the social network has a signicant positive effect on the commit-
ment of users with the social network
H1b: Identication with the social network has a signicant negative effect on social pressure.
H1c: The commitment to the social network has a signicant positive effect on social pressure.
The second block includes the antecedents of the identication with the social network
suggested by Okazaki et al. (2012) with the purpose of obtaining a different perspective
of social networks from electronic Word-of-Mouth (e-WOM). To achieve this purpose, a
study by Brown et al. (2007) that explores e-WOM communication on online communities
was considered. Three main variables are observed in this study: the attachment degree, the
identication and credibility of the source, and the inuence over the e-WOM exchange
of information. The rst two variables were considered by the model of Okazaki et al., in
which the attachment degree can be dened as “the intensity of the interactive and per-
sonalized relationship between and individual and the website”, and the identication as
“the convergence between psychological attributes of the users and the website’s content”
(Okazaki et al. 2012, p. 35). The hypotheses considered in the second block are:
H2a: Attachment degree has a signicant positive effect on the identication with the so-
cial network.
H2b: Entertainment has a signicant positive effect on the attachment degree.
H2c: Perceived usefulness has a signicant positive effect on the attachment degree.
H2d: Social recognition has a signicant positive effect on the attachment degree.
Figure 1
Model by Okazaki et al. (2012)
Source
: Okazaki et al. (2012).
Identification
with the social
network
(IWSN)
Commitment
with the social
network
(CWSN)
Social
pressure
(SP)
Electronic
Word-of-
Mouth (e-
WOM)
Intention to
search
information
(ISI)
Perceive
usefulness
(PU)
Entertainment
(E)
Social
recognition (SR)
Attachment
degree (AD)
Block 2: Proposal about
the background of the
brand community
Block 1: Brand
community model by
Algesheimer et al. (2005)
Block 3: Proposal about
consequences of Brand
community
H2b
H2c
H2d
H1b
H1c
H1a
H3a
H3b
H2a

L.E. Gutiérrez Flórez / M.I. Correa Escobar / A. Henao Restrepo / D. Arango-Botero / A. Valencia-Arias
ISSN: 1131 - 6837 Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 69
Okazaki et al. add two more variables to the third block: e-WOM and the intention
to search information, where the rst one affects the second given that communications
through e-WOM affects the intentional behavior toward the brand. Furthermore, commit-
ment with the social network inuences the e-WOM, because when the users are commit-
ted to a brand, it is more probable that they would transmit information they think interest-
ing to other users. Then, the following hypotheses are suggested:
H3a: Commitment with the social network has a signicant positive effect on the e-WOM.
H3b: The e-WOM has a signicant positive effect on the intention to search information.
4. RESEARCH METHODOLOGY
Given that the purpose of this study is to analyze the inuence of contents in social
networks in the purchase decisions of university students, one of the models contemplated
by Okazaki et al. (2012) was used as reference because it presents well-dened constructs
and relationships between them, aiming to comprehend the basic mechanism of social in-
uence that acts in social networks. Just as Okazaki acknowledges and extends the brand
community model by Algesheimer et al. (2005), contrasting the proposed relations in the
theoretical model and looking for validation of the inuence of social networks, the present
study seeks to adapt this model to the understanding the factors and variables that moti-
vate the use of social networks in the decision-making processes when a user is making a
purchase.
To reach this aim, a quantitative methodological design was carried out through self-ad-
ministered questionnaire, which were applied in physical format to 224 undergraduate stu-
dents from the Engineering Department of the Universidad Nacional de Colombia, Me-
dellín.
According to ComScore (2013), the highest percentage of Internet visitors in Colombia
corresponds to people between 15-24 years old, with a value of 42.9%, followed by 26.8%
for the age group between 25 and 34 years, 16.1% for those aged between 35-44 and 14.2%
for those over 45. All these people, according to the same source, consume their longest
connection time in social media with an average of 5.76 hours-person per month followed
by 3.33 hours for entertainment and 2.88 hours on average for services. This justies the
selection of the sample of university students, who were selected through non-probabilistic
sampling by criterion, to answer a total of 30 questions, with the intention of knowing their
perceptions in a Likert scale of 5 points, in the following areas: i) Advertising of a brand
through social networks (brand community model by Algesherimer et al. 2005, which in-
cludes social pressure, identication with the social network and commitment to the social
network); ii) Brand identication (entertainment, attachment degree, perceived usefulness
of the network and social recognition) and iii) Commitment to the social network (elec-
tronic Word-of-mouth and behavior towards the brand).
In order to validate the proposed model, its constructs and the relationships between
them, convergent and discriminant validations were made, in which some statistics such as
the Bartlett’s sphericity test and the KMO measure were used and analyzed, correspond-
ing to the evaluation of the Cronbach’s alpha, to verify the reliability of the instrument’s

Inuence of social networks on the purchase decisions of university students
Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 ISSN: 1131 - 6837
70
internal consistency, which assumes that different items measure the same construct and
are highly correlated. Finally, the methodology of structural equations was used to evaluate
the adjustment of the model to the data, since it is a reliable methodology for this type of
intention models (Valencia et al. 2016).
5. PROPOSED MODEL FOR THE INTENTION TO USE SOCIAL NETWORKS
TO MAKE A PURCHASE
The model proposed in this article, as already discussed above, is based on the model
presented by Okazaki et al., considering the same three blocks and most of the relations
included in the model.
According to Dholakia et al. (2004), the participation of consumers in social networks
is motivated by several factors, including the intentional value, recognition, and value of
entertainment. Hence, the same three hypotheses are taken from the second block that
Okazaki contemplated for the model proposed in the present article:
H1: Entertainment has a signicant positive effect on the attachment degree.
H2: Perceived usefulness has a signicant positive effect on the attachment degree.
H3: Social recognition has a signicant positive effect on the attachment degree.
Communication and marketing have entered a new era characterized by the interactiv-
ity provided by modern technologies. The protagonism acquired by the user, is framed in
a situation of market liberalization, increased competition and the power of the Internet
in communications. In this way, consumers have become an active part of commercial
communication, selecting or ignoring the messages that interest them, and interacting or
creating new content thanks to the possibilities offered by the Internet (Arcos et al. 2014).
Likewise, the growing development of Web 2.0 technologies and social networks has gen-
erated a wide range of possibilities for both companies and consumers. These possibilities
increase due to the great reception they have had from online users. In addition, companies
seek to create better experiences for consumers in order to increase sales, create customer
loyalty and a stronger brand positioning (Gatautis 2008, Vitkauskaite 2011, cited in Gatau-
tis and Kazakevičiūtė 2012; Cartagena et al. 2017).
This interactivity of online consumers was contemplated in the rst block of the model
proposed by Okazaki et al., when considering the constructs of the brand community mod-
el of Algesherimer et al. (2005). As the objective of the present article is to try to identify
the factors that inuence the intention to use social networks to make a purchase, the pro-
posed model will adopt the same three hypotheses:
H4: Identication with the social network has a signicant negative effect on social pressure.
H5: Identication with the social network has a signicant positive effect on the commitment
with the social network.
H6: Commitment with the social network has a signicant positive effect on social pressure.
In addition to the hypotheses proposed by Okazaki et al. (2012) for this rst block, a
new hypothesis will be proposed, which is extracted from the Theory of Reasoned Action
(TRA), in which according to Fishbein and Ajzen (1975) and Ajzen and Fishbein (1980),

L.E. Gutiérrez Flórez / M.I. Correa Escobar / A. Henao Restrepo / D. Arango-Botero / A. Valencia-Arias
ISSN: 1131 - 6837 Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 71
one of the factors that lead an individual to perform certain behavior is the subjective norm,
which is based on normative beliefs or social factor dened by specic social contexts
(Valencia et al. 2014). Other authors such as Davis et al. (1989) argue that the subjective
norm is determined by normative beliefs or individual perceptions regarding the inuence
of external groups composed of individuals that are important to them, and by individual
motivations to follow a particular social norm.
The inuence of the external groups of which the TRA speaks about, in the context of
the intention to use social networks to make a purchase, can be seen as a kind of social
pressure to carry out such behavior, giving way to the next hypothesis, which is not consid-
ered in the model proposed by Okazaki:
H7: Social pressure has a signicant positive effect on the intention to use a social network
to make a purchase.
Smith et al. (2007) and Akar and Topçu (2011) claim that consumers rely on the rec-
ommendations of other consumers, and that is why sellers increasingly use tools such as
WOM. These same authors assert that the information of close consumers is perceived as
more reliable, impartial, and generates more credibility than the persuasive information
obtained in the publicity controlled by the companies.
Given the importance of the WOM and the information received by the consumers from
the people close to them, the third block will establish the same two hypotheses contem-
plated in the model of Okazaki et al., considering “intention to use social networks to make
a purchase” (variable) instead of “intention to search information”.
H8: Commitment to the social network has a signicant positive effect on electronic word-
of-mouth communications.
H9: The e-WOM has a signicant positive effect on the intention to use a social network
to make a purchase.
Communication over the Internet has ceased to be unidirectional to become two-way
communication. That is, the information is transmitted from the Web to the user and from the
user to the Web. Among the tools that allow such interaction are: wikis, podcasts, tags, social
networks, videos, among others. All of the above, converges to user-created content that has
changed not only what the Web contains, but also how it works (Akar and Topçu 2011).
This interaction between the Web and the user will be considered in the construct “at-
tachment degree”, since according to Okazaki et al., this denes the intensity of the rela-
tionship between an individual and a website, so it is natural to think that the more intense
the relationship, greater the interaction of the former with the latter will be for different
purposes, i.e. the purchase. Hence the construction of hypothesis 10, which was not con-
templated by the model of Okazaki et al., and which relates to the attachment degree with
the intention to use the social network to make a purchase.
H10: The attachment degree has a signicant positive effect on the intention to use a social
network to make a purchase.
Regarding the theoretical considerations discussed above and the adaptations and ex-
tensions made from the model of Okazaki et al., the model proposed in the present article
is illustrated in Figure 2.

Inuence of social networks on the purchase decisions of university students
Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 ISSN: 1131 - 6837
72
Figure 2
Proposed model for the present article
Source: Adaptation of Okazaki et al. (2012).
6. ANALYSIS OF RESULTS
6.1. Validity of the model
For the validation of the adapted and extended model from the one proposed by Okaza-
ki et al. (2012) the SPSS4 software was used, with which convergent and discriminat type
validations were performed. These validations allow identifying, among other things, those
items that correlate more with the set of items that are measuring the construct. According
to Campbell and Fiske (1959, cited in Martínez-García and Martínez-Caro 2009) for the
measures of a given model to be valid, there must exist a high correlation between meas-
ures of the same construct (convergent validity), and that correlation in regard to measures
of other constructs, must be greater (discriminant validity).
A. Convergent Validity
In order to achieve the union or convergence of the model, after nding the standard-
ized factor loads of all the initially considered indicators, it was necessary to eliminate
those corresponding to CWSN1, SP1, E3, E4, AD1, IUSVMP1, SR2 and PU1, since these
reduced the value of the average obtained from the loads of the indicators on each factor,
which according to the literature review, should be higher than 0.7 for all constructs (Hair
et al. 2001), as shown in Table 2, where the indicators mentioned above were effectively
eliminated.
4 Statistical Package for the Social Sciences software package.
Figure 2
Proposed model for the present article
H10
Source: Adaptation of Okazaki et al. (2012).
Identification
with the social
network
(IWSN)
Commitment
with the social
network
(CWSN)
Social
pressure
(SP)
Electronic
Word-of-
Mouth (e-
WOM)
Intention to use a
social network to
make a purchase
(IUSNMP)
Perceive
usefulness
(PU)
Entertainment
(E)
Social
recognition (SR
Attachme nt
degree (AD)
Block 2: Proposal about
the background of the
brand community
Block 1: Brand
community model by
Algesheimer et al. (2005)
Block 3: Proposal about
consequences of Brand
communit y
H1
H2
H3
H4
H6
H5
H8
H9
H7

L.E. Gutiérrez Flórez / M.I. Correa Escobar / A. Henao Restrepo / D. Arango-Botero / A. Valencia-Arias
ISSN: 1131 - 6837 Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 73
Table 2
Convergent validity rened from standardized factor loads
Construct Item Standarized factor
loads
Average of standarized
factor loads
Commitment with the social
network
CWSN2 0.919 0.919
CWSN3 0.919
electronic Word-of-Mouth e-WOM1 0.871 0.871
e-WOM2 0.871
Entertainment
E1 0.819 0.819
E2 0.819
Attachment degree
AD2 0.657
0.726AD3 0.850
AD4 0.671
Identication with the social
network
IWSN1 0.901 0.901
IWSN2 0.901
Intetion to use social network
to make a purchase
IUSNMP2 0.639
0.743IUSNMP3 0.764
IUSNMP4 0.826
Social pressure
SP2 0.874
0.900SP3 0.917
SP4 0.909
Social recognition SR1 0.718 0.707
SR3 0.697
Perceive d usefulness
PU2 0.811
0.750PU3 0.832
PU4 0.606
Source: Own elaboration.
Afterwards, the Bartlett’s sphericity test and the KMO measure are presented in Table
3, since they are the statistics corresponding to the study of the adequacy of the model
to the sample. The rst referred statistic is a test that is used to check the hypothesis that
the correlation matrix obtained is not an identity matrix, meaning there are signicant
inter-correlations among the variables that justify the factorial analysis (De la Fuente and

Inuence of social networks on the purchase decisions of university students
Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 ISSN: 1131 - 6837
74
Justicia 2003). Its p value must be lower than the critical levels 0.05 or 0.01, because if the
critical level is greater than 0.05, the null hypothesis of sphericity cannot be rejected and,
consequently, it cannot be assured that the factorial model is adequate to explain the data
(Uriel and Manzano 2002). Given that the proposed model presents Bartlett’s values equal
to zero, it can be afrmed that there are signicant correlations between the variables.
Moreover, the value of the Kaiser-Meyer-Olkin (KMO) sampling adequacy measure is
dened as an index that compares the magnitudes of the correlation coefcients observed
with the magnitudes of the partial correlation coefcients, and their value is between 0 and
1. This value is used as a measure of sample adequacy, knowing that the low values in this
index discourage the application of this analysis (De la Fuente and Justicia 2003). Kaiser
(1974) characterizes these values on a scale that considers KMO measures closer to 0.90 as
marvelous, to 0.80 as meritorious, to 0.70 as medium, to 0.60 as mediocre, and below 0.50
as unacceptable (Lévy et al. 2006).
Table 3
Convergent KMO validation and Bartlett’s sphericity test
Source: Own elaboration.
Table 3 shows the coefcients provided by the SPSS software for each factor meet the
mentioned criteria, which indicates that it is feasible to perform the data reduction tech-
nique, and thus clarify the reality on the factors that inuence the perceptions of university
students to use social networks as a means of purchasing different goods.
B. Discriminant Validity
Discriminant validity is one of the common criteria for evaluating latent variables in
social sciences. It is stated in this phase that for measures to be valid, those of the same
construct must correlate highly among them, and that correlation must be greater than that
existing among the measures proposed for another construct (Campbell and Fiske 1959,
cited in Martínez-García and Martínez-Caro 2009). In the present research, a discriminant
Factor KMO
value
Barlett
value
Fullled
criteria
Commitment with the social network 0.500 0.00 Yes
Electronic Word-of-Mouth 0.500 0.00 Yes
Entertainment 0.500 0.00 Yes
Attachment degree 0.522 0.00 Yes
Intention to use social network to make a purchase 0.500 0.00 Yes
Identication with the social network 0.590 0.00 Yes
Social pressure 0.733 0.00 Yes
Social recognition 0.500 0.00 Yes
Perceived usefulness 0.590 0.00 Yes

L.E. Gutiérrez Flórez / M.I. Correa Escobar / A. Henao Restrepo / D. Arango-Botero / A. Valencia-Arias
ISSN: 1131 - 6837 Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 75
validity analysis was performed, by checking that the condence interval in the correlation
estimated between each pair of factors did not contain the value of 1 (Anderson and Gerb-
ing 1988). Table 4 shows that all cases satisfy this criterion.
Table 4
Discriminant validity of the measurement model
CWSN e-WOM E AD IWSN IUSNMP SP SR PU
CWSN ….
e-WOM [0,133;0,396] ….
E[-0,132;0,122] [-0,041;0,226] ….
AD [-0,075;0,174] [-0,037;0,236] [-0,061;0,216] ….
IWSN [0,074;0,301] [-0,024;0,255] [-0,116;0,142] [-0,024;0,243] ….
IUSNMP [-0,060;0,202] [-0,055;0,207] [-0,096;0,180] [0,121;0,377] [0,113;0,358] ….
SP [-0,030;0,240] [0,092;0,342] [-0,062;0,192] [0,121;0,362] [0,009;0,270] [0,448;0,655] ….
SR [0,101;0,357] [-0,033;0,239] [0,107;0,368] [-0,124;0,142] [0,068;0,322] [-0,063;0,190] [-0,002;0,244] ….
PU [0,013;0,252] [0,085;0,339] [0,186;0,437] [-0,015;0,255] [0,100;0,362] [0,031;0,305] [0,210;0,441] [0,159;0,399] ….
Source: Own elaboration.
It is veried that the instrument used measures in a high degree what it intends to meas-
ure. The reliability of the internal consistency of the instrument will be estimated using the
Cronbach’s alpha, since it is a reliable tool in which the items (Likert’s scale) are assumed
to measure the same construct and are highly correlated (Welch and Comer 1988). The
internal consistency of the items is greater if the value of Cronbach’s alpha is closer to 1,
since the test reaches positive values between 0 and 1, where 0 indicates total absence of
internal consistency, and 1 indicates total redundancy between items.
George and Mallery (2003) suggest the following recommendations to evaluate Cron-
bach’s alpha coefcients: Coefcient alpha > 0.9 is excellent, Coefcient alpha > 0.8 is
good, Coefcient alpha > 0.7 is acceptable, Coefcient alpha > 0.6 is questionable, Coef-
cient alpha > 0.5 is poor, and Coefcient alpha < 0.5 is unacceptable. However, in the early
phases of the research a reliability value of 0.6 or 0.5 may be sufcient (Nunnally 1967,
cited in Frías-Navarro 2013).
As shown in Table 5, the measurement instrument seems to have a high reliability of the
internal consistency of the measurement scale, since the Cronbach’s alpha of the analyzed
constructs is among the range of values recommended and all factors are considered within
the values recommended by George and Mallery (2003).
Table 5
Reliability index – Cronbach’s Alpha
Factor Cronbach’s alpha
Commitment with the social network 0.921
electronic Word-of-Mouth 0.878
Entertainment 0.828

Inuence of social networks on the purchase decisions of university students
Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 ISSN: 1131 - 6837
76
Attachment degree 0.774
Intention to use social network to make a purchase 0.915
Identication to the social network 0.781
Social pressure 0.929
Social recognition 0.771
Perceived utility 0.794
Source: Own elaboration.
The previous results show the existence of a sustainable factorial model for the analy-
sis, based on the perceptions of university students on the use of social networks as a means
to purchase different goods. The presence of convergent validity and discriminant validity
within the instrument, together with an acceptable reliability, conrms that the instrument
evaluates fundamental variables that directly or indirectly inuence the ideas, expectations,
and motivations in the use of social networks by university students.
C. Model Adjustment
At this point, the proposed structural model was estimated (Figure 3) to evaluate the inten-
tions of social networks usage for purchase purposes. Various hypotheses were collected and its
suitability evaluated, by recognizing its adjustment or not to the model proposed in this paper.
Figure 3
Proposed model of structural equations
Source: Own elaboration.
Before carrying out the global evaluation of the model and determining the degree
of adjustment in relation to the collected data, a conrmatory factor analysis (CFA) was
e5
Figure 3
Proposed model of structural equations
Source
:Own elaboration.
e1
e15
e16
e17
e23
e22
e21
e20
e19
e18
CWSN2
IUSNMP2
IUSNMP3
IUSNMP4
IWSN1
IWSN2
SR1
SR
3
PU2
PU3
PU4
CWSN3
eWOM2
eWOM1
E1
E2
AD2
AD3
AD4
SP3
SP4
SP2
IWSN
CWSN
SP
e–WOM
IUSNMP
PU
E
SR
AD
e2
e3
e4
e5
e6
e7
e8
e9
e14
e10
e11
e12
e13

L.E. Gutiérrez Flórez / M.I. Correa Escobar / A. Henao Restrepo / D. Arango-Botero / A. Valencia-Arias
ISSN: 1131 - 6837 Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 77
Attachment degree 0.774
Intention to use social network to make a purchase 0.915
Identication to the social network 0.781
Social pressure 0.929
Social recognition 0.771
Perceived utility 0.794
Source: Own elaboration.
The previous results show the existence of a sustainable factorial model for the analy-
sis, based on the perceptions of university students on the use of social networks as a means
to purchase different goods. The presence of convergent validity and discriminant validity
within the instrument, together with an acceptable reliability, conrms that the instrument
evaluates fundamental variables that directly or indirectly inuence the ideas, expectations,
and motivations in the use of social networks by university students.
C. Model Adjustment
At this point, the proposed structural model was estimated (Figure 3) to evaluate the inten-
tions of social networks usage for purchase purposes. Various hypotheses were collected and its
suitability evaluated, by recognizing its adjustment or not to the model proposed in this paper.
Figure 3
Proposed model of structural equations
Source: Own elaboration.
Before carrying out the global evaluation of the model and determining the degree
of adjustment in relation to the collected data, a conrmatory factor analysis (CFA) was
e5
performed for each of the constructs (Ranaweera 2016). Some examples of the use of such
analysis for the of constructs’ validation and evaluation of model factor structures can be
found in Bryant et al. (2016), Pavia et al. (2016), Mahler et al. (2016).
It was possible to identify the degree which the indicators reected the construct with
the CFA, through the chi-square test and the following indexes: CFI, TLI, RMSEA and
SRMR. The results are presented in Table 6.
Table 6
Chi-square test and construct t indexes
Construct p-value CFI TLI RMSEA SRMR
PU 0.649 1 1.026 0 0.018
AD 0.218 0.993 0.978 0.048 0.051
SP 0.59 1 1.004 0 0.061
IUSNMP 0.847 1 1.038 0 0.009
Source: Own elaboration.
Table 6 only presents the results for four of the nine constructs, and was obtained after
a preliminary analysis. After determining the initially proposed model, it was required the
following correlations between the residuals of indicators: PU2 and PU3, AD2 and AD3,
AD3 and AD4, SP2 and SP3, IUSNMP2 and IUSNMP3, IUSNMP3 and IUSNMP4. The
modied model is shown in Figure 4. The other ve constructs were not analyzed, since
within the limitations of the study stands out the fact that most of the constructs were orig-
inally associated with 3 or less indicators. At the moment of analyzing the factorial loads
and eliminating those that did not contribute to the global average to exceed 0.7, these ve
constructs were left with only two indicators, which imposes many constraints on the mod-
el and does not allow to considerate the presence of correlated errors (Kenny et al. 1998),
making it difcult to evaluate the overall model adjustment.
Figure 4
Modied model of structural equations
Source: Own elaboration.
Figure 4
Modified model of structural equations
Source
: Own elaboration.
e1
e15
e16
e17
e23
e22
e21
e20
e19
e18
CWSN2
IUSNMP2
IUSNMP3
IUSNMP4
IWSN1
IWSN2
SR1
SR3
PU2
PU3
PU4
CWSN3
eWOM2
eWOM1
E1
E2
AD2
AD3
AD4
SP3
SP4
SP2
IWSN
CWSN
SP
e–WOM
IUSNMP
PU
E
SR
AD
e2
e3
e4
e5
e6
e7
e8
e9
e14
e10
e11
e12
e13

Inuence of social networks on the purchase decisions of university students
Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 ISSN: 1131 - 6837
78
As can be seen in Table 6, the p-values of the chi-square test largely exceed the level of
signicance of 0.05, so it can be concluded that there is a good t in each of the constructs.
It is worth to mention at this point that, like any statistical test, chi-square is sensitive
to the size of the sample (Hox and Bechger 1998). Therefore, the adjustment assessment
is also supported by other indexes indicating a good t if they are above 0.9, for the case
of IFC and TLI (Hair et al. 2001) and below 0.08 and 0.06 for SRMR and RMSEA indices
respectively (Hu and Bentler 1999). As the indices presented in Table 6 are within those
reference values, it can be concluded that the t between the modied model and the ob-
served data is relatively good.
After modifying the initially proposed model by including some correlations between
the residuals of some indicators, as indicated by the CFA, we proceeded with the analysis
of the global model through the implementation of the methodology of structural equa-
tions, which was achieved with the Lavaan package of the software R. From such adjust-
ment the p-value was obtained through the chi-square test and the values of the indices
presented in Table 7 were obtained from this adjustment. Although the p-value seems to
indicate that there is no good t of the overall model, it has been previously indicated that
the chi-square test is sensitive to the sample size, and as the other considered indices ev-
idence that t is good. It can be concluded that this model serves as a basis to understand
the factors that have some inuence, either directly or indirectly on the intentional use of
social networks to make a purchase.
Table 7
Chi-square test and construct t indexes
p-value CFI TLI RMSEA SRMR
Global model 0.0 0.935 0.919 0.048 0.076
Source: Own elaboration.
7. CONCLUSIONS AND IMPLICATIONS OF THE STUDY
7.1. Theoretical and management implications
After carrying out a literature review, having as aim to nd studies about social net-
works and their interactions with the marketing processes, a possibility was found to try
to identify the factors and variables that could have some inuence on the intention to
use social networks to make a purchase. Although the model proposed by Okazaki et al.
(2012), with its main constructs and relations between them is adapted and extended. Two
cause and effect relationships are included. One of them is part of what is indicated by the
Theory of Reasoned Action (inuence of social pressure on the intention to have a deter-
mined behavior), which highly enriches the model of the authors, and is taken as reference
in the analysis of the inuence of social networks, expanding even further its eld of study.

L.E. Gutiérrez Flórez / M.I. Correa Escobar / A. Henao Restrepo / D. Arango-Botero / A. Valencia-Arias
ISSN: 1131 - 6837 Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 79
7.2. Discussion
When adapting and extending the social networking model proposed by Okazaki et al.
(2012), problems with its convergence were presented, and to solve this situation, certain
indicators of the constructs had to be eliminated in order to have standardized factor loads
that contributed to maintain the average for all constructs above 0.7, according to the ref-
erence value found in the literature. During the evaluations of the statistics corresponding
to the Bartlett’s test and the KMO measure, it was possible to verify a correct adaptation of
the model to the sample, thus allowed the further progress towards the understanding of the
factor that inuence the use of social networks for purchasing decisions.
Furthermore, the model validation via discriminant analysis ensured that valid meas-
ures were obtained, since the ones corresponding to the same construct presented a high
correlation between them, being these higher than other measurements of constructs. In
addition, through the evaluation of Cronbach’s alpha, it was possible to verify the high
reliability of the internal consistency of the measurement scale.
Once the correct t of the model to the sample and the internal consistency of the
measurement scale were veried, a conrmatory factor analysis was carried out, for the
constructs that had at least two indicators with the purpose of identifying to what extent
these indicators were reecting the associated construct. Thus, some correlations between
residuals of several indicators were added to nally implement the adjustment methodol-
ogy of the model through structural equations. The results nally indicated that the model
initially proposed with some modications, allows obtaining a knowledge approximation
of the factors that may have some inuence on the intention to use social networks to make
purchases by university students.
7.3. Limitations of the study and future outlines
The previous results show the model adequacy to the sample and the reliability of the
internal consistency of the instrument estimated by Cronbach’s alpha, which justies the
factorial analysis with the aim of nding the factors that guide and motivate the use of
social networks in purchasing decisions. However, it should be noted that at the time of the
respective conrmatory factor analyze for each of the constructs, the obtained results are
limited and do not contribute to the overall model adjustment.
Despite the difculties pointed out above, the evaluation of the overall model adjust-
ment, allowed to conclude that it can serve as a basis in the task of identifying the factors
that may have some inuence on the intention to use social networks to make a purchase,
taking the model proposed by Okazaki et al. (2012) as a reference, to which the relationship
between the social pressure and intention to use social networks to buy constructs were
included based on what is exposed by the Theory of Reasoned Action. This is useful to
further expand the theoretical debate and continue to address future studies that evaluate
the inuence of social networks on different marketing processes, such as the buying and
selling relationships of Web pages or social networks.

Inuence of social networks on the purchase decisions of university students
Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 ISSN: 1131 - 6837
80
7.4. Implications and recommendations for companies
The presented ndings in the present study support the research antecedents reported
in the literature review, conrming the strong inuence of the word-of-mouth channel for
information transmission about a certain product or service, opening up the possibility of
exploiting this information channel to create a competitive advantage due to its low costs,
massive access, and ease of penetration in the target audience. This should motivate com-
panies to contemplate new mechanisms and strategies that allow them dening alternative
approaches in the advertising of their products in social networks. These strategies should
be based on the fact that the surveyed population is very open to receive the contributions
or opinions of their peers (friends, acquaintances or other consumers) to make purchasing
decisions.
After observing the growing acceptance of marketing in social networks and the inu-
ence that such information has on consumer purchasing decisions, it is recommended that
companies have more effort to studying what is happening around this phenomenon, and
to establish how this marketing tool is able to revitalize the relationship between custom-
ers and brands. Consumer behavior could be shaped towards a more accessible attitude
to the purchase through the use of social networks. It is worth studying the impact on the
implementation of higher levels of transparency in the information transmission, which can
generate greater target audience condence towards the advertising that companies provide
in social networks. Likewise, it is recommended to companies that wish to promote their
products through social networks, seek alternative strategies that incorporate transparency
in information, and stimulate word-of-mouth communication among users, generating a
greater impact on the purchase decisions of customers.
8. REFERENCES
Abascal, E. and Esteban, I.G., 2005. Análisis de encuestas. Madrid: ESIC Editorial.
Acar, A.S. and Polonsky, M., 2007. Online social networks and insights into marketing
communications. Journal of Internet Commerce, 6(4), 55-72.
Ajzen, I., 1991. The theory of planned behavior. Organizational behavior and human de-
cision processes, 50 (2), 179-211.
Akar, E. and Topçu, B., 2011. An examination of the factors inuencing consumers’ at-
titudes toward social media marketing. Journal of Internet Commerce, 10 (1), 35-67.
Alvarado, P.A., 2012. Impacto de las redes sociales sobre las variables de decisiones de
los agentes. Universidad Nacional de Colombia, Facultad de Ciencias Económicas,
Colombia. Available at: http://www.fce.unal.edu.co/wiki/images/0/04/Impacto_de_
las_redes_sociales
_en_las_decisiones_de_los_agentes.pdf. [Accessed: 20 December 2014].
Anderson, J.C. and Gerbing, D.W., 1988. Structural equation modeling in practice: A re-
view and recommended two-step approach. Psychological bulletin, 103 (3), 411-423.
Arcos, V.A., Gutiérrez, S.S.M. and Hernanz, R.J.P., 2014. La aplicación empresarial del
marketing viral y el efecto boca-oreja electrónico: opiniones de las empresas. Cuader-
nos de gestión, 14 (1), 15-31.

L.E. Gutiérrez Flórez / M.I. Correa Escobar / A. Henao Restrepo / D. Arango-Botero / A. Valencia-Arias
ISSN: 1131 - 6837 Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 81
Ajzen, I. and Fishbein, M., 1980. Understanding Attitudes and Predicting Social Behavior.
Englewood Cliffs, NJ: Prentice Hall.
Bermudez, J., Chalela, S., Valencia, J. and Valencia, A., 2017. Research Trends in the Study
of ICT Based Learning Communities: A Bibliometric Analysis. Eurasia Journal of
Mathematics, Science and Technology Education, 13 (5), 1539-1562. DOI: 10.12973/
eurasia.2017.00684a
Bolotaeva, V. and Cata, T., 2010. Marketing opportunities with social networks. Journal of
Internet Social Networking and Virtual Communities, 1-8.
Bran, L., Romero, K., Echeverri, L., Peña, J., Vasquez, S., Aguilera, M., Herazo, C. and
Valencia, A., 2017. Information and Communication Technologies Inuence on Family
Relationship. Global Journal of Health Science, 9 (6), 204-213. DOI:10.5539/gjhs.
v9n6p204.
Bryant, C.L., Maarouf, S., Burcham, J. and Greer, D., 2016. The examination of a teacher
candidate assessment rubric: A conrmatory factor analysis. Teaching and Teacher Ed-
ucation, 57, 79-96.
Brown, J., Broderick, A.J. and Lee, N., 2007. Word of mouth communication within online
communities: Conceptualizing the online social network. Journal of interactive mar-
keting, 21 (3), 2-20.
Calvo-Porral, C., Martínez-Fernández, V.A. and Juanatey-Boga, O., 2013. Análisis de dos
modelos de ecuaciones estructurales alternativos para medir la intención de compra.
Revista Investigación Operacional, 34 (3), 230 – 243.
Cartagena, C., Vasquez, A., Benjumea-Arias, M. and Valencia-Arias, A., 2017. Proposed
Model for Measuring Customer Satisfaction with Telecommunications Services. Med-
iterranean Journal of Social Sciences, 8 (2), 15-26. DOI: 10.5901/mjss.2017.v8n2p15.
Cheung, C.M. and Lee, M.K., 2010. A theoretical model of intentional social action in
online social networks. Decision support systems, 49 (1), 24-30.
ComScore, 2013. Futuro Digital Colombia 2013. Available at: http://www.comscore.
com/esl/Prensa-y-Eventos/Presentaciones-y-libros-blancos/2014/2014-Digital-Fu-
ture-in-Focus-Colombia. [Accessed: 12 August 2015].
Davis, F.D., Bagozzi, R. and Warshaw, P., 1989. User acceptance of computer technology:
A comparison of two theoretical models. Management Science, 35 (8), 982–1003.
De la Fuente, J. and Justicia, F.J., 2003. Escala de estrategias de aprendizaje ACRA-Abre-
viada para alumnos universitarios. Electronic journal of research in educational psy-
chology, 1 (2), 139-158.
Dholakia, U.M., Bagozzi, R.P. and Pearo, L.K., 2004. A social inuence model of consum-
er participation in network-and small-group-based virtual communities. International
journal of research in marketing, 21 (3), 241-263.
Fishbein, M. and Ajzen, I., 1975. Belief, attitude, intention, and behavior: An introduction
to theory and research. Reading, MA: Addison-Wesley.
Frías-Navarro, D., 2013. Alfa de Cronbach y consistencia interna de los ítems de un in-
strumento de medida. Available at: http://www.uv.es/~friasnav/AlfaCronbach.pdf. [Ac-
cessed: 10 May 2015].
Gatautis, R. and Kazakevičiūtė, A., 2012. Consumer behavior in online social networks:
review and future research directions. Economics and Management, 17 (4), 1457-1463.

Inuence of social networks on the purchase decisions of university students
Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 ISSN: 1131 - 6837
82
George, D. and Mallery, M., 2003. Using SPSS for Windows step by step: a simple guide
and reference. Boston, MA: Allyn y Bacon.
Hair, J.F., Anderson, R.E., Tatham, R.L. and Black, W.C., 2001. Análisis Multivariante. 5a
ed. Madrid: Prentice Hall Iberia.
Hill, S., Provost, F. and Volinsky, C., 2006. Network-based marketing: Identifying likely
adopters via consumer networks. Statistical Science, 256-276.
Hoyos, J. and Valencia, A., 2012. El papel de las TIC en el entorno organizacional de las
pymes. Trilogía. Ciencia, Tecnología y Sociedad, 7 (1), 105-122.
Hox, J.J. and Bechger, T.M., 1998. An introduction to structural equation modeling. Family
Science Rewiew, 11, 354-373.
Hsu, M.H., Ju, T.L., Yen, C.H. and Chang, C.M., 2007. Knowledge sharing behavior in
virtual communities: The relationship between trust, self-efcacy, and outcome expec-
tations. International journal of human-computer studies, 65 (2), 153-169.
Hu, L.T. and Bentler, P.M., 1999. Cutoff criteria for t indexes in covariance structure
analysis: Conventional criteria versus new alternatives. Structural equation modeling:
a multidisciplinary journal, 6 (1), 1-55.
Kaiser, H.F., 1974. An index of factorial simplicity. Psychometrika, 39 (1), 31-36.
Kalpaklioglu, N.U. and Toros, N., 2011. Viral marketing techniques within online social
network. Journal of Yasar University, 24 (6), 4112-4129.
Kaplan, D., 2008. Structural equation modeling: Foundations and extensions. 2a ed. Thou-
sand Oaks: Sage Publications.
Kenny, D.A., Kashy, D.A. and Bolger, N., 1998. Data analysis in social psychology.
In: Gilbert, D., Fiske, S. and Lindzey, G, eds. The handbook of social psychology.
McGraw-Hill, 233–265.
Kunz, M.B. and Hackworth, B.A., 2011. Are consumers following retailers to social net-
works? Academy of Marketing Studies Journal, 15 (2), 1-22.
Lévy, J.P., Martín, M.T. and Román, M.V., 2006. Modelización con estructuras de covari-
anzas en ciencias sociales. España: Gesbiblo, S.L.
Mahler, C., Giesler, M., Stock, C., Krisam, J., Karstens, S., Szecsenyi, J. and Krug, K.,
2016. Conrmatory factor analysis of the German Readiness for Interprofessional
Learning Scale (RIPLS-D). Journal of interprofessional care, 30 (3), 381-384.
Mangold, W., Faulds, D., 2009. Social media: The new hybrid element of the promotion
mix. Business Horizons, 52 (4), 357-365.
Martínez-García, J.A. and Martínez-Caro, L., 2009. La validez discriminante como criterio
de evaluación de escalas: ¿teoría o estadística? Universitas Psychologica, 8 (1), 27-36.
Maurer, C. and Wiegmann, R., 2011. Effectiveness of advertising on social network sites:
a case study on Facebook. Vienna: Springer.
Miller, R. and Lammas, N., 2010. Social media and its implications for viral marketing.
Asia Pacic Public Relations Journal, 11 (1), 1-9.
Okazaki, S., Benito, N.R. and Martínez, S.C., 2012. Antecedentes de las redes sociales
como canales de comunicación promocional para los jóvenes. Revista Española de
Investigación en Marketing ESIC, 16 (2), 31-48.
Pavia, L., Cavani, P., Di Blasi, M. and Giordano, C., 2016. Smartphone Addiction Inven-
tory (SPAI): Psychometric properties and conrmatory factor analysis. Computers in
Human Behavior, 63, 170-178.

L.E. Gutiérrez Flórez / M.I. Correa Escobar / A. Henao Restrepo / D. Arango-Botero / A. Valencia-Arias
ISSN: 1131 - 6837 Cuadernos de Gestión Vol. 18 - Nº 1 (2018), pp. 61-84 83
Pookulangara, S. and Koesler, K., 2011. Cultural inuence on consumers’ usage of social
networks and its’ impact on online purchase intentions. Journal of Retailing and Con-
sumer Services, 18 (4), 348-354.
Ramírez-Correa, P., 2014. Uso de internet móvil en Chile: explorando los antecedentes
de su aceptación a nivel individual. Ingeniare. Revista chilena de ingeniería, 22 (4),
560-566.
Ranaweera, H.M.B.P., 2016. Perspective of trust towards e-government initiatives in Sri
Lanka. SpringerPlus, 5 (1), 1-11.
Schlereth, C., Barrot, C., Skiera, B. and Takac, C., 2013. Optimal Product-Sampling Strat-
egies in Social Networks: How Many and Whom to Target? International Journal of
Electronic Commerce, 18 (1), 45-72.
Smith, T., Coyle, J.R., Lightfoot, E. and Scott, A., 2007. Reconsidering models of inu-
ence: the relationship between consumer social networks and word-of-mouth effective-
ness. Journal of Advertising Research, 47 (4), 387-397.
Soares, A.M., Pinho, J.C. and Nobre, H., 2012. From social to marketing interactions: The
role of social networks. Journal of Transnational Management, 17 (1), 45-62.
Tanuri, I., 2010. A literature review: Role of social media in contemporary marketing.
Available at: https://issuu.com/virginiahall5/docs/a-literature-review-role-of-social.
[Accessed: 10 January 2012].
Uriel, E. and Manzano, J.A., 2002. Análisis multivariante aplicado. Madrid: Paraninfo.
Uribe, F., Rialp, J. and Llonch, J., 2013. El uso de las redes sociales digitales como her-
ramienta de marketing en el desempeño empresarial. Cuadernos de Administración, 26
(47), 205–231.
Wang, J.C. and Chang, C.H., 2013. How online social ties and product-related risks inu-
ence purchase intentions: A Facebook experiment. Electronic Commerce Research and
Applications, 12 (5), 337-346.
Valencia, A., Benjumea, M. and Rodríguez-Lora, V., 2014. Intención de uso del e-learn-
ing en el programa de Administración Tecnológica desde la perspectiva del modelo de
aceptación tecnológica. Revista Electrónica Educare, 18 (2), 247-264. DOI: 10.15359/
ree.18-2.13.
Valencia, A., González, G. and Castañeda, M., 2016. Structural equation model for study-
ing the mobile-learning acceptance. IEEE Latin America Transactions, 14 (4), 1988–
1992. DOI: 10.1109/TLA.2016.7483544.
Valencia, A., Salazar, O. and Ovalle, D., 2013. Improving the Entrepreneur’s Market Re-
search Strategies Learning Process Using the MaREMAS Environment. In J. M. Cor-
chado et al., eds. Highlights on Practical Applications of Agents and Multi-Agent Sys-
tems: International Workshops of PAAMS 2013, Salamanca, Spain, May 22-24, 2013.
Proceedings. Berlin, Heidelberg: Springer Berlin Heidelberg, pp. 363–374. Available
at: https://doi.org/10.1007/978-3-642-38061-7_34.
Welch, S. and Comer, J., 1988. Quantitative methods for public administration: Techniques
and applications. 2a ed. Fort Worth: Harcourt Brace College Publishers.