DOES SOCIAL NETWORK QUALITY DIFFER BETWEEN FACEBOOK AND INSTAGRAM? APPLICATION OF SNSQUAL MODEL

Although numerous studies investigated service quality in online environment, the social network quality has been inadequately captured by previous empirical research. Thus, the present study focuses on measuring social network quality. Specifically, it aims to examine potential differences in perceived social network quality between two popular social networks, namely Facebook and Instagram. The empirical data are based on gathering primary data using questionnaire based on SNSQUAL model, developed by Phillips et al. (2016). Descriptive and bivariate statistical analysis were conducted using data collected from undergraduate and graduate students who use social networks on regular bases. The study results show significant differences in 16 out of 27 social network quality items, revealing that Instagram’s social network quality was rated significantly higher than Facebook’s. These findings may contribute to the development of service excellence approach that aims to enhance social networks’ performance.


INTRODUCTION
Today, due to the availability of much more data and computer power, information and communication technologies (ICTs) have become essential part of people's lives (Hennig-Thurau et al., 2010;Karanasios & Parker, 2018). As Gërguri -Rashiti et al. (2017) noted: "ICT has become an intrinsic part of everyday life to the extent that, like electricity, modern society could not function in its absence". Social media tools such as Facebook and Instagram cover a prominent role within the society and in different fields, such as tourism (Del Vecchio et al., 2018;Shams & Lombardi, 2016), accounting (Secundo & Lombardi, 2020) and academic entrepreneurship . Those computer-mediated tools allow anyone to create, circulate, share, and exchange information in a variety of formats and with multiple communities (Leonardi & Vaast, 2017).
According to Kaplan and Haenlein (2012), social media are a group of Internet applications based on Web 2.0 technology, that enable users to create and share the content that they had created themselves. Social media include social networking sites, blogs, multimedia platforms, Internet forums, and various virtual communities (Obar & Wildman, 2015). They enable fast communication and interaction, which affects the users' level of being informed and influences their purchase decisions. Due to that, social media are no longer perceived as solely a form of entertainment among the younger population, but they are more and more used by firms to promote their products and services, thus becoming an important online marketing tool.
The data from Global social media research summary (2020) attest to the popularity and widespread usage of social media and social networking sites, according to which 49 per cent of the world population in 2020 was actively using social media. The popularity of social network sites is most clearly observed among young adults with many of them having at least one social network account and logging on to them at least once a day (Duffett, 2015;Gangadharbatla, 2008;Raacke & Bonds-Raacke, 2008;Ting, de Run, & Liew, 2016). The most popular social networking site was Facebook with more than 2.4 billion users (Statista Report, 2020). In 2020, there were 2,141,000 Facebook users in Croatia as of November 2020, with 28 percent of those users being between 25 and 34 years of age (Statista Report, Croatia, 2020).
The importance of conducting research on the quality of social networking sites is visible precisely in the role that social networks play in the society and economics, and in the positive influence that the service quality has over the clients' satisfaction and loyalty. Although there is a number of service quality research in online environment, the research of social network quality is sparse. Therefore, the purpose of this study is twofold. Firstly, to provide a comprehensive review of research on the e-service quality concept and service quality measurement scale in online environment. Secondly, to examine potential differences in perceived social network quality between two popular social networks, namely, Facebook and Instagram. Specifically, the paper aims to address the following questions: Q1: How has research on e-service quality and service quality measurement scale evolved during last two decades?
Q2: How do users (students) perceive the quality of two social networks, namely Facebook and Instagram?
Q3: Are there significant differences between Facebook and Instagram in terms of perceived social network quality?
To authors' knowledge, this paper is the first research done in Croatia related to this topic. It could serve as a good base for further studies related to the impact of ICTs and social media on the business performance.
The paper is structured as follows. In the literature review, overview of the previous research on the e-service quality concept and service quality measurement in online environment are presented. In the methodology section, the research method and sampling are described, while in the findings section, the results of the conducted descriptive and bivariate statistical analyses are presented. The paper ends with conclusion, limitations and recommendation for future studies.

LITERATURE REVIEW
Quality is a central element in both business strategy and academic research. It is a key force leading to delighted customers, firm profitability, and the economic growth of the countries (Golder, Mitra & Moorman, 2012). In order to be competitive in a changing market, companies must improve their quality and innovativeness (Zehir & Sadikoğlu, 2012). Today, the rapid expansion of information and communication technologies has forced companies to concentrate on service quality in the online environment.
The term electronic services or e-services has been defined by Colby and Parasuraman (2003) as services offered by an electronic means (normally Internet) and which refer to transactions begun and to a great extent controlled by the consumer. De Ruyter, Wetzels & Kleijnen (2001) and Mou, Shin & Cohen (2017) define e-service as an interactive content-centred and Internet-based customer service. E -service quality is the extent to which the web site promotes efficient and effective shopping, purchasing and delivering of products and services (Parasuraman, Zeithaml, & Malhotra, 2005). It is a consumer overall opinion and evaluation regarding the excellent e-service delivery in online market (Santos, 2003).
One of the most commonly used instruments for measuring the service quality in various service industries is the SERVQUAL model, developed by Parasuraman et al. (1985Parasuraman et al. ( , 1988. According to them, service quality is a result of comparison of expectation and perception of the received product or service, and it comprises of five main dimensions: tangibles, reliability, responsiveness, assurance, and empathy. With growth of Internet usage and its role in the personal and business lives, the need to evaluate the quality of e-services was also recognized. The literature review shows that several authors mentioned the basic dimensions (elements) of e-service quality in their research. Zeithaml, Parasuraman, & Malhotra (2002) suggested eleven dimensions of e-service quality: reliability, responsiveness, access, flexibility, ease of navigation, efficiency, security/trust, knowing the price, aesthetics, adaptation/personalization. According to Parasuraman et al. (2005), e-service quality comprises efficiency, system availability, fulfilment, privacy, responsibility, responsiveness/contact. Zhang, Huang, He & Wang (2015) determined that service quality is influenced by practicality, accuracy of information, security and functionality. Furthermore, Farida, Suyudi, Nuryuliani, & Hermana (2014) stated that the most common dimensions that appear in the research on websites quality are information quality, privacy/security/insurance, reliability/fulfilment, quality of service and quality of system.
Although SERVQUAL model was developed to measure the quality of "traditional" services, it was also used in the context of measuring the quality of e-service (Li, Tan, & Xie, 2002;Jiang, & Klein, 2002;Santos, 2003;Collier & Bienstock, 2006;Cristobal, Flavián & Guinalíu, 2007). However, since e-service characteristics differ from the "traditional" service whose characteristics are affected by tangible elements of the environment in which it is provided, it was necessary to develop model designed precisely to measure the quality of service in online environment.
For this purpose, the following models were developed: SITEQUAL (Yoo & Donthu, 2001), WebQual (Barnes & Vidgen, 2002;Loiacono et al. 2002), eTailQ (Wolfinbarger & Gilly 2003), E-S-Qual (Parasuraman et al., 2005), hierarchy model of the e-service quality (Blut et al. 2015;Blut, 2016), SNSQUAL (Phillips et al., 2016). Ting et al. (2016) studied the relationship between e-service quality perception, e-satisfaction and e-loyalty in case of Malaysian online shoppers. The study considered a multi-dimensional construct by bringing together efficiency, fulfilment, reliability of E-SERVQUAL, web design from e-TailQ and considered privacy and trust as a single variable to measure e-service quality in B2C (Business to Consumer) space. The E-SERVQUAL scale was also used in higher education environment, airline & hotel reservation websites, internet banking and online bus reservation. Furthermore, Raza et al. (2020) studied the service quality dimensions in Internet banking and their impact on e-customer's satisfaction and e-customer's loyalty in the context of Pakistani users. The study used a modified construct by adding variables such as user friendliness and personal needs to original E-SERVQUAL constructs such as responsiveness, reliability and efficiency. It was observed that all proposed determinants had a direct and positive impact on electronic customer satisfaction, which in turn resulted in customer loyalty. The summary of models for measuring e-service quality and web -site quality in literature is presented in Table 1.

Source: Authors
According to Table 1, a widely used model for measuring e-service quality is E-SERVQUAL. This model was used in different fields (e.g., retail, higher education, tourism, banking), and measures different e-service quality dimensions.
These models measure the e-service quality and the websites quality, whereas SNSQUAL model (Phillips et al., 2016) measures the social networks quality. The model consists of 6 dimensions representing social network quality features: ease of use, trustworthiness, personalization, integration, reliability and information quality on social networks, and was used in the empirical part of the present study.

METHODOLOGY
The literature review was used to present the development of e-service quality and service quality measurement scale during the last two decades. In order to examine last two research questions, this study used the measurement instrument which comprises items for measuring the main research construct, namely social network quality, and items for determining respondents' demographic characteristics. Social network quality was measured using 27-item SNSQUAL scale, developed by Phillips et al. (2016). These items describe ease of use, trustworthiness, personalization, reliability, integration, and information quality on social networks, and were measured with 5-point Likert scale, ranging from "strongly disagree" (as 1) to "strongly agree" (as 5). In addition, items for measuring demographic characteristics of the respondents included gender, age, average time spent on social networks, length of social networks usage, number of social networks' memberships (accounts), gadget used for interaction with social networks, and the purpose of using social networks.
Data were gathered on the population of undergraduate and graduate students who use social networks on regular bases. The questionnaires were distributed during November 2019. A total of 127 questionnaires suitable for conducting data analysis were gathered.
Data analysis includes descriptive and bivariate statistical analyses. The demographic profile of the respondents and the level of perceived social network quality was examined with methods of descriptive statistics, calculating percentages, mean, and standard deviation. The significance of differences in the levels of perceived social network quality between Facebook and Instagram was tested using independent samples t-test.

FINDINGS
In this section, findings of descriptive and bivariate statistical analyses are presented.

1. Respondents' profile
The sample consisted of 64 Facebook and 63 Instagram users. Females accounted for about 55 per cent of the sample. Most of the respondents (61.4 per cent) were between 18 and 20 years of age, followed by the ones between 21 and 23 years of age (21.3 per cent), and those who were older than 24 years of age (17.3 per cent). Respondents were approximately equally distributed regarding the average time spent on social networks per day. They mostly spend on social networks more than 3 hours per day (29.1 per cent). About 27 per cent of them spend 2 to 3 hours, while about one quarter of the respondents spends on social networks between 1 and 2 hours daily. The majority of the students in the sample (56.7 per cent) has been using social networks more than 6 years, and about 30 per cent of them are 5 to 6 year users. Almost half of the respondents (48.8 per cent) have membership (accounts) on 3 -4 social networks, followed by the ones with 1 to 2 memberships (29.9 per cent). About 21 per cent of the respondents have membership on more than five social networks. The majority of the students in the sample (88.2 per cent) use social networks via mobile phone, while others use personal computer. The most common purpose of using social networks is communication (87.4 per cent), followed by information (60.6 per cent), information sharing (40.2 per cent), fun (40.2 per cent), learning (36.2 per cent), and meeting others (21.3 per cent).

Source: Authors
As noted in Table 2, the perceived social network quality mean scores of Facebook and Instagram users ranged from 2.72 to 4.64, and from 3.29 to 4.76, respectively. Both groups of the respondents rated with the lowest score the item "I trust this social network to keep my personal information safe". On the other hand, Facebook users gave the highest score to the item "it is easy to use this social network", while the item "I find this social network easy to use" was rated the highest by the Instagram users.
When comparing social network quality scores between Facebook and Instagram, the analysis indicated higher scores in the Instagram sample for 25 out of 27 items. On the other hand, social network quality scores for item "this social network does not go down often" were higher in Facebook sample. In addition, for one item ("I am able to interact with the social network in order to get information tailored to my specific needs") the social network quality scores between Facebook and Instagram were identical (mean score = 3.78).
The results of independent samples t-test showed that in 16 out of 27 social network quality items significant differences were found between Facebook and Instagram samples. These results indicated that Instagram's social network quality was rated significantly higher in the following items: "it is easy to use this social network to do what I want", "the interface of this social network is easy to use", "the interface of this social network is easy to understand", "I trust this social network keeps my personal information safe", "this social network is trustworthy", "this social network gives the impression that it keeps promises and commitments", "I am satisfied with the reliability of this social network", "this social network seems reliable", "this social network does not seem to have technical problems often", "this social network provides accurate information", "this social network provides relevant information", "this social network provides correct information", "overall, I think this social network provides useful information", "this social network provides reliable information", "I am satisfied with the information that this social network provides", "overall, the information this social network provides is high quality".

CONCLUSION
The present study focused on measuring social network quality. The first part of this paper presents the comprehensive review of research on the e-service quality concept and service quality measurement scales in online environment. In this study social network quality was measured using variables that describe the ease of use, trustworthiness, personalization, reliability, integration, and information quality on social networks. In particular, users' social network quality perceptions of Facebook and Instagram were compared.
According to the study results, respondents displayed neutral attitude regarding trusting social network to keep users' personal information safe. In addition, this feature had the lowest quality level, implying that there is room for improvement of information safety and users' trustworthiness toward social networks. On the other hand, social network's ease of use is the best graded quality feature. This finding is in line with one of the main features of social networks as being user friendly and can explain the widespread usage of social networks.
The results of perceived social network quality comparison showed significant differences regarding the social network quality levels between Facebook and Instagram. The ease of use, trustworthiness, reliability, and information quality are social network quality features that are significantly more recognized on Instagram than Facebook. These results imply that compared to Facebook, Instagram is easier to use, users trust it more regarding the safety of their personal information, it is more reliable, and information that it provides is more accurate, relevant, correct, and useful. Research done by Alhabash and Ma (2017) indicated that college students spent the greatest amount of time on Instagram and Snapchat, then Facebook and Twitter. Furthermore, Trifiro (2018) revealed that Instagram use has the potential to enhance the users' well-being and self-esteem and the interaction with on-campus friends was particularly high on Instagram, compared with the other two platforms such as Facebook and Twitter (Yang and Lee, 2020). While there is a lack of literature on Instagram use, few of the current studies revealed its potential for learning process (AlGhamdi, 2018;Gonulal, 2019).
On the other hand, there were no significantly different social network quality levels between Facebook and Instagram regarding personalization and integration features. These results suggest that the ability to personalize the site and to customize information presentation to users' specific needs, as well as to provide integrated and expended services are similarly recognized by both, Facebook and Instagram users.

RESEARCH IMPLICATIONS
The present study is drawing attention to social network quality by comparing perceived quality of two social networks among students, thus expanding the perspective of academic research in the field of service quality in online environment. The study results will broaden research framework on social network quality measurement.
Furthermore, this study highlights the information safety and users' trustworthiness toward social networks as features with the lowest quality level. Thus, social network administrators could use these results to enhance the features of safety and trust to improve the perceived social network quality that consequently may lead to greater users' satisfaction.
Another study implication is similarities and dissimilarities between the two observed social networks in relation to the perceived quality levels. The distinctiveness of each social network is reflected in the ease of use, trustworthiness, reliability, and information quality. These are the social network quality features that young social network users appreciate more on Instagram than on Facebook. This may indicate social network administrators' way of acting in order to increase the market share, gain new users and retain the present ones.

LIMITATIONS AND FUTURE RESEARCH
Recommendations for future research could be derived from the limitations of present study. One of the shortcomings is a sample size. Although the sample comprises population's characteristics, it should be broadened for conducting a more complex analyses. The present study is limited to a single age group. Although students (and generally young people under 30 years of age) present significant and dominant group of social network users, they are not the only ones that use social networks. Thus, future research should be broadened to other age groups of social network users to be able to conduct a more comprehensive comparison. In addition, other characteristics of social network users might influence their perception of social network quality. Accordingly, future research could investigate differences between different purposes of using social networks, comparing social network quality perceptions between business and private usage. Similarly, additional attention should be given to the effect of using intensity of each social network on the perceived social network quality, since this might reflect the differences in the perceived quality between those using one social network more often than the other.
Another issue is that present study takes into account two social networks. Although Facebook and Instagram are widely used, there are other social networks that are oriented to different users' groups and have respectful market share and growth. Therefore, future research could be oriented on measuring social network quality on these social networks.