Sage Journals: Discover world-class research

Abstract

Facebook long-term care groups have been established by long-term care workers, and few studies have explored the related drivers of participation from the perspective of caregivers. This study applied a mixed-methods approach; the qualitative component conducted a pilot study to explore the drivers of participation in Facebook long-term care groups and found that the use and gratification theory provides a valid approach for explaining these drivers. Subsequently, the quantitative component, based on the use and gratification theory, proposed a conceptual model to examine the effect of these drivers on the social identity of these groups as well as the modulating effects of extraversion and group diversity in terms of age and educational background. The results showed that the most important social identification–related driver underlying participation in these groups was self-discovery. In addition, the relationship among these drivers and social identification was found to be modulated by extraversion, age, and educational background.

Keywords

deep diversity extraversion Facebook support groups group diversity group identification long-term care worker mixed method multi-group analysis surface diversity use and gratification theory

Introduction

Online support groups allow their members to interact unencumbered by the limits of time and space. Prior to the era of social network sites (SNSs), members communicated by writing messages and posting these to chat rooms. In the present SNS era, members of online support groups may interact in cyberspace via multiple formats. Facebook, one of the most popular SNSs, provides a variety of applications on their platform. Facebook Groups is the most useful of these applications, allowing members to create discussion forums based on common interests.¹ The various mutual-help groups set up on the Facebook platform represent a form of computer-mediated social or online social support.²

Many health-related support groups on the Facebook platform allow patients, friends, and families to interact for mutual support on health-related issues.³ Prior studies have investigated the effects of different Facebook support groups (FSGs) that focus on issues such as diabetes, autism, weight loss, sexual health, and smoke cessation,^4–8 with findings showing that FSGs offer positive psychological effects, providing emotional social support and informational social support to their members.^6,9 In addition, other related studies have treated Facebook as a tool for enhancing self-esteem, life satisfaction, mental health, and well-being.¹⁰

However, FSG-related studies have been conducted primarily from the perspective of care recipients such as patients and their family members, with few studies conducted from the perspective of caregivers. Having a sufficient and effective long-term care (LTC) workforce is a critical healthcare management issue in countries with rapidly aging populations such as Taiwan. Facebook long-term care groups (FBLTCGs) have been established by Taiwanese LTC workers, allowing the sharing of long-term care-related knowledge and experience in order to improve the overall care abilities of FBLTCG members and allowing individual members to communicate on relevant issues. LTC workers in Taiwan provide care for a variety of dysfunctions in people, especially in the elderly. These workers, broadly defined, include medical doctors, occupational therapists, physical therapists, nutritionists, registered nurses, licensed practical nurses, certified nursing aides, and family caregivers who provide care in residential, community, or home settings. They provide their service mostly in a standalone format and thus require and benefit greatly from mutual support. LTC workers who participate in FBLTCGs not only share LTC knowledge and related information but also have an outlet to ameliorate their work pressure through mutual support. Aging population is a very serious problem in Taiwan. Caring for the elderly, especially the disabled, is not only increasing dramatically the demand for LTC workers but also necessitating improvements in job quality and satisfaction.

FBLTCGs have been running in Taiwan since 2012. The most important FBLTCGs are currently the “Elder baby home care group” (https://www.facebook.com/groups/340808172669696/), “Long-term care you, me, and they” (https://www.facebook.com/groups/longtermcare95/), and “Home care worker communication zone” (https://www.facebook.com/groups/481651241846449/). As these groups allow their members to share information and accomplish specific tasks,¹¹ it is reasonable to expect that FBLTCGs wield a positive influence on its members in terms of professional capabilities and professional satisfaction. Therefore, increasing participation and improving the coherence among FBLTCG members are issues worth further study.

Method

FBLTCGs are typically initiated by LTC workers. Few FSG studies have been conducted from the perspective of caregivers. This study applied a mixed-methods approach that contained both qualitative and quantitative components.

Qualitative component

A pilot study was conducted to investigate the possible underpinning theory for exploring the drivers of caregiver participation in FBLTCGs. A total of 30 FBLTCG members from the “elder baby homecare group,” “Long-term care you, me, and they,” and “Home care worker communication zone” were interviewed. In total, 10 members who were LTC workers and who volunteered to participate in the interview were recruited from each of the three groups. The interviews were conducted via Facebook Messenger. The semi-structured interviews were designed to understand why participants had joined the FBLTCG and had continued participating in the group.

Quantitative component

The results of the qualitative interviews showed that the important drivers included getting and sharing information, participating in interactive communication, venting grievances, and learning more about LTC as a potential career (please refer to the Data analysis section for details). All of these drivers may be explained under the use and gratification (U&G) theory in terms of purposive value (PV) and interpersonal interconnectivity value (IIV).

U&G theory has been applied widely to explain how media usage satisfies the needs of media users. More recently, U&G has been applied to explore online information behavior related to chat rooms, the Internet, online gaming, online newspapers, and SNS.^11–14 Hence, this theory provides a strong theoretical base for exploring the behaviors of FBLTCG members. Moreover, as age and extraversion may modulate Internet behavior,^15,16 this study takes the roles of these two variables into consideration in order to achieve a more comprehensive understanding of the behavior of these groups.

Theoretical background and hypotheses in the quantitative component

Researchers have linked SNS usage to psychological outcomes such as sense of community and social support.^3,10 Possible mechanisms of these psychological outcomes include the amount of supportive interaction, the quality of interaction, and the perception of available social support.³ However, SNSs include not only virtual communities but social media as well. U&G has previously been applied to investigate the drivers of participation in virtual communities.^11,17 Dholakia et al. proposed that joining a virtual community is a goal-directed action intended to fulfill a set of values. For online patient-support groups, patients may want to collect disease-related information, to discuss or chat, or to encourage other patients.¹⁷

The underpinning assumption of U&G is that cognitive and affective needs motivate individuals to use media, which includes communication media such as virtual communities, social networking sites, online gaming, and so on.^{11,12,18–20} U&G not only explores the choice of media but also reveals the behaviors that come after making a choice.¹⁸ The motivators of U&G cover a broad range of factors that may evolve as the nature of communication media changes and that are derived using diverse theories. For example, Li et al.¹⁹ derived these motivators from motivation theories, hedonic theory, self-determination theory, and social presence theory. The keys motivators of U&G that were identified in previous studies include PV, self-discovery, interpersonal interconnectivity, social enhancement, and entertainment.^{11–13,20,21}

PV refers to values that an individual derives from accomplishing something with a predetermined informational and instrumental purpose, for which the informational value is derived from getting and providing information and the instrumental value is derived from accomplishing specific tasks.^11,21 Initially, informational value and instrumental value were distinct concepts. However, Dholakia et al.¹⁷ integrated the two into a single construct, termed “purposive value,” for more appropriate application in SNS settings. As members of Facebook groups regularly share information within their groups, the information-seeking motivation is an essential factor for encouraging individuals to share information in the SNS.²² In LTC online groups, members are motivated to seek related LTC information in order to improve their caring abilities. Therefore, this study uses PV rather than informational value or instrumental value. Self-discovery value (SDV) refers to the process by which an individual works to understand him/herself through participation in an online group;¹¹ IIV is a benefit that is derived by establishing and maintaining interaction with a group;^11,23 social enhancement refers to the benefits gained by earning acceptance and recognition from group members and from enhancing social status within a group;^11,23 and entertainment value is the fun and enjoyment gained through interacting with other group members.^18,23

These drivers may also be distinguished into three categories of gratification, including utilitarian gratification (utility), social gratification (status), and hedonic gratification (fun).^19,24 Utilitarian gratification comprises information sharing, pursuing achievement, and generating a particular self-image to others;^19,24 social gratification refers to feelings of connectedness with, interaction with, and recognition from a group; and hedonic gratification refers to related perceptions of enjoyment and fun.²⁴

The initial pilot study interviewed 30 members of Taiwan-based LTC online groups to elicit their motivations for participation. The motivators identified included getting and sharing information, venting grievances, communicating with others, and learning more about LTC careers. These motivators all fall within utilitarian and social gratification categories.

Social identification is the sense of self-identifying with a specific group.²⁵ When individuals perceive that they belong to a specific group, they will begin comparing their group to other groups in order to maintain the advantage of belonging to “their” group instead of other groups.²⁶ Social identity (SI) consists of three distinct, but related, components: cognitive, affective, and evaluation.²⁷ These three components may also be used to depict the identification process.^28,29 Cognitive sense is the similarities judgment about a particular social group or about the extent that one’s self-image overlaps with the image of the group, which then produces an in-group preference.^28,29 This process was termed by Bergami and Bagozzi²⁷ as the “depersonalized process.” The cognitive process increases the involvement of an individual in a target group and produces feelings of attachment or kinship.²⁸ Eventually, an evaluation is made, which produces either a favorable or unfavorable attitude toward the target group.

Dholakia et al.¹⁷ investigated participation behaviors in virtual communities (e.g. website bulletin boards, Usenet newsgroups, instant messenger apps, and chat rooms) and reported that higher levels of PV and entertainment value within a particular virtual community yielded a stronger sense of SI in that community. Zhang et al.³⁰ reported that functional value (information value) and hedonic value influence positively the stickiness to a social network. Thus, the inferred assumptions of this study are as follows:

H1. PV affects SI in LTC groups positively.

H2. Entertainment value affects SI in LTC groups positively.

Self-discovery is the process of better understanding oneself, and social interactions within a group setting may more clearly define and elaborate an individual’s personal preferences, tastes, and values.^17,21 Furthermore, SI is the process of achieving self-awareness as a member of a group.²¹ Thus, the higher the level of self-discovery achieved through social interaction, the higher the level of emotional preference that an individual should hold toward the related group(s). Therefore, it is further hypothesized that

H3. SDV affects SI in LTC groups positively.

PV and SDV are both derived from self-reference and referrers to oneself.^11,17 The IIV is derived from receiving social support and feeling friendship or intimacy; the social enhancement value is derived from accepting and finally enhancing social status within a group.^17,21 PV and SDV refer to social gratification, while IIV and social enhancement value refer to social benefit.^11,24,26 Ho and Lin³¹ investigated the behavior of participants in virtual classrooms on the Facebook platform and reported that affective commitment and emotional support both significantly and positively affect the identity of this virtual classroom. Affective commitment includes interactions with other group members, acceptance by these members, and self-perceiving as part of the group, which in turn determine the degree of identification, involvement, and feeling of emotional attachment for this group. Affective commitment is similar to social gratification. Thus, the inferred assumptions of this study are as follows:

H4. IIV affects the SI of LTC groups positively.

H5. Social enhancement value affects the SI of LTC groups positively.

Offline activities (OA), such as making voice phone calls and having face-to-face meetings, are frequent occurrences on the Facebook platform. OA may help group members further develop their sense of belonging, strengthen mutual relationships, and improve mutual understanding.^32,33 Intensive offline interactions may increase the intention of group members to participate in further developing the group.³³ Furthermore, OA may enhance social ties with others and subsequently increase the social capital.³⁴ Thus, the study proposes the following hypothesis:

H6. OA affect the SI of LTC groups positively.

Modulation effects of group diversity and extraversion in the quantitative component

Group diversity refers to the degree of difference among group members.³⁵ Due to its potential effect on group outcomes such as satisfaction and turnover, group diversity is an important issue for organizations.³⁶ Group diversity covers many categories. For example, Milliken and Martins³⁶ categorized group diversity into observable attributes such as age and gender and less-visible attributes such as educational background and personality. Harrison et al.³⁷ proposed surface-level diversity with regard to demographic characteristics such as age, gender, and race and deep-level diversity with regard to psychological characteristics such as attitudes, beliefs, and values. With regard to deep diversity, Shemla et al.³⁸ had similar findings and proposed that personality plays an important role in the perception of difference.

The effects of group diversity may be interpreted through the social categorization and information/decision-making processes.^35,38 The social categorization process refers to the straightforward interpretation of intergroup relationships through, for example, the stereotyping of dissimilar others and the acceptance of similar others as the “in-group.” This perspective is called “viewed diversity” and may impede group performance.^35,38 The information/decision-making perspective postulates that diversity may increase the range of knowledge and expertise and positively influence performance.^35,38

The five-factor model (FFM) is a widely used approach for modeling personality traits. The FFM presumes that most individual differences in human personality may be classified into five broad domains, including extraversion, agreeableness, conscientiousness, neuroticism, and imagination/openness to experience.³⁹ The common scales used to measure personality include NEO Personality Inventory Revised (NEO-PI-R), NEO Five-Factor Inventory (NEO-FFI), and International Personality Item Pool Big Five Markers 50 items (IPIP-BFM-50).⁴⁰ The advantage of IPIP-BFM-50 is that there is no cost to use, and it has demonstrated good psychometric properties.^40,41 Extraversion is associated with the traits of sociability, liveliness, assertiveness, sensation seeking, and being carefree in attitude, among others. In addition, extraversion is one of the personality domains that is most widely understood conceptually and is fundamental to personality research.^16,42

With regard to Internet behavior, researchers report that young adults have a relatively high preference for social media and that introverts prefer to socialize online, whereas extraverts prefer face-to-face interaction.^15,16 Moreover, the group process affects online participant behavior, and similar individuals associating with one another form ties through the mediating influence of sociodemographic factors such as age and education.⁴³ As age is a surface-level diversity attribute and educational background and extraversion are deep-level diversity attributes, the study proposes the following hypotheses:

H7. Age has a modulating effect on the relationship among the model constructs.

H8. Educational background has a modulating effect on the relationship among the model constructs.

H9. Extraversion has a modulating effect on the relationship among the model constructs.

Instruments and participants in the quantitative component

Measurement items for all of the constructs were adopted and modified from a scale that had been applied previously in validated researches.

PV, entertainment value, SDV, maintaining interpersonal connectivity value, and social enhancement value were adapted mainly from Dholakia et al.¹⁷ and Ifinedo.¹¹ OA were adapted from Lin.³² SI was adapted from Tsai and Bagozzi²⁸ and Wang and Sun⁴⁴ and conceptualized as a reflective second-order construct composed of cognitive SI, affective SI, and evaluative SI. In the measurement model analysis, the item loadings for all three of the first-order constructs were above 0.85. All of the measurement items were measured on a 7-point Likert-type scale, ranging from “disagree strongly” (1) to “agree strongly” (7). Extraversion adapted the 10 items from Goldberg’s IPIP-BFM-50 that related to extraversion.⁴⁵

Participants were solicited from the main FSGs in Taiwan related to LTC: “Elder baby home care group,” “Long-term care you, me, and they,” and “Home care worker communication zone.” The study received permission from the administrators of these three groups. Recruitment for the quantitative component used a convenience sampling method, and a link to the questionnaire was posted so that group members could volunteer to join and participate. Recruitment was conducted in two phases: pretest and formal study. All of the participants were assured anonymity and researchers informed them that the questionnaire would not collect any identifiable information, and only information provided by participants who self-identified as LTC workers was retained and used as valid data.

Pretest: The study conducted a pretest of the questionnaire on 30 participants to test the reliability of all of the constructs, the threshold values of which should exceed 0.7. The Cronbach’s alpha values for the eight constructs (PV, entertainment value, SDV, interpersonal interconnectivity, social enhancement, OA, SI, and extraversion) were .942, .805, .851, .835, .775, .932, .824, and .804, respectively, indicating that the measured items had high internal consistency for each construct. Thus, all of the construct items were validated and used in the formal questionnaire.

Formal study: Participants in the formal study were recruited in the same manner as the pilot study. All of the three Facebook LTC support groups that were approached for this study gave permission for their members to participate if interested. Due to this study’s focus on the perspective of caregivers, the authors excluded responses from those who were not LTC workers. The study collected 314 valid responses.

Data analysis

Qualitative components

The pilot study interviewed 30 FBLTCG members who self-identified as being LTC workers. Of these, 26 were certified nursing aides and 4 were family caregivers. These two categories represented the majority of members on the FBLTCG. The interviews were recorded verbatim, and the reasons for participating in the FBLTCG were extracted as post-formed keywords (codes). The frequency of these keywords was as follows: getting and sharing information (18 times), participating in interactive communication (6 times), venting grievances (6 times), and learning more about LTC as a potential career (5 times). The results show that all of the participants were satisfied with their groups and expressed willingness to continue participating in the FBLTCG. All of these reasons (keywords) may be illustrated under U&G theory (see the Method section for details).

Quantitative component

This study used IBM SPSS 22.0 and structural equation modeling (SEM) with partial least square (PLS) approach using SmartPLS version 3.2.6 software for statistical analysis.⁴⁶ PLS is a component-based SEM technique that estimates both the measurement model and the structural model and is suited to assessing the validity of antecedent variables and exploratory studies.⁴⁶

Demographics

The study collected 314 valid questionnaires. A majority of participants were female (259, 82.4%); half (179, 57.0%) held baccalaureate degrees; one-third (107, 34.1%) had less than one year of LTC working experience. In terms of occupation, most participants were certified nursing aides (146, 46.5%), with family caregivers as the second largest group (106, 33.8%; Table 1).

Table 1.

Participant demographics.

Measures		Frequency	Percentage	Measures		Frequency	Percentage
Gender	Male	55.000	17.516	Education	Below high school	20.000	6.369
	Female	259.000	82.484	Education	High school	69.000	21.975
Age	21–25	40.000	12.739	Years of career	Bachelor	179.000	57.006
	26–30	35.000	11.146		Master	46.000	14.650
	31–35	53.000	16.879		Less than 1 year	107.000	34.076
	36–40	41.000	13.057		1–3 years	88.000	28.025
	41–45	44.000	14.013		4–6 years	34.000	10.828
	46–50	55.000	17.516		7–9 years	41.000	13.057
	Above 51	46.000	14.650		More than 9 years	44.000	14.013
Groups	Long-term care you, me, and they	106.000	33.758	Occupation	Certified nursing aide	146	46.497
					Registered nurses	27	8.599
	Elder baby home care group	98.000	31.210		Licensed practical nurse	35	11.146
	Home care worker communication zone	110.000	35.032		Family caregivers	106	33.758

Assessing the measurement model

The measurement model was assessed using convergent validity and discriminant validity. The convergent validity included items loading (criterion > 0.7), Cronbach’s α (criterion > 0.7), composite reliability (CR, criterion > 0.7), average variance extracted (AVE, criterion > 0.5), and discriminant validity.⁴⁷ Regarding discriminant validity, the study compared the square root of the AVE with correlations among constructs, with the square root for the AVE greater than the correlation with other constructs, indicating good discriminant validity. Results showed that factor loading values for all of the items were between .729 and .970, CR values for all of the constructs were between .864 and .976, and AVE values for all of the constructs were between .587 and .932. All of the obtained values exceeded the required minimum value and the discriminant validity was also confirmed (Tables 2 and 3).

Table 2.

Reliability and convergent validity.

Constructs (abbreviation)	Items	Mean	SD	Item reliability (Cronbach’s α)	Convergent validity
					Composite reliability	Average variance extracted	Factor loading
Purposive value (PV)	PV_1	6.408	0.819	0.917	0.935	0.706	0.793
	PV_2	6.315	0.842				0.757
	PV_3	6.325	0.760				0.872
	PV_4	6.331	0.798				0.883
	PV_5	6.344	0.809				0.913
	PV_6	6.322	0.839				0.812
Entertainment value (EV)	EV_1	3.586	1.863	0.761	0.864	0.681	0.889
	EV_2	4.834	1.756				0.729
	EV_3	3.726	1.905				0.848
Self-discovery value (SDV)	SDV_1	5.882	1.106	0.775	0.899	0.816	0.896
	SDV_2	5.720	1.273				0.911
Interpersonal interconnectivity value (IIV)	IIV_1	5.586	1.210	0.921	0.962	0.927	0.964
	IIV_2	5.662	1.139				0.961
Social enhance value (SEV)	SEV_1	4.589	1.430	0.811	0.913	0.839	0.899
	SEV_2	4.882	1.367				0.933
Offline activities (OA)	OA_1	4.551	1.796	0.964	0.976	0.932	0.958
	OA_2	4.261	1.813				0.970
	OA_3	4.430	1.804				0.969
Social identity (SI, second-order construct; consist of cognitive social identity (CSI), affective social identity (ASI), and evaluative social identity (ESI))	SI			0.857	0.894	0.587	SI
	CSI_1	5.475	1.234	0.845	0.928	0.866	0.927
	CSI_2	5.338	1.432				0.934
	ASI_1	5.615	1.183	0.799	0.908	0.832	0.903
	ASI_2	5.637	1.067				0.921
	ESI_1	5.137	1.384	0.769	0.896	0.811	0.921
	ESI_2	4.838	1.577				0.880

Table 3.

Discriminant validity.

	EV	IIV	OA	PV	SDV	SEV	SI
EV	0.825
IIV	0.459	0.963
OA	0.602	0.642	0.966
PV	0.164	0.408	0.093	0.840
SDV	0.285	0.691	0.438	0.457	0.903
SEV	0.669	0.610	0.657	0.054	0.545	0.916
SI	0.540	0.689	0.597	0.371	0.713	0.650	0.766

PV: purposive value; EV: entertainment value; SDV: self-discovery value; IIV: interpersonal interconnectivity value; SEV: social enhance value; OA: offline activities; SI: social identity.

Assessing the structural model and testing the hypotheses

The PLS technique was used to investigate the structural model by providing information about the standardized path coefficients (β) among the constructs, significant value (p value), and the explained construct variances (R²). As hypothesized, the path from PV to SI (H1) was found to be positive and significant, with a path coefficient of 0.201 (t = 5.034, p < 0.001); the path from entertainment value to SI (H2) had strong significant and positive effects, with a path coefficient of 0.275 (t = 4.505, p < 0.001); the paths from SDV to SI (H3) had strong significant and positive effects, with a path coefficient of 0.372 (t = 7.691, p < 0.001); the paths from social enhancement to SI (H5) had strong significant and positive effects, with a path coefficient of 0.134 (t = 2.400, p < 0.017); and the paths from OA to SI (H6) had significant and positive effects, with a path coefficient of 0.120 (t = 2.944, p < 0.003). However, the data did not support H4, as the paths from IIV to SI (H4) were not significant, with a path coefficient of 0.065 (t = 0.984, p = 0.315). The structural model explained 69.2 percent of the variance (R²) in social identification in LTC online groups.

To examine the modulation effect, partial least squares–based multi-group analysis (PLS-MGA) was carried out to test H7, H8, and H9.⁴⁷ PLS-MGA used the respective two data sets to test path coefficients with significantly differences. To conduct PLS-MGA, the study classified participants with an education level of high school or below as the low education group and those with a bachelor’s degree or higher as the high education group; those less than 40 years of age the young group and those 40 years of age or older as the elderly group. In addition, the study used median split to separate age into the young and elderly groups and extraversion to distinguish participants into the introvert group and extravert groups.

With regard to age, the results indicated that two path coefficients of the model had significant differences, with the PV→SI of the young group showing significantly higher path coefficients and the social enhance value (SEV)→SI of the elderly group having higher path coefficients. Regarding educational background, four path coefficients had significant differences: the IIV→SI and OA→SI of the low education group had higher negative path coefficients, and the PV→SI and SDV→SI of the low education group had higher path coefficients. Regarding extraversion, the results indicated that four path coefficients had statistically significant differences: the IIV→SI and OA→SI of the extravert groups had higher path coefficients, and the PV→SI and SEV→SI of the introvert group had higher path coefficients (Table 4).

Table 4.

Multi-group analysis results.

	Age				Education				Extraversion
	Path coefficients G1 vs G2	Difference	P value	Higher results	Path coefficients G1 vs G2	Difference	P value	Higher results	Path coefficients G1 vs G2	Difference	P value	Higher results
EV→SI	.282 vs .070	0.212	0.065	–	.294 vs .273	0.020	0.400	–	.298 vs .134	0.163	0.094	–
IIV→SI	.187 vs .075	0.112	0.334	–	−.315 vs .086	0.401	0.006	High education	−.042 vs .201	0.242	0.030	Extraverts
OA→SI	.081 vs –.016	0.097	0.202	–	−.183 vs .134	0.317	0.022	High education	−.024 vs .307	0.330	0.002	Extraverts
PV→SI	.090 vs .049	0.042	0.666	–	.343 vs .145	0.198	0.045	Low education	.307 vs 116	0.190	0.032	Introverts
SDV→SI	.460 vs .188	0.272	0.007	Younger	.839 vs .372	0.467	0.016	Low education	.457 vs .364	0.093	0.162	–
SEV→SI	.106 vs .569	0.463	0.000	Elder	.022 vs .183	0.161	0.200	–	.237 vs –.076	0.313	0.004	Introverts

Age: G1 = young, G2 = old; years of career: G1 = young, G2 = old; education: G1 = low, G2 = high; extraversion: G1 = introverts, G2 = extraverts; role: G1 = carers, G2 = relatives.

Discussion

U&G theory is an important theory that is widely used to explain the drivers underlying participation in SNSs. Although the drivers identified in previous studies have varied, the major drivers identified include PV, SDV, maintain interpersonal interconnectivity, social enhancement value, and entertainment value,^11,17,21 with other lesser drivers including information search/sharing, achievement, habitual pass time, and enjoyment.^12,13,20 However, these drivers are neither well defined nor widely agreed upon with regard to their effect. For example, some studies, including this study, have defined PV as including instrumental value and informational value,^11,17 while some have included informational value only.^13,20 Achievement value, proposed by Li et al.,¹⁹ is one form of utilitarian gratification that, when driven by a desire to gain power or improve status within a group, may be classified as a social enhancement value. Habitual pass time was proposed by Smock et al.²⁰ to explain the behavior of posting birthday greetings to the “walls” of Facebook friends, which, as a form of “social grooming” on the site, may be classified as an IIV. Enjoyment is a form of hedonic gratification that belongs to entertainment value.¹⁹ Hence, the U&G drivers applied in this study were self-discovery, entertainment value, PV, social enhancement value, and interpersonal interconnectivity.

Regarding the influencing factors (drivers) of the LTC online group’s social identification, the results showed that in terms of U&G drivers, self-discovery, entertainment value, PV, and social enhancement value had positive effects and that interpersonal interconnectivity was not supported. According to a review of related studies, only the effect of entertainment value earned support. Entertainment value incorporates the meanings of fun, relaxation, and passing time.^11,17 Furthermore, SNS has a hedonic facet in terms of its allowing the posting of meme images and animated/funny poses. PV was supported by the data, a result that is consistent with Dholakia et al.¹⁷ However, two other studies that used Facebook users as participants obtained different results, with Smock et al.²⁰ reporting that information sharing positively influenced time spent on Facebook and Ifinedo¹¹ reporting that PV did not influence intention to use SNS. These inconsistent results may be due to the different dependent variables used in these studies and is an issue worth further study. SDV and social enhancement value were supported by the data in this and Ifinedo’s¹¹ study, the participants of which were also Facebook users. Members of LTC online groups may post the related problems that they encounter for group discussion and response. Those who post helpful responses not only show their expertise, enhance their reputation, and improve their status within the group but also enhance their caring ability through the question and answer process. Thus, these processes allow those who post responses to realize social enhancement value and SDV. However, the effect of IIV was not supported by the data. The explanation for this result may be that the participants in this study are network-based groups rather than small groups. The major difference between these two types of groups is that the former holds no expectation that members meet or communicate in person with other members of the group.¹⁷ Dholakia et al.¹⁷ pointed out that the social influence process is different for small groups and network-based groups, with self-referent predominating in network-based groups and group-referent predominating in small groups. Moreover, Dholakia et al.¹⁷ argued that interpersonal interconnectivity should be stronger in small groups than in network-based group because interactions are easier in small groups. The result in this study, related to OA, was in line with the findings of Lin³² and Lai and Chen.³³ OA involve face-to-face forms of interaction, with more frequent interactions forming stronger interpersonal ties.

Furthermore, this study conducted multi-group analyses to test the strength of the relationships between gratification drivers (self-discovery, entertainment value, PV, social enhancement value, and interpersonal interconnectivity) and SI based on respective age, educational background, and extraversion status. PV was shown to exert higher effects on SI in the young group and low education group. This result may be due to the relatively shorter duration of these groups in caring for the elderly or due to diffidence. Thus, these two groups require relatively more information from the LTC groups in order to improve their ability. SDV had a relatively high effect on SI in the elderly group and the low education group. SDV means learning more about oneself or another or obtaining insight through group participation. Therefore, participation in LTC groups tends to benefit the low education group more than the other groups. Moreover, the results of this study revealed that extraverts were inclined toward interpersonal interaction and participating in OA, while introverts were inclined to experience higher effects in the realms of PV and SDV. This result is consistent with previous studies as well as the known characteristics of extraverts. Orchard and Fullwood¹⁶ postulated that Internet consumption patterns are possibly influenced by personality and found that extraverts tend to be sociable, prefer face-to-face interactions, and use social networking as a supplementary method of communicating with friends, while introverts prefer to locate the “real me” through online social communication.

Implications for practice

LTC workers are increasingly important in aging societies. Thus, retaining the current LTC workforce and encouraging new talent to take up LTC careers are crucial issues for healthcare institutions and the government. Certified nursing aides and family caregivers are both entry-level LTC positions that directly perform many of the basic difficult duties of LTC. However, these positions are affected by poor wages, low social status, heavy workloads, considerable occupational stress, and high levels of burnout and turnover.^48,49 Social support outside of work may help ameliorate these dilemmas.⁴⁹ In total, 80 percent of the participants in this study were certified nursing aides and family caregivers. Our findings demonstrated that participation in FBLTCGs provides important social support for this LTC group. Thus, joining FBLTCGs may be one solution for providing social support for LTC workers.

Continuing education is a critical issue for all healthcare professionals, including entry-level LTC workers. The findings of this study show PV (i.e. obtaining LTC-related information) as one of the drivers identified for participating in FBLTCGs, especially among the participants with less formal education. Healthcare institutions and governmental agencies may encourage LTC workers to participant in FBLTCGs as a vehicle for obtaining social support and improving rates of retention. Furthermore, these institutions and agencies may cooperate with FBLTCGs to deliver related education materials to enhance their members’ care skills. The effects of FBLTCG on LTC workers should be investigated and clarified in future studies.

Conclusion

This study explored FBLTCG participation behavior from the perspective of caregivers; used U&G theory to investigate the drivers of FBLTCG member social identification; and considered the modulating roles of age, educational background, and extraversion. The results showed that the most important influence driver of social identification was self-discovery, followed in sequence by entertainment value, PV, social enhancement value, and OA. In addition, the relationship among these drivers and social identification was shown to be modulated by age, educational background, and extraversion. Although OA may generally help FBLTCG group members develop stronger interpersonal ties, the results of this study indicate that these activities may have the most significant effect in those who are more highly educated or who are extraverts, with benefits also applicable to IIV.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Tung-Cheng Lin

References

Park

Kee

Valenzuela

Being immersed in social networking environment: Facebook groups, uses and gratifications, and social outcomes. Cyberpsychol Behav 2009; 12(6): 729–733.

Chen

Choi

AS.

Internet and social support among Chinese migrants in Singapore. New Media Soc 2011; 13(7): 1067–1084.

Ozkaya

LaRose

How does online social networking enhance life satisfaction? The relationships among online supportive interaction, affect, perceived social support, sense of community, and life satisfaction. Comput Hum Behav 2014; 30: 69–78.

Zhang

Sang

Facebook as a platform for health information and communication: a case study of a diabetes group. J Med Syst 2013; 37(3): 9942–9947.

Gerber

Exploring the value of a Facebook support group for parents of children with autism. PhD Thesis, Stellenbosch University, Stellenbosch, South Africa, 2014.

Mohd Roffeei

Abdullah

Basar

SKR.

Seeking social support on Facebook for children with autism spectrum disorders (ASDs). Int J Med Inform 2015; 84(5): 375–385.

Maher

Ryan

Kernot

, et al. Social media and applications to health behavior. Curr Opin Psychol 2016; 9: 50–55.

Dahl

Hales

Turner-McGrievy

GM.

Integrating social media into weight loss interventions. Curr Opin Psychol 2016; 9: 11–15.

Davis

Anthony

Pauls

SD.

Seeking and receiving social support on Facebook for surgery. Soc Sci Med 2015; 131: 40–47.

10.

Indian

Grieve

When Facebook is easier than face-to-face: social support derived from Facebook in socially anxious individuals. Pers Indiv Differ 2014; 59: 102–106.

11.

Ifinedo

Applying uses and gratifications theory and social influence processes to understand students’ pervasive adoption of social networking sites: perspectives from the Americas. Int J Inform Manage 2016; 36(2): 192–206.

12.

Wang

Tsai

HH.

Falling in love with online games: the uses and gratifications perspective. Comput Hum Behav 2010; 26(6): 1862–1871.

13.

Luo

Remus

Uses and gratifications and acceptance of Web-based information services: an integrated model. Comput Hum Behav 2014; 38: 281–295.

14.

Apaolaza

Hartmann

, et al. Shanghai adolescents’ brand interactions on the Chinese social networking site Qzone: a uses and gratifications approach. Rev Esp Investig Marketing 2015; 19(1): 62–70.

15.

Correa

Hinsley

De Zuniga

HG.

Who interacts on the Web? The intersection of users’ personality and social media use. Comput Hum Behav 2010; 26(2): 247–253.

16.

Orchard

Fullwood

Current perspectives on personality and Internet use. Soc Sci Comput Rev 2010; 28(2): 155–169.

17.

Dholakia

Bagozzi

Pearo

LK.

A social influence model of consumer participation in network-and small-group-based virtual communities. Int J Res Mark 2004; 21(3): 241–263.

18.

Pai

Arnott

DC.

User adoption of social networking sites: eliciting uses and gratifications through a means-end approach. Comput Hum Behav 2013; 29(3): 1039–1053.

19.

Liu

, et al. Modeling hedonic is continuance through the uses and gratifications theory: an empirical study in online games. Comput Hum Behav 2015; 48: 261–272.

20.

Smock

Ellison

Lampe

, et al. Facebook as a toolkit: a uses and gratification approach to unbundling feature use. Comput Hum Behav 2011; 27(6): 2322–2329.

21.

Cheung

Lee

MK.

Understanding the sustainability of a virtual community: model development and empirical test. J Inform Sci 2009; 35(3): 279–298.

22.

Syn

SY.

Motivations for sharing information and social support in social media: a comparative analysis of Facebook, Twitter, Delicious, YouTube, and Flickr. J Assoc Inf Sci Tech 2015; 66(10): 2045–2060.

23.

Cheung

Chiu

Lee

MK.

Online social networks: why do students use Facebook?

Comput Hum Behav 2011; 27(4): 1337–1343.

24.

Gan

Understanding the effects of gratifications on the continuance intention to use WeChat in China: a perspective on uses and gratifications. Comput Hum Behav 2018; 78: 306–315.

25.

Casaló

Flavián

Guinalíu

Determinants of the intention to participate in firm-hosted online travel communities and effects on consumer behavioral intentions. Tourism Manage 2010; 31(6): 898–911.

26.

Le Hénaff

Michinov

Le Bohec

, et al. Social gaming is inSIDE: impact of anonymity and group identity on performance in a team game-based learning environment. Comput Educ 2015; 82: 84–95.

27.

Bergami

Bagozzi

RP.

Self-categorization, affective commitment and group self-esteem as distinct aspects of social identity in the organization. Br J Soc Psychol 2000; 39(Pt 4): 555–577.

28.

Tsai

Bagozzi

RP.

Contribution behavior in virtual communities: cognitive, emotional, and social influence. MIS Quart 2014; 38(1): 143–164.

29.

Marino

Vieno

Pastore

, et al. Modeling the contribution of personality, social identity and social norms to problematic Facebook use in adolescents. Addict Behav 2016; 63: 51–56.

30.

Zhang

Guo

, et al. Influence of customer engagement with company social networks on stickiness: mediating effect of customer value creation. Int J Inform Manage 2017; 37(3): 229–240.

31.

Lin

YT.

The effects of virtual communities on group identity in classroom management. J Educ Comput Res 2016; 54(1): 3–21.

32.

Lin

HF.

The role of online and offline features in sustaining virtual communities: an empirical study. Internet Res 2007; 17(2): 119–138.

33.

Lai

Chen

TT.

Knowledge sharing in interest online communities: a comparison of posters and lurkers. Comput Hum Behav 2014; 35: 295–306.

34.

de Zúñiga

Barnidge

Scherman

Social media social capital, offline social capital, and citizenship: exploring asymmetrical social capital effects. Polit Commun 2017; 34(1): 44–68.

35.

Van Knippenberg

Schippers

. Work group diversity. Annu Rev Psychol 2007; 58: 515–541.

36.

Milliken

Martins

LL.

Searching for common threads: understanding the multiple effects of diversity in organizational groups. Acad Manage Rev 1996; 21(2): 402–433.

37.

Harrison

Price

Gavin

, et al. Time, teams, and task performance: changing effects of surface-and deep-level diversity on group functioning. Acad Manage J 2002; 45(5): 1029–1045.

38.

Shemla

Meyer

Greer

, et al. A review of perceived diversity in teams: does how members perceive their team’s composition affect team processes and outcomes? J Organ Behav 2016; 37: S89–S106.

39.

Gow

Whiteman

Pattie

, et al. Goldberg’s “IPIP” Big-Five factor markers: internal consistency and concurrent validation in Scotland. Pers Indiv Differ 2005; 39(2): 317–329.

40.

Strus

Cieciuch

Rowiński

The Polish adaptation of the IPIP-BFM-50 questionnaire for measuring five personality traits in the lexical approach. Roczniki Psychologiczne/Ann Psychol 2017; 17(2): 347–366.

41.

Topolewska

Skimina

Strus

, et al. The short IPIP-BFM-20 questionnaire for measuring the Big Five. Roczniki Psychologiczne/Ann Psychol 2017; 17(2): 385–402.

42.

Gosling

Rentfrow

Swann

Jr.

A very brief measure of the Big-Five personality domains. J Res Pers 2003; 37(6): 504–528.

43.

Malinen

Understanding user participation in online communities: a systematic literature review of empirical studies. Comput Hum Behav 2015; 46: 228–238.

44.

Wang

Sun

Social influence or personal preference? Examining the determinants of usage intention across social media with different sociability. Inform Dev 2016; 32(5): 1442–1456.

45.

Goldberg

International personality item pool, https://ipip.ori.org/New_IPIP-50-item-scale.htm

46.

Ringle

Wende

Becker

JM.

SmartPLS 3. Boenningstedt: SmartPLS Gmbh, 2015, http://www.smartpls.com

47.

Sarstedt

Henseler

Ringle

CM.

Multigroup analysis in partial least squares (PLS) path modeling: alternative methods and empirical results. Adv Int Marketing 2011; 22: 195–218.

48.

Jang

Lee

Zadrozny

, et al. Determinants of job satisfaction and turnover intent in home health workers: the role of job demands and resources. J Appl Gerontol 2017; 36(1): 56–70.

49.

Woodhead

Northrop

Edelstein

Stress, social support, and burnout among long-term care nursing staff. J Appl Gerontol 2016; 35(1): 84–105.

Drivers of participation in Facebook long-term care groups: Applying the use and gratification theory,social identification theory,and the modulating role of group diversity

Abstract

Keywords

Introduction

Method

Qualitative component

Quantitative component

Theoretical background and hypotheses in the quantitative component

Modulation effects of group diversity and extraversion in the quantitative component

Instruments and participants in the quantitative component

Data analysis

Qualitative components

Quantitative component

Demographics

Assessing the measurement model

Assessing the structural model and testing the hypotheses

Discussion

Implications for practice

Conclusion

Footnotes

Declaration of conflicting interests

Funding

ORCID iD

References