How Online Communities Affect Online Community Engagement and Word-of-Mouth Intention

Abstract

The purpose of this research is to examine the impact of online communities on online community engagement and word-of-mouth intention. The current research model was extended and developed into a theoretical model that conceptualizes the relationship between the main concepts to study the influence of utilitarian value, hedonic value, entitativity value, social presence, social support, and community identification on online community engagement, which will therefore affect word-of-mouth intention. Therefore, based on the collected survey questionnaire of 338 responses in Jordan, the results were analyzed using Amos 22. EFA is used to test reliability and validity, analyze the validation of the convergent model and the discriminant validity and analyze the data description. Moreover, this study applied Path Analysis, which was used to check the hypotheses of the studies that were previously developed by researchers. The findings of the seven hypotheses that were conducted to test the research hypotheses indicate that entitativity value, social support, social presence, community identification, and word-of-mouth intention significantly influence online community engagement. However, the utilitarian value and the hedonic value did not influence online community engagement significantly. In addition, the results represent various findings that hold essential implications and accordingly, recommendations are suggested for future marketing research and practices.

1. Introduction

Online communities have grown immensely in their popularity, as was predicted by Eric Schmidt, who is the chairman of Google. Consequently, data have shown that in 2019, 83% of internet users around the world regularly shared large amounts of content, ranging from photos to product reviews online, ranging from 70% of boomers to almost 90% of Gen Z and millennials [1]. Online communities are defined as groups of participants, including persons and/or organizations, which interact together and share goals, interests, or hobbies where this interaction is mediated and supported by technology and driven by group norms and specific protocols [2]. Due to the development of technology and online communities, there is an increase in not only user engagement but also information exchange and other knowledge-sharing activities [3]. The emergence of Web 2.0 technologies, as well as the surge of social media, has led to the development of electronic word of mouth (eWOM), which enabled participants within online communities to exchange information that includes experiences and evaluations [4]. This emergence has led to a new type of communication referred to as social word of mouth [5].

For more than two decades, online communities have existed on the Internet. Existing literature specifies that the reasons that motivate people to join online communities are to seek social support and exchange information [6,7]. Throughout these years, online communities have encouraged participants to come together and learn from each other, discuss real-world issues, and focus on social support, collaborative thinking, and teamwork [5]. With the continuous growth of web and social network usage, it was noted that online communities are seen as an important platform for social interactions across the world as well as an important tool for information exchange. In addition, unexpected uses of online communities include alerting citizens during emergencies, conflicts, and emerging attacks [8]. This indicated that the most essential factor in the sustainability of online communities revolves around user interaction and participation [9]. Participants believe that online communities provide useful and credible word of mouth to the participants [5]. Also, users’ behavior is proven to be influenced by other participants and members of the communities, as well as their motivations to join, such as perceived usefulness, social support, and a sense of belonging [10]. Internet use theories have stipulated entertainment as well as the search for friendships as motivational factors [11].

Finally, it was found that several previous studies concentrated on the impact of online communities that were brand focused, which centered around monetary value and purchase intention concerning E-WOM and online community engagement instead of user-generated online communities [12,13]. Some authors discussed maintaining user participation within an online community, whereas others discussed the importance of content quality and quantity in comparison to others emphasizing that shared beliefs and values play a more evident part in enhancing user participation [14,15]. There are no research studies on the relationship between these three variables together. Therefore, this paper’s objective aims to fulfill the current gap in the previous literature; it is crucial to examine the influence of online communities on online community engagement as well as word-of-mouth intention from the users’ perspective. Thus, this study aims to explore the variables that influence online communities through social media networks in Jordan. Hence, the upcoming questions were formed and will be answered by the end of this research:

RQ1. 

Do online communities influence online community engagement and word-of-mouth intention?

RQ2. 

Do utilitarian value, hedonic value, entitativity value, social support, and social presence influence online communities?

RQ3. 

Do online communities affect word-of-mouth intention?

Finally, the significance of the current research is that it adds and expands the knowledge of the key factors affecting user engagement and participation within online communities.

According to the researchers’ best knowledge, this study is considered the first to examine the influence of online communities on participants’ engagement as well as word-of-mouth intention in Jordan. It investigates the effect of variables (utilitarian value, hedonic value, entitativity value, social support, social presence, online community engagement) on online community engagement, which could directly affect word-of-mouth intention.

The current research includes six independent variables, one mediating variable, and one dependent variable and evaluates their level of significance in online communities; accordingly, many recommendations are suggested to managers and firms.

1.1. Literature Review

1.1.1. Motivation to Join Online Communities

Several terms have been used to describe the online community, such as web community or Internet community; in any case, it is a virtual community where group members meet and interact together through the Internet [16]. The majority of the prior research focused on different forms of motivation, with a larger focus being on the sense of belonging and social factors [10,11,17]. Both group norms and social identity have a significantly positive influence on users’ participation and motivation to join [10]. The group norms in online communities directly influence social identity and lead users to think that they are elected community members [17]. Similarly, other researchers discovered that support and belonging have a larger influence over motivation than ease of use [7,11]. The main motivation to be in an online community initially revolves around information and social support; their findings also suggested that users remain in the community for a longer period due to primarily friendship and entertainment [11]. Furthermore, emotional support results in more user engagement within most online communities than informational support [7].

Other researchers have focused on technological aspects and knowledge exchange and their influence on users to motivate them to join and be involved in online communities [9,18]. For example, due to the development of technology and online communities, there is an increase in not only user engagement but also information exchange and other knowledge-sharing activities [18]. Furthermore, the most essential factor in the sustainability of online communities revolves around user interaction and participation [9]. In contrast, other streams of literature found that geographical differences and invisible organizational culture regarding goals and practices make the user’s intent and motivation to join uncertain [3,9]. This proved to be a challenge for administrators as they need to motivate different types of users to join and exchange knowledge online [9].

1.1.2. Influential Factors behind Word-of-Mouth Intention

The term “word-of-mouth intention” is defined as the ongoing and dynamic information that is exchanged between people concerning a product, service, or brand across online communities and is available to individuals and institutions [19]. Several previous studies revolved around the various factors that influence word-of-mouth intention [20,21,22,23]. Most of them focus on WOM intentions concerning its effect on purchase intention within brand communities and consumer behavior, both online and offline, with a few focusing on WOM intention regarding online communities. Regarding purchase intention, some studies found that the main factors, which influence purchase intention, are trust, the ease of understanding a message concerning word of mouth, and the volume of messages through social media in their ability to influence the purchasing decision. Many authors agree that user experience is the influencing factor behind word-of-mouth and community engagement [24,25]. The main impact behind word of mouth is various personality traits and consumer behavioral attitudes, including tie strength, stickiness, loyalty, and monetary influence [24]. Meanwhile, despite the feelings of warmth and friendship that members might feel with others within a certain community, it still would not be motivating enough to push them to recommend said community to those who are not already members, thus lacking word-of-mouth intentions. Considering social presence is impactful in relation to word-of-mouth influence as it increases customer engagement levels, which they found has an indirect relationship with word-of-mouth communication; their limitations suggested focusing on other social influence variables that could influence community engagement [25]. Other findings suggest that the effect of word of mouth is highly dependent on the sender–receiver relationship; thus, the potential of word of mouth in affecting perceptions and actions is highly dependent on the relationship as well as the richness and strength of the message and its delivery and different personal and situational factors [26].

1.1.3. Maintaining User Participation within an Online Community

There are significant previous studies that aim to analyze the reasons why users socialize within an online community or that discuss the ways user participation is evoked [10,14,15,27,28]. Most studies reported that the popularity of online communities has a huge influence on user participation. For instance, they examined the factors that forecast the popularity of the content of online communities. These factors include the types of posts (entertainment, informational, and social posts), the period of those posts (weekends, peak hours, and time of publication), and the characteristics of the content posted (posts calling for action) [15]. These factors were positively related to the popularity of online communities as well as user interactivity and participation [14,15]. Other researchers agreed that content quality within online communities positively influences online word of mouth as well as user engagement [29,30]. User participation is maintained and enhanced through the quality and quantity of content in an online community. Their findings agreed that users participate and engage more in communities that contain large amounts of information, frequent updates, and active discussions [29]. Users’ desire to share information is enhanced when others engage in conversations that are based on their personal experiences [30]. In contrast, other studies have concluded that efficient user participation is created through identification and internalization, which are more evident in social online communities rather than communities with informational content or remuneration posts. Identification refers to how much users identify with the community and their sense of belonging and attachment [10,27,31,32]. Identification induced the need for users to maintain their social identity, which led to their active participation, while internalization reflected the shared values and goals between the users and their influence [10]. Users become invested in the online community and participate more actively when they believe that their views and goals are encouraged and consistent with those of other users [10,27,31]. Sharing and liking have a positive effect on social capital, while social capital has a positive effect on e-WOM intention [33]. Finally, greenwashing perception has a significant impact on customer brand engagement indirectly through green word of mouth [34].

Despite the emerging stream of literature on online communities, user-centric activities are still not fully understood. Thus, more research is required to fill this gap. Most research on online communities has generally explored the topic from a receiver’s perspective. Further, there has been minimal research revolving around user behavior and their response to user-generated online communities, as most have focused on specific brands or industries. As a consequence, examining user responses towards online communities based on consumer behavior and their influence on engagement and word-of-mouth may assist in achieving a more comprehensive understanding of how users interact, participate, and exchange information and emotional support within those communities. Therefore, this research aims to examine the various influences on Jordanian online communities, taking into consideration the variables that affect said communities, which are utilitarian value, hedonic value, entitativity value, social support, social presence, community identification, online community engagement, and word-of-mouth intention.

1.1.4. Theoretical Framework and Hypotheses Development

In accordance with the previous literature related to online communities, the coming part highlights the theoretical model that conceptualizes the relationship among the main concepts, i.e., the impact of utilitarian value, hedonic value, entitativity value, social presence, social support, and community identification, considering their influences through online community engagement, which will therefore affect word-of-mouth intention.

1.1.5. Utilitarian Value

The term ‘utilitarian value’ is described as the value that an individual receives from the functionality of an online community or a consumption experience of a purchased product [35,36,37]. In this research, utilitarian value refers to the value that a participant obtains from the content quality and functionality of online communities. In other words, users of online communities are concerned with seeking content that is valuable and beneficial, as well as finding a community that contains useful information and is a helpful resource. Participants seek utilitarian benefits as well as hedonic benefits in online communities to varying degrees. Some participants only join communities for their utilitarian values and attributes, especially if their intentions of joining are purpose or task-related [35]. Several previous research studies linked utilitarian value to content quality since it refers to the user’s perception concerning the completeness, accuracy, and relevancy of the information available in online communities [38]. Furthermore, other researchers identified that utilitarian value and content quality act as indicators for understanding and determining community users’ behavior styles [39,40]. Several references described utilitarian value as a motivational factor for joining online communities [41,42,43]. Utility value has a significant positive effect on online community usefulness [43]. As for this current research, it is crucial to study the influence of utilitarian value on online community engagement; hence, the following hypothesis is suggested:

H1. 

Utilitarian value has a significant positive impact on online community engagement.

1.1.6. Hedonic Value

In this research, hedonic value refers to experiences that are mentally stimulating, relaxed, pleasurable, and enjoyable [44]. Brand pages and online communities in previous research suggested that users might derive the personal experience of hedonic value and acquire social interactions where they can communicate and interact with people who share similarities with them [45]. User and customer interaction in band pages offer a source of entertainment and mental stimulation, which highlights the interactivity that classifies social media [46]. Within online communities, there are elements of enjoyment derived through participation with other members and other community admins [47]. Hedonic value has a positive influence on consumer participation via a Facebook fan page [48]. In addition, several studies found that hedonic value significantly influences online communities [49,50]. Participants seek utilitarian benefits as well as hedonic benefits in online communities to varying degrees [35]. Engagement is an accumulation of utilitarian value, consumers’ experiences, hedonic value, and social-psychological value [51]. However, customer brand engagement creates higher hedonic values compared to utilitarian values [52]. As for this research, it is important to investigate the influence of Hedonic value with a focus on online community engagement; hence, the next hypothesis is developed:

H2. 

Hedonic value has a significant positive impact on online community engagement.

1.1.7. Entitativity Value

Entitativity value describes the utility gained by the user of an online community page, such as varying advantages and benefits by remaining as an ongoing entity, single meaningful, as well as remaining linked to others systematically as a clear unit or entity [53]. In this research, the term entitativity refers to the group perception as an entity, formed by its attendant participants. Network entitativity is based on the engagement and interaction level among group members [54]. Community entitativity affects motivations as well as attitudes and stimulates productive sociability and community outcomes [55]. Based on researchers’ knowledge, there is not a sufficient amount of previous research related to entitativity and online communities, as it is mainly related to social networks in general. Therefore, in the current research, it is essential to study the influence of entitativity value on online community engagement; so, the next hypothesis is suggested:

H3. 

Entitativity value has a significant positive impact on online community engagement.

1.1.8. Social Support

The term “social support” refers to an individual’s sense of being responded to, helped, and supported by others [56]; it was also described as multidimensional and categorized into emotional support and informational support [5,6,57]. Informational support is described as responses or messages, which include useful information such as guidance, advice, recommendations, or feedback actions. Meanwhile, emotional support is defined as expressions or responses that include concern, caring, sympathy, and empathy [6,58,59]. In the current research, the term social support describes emotional comfort given to members within online communities through non-verbal online messages that convey emotion and information or that help to reduce uncertainty or stress.

Previous studies contributed to the literature by exploring the value of social support in enhancing the user’s intention to become a part of an online community. For instance, emotional support resulted in more user engagement within most online communities than informational support [5,57]. Moreover, other studies emphasized how the presence of social support in online communities where people share similar interests, share ideas, and discuss experiences has been associated with an increase in credibility via word of mouth [7]. As for the current study, it is crucial to investigate the influence of social support on online community engagement. Therefore, the following hypothesis is suggested:

H4. 

Social support has a significant positive impact on online community engagement.

1.1.9. Social Presence

The term “social presence” indicates the degree to which an individual presumes the participants’ presence in the online community [60]. Other researchers identified it as a measure of the ability of individuals to project emotional presence and experience it with other individuals in online interactions [61,62]. In this research, the term social presence concerns the need for participants of virtual communities to perceive each other as real persons [63].

It was agreed by researchers that social presence is key in online communities since many individuals need to feel that they are connected to others to share ideas and experiences and exchange opinions [63,64]. Taking this idea further, most researchers agree that user interactions and behaviors impact engagement and participation. Discussion about how friendly and humanizing interactions between members of a community influence participants positively, therefore, increasing their social presence by motivating them to become more active [65]. As for this research, it is crucial to study the impact of social presence on online community engagement; hence, the following hypothesis is introduced:

H5. 

Social presence has a significant positive impact on online community engagement.

1.1.10. Community Identification

The term “community identification” refers to a connection between a member and a specific group that relays the deep emotional commitment of the member to that community [66]. Community identification has been discussed by a limited number of studies over the years, and it is referred to as the process by which users identify themselves with a specific community with which they most identify, which also helps them in being distinct from other users [67]. The previous literature indicated that community identification, also referred to as social identity, affects the participation and engagement of a social network user in online communities [68]. For the current research, community identification is viewed as a connection between a member and a specific group, relying on the deep emotional commitment of the member to that community [66].

Furthermore, some researchers agree on the huge impact that a web community may have on its participants [30,69,70,71]. For example, some suggest that participants who strongly identify with a community and have developed a sense of attachment to it are more concerned about the central subject or theme of discussion in the community and, therefore, are greatly committed to it [69,70]. Community identification is the primary basis of a participant’s continuous use intention and is a strong foundation for the development of interpersonal trust and loyalty [71]. Community identification exerts a positive influence on offline and online word of mouth [30]. As for this research, it is crucial to study the influence of community identification on online community engagement; hence, the upcoming hypothesis is formulated:

H6. 

Community Identification has a significant impact on online community engagement.

1.1.11. Online Community Engagement

The term “online community engagement” is defined as the intrinsic feeling between community members to interact and cooperate [72]. Interesting information is acquired by the user and displayed through engagement behaviors like views, likes, comments, and shares [73,74]. Moreover, there is an important contribution to social networks that helps expand the community definition based not only on political, geographical, or religious similarities but also on other similarities like sharing the same interest with other members in the online community [75]. Additionally, one study indicated that online community engagement is very important for active participation within online communities [68]. Furthermore, some studies referred to online community engagement as a dynamic interaction between users and rich generated content that is full of videos, pictures, and audio as “community 2.0” [76]. In addition, another study proposed that online community engagement increases when there is a motivation for community members to engage in collective actions, share what they know, and give enough time to the virtual community [67].

Most studies agreed on online user engagement in connection with online communities with respect to content creation (pictures, posting comments, or videos) and responding to online content such as sharing and liking marketers’ created content or other community members [73,74,77,78]. Hence, this process of engagement directly influences word-of-mouth intention through users’ active participation. As for this study, it was crucial to investigate the impact of online community engagement on word-of-mouth intention. Hence, the next hypothesis is suggested:

H7. 

Online community engagement has a significant positive impact on word-of-mouth intention.

In summary, this section reviewed the previous literature related to the impact of online communities on online community engagement as well as word-of-mouth intention. The gaps in the literature have been identified and discussed in terms of the correlation between online community engagement and the related variables that influence users’ behavior toward online communities. The following sections discuss the need for further exploration of variables that may influence participants’ intentions toward virtual communities. Six independent variables were identified (utilitarian value, hedonic value, entitativity value, social support, social presence), as well as a mediating variable (online community engagement), and lastly, a dependent variable (word-of-mouth intention). The reasons behind choosing these variables have been discussed; moreover, the explanations for the potential impact of the variables on online community users were discussed.

1.1.12. Research Model

The current research model (see Figure 1) was developed based on a research model which included social presence, social support, online community engagement, as well as word-of-mouth intention [25]. This model was later extended based on multiple studies that included hedonic value, community identification, utilitarian value, and word of mouth [79]. Entitativity value acts as a driver of participants’ feedback and collaboration among members in the online community. Entitativity value positively impacts online community engagement towards the online community [38].

2. Materials and Methods

This study followed a quantitative research method to reach a larger audience and create a better response rate since the individual has enough time and less pressure to fill out the survey [80]. It provides a straightforward analysis, and as a result, when you collect and interpret your data, the presented findings are less open to error and subjectiveness [81]. Moreover, this approach guarantees valid and reliable results and could be generalized to a large sample or population, and generates findings that are less open to error and subjectiveness [82]. Moreover, it also requires precise empirical design, which enables anyone to test and copy the results, making the gathered data more reliable [83]. That is why quantitative approach results are valid, reliable, and can be generalized to a large sample or population. [82].

This research uses an online survey to test the influence of online communities on community engagement as well as word-of-mouth intention. Using this kind of survey allows for various participants, collects high-quality data, and employs different instrumentation methods [84,85]. An online survey is the most appropriate procedure for this research, as surveys are a pragmatic way to collect data; they also offer a way to gather substantial amounts of data on any subject from a wide-ranging audience [86]. Moreover, this type of survey allows for various methods to find respondents, collect data, and employ different instrumental methods [84,85].

The main purpose of this survey is to obtain data regarding the sample’s characteristics of interest comparatively quickly [87]. The survey was designed on Google Forms, which is an online survey template for personal use. It was published on social media tools (Facebook, WhatsApp, Instagram, and Twitter), which proved the most convenient and easiest way of collecting primary data. The survey, using a Likert scale, included various questions that ranged from the participant’s gender and age to their experiences with opinions on online communities. The data analysis was tested by using AMOS since it is the best statistical technique to test the research model.

2.1. Survey Instrument

In this study, a questionnaire was prepared to collect data, measure the effect of each variable, and verify the research hypotheses. A survey was conducted with 28 questions, testing each variable. Based on this, the survey was designed on Google Forms, which is an online survey template for personal use, in which documents and spreadsheets can be created, edited, and stored online. The survey was organized in this manner; the first two questions were demographic questions posed before the start of the survey (Gender, Age). In the second part, questions 1–3, examined the utilitarian value; questions 4–6 examined the hedonic value, questions 7–9 tested the entitativity, questions 10–15 tested Social Presence, questions 16–18 examined Social Support, questions 19–21 examined Community Identification, questions 22–24 tested Online Community Engagement, and questions 25–28 examined word-of-mouth intention. From question number 1 to question number 28, a Likert scale of five grades ranged from strongly unimplemented to strongly implemented.

The researchers guaranteed the clarity and accuracy of the questions to avoid any misunderstanding that may have occurred when the participants filled out the survey. The survey was organized in a way that guaranteed understanding on the part of the participant. It was also tested by the researcher’s supervisor and modified based on the given feedback.

2.2. The Scales and Response Format

The survey questions were developed on existing scales and measures derived from the existing literature. Questions in the survey were measured according to a five-point Likert scale sorted from “strongly agree” to “strongly disagree” and included a “neither” option to help respondents who preferred to remain undecided. The construct of the research was as follows: three items measured Utilitarian Value [38]. Three items also measured Hedonic Value [88], while Entitativity Value was measured by three items [89]. Social Presence was measured through six items [90]. In addition, Social Support was measured by three items [91]. Community Identification was measured by three items [92]. Four items measured online Community Engagement [90]. Lastly, word-of-mouth intention was measured by using two items [93,94]. For more details, see Appendix A.

2.3. The Population of the Study

The current research considers the impact of online communities on online community engagement as well as word-of-mouth intention. The research population consisted of social media users and online community participants, including both males and females of various age groups in Jordan, who chose to fill out the survey. Therefore, every user needed to understand what factors affect their decision to engage within an online community and how this would be reflected in their participation and word-of-mouth intentions, whether positive or negative.

2.4. Sample of the Study

A convenience sampling method was used to collect the data via an online research survey. This sampling technique included specifying the population to be surveyed. The main reason behind using the convenience sampling technique was that it is inexpensive and is an easier option for collecting data when compared to other sampling techniques [95]. Moreover, this technique allows data to be collected in a short period of time and is considered the most useful for hypothesis generation and testing [96]. The sampling strategy that was used was judgmental, where the targeted audiences were members of existing online communities. This sampling technique was used because the sample members were chosen based on their knowledge and judgment. Moreover, judgmental sampling results in more accurate data with minimum errors since the members of the sample are not randomly chosen [97]. An average of 338 responses were used to define the sample size of the current research. To justify that the sample size conducted for the current research survey was within the required range, some researchers indicated that the lower limit would be five respondents per each variable to be analyzed, whereas the most acceptable way of determination is ten samples for one variable [98].

The survey was published on four social media platforms (Facebook, WhatsApp, Twitter, and Instagram). This proved to be the most convenient and easiest way of collecting the primary data needed for the research. The published post included a summary of the purpose of the research, as well as the survey link, which directed participants to the Google Form where they could fill out the online survey.

2.5. Data Collection

A total of 338 valid surveys were obtained and distributed in Jordan; the results were analyzed using Amos 22. EFA was used to construct validity and reliability analysis, analyze the validation of the model convergence and discriminant validity, and analyze the descriptive statistics. Moreover, the study applied Path Analysis to check the hypotheses of the studies that were previously developed by researchers. The argumentation behind testing the research model using AMOS is that the study model includes a high-order construct that contains first-order indicators; so, AMOS is the most suitable statistical method for checking the research model.

3. Results

3.1. Sample Characteristics

A cross-sectional design was used to collect the required data, which was published online through Google Forms. A total of 338 surveys were collected, and all surveys are valid for statistical analysis.

Table 1. Demographics Analysis.

	Gender	Frequency	%
Gender	Female	215	63.6
	Male	123	36.4
	Total	338	100.0
Age	15–25	178	52.7
	26–36	67	19.8
	37–47	56	16.6
	48+	37	10.9
	Total	338	100.0

shows respondents’ demographic characteristics along with gender and age.

The researchers distributed online to various social media platforms by sharing the survey on Facebook and online communities on Facebook, as well as WhatsApp, Twitter, and Instagram. From the data shown above in Table 1, it is seen that there were 215 females (63.6%) and 123 males (36.4%). As for the ages, the majority were 15–25 years old, which included 178 people (52.7%), and those that were 26–36 years old totaled 67 people. The rest of the sample varied in age from 37–47 years and totaled only 56 people (16.6%), and those aged above 48 years totaled 37 people (10.9%).

3.2. Reliability Test and Validation of Model

3.2.1. Exploratory of Factorial Analysis (EFA)

EFA is a statistical analysis “that accounts for common variance among a set of items by their linear relations to latent dimensions” [99]. Some people use the Congeneric Latent Construct Estimator to address the problem of estimating latent unidimensional constructs [100]. Factor analysis and sum scoring both are used for latent variables, though they are different [101]. However, in this study, the standard that defined the rule of thumb when performing an EFA was factor loadings higher than 0.60 and cross-loadings lower than 0.30. When applying these standards, no questions were ignored, and for all twenty-eight items of the eight constructs, the factor loads were above the threshold (0.60) [102,103]. Moreover, the Kaiser–Meyer–Olkine (KMO) value indicates harmony and coherence, and the inter-relationship of the data was KMO = 0.949, df = 378, p < 0.05 = 0.000, which is higher than the threshold of 70% [104,105,106]. Furthermore, the Bartlett test for sphericity was χ² = 6814.632, df = 253, p < 0.05 = 0.000. Hence, data suitable for Path Analysis could be judged and reported (Outer Loading Matrix), as shown in

Table 2. Outer Loading Matrix.

	1	2	2	3	5	6	7	8
CI1	0.905
CI2	0.900
CI3	0.912
EV1		0.860
EV2		0.838
EV3		0.792
HV1			0.861
HV2			0.880
HV3			0.886
OCE1				0.890
OCE2				0.901
OCE3				0.822
SP1					0.783
SP2					0.785
SP3					0.837
SP4					0.809
SP5					0.814
SP6					0.698
SS1						0.912
SS2						0.919
SS3						0.866
UV1							0.896
UV2							0.845
UV3							0.889
WOM 1								0.848
M 2								0.894
WOM 3								0.913
WOM 4								0.845

Utilitarian Value (UV); Hedonic value: (HV); Entitativity Value (EV); Social Presence: (SP); Social Support: (SS); Community Identification: (CI); Online Community Engagement (OCE); Word-of-mouth intention: (WOM). Kaiser–Meyer–Olkine = 0.949; df = 378; Sig = 0.000. Bartlett = 6814.632; df = 378; Sig = 0.000.

. However, there was a need to verify the variables’ validity (convergent validity, reliability analysis, and discriminant validity).

3.2.2. Reliability Analysis

Reliability analysis “is a psychometric property of a specific sample’s responses to a measure administered under specific conditions” [107]. Cronbach’s Alpha indicates internal consistency [108,109]; it is biased on reliability compared to composite reliability [110]. The advised starter of CR (composite reliability) is 0.80, and for (α), the Cronbach alpha is 0.70 [111].

The present study shows that (α) Cronbach’s alpha is more than 70% for all scales (e.g., 0.775–0.850), and the Composite Reliability (CR) of every construct is more than 80% (0.869–0.932). On the other hand, convergent validity through the Average Variance Extracted (AVE) indicates the agreement between concepts (i.e., a latent construct), so it is a suitable measuring instrument (i.e., several elements that estimate the latent construct) [112]. This type of validity is determined via the extracted average variance that described the shared average variance between variable and variable components [113].

Table 3. Reliability, Validity, and Descriptive Analysis.

Variables	Mean	AVE	CR	α	Std. Dev.
Community Identification (CI)	3.167	0.820	0.932	0.891	1.128
Entitativity Value (EV)	2.334	0.690	0.869	0.775	0.879
Hedonic Value (HV)	3.857	0.767	0.908	0.848	0.913
Online Community Engagement (OCE)	3.436	0.760	0.905	0.841	1.111
Social Presence (SP)	3.398	0.622	0.908	0.879	0.911
Social Support (SS)	2.401	0.808	0.927	0.882	0.994
Utilitarian Value (UV)	3.167	0.769	0.909	0.850	1.128
Word-of-mouth Intention (WOM)	2.334	0.766	0.929	0.898	0.879

Notes: Values of means, 5 = strongly agree and 1 = strongly disagree. To determine the agreement of answers of the sample members of the study on each variable, “If the arithmetical mean has a value between 1 and 1.79, the result of the axis strongly disagrees., A value between 1.80 and 2.59, the result of the axis disagrees. With a value between 2.60 and 3.39, the result of the axis is neutral. A value between 3.40 and 4.20, the result of the axis is agreed. Finally, if the value is between 4.21 and 5, the result of the axis is strongly agreed”.

shows that the AVE of the eight variables was higher than the threshold (60%) [114]; they ranged from 0.622 to 0.820, which indicates that the scales’ items are related; so, the study’s instrument is valid.

In addition, Table 3 shows a descriptive analysis of the respondents’ answers. The average for Community Identification is 3.167, Entitativity Value is 2.334, Online Community Engagement is 3.436, Social Presence is 3.398, Social Support is 2.401, Utilitarian Value is 3.167, and Word-of-mouth Intention is 2.334.

3.2.3. Model Validation (Convergent and Discriminant Validity)

Before conducting structural analysis, the model needed to pass some assumptions to check the model’s suitability; the results for this were as follows (see Table 3 and

Table 4. Model Fitness.

Index	Default Model	Rule of Value Index	Reference	Results
Minimum discrepancy (X2/df)	3.123 (p 0.05 ≥ 0.021)	X2/df < 5 (p > 0.05)	[115]	Suitable
Comparative fit index (CFI)	0.872	CFI > 80%	[116]	Suitable
Normed fit index (NFI)	0.937	NFI > 90%	[117]	Suitable
Root mean square error of approximation (RMSES)	0.062	RMSES < 0.10	[118]	Suitable

):

The confirmatory factor analysis findings show that all results are within the recommended threshold, as the results of previous tests indicated the possibility of relying on structural equation models in the study’s hypotheses. Also, the results in Table 3 and Table 4 show that all results pass the suggested thresholds and indicate model correctness (fit) and suitability; also, the discriminant validity had to be tested, which indicates whether the level of items is related or not related [111]. Discriminant validity is assured by comparing and contrasting the correlation coefficients of the construct and the square roots of the AVE, which must be more than other constructs [119].

Table 5. Discriminant Validity Matrix.

	CI	EV	HV	OCE	SP	SS	UV	WO M
CI	0.906
EV	0.522	0.830
HV	0.504	0.509	0.876
OCE	0.795	0.459	0.463	0.872
SP	0.693	0.640	0.535	0.743	0.789
SS	0.589	0.542	0.538	0.498	0.636	0.899
UV	0.536	0.502	0.528	0.475	0.555	0.549	0.877
WOM	0.663	0.501	0.493	0.670	0.623	0.573	0.615	0.875

indicates that the construct correlation coefficient is less than the square roots of the AVE. Therefore, this proves that the validity discriminant of the study model is satisfactory; so, the results can be generated and disseminated to the research population.

3.3. Testing Hypotheses

To check the suggested model, Path Analysis via structural equations was used. This test is capable of analyzing complicated casual models and thus is regarded as the modeling of the covariance-based structural equation [112]. When the structural equation is used in testing a model, four values should be revealed: p-value and path coefficients (β), and this will clarify the effect size (f²) throughout the study [102]; moreover, it will clarify the variation in a variable by R² [98,120].

Table 6. Tested Hypotheses.

Hypotheses	Effect Size (f2)	Path Coefficients (β)	t-Value	Std. Error	p-Value p < 0.05	R2	Decision
UV…OCE	0.000	0.001	0.031	0.038	0.975	0.648	H1: Rejected
HV…OCE	0.000	0.041	1.257	0.036	0.209		H2: Rejected
EV…OCE	0.027	0.108	3.340	0.037	0.000		H3:Accepted
SP…OCE	0.257	0.481	14.874	0.037	0.000		H4:Accepted
SS…OCE	0.024	0.097	3.014	0.033	0.003		H5:Accepted
CI…OCE	0.523	0.628	19.438	0.029	0.000		H6:Accepted
OCE…WOM	0.386	0.638	15.218	0.043	0.000	0.407	H7:Accepted

Effect size F²: > 0.02 = No effect F²: (0.02 to 0.15) = Small effect. F²: (0.16 to 35) = Moderate effect. F²: ≥ 0.36 = Large effect [102]. R²: > 0.19 = Reject. R²: (0.20 to 0.33) = Weak. R²: (0.34 to 0.50) = Moderate. R²: ≥ 0.51 = Strong [98,120].

and Figure 2 show the cut-off points of the tests. The Structural Equation Analysis results show:

H1: “Utilitarian value has a significant positive impact on online community engagement”.

Hypothesis 1 is rejected (Path Coefficients (β) = 0.001; t-value = 0.031; p < 0.05; = 0.975).

H2: “Hedonic value has a significant positive impact on online community engagement”. Hypothesis 2 is rejected (Path Coefficients (β) = 0.041; t-value = 1.257; p < 0.05; = 0.209).

H3: “Entitativity value has a significant positive impact on online community engagement “. Hypothesis 3 is accepted (Path Coefficients (β) = 0.108; t-value = 3.340; p < 0.05; = 0.000); this is counted as a small effect (f² = 0.027), so increased entitativity value leads to increased online community engagement.

H4: “Social presence has a significant positive impact on online community engagement”. Hypothesis 4 is acceptable (Path Coefficients (β) = 0.481; t-value = 14.874; p < 0.05; = 0.000); this is counted as a medium effect (f² = 0.257), so increased social presence leads to increased online community engagement.

H5: “Social support has a significant positive impact on online community engagement”. Hypothesis 5 is accepted (Path Coefficients (β) = 0.097; t-value = 3.014; p < 0.05; = 0.003); this effect is counted as small (f² = 0.097), so increased social support leads to increased online community engagement.

H6: “Community identification has a significant positive impact on online community engagement”. Hypothesis 6 is accepted (Path Coefficients (β) = 0.628; t-value = 19.438; p < 0.05; = 0.000); this is counted as a small effect (f² = 0.523), so increased community identification leads to increased online community engagement.

The variation in online community engagement that was presented by four online community characteristics represented by entitativity value, social support, social presence, and community identification was strong (R² = 0.648). This value points to the process of interpretation and prediction of online community engagement being reliable.

H7: “Online community engagement has a significant positive impact on Word-of-mouth intention”. This hypothesis is accepted (Path Coefficients (β) = 0.638; t-value = 15.218; p < 0.05; = 0.000); this is counted as a large effect (f² = 0.386), so increased online community engagement leads to increased word-of-mouth intention.

Finally, the credibility variance of word-of-mouth intention presented by online community engagement was moderate (R² = 0.407). However, this value is reliable for the prediction process and interpretation of word-of-mouth intention.

In summary, Path Analysis was used to check the hypotheses of the study; it was concluded that the hypotheses regarding Hedonic Value and Utilitarian Value were rejected, while the hypotheses regarding Entitativity value, Social Support, Social Presence, Community Identification, and Online community engagement were accepted.

4. Discussion

The purpose of this research is to examine the impact of online communities on online community engagement and word-of-mouth intention in Jordan. In this section, the findings related to the hypotheses will be debated. To explore the impact of the online community on users’ engagement, the data were collected and analyzed to reach the primary results for the following study. A comprehensive model was developed that included the following independent variables: utilitarian value, hedonic value, entitativity value, social support, and social presence, as well as a mediating variable (online community engagement), and lastly, a dependent variable (word-of-mouth intention).

The results of the path analysis suggest that the variance in online community engagement is explained by four online community characteristics (entitativity value, social support, social presence, and community identification), which were strong (R² = 0.648). The empirical outcomes submitted strong evidence for the explanatory strength of the current model. In particular, this research found that the four variables (entitativity value, social support, social presence, and community identification) have a direct influence on online communication engagement value that in turn will affect word-of-mouth intention. However, the empirical outcomes also showed that both hedonic value and utilitarian value neither have an impact on online community engagement nor word-of-mouth intention. Moreover, the standardized coefficients (β) are different for each independent variable. Communication identification (β = 0.628) had the largest positive influence on online community engagement, while social presence value had a moderately positive effect on online community engagement (β = 0.481), while social support (β = 0.097) and entitativity value (β = 0.027) had a small impact on online community engagement. Moreover, the mediating variable (online community engagement) had a considerably large influence on the independent variable (word-of-mouth intention) (β = 0.638).

Research Hypotheses

The utilitarian value was hypothesized to have a significant positive influence on online community engagement. Previous studies agreed that there is a positive correlation between utilitarian value and online communities [42,43]. Utility value significantly and positively affects online community usefulness [43]. However, the results indicated that the hypothesis is rejected, which means that utilitarian value did not have a significant positive impact on online community engagement. A reasonable justification for this is that utilitarian appeals distract users from interacting within the online community for social and entertainment purposes and are often viewed as an attempt to control them, which makes them reduce their inherent motivations to engage [78,121]. The hedonic value was suggested to have a positive influence on online community engagement. The findings indicated that hedonic value did not have a significant positive influence on online community engagement. Concerning online communities, to the best of the researcher’s knowledge, there were no studies that explored the impact of hedonic value on online communities; thus, the knowledge regarding such relationships has been enhanced by the current research. The impact of hedonic value was mostly studied concerning interactions within a social network’s context in general, and it was found that there is a positive impact between hedonic value and social networks. The current study’s result contradicts some previous studies’’ results related to the effect of hedonic value on social media users [47,48,49,50].

The entitativity value was hypothesized to have a significant positive effect on online community engagement. The analysis findings show that the entitativity value has a significant positive impact on online community engagement; this is a consistent result, with network entitativity strongly based on the interaction levels among members [54]. Entitativity influences attitudes and motivations [55].

Social presence was hypothesized to have a significant positive impact on online community engagement. The results indicate that social presence significantly positively influences online community engagement. Most researchers have identified it as a measure of the ability of individuals to project emotional presence and experience it with other individuals in online interactions, which positively influences community engagement and participation [61,62]. For instance, regarding the online community, the present study indicates that online community engagement has a significant influence on social presence [63]. Moreover, they discussed how friendly and humanizing interaction between members of a community influences participants positively, therefore increasing their social presence by motivating them to become more active [65].

Social support was hypothesized to have a significant positive impact on online communication engagement. The results found that society significantly positively influences online community engagement; this result is consistent with previous studies that emphasized how the presence of social support that affects online communities where people share similar interests, share ideas, and discuss experiences has had a positive influence on how participants engage and participate [7,122]. Moreover, the study results are consistent with previous results on the positive correlation between social support and community participation as well as being the biggest motivation for joining a community [5,6,56,58,59].

Community identification was hypothesized to have a significant positive impact on online community engagement. The results found that community identification significantly and positively affects online community engagement. Concerning community identification, the present study found that it has a significant positive impact on online community engagement. This is matching with those who suggested that participants who strongly identify with a community and have developed a sense of attachment to it are more concerned about the central subject or theme of discussion in the community and, therefore, are greatly committed to it [30,71]. Community identification is the primary basis of a participant’s continuous use intention and is a strong foundation for the development of interpersonal trust and loyalty [71].

Online community engagement was hypothesized to have a significant positive influence on word-of-mouth intention. This result is consistent with previous studies showing that online community engagement increased when community members are motivated to share their knowledge, engage in collective actions, and give enough time to others in a virtual community. Hence, this proved that engagement directly influences word-of-mouth intention through users’ active participation [67]. Another study agreed that community engagement impacts word-of-mouth intention in a way such that it becomes the most influential information source on the internet [123]. Furthermore, several studies agreed that online community engagement is important for active participation within online communities and credible word-of-mouth intention [68].

5. Conclusions

Throughout the past decades, online communities have grown immensely in their popularity, and most of the world has internet access [124]. Furthermore, the continuous growth of the web and social networks has led to the emergence of online communities as important platforms for social interactions. Moreover, online communities have encouraged participants to come together and learn from each other, discuss real-world issues, and focus on social support, collaborative thinking, and teamwork [7]. Thus, the current research considered exploring the effect of online communities on online community engagement as well as word-of-mouth intention by answering the previously stated research questions to reach the study objective. The suggested model has made a significant theoretical contribution. The authors believe it has explained and predicted the first analysis of the factors influencing online communities regarding online community engagement as well as word-of-mouth intention. This contribution related to the study was achieved through the comprehensive extended model. The findings show that entitativity value, social support, social presence, community identification, and word-of-mouth intention significantly influence online community engagement. However, unexpected results show that the utilitarian value and the hedonic value did not affect online community engagement significantly. Finally, the study’s findings indicated that the suggested model proves to exhibit adequate explanatory power and verifies its helpfulness and strong contribution in predicting what elements affect online communities. This study reported the analysis findings to address the hypotheses of the research and any supplementary explanations, including the results’ implications, and limitations of the research, as well as recommendations and suggestions for future researchers.

5.1. Theoretical Implications

From a theoretical perspective, this research expands and adds more insights to the general knowledge of the most important factors that influence online communities that will thereby affect online community engagement. Consequently, the current study applied proven theories and concepts in online community research and expanded the theoretical correlations between the concepts related to the research model. Particularly, this model increases our understanding of the most important factors that affect user participation and engagement in online communities, which in turn will affect word-of-mouth intention in Jordan by combining parts and elements from the online community literature. The model has been extended based on a model by [25] to cover all the aspects of this study.

This research is considered the first to examine the relationship between online community engagement and word-of-mouth intention in Jordan. Moreover, it is also the first to test the impact of such variables (utilitarian value, hedonic value, entitativity value, social presence, social support, and community identification) on online community engagement, which in turn will affect word-of-mouth intention behavior. This study’s results provide new information concerning the online community engagement context from a user perspective, and it examined users’ behaviors and responses to user-generated online communities.

Theoretically, the outcome of the research adds new perspectives and insights to the sphere of online communities through various variables that could influence engagement and word of mouth within online communities in a generalized way, in addition to the knowledge added to better understanding the user’s behavior. Therefore, this research may be considered the first to assess the research model variables in the context of user-generated online communities where brands and monetary value are not involved.

It was found that social presence has the biggest positive impact on participation and engagement, whereas community identification, social support, and entitativity value had a less apparent positive impact. Thus, this proved that social presence, also referred to as social identity, has a positive impact on the participation and engagement of a social network user in online communities [68]. Moreover, it was discovered that online community engagement has an evident, positive influence on word-of-mouth intention. Hence, it is emphasized that active online community engagement in online communities where people share similar interests, share ideas, and discuss experiences increases word-of-mouth and credibility [7].

5.2. Practical Implications

From a practical perspective, this research studies the influences that impact online communities, and according to the research results, expands the existing research surrounding this topic and could help guide online community administrators and those within the industry on deciding on influential factors regarding their strategies. According to this study’s results, it was found that most people become engaged in online communities mainly for social aspects; hence, it is suggested that researchers pursue further study regarding those aspects to expand their understanding of online social support. Moreover, this could help online community admins to enhance their understanding of user behavior, hence allowing them to elicit members’ engagement and participation [7].

In addition, it was found that community identification had the biggest impact on online community engagement. Based on this result, it is recommended that social online communities, as well as brand community administrators, create relevant, compelling content that members would find easy to identify with [71]. Moreover, companies or brands that have online communities are advised to implement segmentation strategies to support positive behaviors and enhanced participation from participants. Furthermore, it was found that social presence impacts online community engagement; thus, it is recommended that social online community creators and brand community administrators focus on creating interactions that are humanizing and that project emotional presence to attract targeted members ([62]. Finally, it was discovered that online community engagement impacts word-of-mouth intention; hence, it is suggested that companies or brands that own online communities focus on maximizing socialization in both social presence and support to enhance engagement, which directly affects positive word-of-intention [26].

5.3. Limitations, Recommendations, and Future Research

This study has a few limitations that should be noted. First, the research was cross-sectional and assessed the respondents’ perception at a specific and limited period. Second, this study discussed the broad concept of online communities, mainly focusing on user-centric activities, and did not consider specific products or brands, which therefore lead to more generalizable results. Thirdly, the current research does not focus on a specific type of online community and social media, which led to more general results. Lastly, the number of female respondents was more than that of male respondents.

Furthermore, for future research, it would be useful to conduct a study on the impact of online communities in other scopes. The current study has many limitations that could help future studies gain more credible results. First, we had chosen online community engagement as a mediating variable so that researchers could test the chosen independent variables on different mediating variables for more comprehensive research. Additionally, this study included a dependent variable, which was word-of-mouth intention, while future studies can replace it with another variable that could be more beneficial. Moreover, the focus of this study was on the local scale in Jordan; therefore, we recommend that future studies conduct their research in another country to allow cross-country validation. Many factors influence online community engagement. This study only discussed the following variables: utilitarian value, hedonic value, entitativity value, social presence, social support, community identification, online community engagement, and word-of-mouth intention, and it only considered their impact on online communities; hence, future researchers could include other variables in their study, such as trust, participation, and content quality.