Engaging with Facebook Health Support Groups among Australian Culturally and Linguistically Diverse Populations

Abstract

Objectives

This study aims to explore the key factors that enhance engagement in Facebook health support groups among Australian culturally and linguistically diverse (CALD) communities.

Methods

A cross-sectional online survey was conducted using convenience sampling. A total of 1,145 CALD participants residing in New South Wales, Australia, were initially recruited. From this sample, 150 participants who self-reported regular engagement with Facebook health support groups were included in the final analysis. A pilot test (n = 30) demonstrated strong internal consistency (Cronbach’s alpha >0.70). Data collection involved a structured questionnaire employing a 7-point Likert scale to assess factors such as motivation, trust, perceived support (received and given), social connectedness, and sense of virtual community.

Results

Motivation and trust significantly influenced both support dynamics and the perceived sense of virtual community. The sense of virtual community, in turn, strongly predicted engagement in Facebook health support groups. Interestingly, social connectedness alone was not a significant predictor of engagement.

Conclusions

Fostering a strong sense of virtual community appears to be a critical factor in encouraging sustained engagement in digital health platforms among CALD populations.

I. Introduction

Australia is among the most culturally and linguistically diverse (CALD) nations, with nearly half its population being first- or second-generation migrants [1]. Although digital access is widespread, many CALD communities remain digitally excluded due to socio-economic barriers, limited English proficiency, cultural norms, and low digital literacy [2]. This digital divide contributes to inequities in accessing online health information and support. Social media platforms, particularly Facebook, play an important role in fostering health communication by enabling CALD individuals to seek advice, share experiences, and build a sense of belonging [3]. Nonetheless, engagement patterns remain under-explored—especially regarding how psychological and social factors influence sustained participation. Previous studies have examined consumer engagement frameworks broadly but have rarely focused on CALD groups within digital health contexts [4]. This study addressed this gap by investigating the key drivers of engagement among CALD individuals in Facebook health support groups. More specifically, this study drew on social capital theory (SCT), which emphasises trust, reciprocity, and resource exchange, and social identity theory, which highlights belonging and shared identity [5,6]. Combined, these frameworks explain how motivation, trust, support (given and received), connectedness, and sense of virtual community shape engagement (Figure 1). In this research, we use an online survey to examine the following question:

Figure 1.

Theoretical framework.

Research Question: To what extent do motivation, trust, perceived social support (given and received), sense of virtual community, and social connectedness influence the engagement of CALD communities in Facebook health support groups in New South Wales (NSW), Australia?

Social capital is categorized into bonding, bridging, and linking [7]; collectively, these forms facilitate collective action, provide emotional support, promote trust, and enable the sharing of resources. Social identity theory encompasses attitudes, belonging, and categorization [8]; this study emphasises belonging to address CALD individuals’ needs for digital community.

In our model, bonding arises from motivation and support, fostering closeness; bridging stems from sense of virtual community and connectedness, broadening social ties; and linking derives from trust in credible health information. Together, these elements foster action, support, trust, and sharing, with motivation reflecting exchange, reciprocity, and belonging. Building on these theoretical foundations, we propose two hypotheses:

• - Hypothesis 1 (H1): Joining motivation (JM) in an online community or group has a positive and significant impact on giving support (GS) online posts.

• - Hypothesis 2 (H2): JM in an online community or group has a positive and significant impact on receiving support (RS) online posts.

In SCT, Trust (TR) is central to both bonding and bridging capital, as it shapes individuals’ willingness to GS and RS. Trust builds strong relationships, facilitates resource sharing, and reinforces group identity. Therefore, we propose the following hypotheses:

• - Hypothesis 3 (H3): TR in an online community or group has a positive and significant impact on GS.

• - Hypothesis 4 (H4): TR in an online community or group has a positive and significant impact on RS.

Sense of virtual community (SOVC) aligns with social identity theory, which emphasizes belonging and group identification. Within this framework, SOVC represents the digital expression of these in-group ties. GS and RS in online communities enhance members’ sense of belonging and value, reinforcing social identity and participation. Based on this reasoning, we propose:

• - Hypothesis 5 (H5): GS online has a positive and significant impact on SOVC.

• - Hypothesis 6 (H6): RS online has a positive and significant impact on SOVC.

In SCT, Connectedness (CON) encompasses bonding ties that foster belonging and support, as well as bridging ties that expand access to diverse resources. This dual role underscores the importance of networks for resilience and advancement, with online interactions reinforcing supportive virtual relationships. Accordingly, we propose:

• - Hypothesis 7 (H7): GS online has a positive and significant impact on social CON.

• - Hypothesis 8 (H8): RS online has a positive and significant impact on social CON.

SCT also posits that social capital drives engagement. Bonding ties provide emotional support, bridging ties expand perspectives and resources, and linking ties with professionals or organisations build confidence. Together, network structure and the SOVC are key determinants of EN in online health groups. Thus, we propose:

• - Hypothesis 9 (H9): SOVC has a positive and significant impact on EN.

• - Hypothesis 10 (H10): CON has a positive and insignificant impact on EN.

Overall, the framework suggests that engagement is influenced by SOVC and CON, which in turn are shaped by support (GS and RS), JM, and TR (Figure 1).

II. Methods

1. Study Participants and Sampling Technique

This study employed an online survey using the convenience sampling method, in which participation was entirely voluntary and contingent upon participants’ willingness to take part (Table 1).

Table 1.

Demographic information of the participants

Characteristic	n (%)
Ethnic background
Arabic	74 (49.3)
Indian	68 (45.3)
Other	8 (5.4)
Sex
Male	75 (50.0)
Female	67 (44.7)
Prefer not to state	8 (5.3)
Year of birth
1954–1973	14 (9.3)
1974–1993	112 (74.7)
>1993	24 (16.0)
Marital status
Married	106 (70.7)
Not married	44 (29.3)
Living status
Living alone	35 (23.3)
Living with family	114 (76.0)
Other	1 (0.7)
Education
Didn’t complete high school	4 (2.7)
High school/Technical college	35 (23.3)
University	111 (74.0)
Religion
Islam	106 (70.7)
Christianity	17 (11.3)
Hinduism	14 (9.3)
Other	13 (8.7)
Employment
Employed	102 (68)
Not employed	48 (32)

2. Sample Size and Recruitment Method

Data were collected over 6 months (February–July 2022) using a multi-modal approach. The online survey was distributed via Facebook, University of Wollongong student networks, CALD groups, forums, and other social media platforms. QR codes on flyers and paper surveys were provided to enhance accessibility. To ensure inclusivity, the survey was available in English, Arabic, and Chinese. The sample size for this study was determined using Cochran’s formula [9]:

$$n=\frac{P\hspace{0.17em}(1-P)\hspace{0.17em}{Z}^2}{e^2}$$

We calculated the required sample size as follows:

$$n=\frac{0.5\hspace{0.17em}(1-0.5){1.96}^2}{{0.05}^2}=384.16$$

It is advisable to enlarge the sample size by 30% to mitigate the impact of potential nonresponse [10], Therefore, we adjusted our target sample size as follows: adjusted target = 499.40 ≈ 500

Of the 1,145 CALD participants who accessed the survey, only those who answered “Yes” to the screening question on using Facebook health support groups were eligible (n = 192). Of these, 42 were excluded due to more than 5% missing data, leaving a final sample of 150. This number was adequate for partial least squares structural equation modelling (PLS-SEM) analysis, as prior research supports samples of at least 150 for models with up to seven constructs [11]. The structured survey comprised seven core constructs—JM, TR, GS, RS, SOVC, CON, and EN—each grounded in established theories and validated scales from previous research. Full measurement items of each construct are described in Table 2.

Table 2.

Full measurement items of each construct

Construct	Survey items
Joining motivation (JM) [12]	I joined the Facebook health support group to receive sympathy and understanding from others.
	I joined the Facebook health support group to receive emotional support.
	I joined the Facebook health support group to discuss my feelings with other people who share a similar experience.
	I joined the Facebook health support group because I did not feel I had enough support from other friends and family members.
	I joined the Facebook health support group because I wanted to find out information about my health status.
Trust (TR) [13]	I feel that the Facebook health support community is reliable.
	I feel that the Facebook health support community is honest.
	I feel that the Facebook health support group is likely to care for members’ welfare.
	I feel that the Facebook health support group is competent to help members enhance their knowledge.
	I feel that the Facebook health support community is reliable.
	I feel that the Facebook health support community is honest.
	I feel that the Facebook health support group is likely to care for members’ welfare.
	I feel that the Facebook health support group is competent to help members enhance their knowledge.
Giving support (GS) [12]	How often have you provided emotional support to other Facebook health support group members?
	How often have you discussed your feelings with other Facebook health support group members?
	How often you have shared information with other Facebook health support group members?
Receiving support (RS) [12]	How often have you received emotional support from other Facebook health support group members?
	How often have you received information from other Facebook health support group members?
Sense of virtual community (SOVC) [14]	I think the Facebook health support group is a good place for me to be a member.
	I feel Facebook health support group members and I share the same values.
	I feel Facebook health support group members and I share the same expectations
	I feel that I can recognize the names of most members of the Facebook health support group.
	I feel that I am at home when I join the Facebook health support group.
	I feel that my problems can be solved by the members of the Facebook health support group.
	I feel It is very important to me to be a member of the Facebook health support group.
	I feel that I will stay in the Facebook health support group for a long time.
	I anticipate how some members will react to certain questions or issues in Facebook health support group.
	I feel that I am getting a lot out of being in the Facebook health support group.
	I have had questions that the Facebook health support group has answered.
	I have gotten support from the Facebook health support group.
	I feel that I can make friends by participating in the Facebook health support group.
	Some members of the Facebook health support group are very friendly and supportive to help others.
	I feel obligated to help other members in the Facebook health support group.
	I like the Facebook health support group as it keeps me connected with friendly and supportive people.
	I feel that Facebook health support group mean a lot to me.
Connectedness (CON) [15]	As a member of the Facebook health support group, I feel more connected with the world around me.
	I feel that I really belong to the Facebook health support group.
	I feel close to the members of the Facebook health support group.
	I have sense of togetherness with my peers in Facebook health support group.
	I feel related to anyone in the Facebook health support group.
	As a member of the Facebook health support group, I have a sense of connectedness with society.
	Among my friends in the Facebook health support group, I feel there is a sense of brother/sisterhood.
	I participate with the Facebook health support group to have the feeling of connectedness.
Engagement (EN) [16]	The Facebook health support group is easy to use.
	I can easily find content on the Facebook health support group.
	I can easily post comments or content onto the Facebook health support group.
	I can easily access the Facebook health support group using the smart devices.
	I learn to use the Facebook health support group quickly.
	I feel the Facebook health support group is useful.
	I feel the Facebook health support group helps me to connect with other users.
	I feel the Facebook health support group is fun to use.
	I feel the Facebook health support group helps me to be more socially active.
	I feel the Facebook health support group helps me to adopt a healthy lifestyle.
	I feel satisfied with the Facebook health support group I found.
	I found the Facebook group safe to use.
	I feel I would recommend the Facebook health support group to friends.
	I would like to continue using the Facebook health support group.

• 1) Joining motivation (JM): This construct reflects participants’ reasons for joining Facebook health support groups, including emotional expression, empathy, and information-seeking. Items were adapted from Welbourne et al. [12].

• 2) Trust (TR): This construct captures perceptions of Facebook health groups as reliable, honest, competent, and caring, using items adapted from Hsu et al. [13].

• 3) Giving support (GS): This construct measures the extent to which participants provide emotional and informational support, reflecting bonding social capital. Adapted from Welbourne et al. [12].

• 4) Receiving support (RS): This construct reflects the degree to which individuals benefit from the community through emotional or informational support.

• 5) Sense of virtual community (SOVC): SOVC was measured using an adapted version of Blanchard’s validated scale [14]. It encompasses elements such as membership, shared values, recognition, and emotional connection.

• 6) Connectedness (CON): Based on Lee and Robbins [15], this scale measures feelings of belongingness, affiliation, and interpersonal closeness with group members and the broader community.

• 7) Engagement (EN): EN was assessed using a modified version of the Facebook Engagement Scale developed by Naslund et al. [16]. Items capture behavioural (frequency of use), cognitive (ease of use), and emotional (satisfaction and perceived usefulness) dimensions of platform interaction.

3. Pilot Study

A pilot test involving 30 participants confirmed strong reliability and variability. Cronbach’s alpha ranged from 0.844 (GS) to 0.954 (SOVC), with other constructs also demonstrating high reliability: JM (0.850), TR (0.925), RS (0.868), CON (0.949), and EN (0.926). Standard deviations (SDs) ranged from 1.13 to 1.61 (M = 1.43). The overall mean α = 0.90 indicated solid internal consistency, supporting readiness for full analysis.

4. Measurement

Data were collected using Likert-type scales. Most constructs applied a 7-point scale, while give and receive support items used a 6-point scale to eliminate the neutral midpoint and encourage clearer responses. All items were standardised before analysis. Reliability remained strong, with Cronbach’s alpha exceeding 0.84 for all constructs.

5. Ethical Considerations

Approved by the University’s Human Research Ethics Committee (No. 2021/283), the study ensured voluntary, anonymous participation with electronic consent, no personal identifiers, and secure data storage in accordance with institutional guidelines.

6. Data Analysis

We used PLS-SEM with SmartPLS 4.0 (https://smartpls.com/) to assess the measurement and structural models. Descriptive statistics and normality checks were performed for the study variables using SPSS (IBM, Armonk, NY, USA). Reliability and validity were evaluated through Cronbach’s alpha, composite reliability (CR), average variance extracted (AVE), and heterotrait–monotrait ratio (HTMT). R² and Q² values were used to confirm the model’s predictive strength.

III. Results

Of the 192 survey responses, 42 with more than 5% missing data were excluded, leaving 150 complete cases. The data met normality assumptions, with skewness (−2 to +2) and kurtosis (−10 to +10) within acceptable ranges [17].

Table 1 presents the demographic characteristics of the 150 participants included in the final analysis. Participants were predominantly Arabic (49.3%) and Indian (45.3%), university-educated (74%), employed (68%), and married (70.7%), reflecting a culturally diverse CALD population engaged in Facebook health support groups.

1. Characteristics of the Participants

Using 2022 as the reference year, birth years were converted to age. Participants had an average age of 37.7 years (SD = 9.5; range 24–58.5), spanning early to late middle adulthood (Table 3).

Table 3.

Descriptive statistics for participants’ age (n = 150)

	Mean ± SD	Min	Max
Age (yr)	37.7 ± 9.5	24.0 (approx.)	58.5 (approx.)

2. Descriptive Statistics of the Variables

All constructs satisfied normality assumptions. Among them, EN scored the highest, while GS and RS were comparatively lower, indicating varied levels of interaction (Table 4).

Table 4.

Descriptive statistics of variables

Variable	n	Min	Max	Mean ± SD	Skewness	Kurtosis
Joining motivation (JM)	150	1.00	7.00	4.65 ±1.32	−0.42	−0.69
Trust (TR)	150	1.00	7.00	4.96 ±1.14	−0.93	0.99
Giving support (GS)	150	1.00	6.00	3.55 ±1.30	−0.47	−0.46
Receiving support (RS)	150	1.00	6.00	3.71 ±1.34	−0.66	−0.34
Sense of virtual community (SOVC)	150	1.00	6.71	4.75 ±1.20	−1.16	1.31
Connectedness (CON)	150	1.00	7.00	4.77 ±1.28	−0.74	0.03
Engagement (EN)	150	1.00	7.00	5.04 ±1.08	−1.08	1.67

3. Measurement Model Assessment

Results for factor loadings, reliability, and convergent validity are collectively summarised in Table 5.

Table 5.

Factor loading, reliability, and convergent validity

Construct	Convergent validity		Consistency reliability
	Outer loading (≥0.60)	AVE (>0.50)	Cronbach’s alpha (>0.7)	Composite reliability (>0.7)
Connectedness (CON)			0.959	0.965
CON_1	0.882	0.776
CON_2	0.9
CON_3	0.889
CON_4	0.881
CON_5	0.886
CON_6	0.87
CON_7	0.871
CON_8	0.87
Engagement (EN)			0.943	0.949
EN_1	0.750	0.572
EN_2	0.729
EN_3	0.748
EN_4	0.69
EN_5	0.752
EM_6	0.82
EN_7	0.765
EN_8	0.735
EN_9	0.75
EN_10	0.77
EN_11	0.805
EN_12	0.701
EN_13	0.806
EN_14	0.76
Giving support (GS)			0.900	0.937
GS_1	0.876	0.833
GS_2	0.937
GS_3	0.925
Joining motivation (JM)			0.868	0.906
JM_1	0.762	0.659
JM_2	0.886
JM_3	0.886
JM_4	0.8
JM_5	0.71
Receiving support (RS)			0.884	0.945
RS_1	0.955	0.895
RS_2	0.937
Sense of virtual community (SOVC)			0.968	0.971
SOVC_1	0.797	0.665
SOVC_2	0.821
SOVC_3	0.799
SOVC_4	0.657
SOVC_5	0.831
SOVC_6	0.798
SOVC_7	0.84
SOVC_8	0.814
SOVC_9	0.774
SOVC_10	0.879
SOVC_11	0.831
SOVC_12	0.836
SOVC_13	0.844
SOVC_14	0.801
SOVC_15	0.791
SOVC_16	0.866
SOVC_17	0.861
Trust (TR)			0.951	0.959
TR_1	0.854	0.744
TR_2	0.879
TR_3	0.827
TR_4	0.845
TR_5	0.863
TR_6	0.908
TR_7	0.846
TR_8	0.876

AVA: average variance extracted.

4. Factor Loading

The study analysed 57 reflective indicators, with each item required to have a factor loading exceeding 0.60 for inclusion, as per established standards [18]. All items met this criterion, justifying their further analysis.

5. Reliability Analysis

CR and Cronbach’s alpha were used to assess internal consistency. Following the criteria established by Hair et al. [19], all constructs exhibited acceptable Cronbach’s alpha values, ranging from 0.868 for JM to 0.968 for SOVC [20]. Corresponding CR values ranged from 0.906 for JM to 0.971 for SOVC, confirming strong construct reliability.

6. Convergent Validity

All reflective measurement model AVE values exceeded the minimum acceptable threshold of 0.50, as recommended by Hair et al. [19], supporting the convergent validity and reliability of the study measures.

7. Discriminant Validity

• 1) Fornell-Larcker criterion: The results indicated that the square root of the AVE for each construct was higher than its correlations with all other constructs, thereby establishing discriminant validity [19] (Table 6).

• 2) HTMT ratio: As shown in Table 7, all HTMT ratios among constructs were below the threshold of 0.90, further confirming discriminant validity [19] (Table 7).

• 3) Structural model: Blindfolding (omission distance = 7) produced Q² values greater than zero. Combined with R² results (Table 7), these findings indicate the model demonstrates predictive relevance (Figure 2).

Table 6.

Fornell-Larcker values for each of the items in the construct

	JM	TR	GS	RS	CON	SOVC	EN
JM	0.812
TR	0.618	0.863
GS	0.605	0.599	0.913
RS	0.607	0.564	0.783	0.946
CON	0.722	0.744	0.585	0.570	0.881
SOVC	0.749	0.787	0.601	0.621	0.834	0.816
EN	0.424	0.611	0.333	0.387	0.551	0.635	0.757

Bold values on the diagonal represent the square root of the average variance extracted (AVE) for each construct. Discriminant validity is assessed using the Fornell–Larcker criterion, whereby the square root of AVE should be greater than the corresponding inter-construct correlations.

JM: Joining motivation, TR: Trust, GS: Giving support, RS: Receiving support, CON: Connectedness, SOVC: Sense of virtual community, EN: Engagement.

Table 7.

Heterotrait-monotrait ratio

	CON	EN	GS	JM	RS	SOVC	TR
CON
EN	0.565
GS	0.629	0.352
JM	0.792	0.453	0.679
RS	0.613	0.410	0.875	0.688
SOVC	0.865	0.651	0.645	0.816	0.667
TR	0.781	0.640	0.644	0.677	0.609	0.819

CON: Connectedness, EN: Engagement, GS: Giving support, JM: Joining motivation, RS: Receiving support, SOVC: Sense of virtual community, TR: Trust.

Figure 2.

Structural model.

8. Path Coefficients

The significance of path coefficients was assessed using t-statistics obtained through bootstrapping. Each hypothesis was accepted or rejected based on path coefficient outcomes.

• - H1: JM in an online community or group has a positive and significant impact on GS online posts.

The results revealed a significant (B = 0.380, t = 4.673, p = 0.000) impact of JM on GS. Hence, H1 was supported.

• - H2: JM in an online community or group has a positive and significant impact on RS online posts.

There was a significant (B = 0.418, t = 5.076, p = 0.000) impact of JM on RS. Hence, H2 was supported.

• - H3: TR in an online community or group has a positive and significant impact on GS.

The results revealed a significant (B = 0.364, t = 4.369, p = 0.000) impact of TR on GS. Hence, H3 was supported.

• - H4: TR in an online community or group has a positive and significant impact on GS.

There was a significant (B = 0.306, t = 3.832, p = 0.000) impact of TR on RS. Hence, H4 was supported.

• - H5: GS online has a positive and significant impact on SOVC.

The results revealed a significant (B = 0.297, t = 2.407, p = 0.008) impact of GS on SOVC. Hence, H5 was supported.

• - H6: RS online has a positive and significant impact on SOVC.

There was a significant (B = 0.357, t = 3.194, p = 0.001) impact of GS on CON. Hence, H6 was supported.

• - H7: GS online has a positive and significant impact on social CON.

The results revealed a significant (B = 0.388, t = 3.319, p = 0.000) impact of GS on CON. Hence, H7 was supported.

• - H8: RS online has a positive and significant impact on social CON.

There was a significant (B = 0.290, t =2.549, p = 0.005) impact of RS on CON. Hence, H8 was supported.

• - H9: SOVC has a positive and significant impact on EN.

The results revealed a significant (B = 0.577, t =5.914, p = 0.000) impact of SOVC on EN. Hence, H9 was supported.

• - H10: CON has a positive and insignificant impact on EN.

There was a positive but insignificant (B = 0.070, t =0.694, p = 0.244) impact of CON on EN. Hence, H10 was not supported.

9. Further Analysis: Assessing the Most Significant Factor on EN

Additional analysis was conducted to identify which factor exerted the strongest influence on EN. SOVC was found to be the most significant predictor, demonstrating a robust positive effect (B = 0.472, t = 4.109, p = 0.000) (Table 8).

Table 8.

Results for the hypotheses

Hypotheses	Coefficient	SD	T statistics	p-value	Q2predict	R2
H1: JM → GS	0.380	0.081	4.673	0.000	CON: 0.503 EN: 0.229 GS: 0.429 RS: 0.411 SOVC: 0.554	CON: 0.374 EN: 0.405 GS: 0.448 RS: 0.426 SOVC: 0.419
H2: JM → RS	0.418	0.082	5.076	0.000
H3: TR → GS	0.364	0.083	4.369	0.000
H4: TR → RS	0.306	0.080	3.832	0.000
H5: GS → SOVC	0.297	0.123	2.407	0.008
H6: GS → CON	0.357	0.112	3.194	0.001
H7: RS → SOVC	0.388	0.117	3.319	0.000
H8: RS → CON	0.290	0.114	2.549	0.005
H9: SOVC → EN	0.577	0.098	5.914	0.000
H10: CON → EN	0.070	0.100	0.694	0.244

JM: Joining motivation, TR: Trust, GS: Giving support, RS: Receiving support, CON: Connectedness, SOVC: Sense of virtual community, EN: Engagement, SD: standard deviation.

IV. Discussion

The findings presented in this section directly address the study research question.

• - H1: JM IN an online community or group has a positive and significant impact on GS online posts.

The findings support prior research indicating that JM significantly increases GS within virtual health communities [21]. Evidence from studies of diabetes forums demonstrates that psychosocial needs—often unmet in traditional healthcare settings—drive participation [21]. For CALD individuals, such spaces are especially valuable as they provide culturally and linguistically relevant information, opportunities for cultural exchange, and connections with others sharing similar experiences [7]. Such resources aid integration, boost confidence, and promote empowerment.

• - H2: JM in an online community or group has a positive and significant impact on RS online posts.

The results for H2 align with previous research emphasising the significant and positive effect of JM on RS [22]. Virtual support groups have been recognised as vital sources of emotional and psychological assistance, effectively reducing feelings of isolation [23]. These online platforms also enable members to learn coping mechanisms and strategies for managing health-related challenges [24], while offering social interaction and a sense of community. For CALD individuals, motivation to seek and exchange support drives greater participation and strengthens their engagement within online communities.

• - H3: TR in an online community or group has a positive and significant impact on GS.

• - H4: TR in an online community or group has a positive and significant impact on RS.

The results for H3 and H4 are in alignment with various studies [25–28], suggesting that TR significantly and positively influences both GS and RS. Previous research has established that TR in virtual health information substantially shapes browsing, communication, and sharing behaviours [25], echoing the findings for H3. Further evidence supports H3 and H4 by highlighting the importance of TR in health data resources among CALD communities in Australia [29]. TR functions as a critical mediator of EN, reinforcing participation and reciprocity within online networks [29].

• - H5: GS online has a positive and significant impact on SOVC.

• - H6: RS online has a positive and significant impact on SOVC.

The results for H5 and H6 align with earlier studies [30,A1–A3], confirming the significant and positive influence of both GS and RS on developing a SOVC. Prior research [30] has demonstrated that individuals who integrate Facebook groups into their daily routines—receiving reinforcement and validation from peers—experience a heightened sense of online belonging. Likewise, mutual aid and shared identity within health networks enhance users’ perceptions of community [A3]. These findings underscore that informational and emotional support strengthen patient autonomy and reinforce intentions to remain active within these communities.

• - H7: GS online has a positive and significant impact on social CON.

• - H8: RS online has a positive and significant impact on social CON.

The conclusions drawn for H6 and H8 corroborate several existing studies [A4–A7]. Factors such as frequency of contact, level of activity, and the extent of social ties are established indicators of social CON [A4]. This aligns with H6, which suggests that providing support through active engagement fosters stronger social bonds. Similarly, the positive effect of receiving culturally relevant support reinforces H8, indicating that culturally tailored communication and empathy enhance participants’ sense of connectedness within virtual health spaces.

• - H9: SOVC has a positive and significant impact on EN.

Previous research has suggested that active community participation promotes group-oriented behaviours and fosters belonging [A8,A9]. This is particularly relevant for CALD populations, where the degree to which a virtual community resonates with personal or cultural identity can heighten engagement. The supportive and inclusive nature of these online communities further encourages participation. The ability to share experiences and seek advice in a non-judgemental environment equips CALD individuals with strategies for managing health conditions and navigating systemic challenges, thereby strengthening engagement.

• - H10: CONC has a positive and insignificant impact on EN.

The findings for H10 diverge from existing research, indicating that CON does not significantly influence EN within Facebook health support groups among CALD populations in Australia. According to prior studies [A10], EN comprises three key components: cognition, emotion, and behaviour. Cognition relates to mental investment in an activity; emotion refers to affective connection; and behaviour encompasses commitment, persistence, and active participation [A11].

The limited effect of CON on engagement may be explained by several contextual factors. Language barriers can restrict comprehension and communication. Cultural norms—particularly collectivist tendencies—may lead individuals to prefer offline forms of support and avoid sharing personal health concerns publicly. Low digital literacy and restricted access to technology further limit participation, while privacy concerns may discourage disclosure. Together, these factors may weaken the link between connectedness and observable engagement in digital platforms.

SOVC plays a pivotal role in facilitating support and engagement among CALD populations [A12]. Strengthening this sense of community requires co-designed content, multilingual accessibility, interactive participation, recognition of contributions, and skilled moderation to cultivate inclusive and responsive Facebook health support environments.

This study advances understanding of CALD engagement within Facebook health groups but has certain limitations. The cross-sectional design restricts causal inference, and the sample was limited to specific linguistic groups. Future research should employ more diverse, longitudinal, and culturally tailored approaches to capture changes in engagement and community dynamics over time.

Future studies should refine conceptualisations of EN, clarify potential overlap between SOVC and CON, and incorporate behavioural analytics to complement self-reported data. Longitudinal and culturally responsive methodologies will be essential for capturing the evolving nature of digital participation among CALD communities.

Appendix 1.

hir-2026-32-1-14-Appendix.pdf ^{(44.3KB, pdf)}