Computers in Human Behavior 59 (2016) 358e367
Contents lists available at ScienceDirect
Computers in Human Behavior journal homepage: www.elsevier.com/locate/comphumbeh
Full length article
Flow in context: Development and validation of the flow experience instrument for social networking Puneet Kaur a, Amandeep Dhir b, Sufen Chen c, Risto Rajala a, * a
Aalto University, Department of Industrial Engineering and Management, Finland Aalto University, Department of Computer Science and Engineering, Finland c National Taiwan University of Science and Technology, Graduate Institute of Digital Learning and Education, Taipei, Taiwan b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 15 April 2015 Received in revised form 13 January 2016 Accepted 9 February 2016 Available online xxx
Flow theory is a popular theoretical framework for understanding the underpinnings of the prolonged use of information systems. While there has been an increasing interest in examining flow experience during nearly four decades, the concept of flow still suffers from various limitations concerning its use as a measurable construct in empirical research. To address these limitations, the present study developed a 26-item instrument for examining flow experience in the domain of social networking services. A crosssectional survey was administered to 804 Facebook users. The development and validation process consisted of exploratory and confirmatory factor analyses, second-order factor analysis, and examination of instrument validity and reliability. The developed instrument represents six components of flow experience: skill, machine interaction, social interaction, playfulness, concentration and enjoyment. The developed instrument possesses good model fit and high validity and reliability. This paper discusses the uses and limitations of the instrument in the examination of users' experiences of social networking services, and suggests avenues for future research on the topic with a special focus on research on usercentric innovations in online service. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Flow experience Social networking Continued use Measurement scale
1. Introduction People sometimes report performing certain activities just for the sake of intrinsic enjoyment. Prior literature has termed such a state as being in flow. The flow experience is defined as “the state in which people are so involved in an activity that nothing else seems to matter; the experience itself is so enjoyable that people will do it even at great cost, for the sheer sake of doing it” (Csikszentmihalyi, 1990:4). The outcome of flow experience provides such intrinsic enjoyment that people are ready to perform the same actions repeatedly. This is a familiar psychological state for individuals performing sports activities (Marsh & Jackson, 1999) and playing di, Nagy, Solte sz, Mo zes, & Ola h, 2013). games (Kiili, 2006; Magyaro According to the flow theory, a flow experience possesses a variety of dimensions including balance of skills and challenges, clear goals, instant feedback, focused attention, perceived control, combination of action and awareness, time distortion, loss of self-
* Corresponding author. E-mail addresses: puneet.kaur@aalto.fi (P. Kaur), amandeep.dhir@aalto.fi (A. Dhir),
[email protected] (S. Chen), risto.rajala@aalto.fi (R. Rajala). http://dx.doi.org/10.1016/j.chb.2016.02.039 0747-5632/© 2016 Elsevier Ltd. All rights reserved.
consciousness, and autotelic experience (Csikszentmihalyi, 1990). In the past two decades, conceptualization around flow experience has undergone extensive research. Different types of research methodologies have been practiced around flow experience including experimental, qualitative, and quantitative and experience sampling (Delle Fave, Massimini, & Bassi, 2011; Guo & Poole, 2009). In addition, researchers have investigated the role of flow experience in diversified fields of operation. This includes computer-mediated communications (Webster, Trevino, & Ryan, 1993), humanecomputer interaction (Hoffman & Novak, 1996; Novak, Hoffman, & Yung, 2000; Schaik & Ling, 2003, 2007), mobile instant messaging (Zhou & Lu, 2011), online shopping (Guo & Poole, 2009; Koufaris, 2002); online banking (Lee, Kang, & McKnight, 2007), online games (Chou & Ting, 2003; Hsu & Lu, 2004; Lee & Tsai, 2010), social networking services (SNS) (Chang & Zhu, 2012; Qi & Fu, 2011; Wu & Wang, 2011; Zhou, Li, & Liu, 2010), sports activities (Jackson, Kimiecik, Ford, & Marsh, 1998; Jackson & Marsh, 1996; Jackson, Martin, & Eklund, 2008) and web navigation (Schaik & Ling, 2003). Furthermore, prior flow experience literature has recommended flow theory as an appropriate theoretical framework for understanding user behaviors in
P. Kaur et al. / Computers in Human Behavior 59 (2016) 358e367
online environments (Chang & Zhu, 2012; Huang, 2003; Novak et al., 2000). Although flow theory has been well researched and utilized during the last four decades, it has several limitations. Ambiguity still prevails in both the conceptualization and operationalization of the flow concept. In particular, we have located three main limitations based on our review that hamper the use of the construct in examining user behavior in the context of contemporary social networking services, described as follows. First, the existing literature provides empirical research with incomplete conceptualizations of flow (Finneran & Zhang, 2002; Guo & Poole, 2009). Csikszentmihalyi (1990) gave three preconditions for entering into flow experience, namely balance of skill and challenge, clear goals, and instant feedback. It should be noted that dimensions refer to components of flow experience. In comparison, preconditions refer to the prerequisites or requirements for entering into flow. Earlier studies are either missing one or more of the preconditions as suggested by the original flow model proposed by Csikszentmihalyi (1990). Similarly, most of the prior work has missed one or more of the dimensions of the flow experience (Guo & Poole, 2009), with only a few empirical studies having considered all dimensions of flow experience according to the original model (e.g., Chan & Ahern, 1999; Chan & Repman, 1999; Chen, 2006; Chen & Nilan, 1999). Second, confusion persists regarding the dimensionality of the flow experience. For example, some studies portray flow as unidimensional (Hoffman & Novak, 1996; Novak et al., 2000), while others consider it a multi-dimensional concept (Ghani, Supnick, & Rooney, 1991; Hsu & Lu, 2004; Huang, 2003; Lu, Zhou, & Wang, 2009). Third, the psychometric properties of the flow construct are unknown. Most of the studies have reported only the Cronbach's alpha as a measure of internal reliability (e.g. Guo & Poole, 2009; di et al., 2013; Wesbster, 1989), while convergent and Magyaro discriminant validity and other forms of instrument reliability are rarely examined. These limitations have negative implications for the flow experience research. Due to the conceptual level confusion in flow theory, researchers tend to select the most often used flow measures (Kwak, Choi, & Lee, 2014). By following the mainstream, it is likely that researchers choose flow measures that represent incomplete dimensions as flow experience, adding bias to the research. Incompleteness of the measures used to estimate flow experience might confuse further development of the phenomenon itself. To address these limitations, the present study proposes a comprehensive instrument for measuring a multi-dimensional viewpoint of flow experience in SNS. We drew on the existing literature to develop a psychometrically valid instrument to investigate flow experience in the SNS context. In addition, we validated the instrument through an empirical study of 804 adolescent Facebook users. We also performed a number of statistical analyses to ensure the validity and reliability of the established instrument, and discuss its implications for future research and practice. 2. Background for research 2.1. Flow experience instrument Despite the overwhelming use and popularity of the flow theory, only a few attempts have been made to develop an instrument for examining flow experience in the context of information systems use. Among the few to address this issue are Csikszentmihalyi and Csikszentmihalyi (1988), who developed the “flow questionnaire”, the first instrument for measuring user flow experience of
359
diverse voluntary activities such as rock climbing and dancing. It consists of open and closed ended questions. Also, Wesbster (1989) developed an 11-item Intensity of Flow Scale (IFS) based on Csikszentmihalyi's (1975) checklist for measuring flow by users' understanding of the phenomenon of playfulness in computer usage. This was followed by active empirical research aimed at developing a flow instrument for the domain of sports and physical activity (Jackson & Eklund, 2002; Jackson et al., 1998; Jackson & Marsh, 1996; Jackson et al., 2008; Jackson & Roberts, 1992). di et al. (2013) developed a 20-item instrument Recently, Magyaro based on the existing instruments: the Flow State Questionnaire of the Positive Psychology Lab (PPL-FSQ). Apart from developing new instruments, numerous attempts have been made to validate and adapt the existing instruments. The flow questionnaire by Csikszentmihalyi and Csikszentmihalyi (1988) has been extended and validated in different languages including English, Italian, Portuguese, French and Spanish (Delle Fave et al., 2011). Davis and Wiedenbeck (2001) modified the IFS into a 9-item instrument for understanding users' flow experience in the context of web navigation; this version had high internal consistency compared with the original IFS. Schaik and Ling (2003, 2007) used it to examine users' flow experience of various forms of human interaction in the contexts of online shopping and websites. Schaik and Ling (2012a, 2012b) investigated web navigation fields using the original instrument developed by Jackson and Marsh (1996) in physical activity and its adapted version in the domain of information systems by Guo and Poole (2009). Kiili (2006) also adapted the Jackson and Marsh (1966) instrument for the educational gaming domain by reducing the 36-item instrument to 23 items. Guo and Poole (2009) made a similar attempt by developing a 30-item instrument based on the instruments of Jackson and Marsh (1996) and Agarwal and Karahanna (2000). The aforementioned flow experience instruments suffer from various limitations. First, most of these instruments are now decades old (Davis & Wiedenbeck, 2001; Jackson & Marsh, 1996; Wesbster, 1989). User preferences change with the ever-changing technology, which creates the need to update the existing instruments in the context of the emerging technologies. Second, most of the developed instruments have unknown psychometric properties, as the studies only reported the Cronbach's alpha values (see Table 1). Third, most of the existing instrument development and validation was based on small samples consisting of university students. Fourth, to the best of our knowledge, there is no specific instrument for measuring users' flow experience of SNS. To address these gaps in the development and validation of flow experience instruments, the present study developed and validated a 26-item flow experience instrument for SNS using a large sample of Facebook users. Kaur, Dhir, Chen, and Rajala (2016) utilized the same instrument to investigate the online regret experience while in flow during SNS usage. Additionally, this developed instrument will also bring clarity to the overall flow experience concept. For example, as mentioned previously, so far studies have mirrored the existing research by considering the components most often used for examining flow experience. Such research practices hinder unbiased development of the flow experience concept. In comparison, the current research investigated the relevance of a large number of constructs for the purpose of measuring users' flow experience while using SNS. 2.2. Flow experience and SNS The widespread popularity of SNS has motivated researchers to investigate flow experience in a variety of social software settings. Hoffman and Novak (2009) suggested that SNS are potential platforms for investigating flow experience considering the large
360
P. Kaur et al. / Computers in Human Behavior 59 (2016) 358e367
Table 1 A Review of prior flow instruments. Author (Year)
Study context
Sample
Instrument
Instrument development
Guo and Poole (2009),a
Online shopping
354 university students (211 females and 143 males) United States
The 30-item instrument adapted from Jackson and Marsh (1996), Agarwal and Karahanna (2000)
Schaik and Ling (2003)
Online shopping
90 undergraduate students (74 females and 16 males with Mage ¼ 24)
Schaik and Ling (2007)
Web sites
Experiment A: 103 undergraduate students (84 females & 19 males & Mage ¼ 25 years) Experiment B: 127 undergraduate students (100 females and 27 males & Mage ¼ 23 years)
Schaik and Ling (2012a)
Web navigation
127 undergraduate students (102 females & 25 males & Mage ¼ 22.87 years)
Davis and Wiedenbeck (2001) Schaik and Ling (2012b)
Computer software 173 undergraduate students (106 males and 67 females & Mage ¼ 21.38 years)
B-SC (a ¼ .74), GC (a ¼ .93), F (a ¼ .88), PS (a ¼ .84), PC (a ¼ .75), C (a ¼ .90), PCon (a ¼ .90), M (a ¼ .73), TT (a ¼ .75), TS (a ¼ .82), AE (a ¼ .91) Model fit: (CFI ¼ .91, RMSEA ¼ . 06) Flow instrument for Likert scale (a ¼ .81) & analogue instrument (a ¼ .83) formats Experiment A: Flow instrument for Likert instrument (a ¼ .82) & visual analogue (a ¼ .74) formats Experiment B: Flow instrument for radio buttons (a ¼ .70 & a ¼ .88) and drop down list (a ¼ .81 & a ¼ .90) B-SC (CR ¼ .96), GC (CR ¼ .94), F (CR ¼ .93), C (CR ¼ .94), Con (CR ¼ .97), M (CR ¼ .87), TT (CR ¼ .91), TS(CR ¼ .92), AE (CR ¼ .95) Flow instrument (a ¼ .81)
Web navigation
di et al. Gaming Magyaro (2013)
Kiili (2006),a
Educational gaming
Wesbster (1989) Playfulness in computer usage Agarwal and Karahanna (2000)
Worldwide web
9-item flow instrument taken from Davis and Wiedenbeck (2001) 9-item flow instrument taken from Davis and Wiedenbeck (2001)
30-item flow instrument based on Guo and Poole (2009)
9-item version of Intensity of Flow Scale (IFS) from Wesbster (1989) 114 undergraduate students Experimental research (i)B-SC (CR ¼ .90), (ii) M (CR ¼ .92), (91 females & 23 males with Mage ¼ 22.66 years) GC (CR ¼ .93), F (CR ¼ .94), C (CR ¼ .93), coupled with survey 36-item FSS from Jackson Con (CR ¼ .95), LSC (CR ¼ .88), and Marsh (1996) TT (CR ¼ .92), AE (CR ¼ .91) Study A: 214 university students 20-item Flow State Study 1: 40-item instrument (a ¼ .39), (139 females & 75 males & Mage ¼ 22.29 years) Questionnaire of the 2 Factors e Flow-Anxiety factor (a ¼ .72), Positive Psychology Lab Study B: 250 university students Flow-Boredom factor (a ¼ .31) (PPL-FSQ) (84 females & 166 males & Mage ¼ 22.45 years) Study 2: 23-item instrument (a ¼ .84), 40-item instrument 2 Factors e B-SC (a ¼ .92), Ab (a ¼ .91) (based on instruments revealed in the previous literature) reduced to a 23-item then a 20-item instrument (Validated for reliability) Flow experience (a ¼ .74) four constructs: 23 item Flow scale for N ¼ 221, native Finnish speakers games (FSG) is a (56% males & 44% females with 72% C (a ¼ .75), TD (a ¼ .82), AE (a ¼ .87), condensed & contextualized within 21e30 years age range) LSC (a ¼ .57) Finland Flow antecedents (a ¼ .71), four constructs: form of FSS (Jackson & Marsh, 1996) Ch (a ¼ .43), GC (a ¼ .49), F (a ¼ . 55), Con (a ¼ .39), P (a ¼ .78) Organization employees Intensity of flow scale (a ¼ .74) 11-item survey based on (N ¼ 43 with 33 females & 10 males the checklist provided by with Mage ¼ 37.9 years) Csikszentmihalyi (1975) 270 University students 20-item instrument of Temporal dissociation (CR ¼ .93), (153 males & 117 females with Mage ¼ 22.9 years) focused immersion (CR ¼ .88), cognitive absorption (Webster et al., 1993) heightened enjoyment (CR ¼ .93), control (CR ¼ .83), curiosity (CR ¼ .93)
Note: Balance of skill and challenge (B-SC), Goal clarity (GC), feedback (F), Perceived skill (PS), perceived, challenge (PC), Concentration (C), Perceived control (PCon), Mergence of action & Awareness (M), Transformation of time (TT), Transcendence of self (TS), Autotelic experience (AE), Control (Con), Loss of self-consciousness (LSC), Absorption (Ab), Work enjoyment (WM), Intrinsic work motivation (IWM), Action-awareness (AA), Time Distortion (TD), Challenge (Ch), Playability (P), Mean age (Mage).The model fit was reported only by Guo and Poole (2009). a Represents the papers which reported the country of the participants.
amount of user time spent on them. However, the research examining flow experience in SNS is still young: only a few empirical studies have been carried out so far (See Table 2). To the best of our knowledge, there are only six studies examining the flow theory in the domain of SNS: three considered flow to be a uni-dimensional construct (Chang, 2013; Chang & Zhu, 2012; Qi & Fu, 2011), while the other three regarded it as a multidimensional concept (Kwak et al., 2014; Wu & Wang, 2011; Zhou et al., 2010). The studies regarding flow as multidimensional considered perceived enjoyment and concentration as part of the flow experience. The others considered constructs of control, social interaction, telepresence, time distortion, curiosity and escape. Consistent with Guo and Poole (2009), we observed that the chosen literature addresses incomplete flow models. The utilized flow experience instruments do not address the flow experience
aspects in the domain of SNS. This creates the need for developing a comprehensive instrument specifically for measuring flow experience on SNS. Such an instrument can help determine the components of flow experience in the SNS. Furthermore, it will augment understanding of the differences in SNS user behavior across groups, cultures and populations. The present study developed a comprehensive 26-item flow instrument specifically for Facebook, a popular SNS that presently has 1.39 billion monthly active users of whom 890 million are daily active users (Facebook Newsroom, 2015). Prior literature has suggested that Facebook is a valuable platform for understanding users' psychological and psychosocial behavior in experimental as well as naturalistic conditions (Wilson, Gosling, & Graham, 2012). Prior research examining flow experience in SNS revealed that almost all empirical studies are examined with either
P. Kaur et al. / Computers in Human Behavior 59 (2016) 358e367
361
Table 2 Prior Literature on flow experience research in SNS. Author (Year)
Sample
Instrument
Zhou et al. (2010) Wu and Wang (2011) Qi and Fu (2011) Chang and Zhu (2012) Chang (2013) Kwak et al. (2014)
305 Chinese university students using mobile SNS (57.4% males)
Perceived enjoyment (CR ¼ .87; a ¼ .87), perceived control (CR ¼ .80; a ¼ . 79), attention focus (CR ¼ .92; a ¼ .92) Perceived enjoyment (CR ¼ .85), concentration (CR ¼ .80), social interaction (CR ¼ .87), escape (CR ¼ .91)
342 respondents aged 16e46 years using SNS (53.8% females) with Mage ¼ 26 years 194 Chinese respondents (50% male) 86.1% respondents aged 20e24 years. 283 Chinese SNS users (51% males) aged 18e60 years
The a ranged between .80 and .90 No clear information but it is between .80 and .90
358 Taiwanese university Facebook game users (58% males) aged 18e27 Flow experience (CR ¼ .85) years 263 Korean Facebook use (58.6% males & 41.6% females with age ranging Focused attention (CR ¼ .92; a ¼ .87), enjoyment (CR ¼ .97; a ¼ .96), curiosity from 50 with 80% of the respondents in their twenties & thirties) (CR ¼ .97; a ¼ .96), telepresence (CR ¼ .89; a ¼ .84), time distortion (CR ¼ .93; a ¼ .89), Flow (CR ¼ .85; a ¼ .78)
undergraduate or university students. Furthermore, Facebook, currently the most used SNS, lacks extensive research. To address these limitations, the present study examined the flow experience of adolescent (12e18 years) Facebook users. According to the recent statistics, 94% of adolescents have a presence on Facebook (Sterling, 2013) with India having the second largest FB user base in the world and the most adolescent Facebook users (Prabhudesai, 2013). 2.3. Adolescent online users Adolescents are avid technology users and early adopters of any new goods and services in the market (Lapowsky, 2014). Additionally, Lapowsky (2014) found that adolescents play a significant role in influencing the purchase and adoption related decisions of their family, friends and peers. Adolescents have been found to be loyal to the market rather than to any particular brand. Hence, they represent an interesting and significant customer segment for organizations (Lapowsky, 2014). Due to this, it is important to incorporate adolescents' opinions and feedback for the ideation, design and implementation of future products and services. Despite the importance of adolescents, they have received little €ntyma €ki & Riemer, 2014; attention in the prior research (Ma €ntyma €ki & Salo, 2011). Ma Lapowsky (2014) stated that the most appropriate way to be in contact with adolescents and to influence their decisions and choices is via their preferred platform. In this regard, SNS are the most preferred online services among adolescents. For example, Madden (2012) reported that 81% of adolescents are using SNS in United States. Among the different SNS, Facebook is the most popular among adolescents (Hofstra, Corten, & Tubergen, 2015). In 2012, Madden (2012) found that among 81% of the adolescent SNS users 94% were using Facebook. Recently, Lenhart (2015) found that 71% of adolescents were still using Facebook. However, most prior studies on computer-mediated environments including Facebook originated from the Western world and were chiefly administered to United States (US)-only samples (Dhir, Chen, & Chen, 2015; Dhir, Kaur, Lonka, & Nieminen, 2016). Consequently, many researchers have argued that future research should examine Facebook related use behavior from other cultural and geographical contexts (Caers et al., 2013; Peters, Winschiers-Theophilus, & Mennecke, 2015; Rains & Brunner, 2015). This is consistent with the recent observation of Facebook.com that over 83.5% of Facebook users are from countries other than the US (Facebook Newsroom, 2015). To address this gap, the present study has investigated the flow experienced by adolescent Facebook users in India while using Facebook.
3. Methodology 3.1. Instrument development First, the review of the existing literature included all the conceptual and empirical flow experience research conducted from the 1990s to 2013, as well as the literature concerning instruments developed for examining flow experience. Second, the reviewed literature was revisited again for the purpose of selecting constructs representing flow experience. The process resulted in selecting 14 flow constructs: skill, curiosity, escape, control, machine interaction, social interaction, telepresence, challenge, exploratory behavior, playfulness, enjoyment, concentration, intrinsic interest, and time distortion. These were repeatedly used for measuring flow experience in the prior information systems literature. Third, it was decided to drop the intrinsic interest construct because it overlaps the enjoyment items, which cover user experience more broadly. Moreover, enjoyment has been considered an important component of the flow theory (Csikszentmihalyi, 1990; Guo & Poole, 2009). This led to the deletion of 4 items, giving a total of 84 items representing 13 constructs. Fourth, this pool of items was evaluated using a sample of 804 Facebook users. 3.2. Data collection During December 2013, 804 adolescent Facebook users from eight private junior and senior high schools in Northern India participated in the study. The data were collected through a researcher-administered paper-and-pencil survey. The effective sample size included 575 (71.5%) males and 228 (28.4%) females. The respondents' ages ranged from 13 to 17 years (mean age ¼ 14.47 and SD ¼ 1.13). The details of the participants' demographic characteristics are presented in Table 3. The targeted population was reached through their respective educational institutions. First, a random sample of schools was chosen and contacted via phone calls. Face-to-face meetings followed this where the schools were informed of the study objectives, time requirements, expected outcomes and procedures. Upon receiving a positive response, the schedule for the study was set with the school authorities. Moreover, the study was advertised in the school through school management and teachers. The advertisement clearly mentioned the objectives, time requirements, targets and expected outcomes. The skills and vocabulary level of the target population were considered while designing the survey. A pilot test was conducted with 30 users representing the target
362
P. Kaur et al. / Computers in Human Behavior 59 (2016) 358e367
4.1. Exploratory factor analysis
Table 3 Demographic distributions of study participants. Demographic Variable
Category
Frequency (Percentage)
Gender
Males Females 13 14 15 16 17 Less than 1 year 1e2 years 2e3 years From 3 to 4 years More than 4 years Less than 1 h 1e3 h More than 3 h Very active Active Neutral Less active Don't use much
575 228 182 218 245 112 39 258 239 150 97 47 504 220 68 90 256 241 148 48
Age
Facebook experience
Daily time spent on Facebook
Facebook activity
(71.5) (28.4) (22.6) (27.1) (30.5) (13.9) (4.9) (32.1) (29.7) (18.7) (12.1) (5.8) (62.7) (27.4) (8.5) (11.2) (31.8) (30.0) (18.4) (6.0)
population. They were instructed to provide suggestions about the difficulty level (e.g., words, in general, difficult to understand) and any language related issues in the survey. Based on the feedback, corrections were made to the survey, most of which related to the use of vocabulary that was difficult for adolescents to comprehend. The data collection took place in two phases. First, a topic of interest (recommended by the school authorities) was presented and discussed with the students. The topic addressed the advancements in the field of information communication technology. This was done in order to establish rapport with the students and motivate them to take part in the study. This informative session served as an incentive for the participants. Second, all students who were using Facebook actively or had experience of using Facebook were invited to participate in the second session. The students were again reminded of the study objectives and expected outcomes. Following this, those interested in participating in the study were asked to fill out the questionnaire. Participation in the workshop was completely voluntary: i.e. participants had freedom to quit at any time. The survey was in English since all the participating schools used English as the language for instruction. School management granted permission to conduct the study. As prior literature has suggested (Morrow & Richards, 1996), giving clear instructions to the respondents to not reveal any identifying information (e.g., mobile number, name or email address) ensured participant anonymity.
4. Results The data analysis was performed using SPSS 22.0 and AMOS 21.0. The collected data were normally distributed since the skewness and kurtosis values were within the recommended threshold of ±1 (Byrne, 2001; George & Mallery, 2003; Hair, Anderson, Tatham, & Black, 1998). The calculated z-scores for the measurement items were less than the suggested threshold value of 3.29, indicating the absence of outliers in the data set. The missing values were imputed using the maximum likelihood algorithm. The collected sample of 804 was randomly split into two data sets: Sample A (n ¼ 382) and Sample B (n ¼ 422). Sample A was used for exploring the factor structure by performing exploratory factor analysis (EFA). Sample B was used to confirm the obtained factor structure using confirmatory factor analysis (CFA) and second-order CFA.
The main objective of this study was to develop a new instrument for examining flow experience in SNS. Drawn from prior literature, the flow constructs from various studies might have overlapping concepts in the SNS context. Hence no a priori structure exists for instrument development; therefore, first the factorial structure for the instrument was examined using EFA. Sample A returned statistically significant Bartlett's (1954) test of Sphericity (X2 ¼ 5093.49, df ¼ 325, p < .01) and a very good value (.91) for the Kaiser-Meyer-Olkin (Kaiser, 1970) measure of sampling adequacy. Both tests suggested that the sample was fit for performing EFA. The EFA was performed using maximum likelihood with Varimax rotation where the minimum threshold value for the factor loading was set to .40. The instrument items that did not satisfy this threshold were deleted, and the process was performed until a stable set of factors was obtained. The six factor solution explained 67.69% variance in the flow experience in SNS use. This process resulted in 26 items representing six factors with excellent internal consistency (a ¼ .92) (see Table 4). The six components of flow experience were skill, machine interaction, social interaction, enjoyment, concentration and playfulness (Table 5). 4.2. Confirmatory factor analysis We performed confirmatory factor analysis (CFA) in which the six-factor instrument resulted in a good model fit (X2/df ¼ 1.97, CFI ¼ .95, TLI ¼ .94, RMSEA ¼ .05, SRMR ¼ .04) (Browne and Cudeck, 1992; Hu & Bentler, 1999; Kline, 2010) (see Table 6). 4.3. Second-order confirmatory factor analysis The second order CFA was performed for investigating the existence of a second order factor (Parasuraman, Zeithaml, & Malhotra, 2005; Wu, Tao, Yang, & Li, 2012). Prior research has argued that a second order construct is considered better compared to first order factors (Chen, Sousa, & West, 2005). The current research fulfills both criteria for performing second order CFA. First, the first order factors are correlated with each other and range from .28 to .67. Second, all constructs have been used in the prior research for measuring flow experience so, theoretically, the six constructs have the possibility to be represented by the second order factor entitled flow experience (Chen et al., 2005). The second order CFA also resulted in an adequate model fit (X2/df ¼ 2.23, CFI ¼ .94, TLI ¼ .93, RMSEA ¼ .05, SRMR ¼ .06) (Schermelleh-Engel & Moosbrugger, 2003) (see Table 6). Furthermore, based on the retrieved factor loadings, enjoyment (.84) contributed the most to the flow experience, followed by playfulness (.80), social interaction (.73), skill (.61), and machine interaction (.60). The construct of concentration (.56) contributed the least to the second order factor. 4.4. Validity and reliability Establishing validity and reliability is crucial for instrument development. It confirms the genuineness, soundness and applicability of the developed instrument. We assured the content validity of the instrument by drawing the instrument items primarily from the existing flow literature and adapting them to the context under study. By grounding the instrument development on the existing literature, we acknowledge the review of the items by the worldwide audience of international researchers and practitioners. Also, the face validity of the variables was examined by running a pilot study with 30 active adolescent FB users. The main objective was to ascertain the existence of any ambiguous terms or issues which might compromise the understanding and meaningfulness
P. Kaur et al. / Computers in Human Behavior 59 (2016) 358e367
363
Table 4 Exploratory factor analysis of the Flow instrument (n ¼ 382). Factor name
Skill
Skill 1 Skill 2 Skill 3 Skill 4 Machine interaction 1 Machine interaction I2 Machine interaction 3 Machine interaction 4 Social interaction 1 Social interaction 2 Social interaction 3 Enjoyment 1 Enjoyment 2 Enjoyment 3 Enjoyment 4 Concentration 1 Concentration 2 Concentration 3 Concentration 4 Concentration 5 Playfulness 1 Playfulness 2 Playfulness 3 Playfulness 4 Playfulness 5 Playfulness 6 Eigen value % Variance explained
a
Machine interaction
Social interaction
Enjoyment
Concentration
Playfulness
.76 .58 .64 .69 .60 .68 .76 .74 .67 .55 .67 .60 .68 .69 .71 .62 .75 .80 .70 .58
1.34 5.16 .82
1.71 6.56 .82
1.10 4.24 .77
1.93 7.41 .86
2.74 10.52 .86
.67 .74 .68 .76 .61 .65 8.79 33.80 .90
Note: The percentage of total variance explained was 67.69%.
Table 5 Components of flow experience. Components
Definition
Reference
Skill
The user competencies for performing tasks or activities on Facebook (e.g., finding information, knowledgeable about using Facebook). The speed of information processing on Facebook.
Novak et al. (2000), Koufaris (2002) Novak et al. (2000), Huang (2003)
Machine interaction Social interaction Playfulness Enjoyment
The possibility of establishing and maintaining online social relationships with other FB users.
Wu and Wang (2011)
An experiential user state of happiness, excitement, satisfaction, and hope. A personally pleasurable state that includes the intrinsic value attained by using Facebook.
Concentration
The state of complete immersion in using Facebook.
Chou and Ting (2003) Ghani et al. (1991), Wu and Wang (2011), Agarwal and Karahanna (2000) Ghani et al. (1991), Wu and Wang (2011), Moon and Kim (2001)
of the used constructs. Discriminant validity was investigated in order to assure that the theoretically different constructs were not related (Anderson & Gerbing, 1988). Prior literature suggests tests for assuring discriminant validity. The tests include that correlations between each pair of the study constructs should be smaller than .80 (Campbell & Fiske, 1959). Also, the correlations of the underlying constructs with other constructs should be smaller than the square root of the average variance extracted (AVE) for a particular construct (Chin, 1998; Fornell & Larcker, 1981). Moreover, the AVE values for all the constructs should be greater than their maximum shared variance (MSV) values and average shared variance (ASV) (Barclay, Higgins, & Thompson, 1995). The study sample complied with all the aforementioned statistical tests, thus confirming the discriminant validity (see Table 7). Convergent validity examines the actual relatedness of the theoretically similar constructs. It was ensured using the following statistical tests: First, we examined that the composite reliability (CR) values of all the study constructs were greater than or equal to
.70 (Fornell & Larcker, 1981; Molina, Montes, & Ruiz-Moreno, 2007;Nunnally, 1978 ). Second, we checked that the AVE values were greater than .50, and always smaller than their corresponding CR values (Fornell & Larcker, 1981). Third, we ensured that the item loadings for each construct were above .50 (Anderson & Gerbing, 1988). The study sample confirmed all the aforementioned statistical tests, thus ensuring the convergent validity (see Table 6 & Table 7). Factorial validity examines the degree to which a given measurement instrument produces a stable and recoverable factorial structure. In the present study, the EFA of Sample A resulted in a six-factor solution. Sample B confirmed this factorial structure. This ensured the presence of factorial validity for the developed instrument. Instrument reliability aims at establishing the low measurement error of the proposed instrument (Cronbach, 1951). In the present study, different types of statistical tests confirmed instrument reliability. We investigated the internal reliability of all the study constructs to ensure construct reliability. It was evaluated by
364
P. Kaur et al. / Computers in Human Behavior 59 (2016) 358e367
Table 6 First-order and second-order CFA (n ¼ 422). Factor name
Survey items
CFA (n ¼ 422)
2nd Order CFA (n ¼ 422)
Skill 1 Skill 2 Skill 3 Skill 4 Machine interaction Machine interaction Machine interaction Machine interaction Social interaction 1 Social interaction 2 Social interaction 3 Enjoyment 1 Enjoyment 2 Enjoyment 3 Enjoyment 4 Concentration 1 Concentration 2 Concentration 3 Concentration 4 Concentration 5 Playfulness 1 Playfulness 2 Playfulness 3 Playfulness 4 Playfulness 5 Playfulness 6 X2/df CFI TLI RMSEA
I am extremely skilled/master in using FB I know how to find the required information on FB I know more about using FB than most users I consider myself knowledgeable about FB use The pages I visit on FB load quickly Everything on FB downloads fast FB uploads fast FB operates at high speed Using FB enables me to share my opinions with others Using FB enables me to develop relationships with others Using FB enables me to know new friends FB is fun to use Using FB is interesting to use Using FB is exciting It is enjoyable to use FB When using FB, I never think about other things When using FB, I am not aware of distractions/disturbances When using FB, I am not aware of happenings around me I get fully immersed/absorbed while using FB When using FB my attention is focused I experience the highest happiness when using FB I experience the highest excitement when using FB I experience the highest satisfaction when using FB I experience the highest hopefulness when using FB I experience the highest relaxation when using FB I experience the highest enjoyment when using FB 3.0 .92 .92 .05
.73 .78 .79 .69 .71 .78 .79 .77 .72 .75 .78 .73 .81 .81 .81 .72 .77 .79 .79 .69 .79 .83 .79 .77 .73 .74 1.97 .95 .94 .05
.73 .77 .79 .74 .70 .78 .79 .77 .70 .76 .79 .75 .80 .81 .81 .72 .77 .79 .79 .69 .78 .84 .79 .78 .74 .75 2.23 .94 .93 .05
1 2 3 4
Table 7 Instrument validity and reliability for Sample A, Sample B and the complete sample. Flow experience
Skill Machine interaction Social interaction Enjoyment Concentration Playfulness
Sample A (n ¼ 382)
Sample B (n ¼ 422)
Complete sample (N ¼ 804)
CR
AVE
MSV
ASV
CR
AVE
MSV
ASV
CR
AVE
MSV
ASV
.82 .82 .78 .86 .86 .90
.52 .53 .54 .61 .55 .60
.27 .21 .45 .45 .37 .41
.19 .14 .25 .28 .16 .26
.85 .85 .79 .87 .87 .90
.58 .58 .56 .63 .57 .61
.25 .28 .45 .45 .37 .43
.20 .19 .25 .32 .17 .30
.83 .83 .78 .87 .86 .90
.55 .56 .55 .62 .56 .61
.26 .23 .46 .46 .37 .42
.20 .17 .26 .30 .16 .28
Note: Composite reliability (CR), Average variance extracted (AVE), Maximum shared variance (MSV), Average shared variance (ASV).
measuring the Cronbach's alpha (a), which should be greater than .70 (DeVellis, 2003; Nunnally, 1978). The a value for the study constructs ranges between .77 and .90, which shows that study constructs possess sufficient internal reliability (see Table 4). Also, we addressed the internal reliability of the instrument as a whole (Cronbach & Meehl, 1955; Nunnally & Bernstein, 1994) for the internal consistency of the instrument. The 26-item instrument resulted in an excellent a value of .92, which showed that the developed instrument is internally consistent. Finally, the composite reliability (CR) is generally found to be a stronger measure of internal consistency as compared with Cronbach's alpha values with the recommended threshold value of .70 (Raykov, 1998). The present study results found that all flow constructs had a CR value greater than .70 (see Table 7). 5. Discussion and conclusions The present study sought to deepen the current understanding of the components of the flow experience in SNS use in order to establish an instrument to measure flow in subsequent analyses. The 26-item developed instrument consists of six components: skill, machine interaction, social interaction, concentration,
enjoyment and playfulness. The present study focused exclusively on adolescent SNS users. To the best of our knowledge, the developed instrument is the first exhaustive attempt to measure the flow experience of SNS users. Furthermore, the 26-item instrument also addresses the prior literature gap of examining flow experience among adolescents. The present study has developed a flow experience instrument in the context of the popular SNS of Facebook. This instrument also addresses the gap in the existing literature by providing an up-to-date instrument, which is both valid and reliable, specifically for measuring flow experience in SNS. The existing literature has shown that flow theory has been extensively used in different domains from various perspectives in the last four decades. However, there have been few efforts to reexamine, re-validate and consolidate the findings of the present flow theory literature. Consequently, there is a lack of clarity regarding the conceptualization of flow experience and this has greatly hampered empirical research on flow in the context of information systems use. The absence of any comprehensive instrument for measuring flow experience has been evident, specifically in the SNS context. Also, there has been a need for updated instruments capturing the latest user preferences across several areas of online social behavior. Of note, the lack of complete reporting of
P. Kaur et al. / Computers in Human Behavior 59 (2016) 358e367
the psychometric properties of the used instruments and their instruments has influenced the validity and reliability of research on the topic. The research on flow in the area of information systems use has suffered from small sample sizes. The current study addressed these limitations by developing and validating an instrument for measuring flow experience in the SNS domain. We acknowledge that flow theory aims at capturing the intrinsic motivational aspects of users' behavior. The information on intrinsic aspects of user behavior is relevant for understanding strategies that can self-motivate users to perform actions without external influence. It is interesting that adolescents value skill, machine interaction, social interaction, enjoyment, concentration and playfulness among other considered components. This means that these components enable adolescents to enter into flow while using SNS. Adolescents are relatively new users of Facebook. Due to this, it is important for them to possess and acquire the required skill set for effective usage of Facebook. Furthermore, anecdotal evidence suggests that the possession of skills by adolescents is one of the ways of enhancing their prestige among their peers. Machine interaction addresses the speed related aspects of Facebook usage. The response from Facebook is one form of feedback from the system. The prior flow literature has reported that receiving timely feedback is an essential component for entering into flow experience (Csikszentmihalyi, 1990; Guo & Poole, 2009). The prior literature has revealed that retention of the existing relationships and establishment of new relationships is one of the motivations behind using Facebook (e.g., Ellison, Steinfield, & Lampe, 2007; Lin & Lu, 2011). Consistent with this trend, adolescents might also be finding social interaction to be valuable in the context of their flow experience on SNS. The prior flow literature has shown enjoyment and concentration to be valuable components of users' flow experience (e.g., Agarwal & Karahanna, 2000; Csikszentmihalyi, 1990). The findings suggest that enjoyment and concentration are essential for entering into flow experience. This finding has been validated in the case of adolescents. Finally, in the context of computer-mediated platforms, young users are inclined towards entertainment and leisure (e.g., Sheldon, 2008). This might provide a sense of excitement, satisfaction and hope to the adolescents. Due to this, playfulness is an important component for adolescents to experience flow while using Facebook. The results of the second order CFA represent that the retrieved six components reflect the holistic phenomenon of flow experience on Facebook which is multi-dimensional in nature. On the other hand, the constructs of control, curiosity, escape, telepresence, exploratory behavior, challenge, and time distortion are found to have no role in adolescents' flow experience in the context of SNS. It might be possible that these constructs are valuable for adolescents for using Facebook. However, the present study results indicate that they are not part of their flow experience on Facebook. 5.1. Implications for theory and practice The instrument presented in this paper has implications for both researchers and practitioners interested in the user experience of a service. Researchers intending to measure users' flow experience in the context of SNS can employ the proposed instrument to improve the accuracy of their research. While a psychometrically sound instrument is an essential prerequisite for accurate measurement, reliable measures may also open up new opportunities for new research initiatives and for the use of the results. Practitioners can use the proposed instrument to investigate the extent of flow experience in their provided services. This can provide organizations with ways to improve their existing and new services. Furthermore, as the developed instrument added reliability of the measurement, it can also assist the development of
365
new information systems: for example, guidance in deciding which factors to focus on when designing new information systems for adolescents while ensuring the experience of flow. The developed instrument also provides feedback to those organizations intending to use online information systems (e.g., social media platforms) for user-centric innovation. Many organizations are increasingly using social media based communities for developing close relationships with their customers and involving them in their processes for improving and innovating new products and services (Bolotaeva & Cata, 2010; Ding, Phang, Lu, Tan, & Sutanto, 2014; Palmer & Koeing-Lewis, 2009; Park & Kim, 2014; Richter, Riemer, & vom Brocke, 2011). Ensuring active user participation and its retention has been a potential challenge for practicing usercentric innovation via brand communities (Deloitte, 2009; Ding et al., 2014; Habibi, Laroche, & Richard, 2014). These two factors are critical to sustaining and succeeding in any online brand community. Organizations can use the proposed flow instrument to monitor users' flow experience. Accordingly, they can modify the platform to enhance flow. Moreover, the existing literature empirically shows that flow can positively contribute towards user satisfaction and their continuation intentions (Hausman & Siekpe, 2009; Hu & Kettinger, 2008; Koufaris, 2002; Lee, 2010; Lee & Tsai, 2010). The repetitive behavior induced in the users due to the intrinsic enjoyment they experience as a result of flow can help organizations build a loyal network of users as well as retain and encourage the active participation of existing users. The findings of the present study have contextual implications: specifically, providing insights into adolescent behavior in the online social context. It highlights the relative importance of components required by adolescents to enter into flow. The findings can also be used for developing the contingencies for favorable educational technologies and information systems for adolescent users. The educational technology and information systems with their design grounded on the findings of this study might have a positive influence on adolescents' intention to continue their use for educational purposes. This might eventually enhance their learning and academic performance. 5.2. Limitations and directions for future work The developed instrument does have its limitations. First, our 26-item flow experience instrument varies from the original flow model. This instrument does not address the preconditions of clarity of goals and unambiguous feedback. Furthermore, it does not consider the precondition of balance of skill and challenge in its original form. However, the instrument is consistent with the prior literature, since it considers skill and challenge as separate dimensions of flow rather than their balance. Second, responses to the current survey were based on the users' reporting on past activities, therefore possibly compromising the reliability of the results (Kiili, 2005). Future research should evaluate the validity of the findings of this research by investigating the same instrument using different approaches, for example, using the instrument presented in this study in the experience sampling approach to examine users' flow experience. Third, the established flow instrument has thus far been validated with Indian adolescent Facebook users. Therefore, the applicability of this instrument in other cultures and validity for other age groups (e.g., adults) is currently unknown. Future work would require further validation of the proposed instrument even in the Indian context with users from different age groups as well as with different cultures and different age groups. Finally, the context of the instrument development is “Facebook based Flow experience”. Congruent with previous flow literature, flow experience constructs vary according to the platforms used (e.g., Wu, Wang, & Tsai, 2010). Hence, it is
366
P. Kaur et al. / Computers in Human Behavior 59 (2016) 358e367
plausible to assume that the “flow instrument” might change when other information systems and social media platforms are considered. Therefore, future work should evaluate its application with various information systems and social media platforms. We call for future work to validate the proposed tool in other domains as well. The generalizability of the results should also be achieved by conducting a series of longitudinal and repeated cross-sectional studies. Acknowledgment The financial support by the FIMECC FutIS research project of the Finnish Funding Agency for Technology and Innovation is gratefully acknowledged. We also acknowledge the support received from the Academy of Finland, Mind the Gap (Project Number 1265528) and Academy of Finland Researcher's mobility grant (Decision No. 290822, 278832, 290038, 277571). References Agarwal, R., & Karahanna, E. (2000). Time flies when you're having fun: cognitive absorption and beliefs about information technology usage. MIS Quarterly, 24(4), 665e694. Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: a review and recommended two-step approach. Psychological Bulletin, 103, 411e442. Barclay, D. W., Higgins, C., & Thompson, R. (1995). The partial least squares (PLS) approach to causal modeling: personal computer adaptation and use as illustration. Technology Studies, 2(2), 285e309. Bartlett, M. S. (1954). A note on multiplying factors for various chi-square approximations. Journal of the Royal Statistical Society, 16, 296e298. Bolotaeva, V., & Cata, T. (2010). Marketing opportunities with social networks. Journal of Internet Social Networking and Virtual Communities, 8. Article ID 109111. Browne, M. W., & Cudeck, R. (1992). Alternative ways of assessing model fit. Sociological Methods and Research, 136e162. Sage publishers, Newbury Park CA. Byrne, M. B. (2001). Structural equation modeling with AMOS: Basic concepts, applications, and programming. New York: Routledge. Caers, R., De Feyter, T., De Couck, M., Stough, T., Vigna, C., & Du Bois, C. (2013). Facebook: a literature review. New Media & Society, 15(6), 982e1002. Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin, 56(2), 81. Chan, T. S., & Ahern, T. C. (1999). Targeting motivation-adapting flow theory to instructional design. Journal of Educational Computing Research, 21, 151e163. Chang, C.-C. (2013). Examining users' intention to continue using social network games: a flow experience perspective. Telematics and Informatics, 30(4), 311e321. Chang, Y. P., & Zhu, D. H. (2012). The role of perceived social capital and flow experience in building users' continuance intention to social networking sites in China. Computers in Human Behavior, 28(3), 995e1001. Chan, T. S., & Repman, L. (1999). Flow in web based instructional activity: an exploratory research project. International Journal of Educational Telecommunications, 5, 225e237. Chen, H. (2006). Flow on the net-detecting web users's positive affect and their flow states. Computers in Human Behavior, 22, 221e233. Chen, H., & Nilan, M. (1999). Digital format of experience sampling method transformation, implementation and assessment. AMCIS, 692e694. Chen, F. F., Sousa, K. H., & West, S. G. (2005). Testing measurement invariance of second-order factor models. Structural Equation Modeling, 12(3), 471e492. Chin, W. W. (1998). Issues and opinion on structural equation modeling. MIS Quarterly, 22(1), 7e16. Chou, T.-J., & Ting, C.-C. (2003). The rolf of flow experience in cyber-game addiction. CyberPsychology and Behavior, 6(6), 663e675. Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16, 297e334. Cronbach, J. L., & Meehl, E. P. (1955). Construct validity in psychological tests. Psychological Bulletin, 52(4), 281e302. Csikszentmihalyi, M. (1975). Play and intrinsic rewards. Journal of Humanistic Psychology, 15, 41e63. Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience (p. 336). New York: HarperCollins. Csikszentmihalyi, M., & Csikszentmihalyi, I. (Eds.). (1988). Optimal experience: Psychological studies of flow in consciousness (p. 432). New York: Cambridge University Press. Davis, S., & Wiedenbeck, S. (2001). The mediating effect of intrinsic motivation, ease of use and usefulness perceptions on performance in first-time and subsequent computer users. Interaction with Computers, 13(5), 549e580. Delle Fave, A., Massimini, F., & Bassi, M. (2011). Instruments and methods of flow research. In Psychological selection and optimal experience across cultures, social
empowerment through personal growth (pp. 59e87). Springer. Deloitte. (2009). Tribalization of business study - Transforming companies with communities and social media. Retrieved from http://www.deloitte.com/assets/ Dcom-United-States/Local%20Assets/Documents/TMT_us_tmt/us_tmt_TribofBu sFlipBook_100609.pdf Accessed 30.07.14.. DeVellis, R. F. (2003). Scale development: theory and applications. Applied Social Research Methods, 1e216. Sage Publications, Thousand Oaks, California. Dhir, A., Chen, G. M., & Chen, S. (2015). Why do we tag photographs on Facebook? Proposing a new gratifications scale. New Media & Society, 1e20. Dhir, A., Kaur, P., Lonka, K., & Nieminen, M. (2016). Why do adolescents untag photos on Facebook? Computers in Human Behavior, 55(B), 1106e1115. Ding, Y., Phang, C. W., Lu, X., Tan, C. H., & Sutanto, J. (2014). The role of marketerand user-generated content in sustaining the growth of a social media brand community. In Proceedings of 47th Hawaii international conference on system sciences (HICSS '14) (pp. 1785e1792). Washington, DC, USA: IEEE Computer Society. Ellison, N. B., Steinfield, C., & Lampe, C. (2007). The benefits of Facebook “Friends”: social capital and college students' use of online social network sites. Journal of Computer-Mediated Communication, 12, 1143e1168. Facebook Newsroom. (2015) Accessed 10.04.15. http://newsroom.fb.com/company-i nfo/ Finneran, C. M., & Zhang, P. (2002). The challenges of studying flow within a computer-mediated environment. In Eighth Americas conference on information systems, Paper 146. Fornell, C., & Larcker, D. (1981). Structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39e50. George, D., & Mallery, P. (2003). SPSS for windows step by step: A simple guide and reference 11.0 update (4th ed.). Boston, MA: Allyn & Bacon. Ghani, J. A., Supnick, R., & Rooney, P. (1991). The experience of flow in computermediated and in face-to-face groups. In International conference on information systems, 18e30, December, New York, USA. Guo, Y. M., & Poole, M. S. (2009). Antecedents of flow in online shopping: a test of alternative models. Information Systems Journal, 19(4), 369e390. Habibi, M. R., Laroche, M., & Richard, M. O. (2014). The roles of brand community and com-munity engagement in building brand trust on social media. Computers in Human Behavior, 37, 152e161. Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data analysis. UK: Prentice. Hausman, A. V., & Siekpe, J. S. (2009). The effect of web interface features on consumer online purchase intentions. Journal of Business Research, 62(1), 5e13. Hoffman, D. L., & Novak, T. P. (1996). Marketing in hypermedia computer-mediated environments: conceptual foundations. Journal of Marketing, 60(3), 50e68. Hoffman, D. L., & Novak, T. P. (2009). Flow online: lessons learned and future prospects. Journal of Interactive Marketing, 23, 23e34. Hofstra, B., Corten, R., & Tubergen, F. V. (2015). Who was first on Facebook? Determinants of early adoption among adolescents. New Media & Society, 1e19. Hsu, C.-L., & Lu, H.-P. (2004). Why do people play on-line games? An extended TAM with social influences and flow experience. Information & Management, 41(7), 853e868. Huang, M.-H. (2003). Designing website attributes to induce experiential encounters. Computers in Human Behavior, 19(4), 425e442. Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1e55. Hu, T., & Kettinger, W. J. (2008). Why people continue to use social networking services: developing a comprehensive model. In Twenty ninth international conference on information systems proceedings, paper 89. Jackson, S. A., & Eklund, R. C. (2002). Assessing flow in physical activity: the flow state scale-2 and dispositional flow scale-2. Journal of Sport & Exercise Psychology, 24, 133e150. Jackson, S. A., Kimiecik, J. C., Ford, S. K., & Marsh, H. W. (1998). Psychological correlates of flow in sport. Journal of Sport & Exercise Psychology, 20, 358e378. Jackson, S. A., & Marsh, H. W. (1996). Development and validation of a scale to measure optimal experience: the flow state scale. Journal of Sport and Exercise Psychology, 18(1), 17e35. Jackson, S. A., Martin, A. J., & Eklund, R. C. (2008). Long and short measures of flow: examining construct validity of the FSS-2, DFS-2, and new brief counterparts. Journal of Sport and Exercise Psychology, 30, 561e587. Jackson, S. A., & Roberts, G. C. (1992). Positive performance states of athletes: toward a conceptual understanding of peak performance. The Sport Psychologist, 6, 156e171. Kaiser, H. A. (1970). A second generation little jiffy. Psychometrika, 35, 401e415. Kaur, P., Dhir, A., Chen, S., & Rajala, R. (2016). Understanding online regret experience using the theoretical lens of flow experience. Computers in Human Behavior, 57, 230e239. Kiili, K. (2005). On educational game design: Building blocks of flow experience. Doctoral Dissertation. Tampere University of Technology, Publication 571. Kiili, K. (2006). Evaluations of an experiential gaming model. Human Technology, 2(2), 187e201. Kline, R. B. (2010). Principles and practice of structural equation modeling (3rd ed.). New York, NY: Guilford Press. Koufaris, M. (2002). Applying the technology acceptance model and flow theory to online consumer behavior. Information Systems Research, 13(2), 205e223. Kwak, K. T., Choi, S. K., & Lee, B. G. (2014). SNS flow, SNS self-disclosure and post hoc interpersonal relations change: focused on Korean Facebook user. Computers in
P. Kaur et al. / Computers in Human Behavior 59 (2016) 358e367 Human Behavior, 31, 294e304. Lapowsky, I. (2014). Why teens are the most elusive and valuable customers in tech. http://www.inc.com/issie-lapowsky/inside-massive-tech-land-grab-teenagers. html Accessed 08.10.15. Lee, M.-C. (2010). Explaining and predicting users' continuance intention toward elearning: an extension of the expectationeconfirmation model. Computers & Education, 54, 506e516. Lee, K. C., Kang, I. W., & McKnight, D. H. (2007). Transfer from offline trust to key online perceptions: an empirical study. IEEE Transactions on Engineering Management, 54(4), 729e741. Lee, M.-C., & Tsai, T.-R. (2010). What drives people to continue to play online games? An extension of technology model and theory of planned behavior. International Journal of Human-Computer Interaction, 26(6), 601e620. Lenhart, A. (2015). Teens, social media & technology overview 2015. Pew Research Center. http://www.pewinternet.org/files/2015/04/PI_TeensandTech_Update 2015_0409151.pdf Accessed 08.10.15. Lin, K. Y., & Lu, H.-P. (2011). Why people use social networking sites: an empirical study integrating network externalities and motivation theory. Computers in Human Behavior, 27, 1152e1161. Lu, Y., Zhou, T., & Wang, B. (2009). Exploring Chinese users' acceptance of instant messaging using the theory of planned behavior, the technology acceptance model, and the flow theory. Computers in Human Behavior, 25(1), 29e39. Madden, M. (2012). Privacy management on social media sites. Pew Internet Report http://www.isaca.org/Groups/Professional-English/privacy-data-protection/ GroupDocuments/PIP_Privacy%20mgt%20on%20social%20media%20sites%20Fe b%20201 2.pdf Accessed 08.07.15.. di, T., Nagy, H., Solte sz, P., Mo zes, T., & Ola h, A. (2013). Psychometric Magyaro properties of a newly established flow state questionnaire. The Journal of Happiness and Well-Being, 1(2), 85e96. €ntym€ Ma aki, M., & Riemer, K. (2014). Digital natives in social virtual worlds: a multimethod study of gratifications and social influences in Habbo Hotel. International Journal of Information Management, 34(2), 210e220. €ntym€ Ma aki, M., & Salo, J. (2011). Teenagers in social virtual worlds: continuous use and purchasing behavior in Habbo Hotel. Computers in Human Behavior, 27, 2088e2097. Marsh, H. W., & Jackson, S. A. (1999). Flow experience in sport: construct validation of multidimensional, hierarchical state and trait responses. Structural Equation Modeling: A Multidisciplinary Journal, 6(4), 343e371. Molina, L. M., Montes, J. L., & Ruiz-Moreno, A. (2007). Relationship between quality management practices and knowledge transfer. Journal of Operations Management, 25(3), 682e701. Moon, J.-W., & Kim, Y.-G. (2001). Extending TAM for a world-wide-web context. Information & Management, 38(4), 217e230. Morrow, V., & Richards, M. (1996). The ethics of social research with children: an overview. Children & Society, 10(2), 90e105. Novak, T. P., Hoffman, D. L., & Yung, Y.-F. (2000). Measuring the customer experience in online environments: a structural modelling approach. Marketing Science, 19(1), 22e42. Nunnally, J. C. (1978). Psychometric theory. New York: McGraw-Hill. Nunnally, C. J., & Bernstein, H. I. (1994). Psychometric theory (3rd ed.). New York: McGraw-Hill, Inc. Palmer, A., & Koeing-Lewis, N. (2009). An experiential, social network-based approach to direct marketing. Direct Marketing: An International Journal, 3(3), 162e176. Parasuraman, A., Zeithaml, V. A., & Malhotra, A. (2005). E-S-QUAL: a multiple-item scale for assessing electronic service quality. Journal of Service Research, 7(3), 213e233.
367
Park, H., & Kim, Y. K. (2014). The role of social network websites in the consumerebrand relationship. Journal of Retailing and Consumer Services, 21, 460e467. Peters, A. N., Winschiers-Theophilus, H., & Mennecke, B. E. (2015). Cultural influences on Facebook practices: a comparative study of college students in Namibia and the United States. Computers in Human Behavior, 49, 259e271. Prabhudesai, A. (2013). Facebook stats: More Indian teenage user on Facebook than USA. http://trak.in/tags/business/2013/08/03/facebook-stats-indian-teenage rs-on-facebook-usa-03082013/ Accessed 10.04.15. Qi, Y., & Fu, C. (2011). The effects of flow and attachment on the e-Loyalty of SNS websites. In International conference on management and service science, Wuhan (pp. 1e6). Rains, S. A., & Brunner, S. R. (2015). What can we learn about social networking sites by studying Facebook? A call and recommendations for research on social network sites. New Media & Society, 17, 114e131. Raykov, T. (1998). Coefficient alpha and composite reliability with interrelated nonhomogeneous items. Applied Psychological Measurement, 22(4), 375e385. Richter, D., Riemer, K., & vom Brocke, J. (2011). Internet social networking: research state of art and implications for enterprise 2.0. Business and Information Systems Engineering, 3(2), 89e101. Schaik, P., & Ling, J. (2003). Using on-line surveys to measure three key constructs of the quality of human-computer interaction in web sites: psychometric properties and implications. International Journal of Human-Computer Studies, 59(5), 545e567. Schaik, P., & Ling, J. (2007). Design parameters of rating scales for Web sites. ACM Transactions on Computer-Human Interaction, 14(1). Article 1. Schaik, P. V., & Ling, J. (2012a). A cognitive-experiential approach to modelling web navigation. International Journal of Human-Computer Studies, 70(9), 630e651. Schaik, P. V., & Ling, J. (2012b). An experimental analysis of experiential and cognitive variables in Web navigation. Human-Computer Interaction, 27(3), 199e234. Schermelleh-Engel, K., & Moosbrugger, H. (2003). Evaluating the fit of structural equation models: tests of significance and descriptive goodness-of-fit measures. Methods of Psychological Research Online, 8, 23e74. Sheldon, P. (2008). Student favorite: Facebook and motives for its use. Southwestern Mass Communication Journal, 23(2), 39e55. Sterling, G. (2013). Pew: 94% of teenagers use Facebook, have 425 Facebook friends, but Twitter & Instagram adoption way up. http://marketingland.com/pew-the-avera ge-teenager-has-425-4-facebook-friends-44847 Accessed 10.04.15. Webster, J., Trevino, L. K., & Ryan, L. (1993). The dimensionality and correlates of flow in human-computer interactions. Computers in Human Behavior, 9, 411e426. Wesbster, E. J. (1989). Playfulness and computers at work. Doctoral Dissertation. Graduate School of Business Administration, New York University. Wilson, R. E., Gosling, S. D., & Graham, L. T. (2012). A review of Facebook research in the social sciences. Perspectives of Psychological Science, 7(3), 203e220. Wu, Y. L., Tao, Y. H., Yang, P. C., & Li, C. P. (2012). Development and validation of a scale to measure blog service quality. Journal of e-Business, 14(1), 211e232. Wu, H.-L., & Wang, J.-W. (2011). An empirical study of flow experience in social network sites. In Proceedings of Pacific Asia conference on information systems, paper 215. Wu, J. H., Wang, S. C., & Tsai, H. H. (2010). Falling in love with online games: the uses and gratifications perspective. Computers in Human Behavior, 26(6), 1862e1871. Zhou, T., Li, H., & Liu, Y. (2010). The effect of flow experience on mobile SNS users' loyalty. Industrial Management and Data Systems, 110(6), 930e946. Zhou, T., & Lu, Y. (2011). Examining mobile instant messaging user loyalty from the perspectives of network externalities and flow experience. Computers in Human Behavior, 27(2), 883e889.
