More Photos From Me to Thee: Factors Influencing the Intention to Continue
Sharing Personal Photos on an Online Social Networking (OSN) Site among Young
Adults in the Netherlands
Ardion D. Beldada and Sabrina M. Hegnerb
aDepartment of Communication Science, Faculty of Behavioural, Management, and Social Sciences, University of Twente, Enschede, The Netherlands;bDepartment of Economic Psychology, Faculty of Business and Health, University of Applied Sciences Bielefeld, Bielefeld, Germany
ABSTRACT
The current study investigated the factors influencing photo sharing continuance intention of Dutch Facebook users aged 18 to 25 years old. The focus was specifically on personal photos (those that include the person sharing), as their disclosure is expected to result in privacy-related issues for the individual sharing. An online survey with 473 respondents was implemented to test the various research hypotheses and to address the question on whether or not the impact of several disclosure predictors differs between men and women. Results of structural modeling analysis (multigroup analysis) reveal that for both male and female users, self-presentation-related and communication-related benefits and habit positively influence their photo sharing continuance intention. Furthermore, descriptive social norms and compe-tence-based trust in Facebook significantly contribute to female users’ repeat photo sharing intention.
1. Introduction
Active participation in Facebook requires users to disclose various types of personal information. Information disclosed to the plat-form could either be static (inplat-formation that normally does not change; for example, name, date of birth) or non-static (informa-tion that frequently changes; for example, updates of daily activ-ities, thoughts on certain issues, photographs; Beldad, 2015). Photographs of Facebook account owners are normally shared on a massive scale either to manage an impression or to commu-nicate using visual cues to a broader audience. Citing Van House, Davis, Ames, Finn, and Viswanathan (2005), Van House (2007) noted that photographs, in general, perform four important social functions, namely (a) construction of narratives about the self, (b) maintenance and display of relationships, (c) self-presentation, and (d) self-expression.
While the benefits of sharing photographs on Facebook are aplenty (e.g., self-presentation, entertainment, and communica-tion), the risks associated with their disclosure also abound. In her book I Know Who You Are and I Saw What You Did: Social Networks and the Death of Privacy, Andrews (2012) reported real-life cases of people who experienced the negative ramifications of their decision to post photos on Facebook—ranging from legal battles over child custody to employees losing their jobs.
Despite the unwarranted negative effects of photo sharing on Facebook users’ privacy sphere, one still wonders why the posting of photos on a massive scale proceeds unabated. One report estimated that people share and upload approximately 1.8 billion photos a day (Knibbs,2014). Can one assume, then, that those who post photos on Facebook are not concerned
about their personal privacy at all? Or are they just swayed by the benefits associated with photo sharing, despite having information privacy concerns?
It should be noted that the term“photograph” used through-out this primarily article refers to a more personal type—one that always includes the person sharing either alone or with other individuals (e.g., friends, family) and who may or may not be engaged in a particular activity (e.g., vacationing, visiting a restaurant) or be capturing a special moment or event. This emphasis is critical since the sharing of a “more personal” photograph is expected to result in privacy-related issues for the individual sharing compared to the posting of more neutral photos (e.g., pictures of celebrities, scenic pictures).
Beldad, De Jong, and Steehouder (2011a) argue that the process involved in the decision on whether or not to share or continue sharing personal information (under which photo-graphs could be categorized) may not really be that simple since various factors contribute to that decision. This point prompts the questions on what really determines the decision of online social networking (OSN) site users (in this study, Facebook) to continue sharing their photographs on the platform and on whether or not the factors influencing photo sharing continu-ance intention differ between male and female users.
2. Theoretical Framework 2.1. Photo Sharing on Facebook
Although studies (e.g., Christofides, Muise, & Desmarais,2011; Hollenbaugh & Ferris,2014; Kisekka, Bagchi-Sen, & Rao,2013)
CONTACTArdion D. Beldad, PhD, a.d.beldad@utwente.nl Department of Communication Science, Faculty of Behavioural, Management, and Social Sciences, University of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands.
2017, VOL. 33, NO. 5, 410–422
http://dx.doi.org/10.1080/10447318.2016.1254890
© 2017 The Author(s). Published by Taylor & Francis
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.
into the factors influencing personal information disclosure on OSN sites abound, the mechanism behind the disclosure of photos, as a specific type of information, is not yet fully under-stood. Thus far, research into photo sharing on Facebook has concentrated on an array of themes such as face/frame ratio for Facebook profile picture (Huang & Park, 2013), decisions on when to make posted photos private or public (Litt & Hargittai, 2014), and the contents and the amount of photos shared (Hum, Chamberlin, Hambright, Portwood, & Schat, 2011). The mechanism behind photo sharing (as something different from sharing person information, in general) certainly merits atten-tion since the decision to share various types of personal infor-mation (e.g., photographs, daily activities, thoughts on certain issues) might have varying privacy implications for the ones sharing, just as the determinants of their disclosure would expectedly be different (Beldad,2015).
2.2. The Impact of Benefits
The notion that people willingly share their personal infor-mation online due to the benefits that can be derived from the act resonates in the literature on information disclosure and online privacy (Beldad, Van Der Geest, De Jong, & Steehouder, 2012). These benefits could range from com-munication and relationship maintenance (Ellison, Steinfield, & Lampe, 2007; Krasnova, Spiekermann, Koroleva, & Hildebrand,2010) to new relationship creation (Krasnova et al., 2010). From a purely calculative point of view, people will not hesitate to share their personal infor-mation, despite privacy risks, if the perceived value of the benefits derived from the disclosure act outweighs the expected costs of disclosure (Berendt, Gunther, & Spiekermann, 2005; Culnan & Bies, 2003; Norberg & Dholakia, 2004; Olivero & Lunt, 2004).
A recent study by Beldad (2015) reported that, in gen-eral, the benefits of disclosure prompt users to share non-static information (e.g., photos, statements of current activ-ities) on OSN sites. Specifically, however, as evidenced by a few studies, relationship maintenance (Hollenbaugh & Ferris,2014,2015; Krasnova et al.,2010) and the enjoyment one derives from the disclosure act significantly drive peo-ple’s information disclosure behavior (Krasnova et al.,2010; Ng, 2014).
Moreover, although the study by Krasnova et al. (2010) did not find a direct effect of self-presentation on disclo-sure, Ng’s (2014) research revealed that a statistically sig-nificant positive relationship exists between presentation and disclosure. More importantly, still, self-presentation has been identified as an important factor influencing the frequency of online photo sharing (Hunt, Lin, & Atkin, 2014). These findings undergird the first set of research hypotheses:
Hypothesis 1: (a) Entertainment-related benefits, (b) self-pre-sentation-related benefits, and (c) communica-tion-related related benefits positively influence the intention to continue posting photos on Facebook of users aged 18 to 25 years.
2.3. The Impact of Trust in Facebook
The risks to one’s information privacy when using Facebook could be attributed either to the actions of both the parties receiving the information (Facebook and contacts) or to those entities with the technical proficiency to gain unauthorized access to shared information (Beldad,2015). With the reality of risks comes the need for trust, as the latter is only relevant when people are uncertain about the consequences of their exchanges with others (Lewis & Weigert, 1985). For this study, the frequently cited operationalization of trust as“the willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the trustor, irrespec-tive of the ability to monitor or control that other party” (Mayer, Davis, & Schoorman,1995, p. 712), will be used.
The effect of trust on people’s willingness to disclose per-sonal information has been reported in several studies. Mesch (2012) found that online trust is associated with the disclosure of personally identifiable information. However, considering the sources of disclosure-related risks, it makes sense to dif-ferentiate the impact of trust in two targets (Facebook and online network members), which, according to previous stu-dies (e.g., Krasnova, Veltri, & Günther, 2012; Lo & Riemenschneider, 2010), positively influence people’s perso-nal information disclosure behavior on OSN sites.
Since information privacy risks could be attributed both to the actions of Facebook and to those of external parties, trust in the site must be operationalized based on McLain and Hackman’s (1999) view of trust as a two-dimensional construct: competence-based trust and character-based trust. In an online setting, while the first dimension (the extent to which a site possesses the technical expertise to protect users’ privacy) is expected to counter people’s concern about third-party infor-mation abuse, the second dimension (the extent to which the site will treat users’ information in an ethical way) could reduce concerns regarding the propensity of a site to exploit users’ personal information (Beldad, De Jong, & Steehouder,2011b). Based on these points, another set of hypotheses is advanced:
Hypothesis 2: (a) Competence-based trust and (b) character-based trust in Facebook positively influence the intention to continue posting photos on Facebook of users aged 18 to 25 years.
2.4. The Impact of Trust in Facebook Contacts
Trust in one’s online network contacts also matters in the decision to share personal information on Facebook. The pivotal role of this variable in disclosure intention or beha-vior has been confirmed in a study into the sharing of location-based information (Beldad & Kusumadewi, 2015). Expectedly, the importance of trust in network contacts could be heightened by users’ concerns about the negative consequences of location information sharing (e.g., infor-mation as a source of gossip; Fusco, Michael, & Michael, 2010) or to the general apprehension that shared location
information could be misappropriated by their contacts (Zhao, Lu, & Gupta, 2012).
Since it is impossible for Facebook site users to know how their shared personal information will be used by their contacts, the former can only trust that the latter will not misappropriate those information (Krasnova et al.,2010). These points result in the next hypothesis:
Hypothesis 3: General trust in Facebook contacts positively influences the intention to continue posting photos on Facebook of users aged 18 to 25 years.
2.5. The Impact of Social Norms
Reviewing various empirical research on social influence, Cialdini and Goldstein (2004) underscored the pivotal role of social norms in nudging people to perform various forms of behavior in different contexts. Social norms, they added, could be differentiated into two, namely injunctive norms (referring to what people typically approve or disapprove) and descriptive norms (referring to what people typically do). Currently, most research into the use of OSN sites (e.g., Braun,2013; Chen, 2014; Cheung & Lee, 2010) and the dis-closure of personal information on these sites (Chang & Chen, 2014; Van Gool, Van Ouytsel, Ponnet, & Walrave, 2015; Varnali & Toker, 2015) has taken a narrow view on the role of social influence by just looking at the impact of injunctive social norms (alternatively labeled as subjective norms) on the two behaviors mentioned. Nonetheless, several studies have confirmed the positive effect of injunctive norms on personal information disclosure intention (Chang & Chen, 2014; Van Gool et al., 2015) and behavior (Varnali & Toker, 2015) on OSN sites.
Thus far, only a handful of studies (e.g., Chen, Yen, & Hwang, 2012; Cheung, Chiu, & Lee, 2011) have reported on the positive effect of descriptive social norms, also referred to as “critical mass” (Chen, Yen, & Hwang,2012), on the use of an OSN site. One thing that still has not been clearly understood is the extent to which people’s awareness of the information sharing behavior of their contacts (descriptive norm) would impact their intention to share information on an OSN site.
In line with Cialdini and Goldstein’s view of the influence of the two social influence dimensions on behavioral intention or behavior, this research postulates that people’s photo shar-ing continuance intention is a function of both injunctive and descriptive social norms. Hence, the fourth hypothesis is advanced:
Hypothesis 4: (a) Injunctive social norms and (b) descriptive social norms positively influence the intention to continue posting photos on Facebook of users aged 18 to 25 years.
2.6. The Impact of Habit
The first three clusters of personal information behavior pre-dictors just discussed are virtually hedged on the rational
mechanism governing human actions. From a purely social exchange perspective, one’s decision to disclose valued infor-mation is anchored on the belief that disclosure will be ben-eficial for the individual disclosing (Beldad et al., 2011a). Additionally, the effects of social influence and attitude (e.g., trust) on behavioral intention have been heavily emphasized in rational theories of human behavior (e.g., Theory of Planned Behavior; Ajzen,1991).
However, one should recognize the fact that purely non-rational factors could also spur an individual to perform a particular behavior. As Kahneman (2011) claimed, human actions emerged from two systems—with system 1 invol-ving the automatic and quick performance of an act with less or no effort and system 2 requiring a thorough mental deliberation prior to one’s engagement in an action. For instance, a person’s buying behavior may not always be a result of a more rational process of reflection but might also be activated by impulse (Strack & Werth, 2006). Indeed, as Strack and Deutsch (2004) contend, people may act in a certain way mindlessly or automatically, with-out paying attention to the merit of that action’s outcome, just as people might engage in an action that they have been repeatedly pursuing in the past. The last point echoes Ajzen’s (2002) view on the significance of habits in shaping behavior. Verplanken and Aarts (1999) define habits as “learned sequences of acts that have become automatic responses to specific cues, and are functional in obtaining certain goals or end-states” (p. 104).
For some people, using Facebook is a habitual act (Burke, Kraut, & Marlow, 2011; Debatin, Lovejoy, Horn, & Hughes, 2009), as a consequence of its being embedded into their daily routines and rituals (Debatin et al., 2009). Certain activities, for instance, writing on contacts’ Facebook walls, emerge from people’s habits (Smock, Ellison, Lampe, & Wohn, 2011). Additionally, some people also post their photos on Facebook out of habit and without thinking of the reasons behind their decision to post (Wang et al., 2011). What is apparent from these findings is that people’s decision to share information about themselves, regardless of type, is not fre-quently predicated on a “rational” consideration but could also be triggered by nonrational factors such as habit. Hence, the last hypothesis is proposed:
Hypothesis 5: The habit of sharing information online posi-tively influences the intention to continue posting photos on Facebook of users aged 18 to 25 years.
2.7. The Moderating Role of Gender?
Men and women have been found to substantially differ in their levels of online privacy. More specifically, female inter-net users are known to be more concerned about online privacy violations than their male counterparts (Cho, Rivera-Sanchez, & Lim,2009; O’Neil, 2001; Sheehan, 1999; Youn & Hall, 2008). Additionally, women perceive more information sharing-related risks than men do (Youn & Hall,2008).
When looking at the pattern of their personal information disclosure, however, heightened privacy concerns among women would certainly be the explanation for their inclination to share less personal information on OSN sites (Ji, Wang, & Zhu, 2014) and for their penchant for highly private OSN site profiles (Kisekka et al., 2013; Thelwall, 2008) when compared to men. Differences in the types of information men and women share on OSN sites have also been noted, with the former more likely to share their mobile phone number, home address, and instant messaging screen names than the latter (Taraszow, Aristodemou, Shitta, Laouris, & Arsoy,2010).
Interestingly, however, women tend to share more photos on an OSN site and to update their wall posts more than men do (Rui & Stefanone,2013). Additionally, concerning themes highlighted in profile photos, gender differences have been noted, with men’s profile photos focusing on status and risk-taking behavior and women’s photos focusing on familial relations and emotional expression (Tifferet & Vilnai-Yavetz, 2014). Furthermore, in terms of photo orientation, women prefer using portrait photo-graphs for their profile pictures, while men frequently use full-body shots that show the environment and other people (Haferkamp, Eimler, Papadakis, & Kruck, 2012). Variations in the amount and the types of photos shared on an OSN site between men and women, therefore, prompt the question:
“To what extent do the factors that influence the intention to continue posting photos of male and female Facebook users aged 18 to 25 years differ?”
2.8. The Research Model
Figure 1 shows the research model that will be tested in this research. Based on the theoretical discussion above, nine factors are hypothesized to positively influence the intention to continue posting photos on Facebook, just as it is surmised that the impact of those factors on the intention of interest might substantially differ between male and female users.
3. Methodology
3.1. Design and Procedure
The hypotheses proposed for this study were tested using data collected through an online survey for 2 weeks. A link to an online questionnaire was sent to Dutch Facebook users. However, the fact that people from the age cluster of 18 to 25 years old still constitute majority of Facebook users (Pew,2015) provided the rationale to focus on respondents from the above-mentioned age category. In the Netherlands, specifically, over 80% of people aged 18 to 25 years old are reported to be active users of OSN sites such as Facebook (Statistics Netherlands, 2013).
A snowball sampling with the help of 57 students who followed the pre-master communication science program of a Dutch university was employed to gather data. Although the sampling approach has its share of limitations (e.g., sample bias; Bhutta, 2012), it also enabled the researchers to collect
Entertainment-related benefits (ENT) Self-presentation-related benefits (SPR) Communication-related benefits (COM) Competence-based trust in Facebook (TCO) Character-based trust in Facebook (TCH)
General trust in Facebook contacts (TFC)
Injunctive social norms (INJ)
Descriptive social norms (DES)
Habit (HAB)
Intention to continue sharing photos (INT) GENDER B E N E F I T S T R U S T S O C I A L N O R M S H1a H1b H1c H2a H2b H3 H4a H4b H5
data within a reasonable timeframe and with less cost. Furthermore, to enable the widespread dissemination of the online questionnaire, students did not only approach indivi-duals within their social networks but also posted the link to the online survey on popular OSN sites in the Netherlands. After a two-week data collection, completed and useful ques-tionnaires from 473 Facebook users who belonged to the age category of interest were obtained. The system used for designing the online questionnaire prevented respondents from failing to respond to some items; hence, the entire dataset did not have missing data.
3.2. Respondents
Of the 473 respondents who completed the online question-naire for analysis, 179 (37.8%) were male, while the remaining 294 (62.2%) respondents were female. The average age of research respondents was 22.18 (SD = 1.89). In terms of educational qualification, most of the respondents were highly educated, as 372 (79%) have obtained either an applied science bachelor degree (the Dutch hogerberoepsonderwijs or higher professional education) or a university-level bachelor. This somehow reflects what is currently known about the educational level of OSN users in the Netherlands—as close to 75% of Dutch individuals who use any OSN site in 2013 possessed higher education (Statistics Netherlands,2014).
Additionally, approximately 85% (n = 402) of the 473 respondents had been Internet users for more than 10 years at the time of the study. In terms of Facebook use, 163 respondents had been using the site for 4 years and another 183 respondents had been Facebook users for 3 years. The average number of contacts respondents had in their Facebook network was 220 (maximum 250, minimum 25; SD = 50.34).
3.3. Research Instrument
The dependent variable“intention to continue sharing photo on Facebook” (INT) was measured with six newly formulated statements. For this construct, the emphasis was on whether or not respondents were intending to continue sharing photos of themselves, of themselves with their family and friends, of themselves on a holiday location or visiting an “interesting” site or spot, and of themselves engaged in a certain activity. The choice for these photograph types was based on research results that younger users share photographs that depict rela-tionships by featuring themselves with family and friends (Mendelson & Papacharissi, 2011; Siibak, 2009), photos of important moments of their lives (e.g., being on holidays; Siibak, 2009), photos of users engaged in certain activities (e.g., driving a car; Siibak,2009), and photos depicting solely the users (Mendelson & Papacharissi,2011).
The first predictor“perceived benefits of sharing” was split into three dimensions, namely entertainment-related (ENT), self-presentation-related (SPR), and communication-related (COM) benefits. Three items by Baek, Holton, Harp, and Yaschur (2011) were slightly modified to measure ENT, while two items by Walther, Slovacek, and Tidwell (2001)
were also substantially reformulated to measure SPR. COM was measured with three items by Krasnova et al. (2010).
“Trust in Facebook” was conceptualized as a two-dimen-sional construct: competence-based trust (TCO) and charac-ter-based trust (TCH; Beldad et al., 2011a). Two items were used to measure the first trust dimensions, while three items were selected for the second dimension. The five items were all derived from the scale by Beldad et al. (2011b) and were substantially modified to fit the research context. “General trust in Facebook contacts” (TFC) was measured with three originally formulated items.
Items by Qin, Kim, Hsu, and Tan (2011) were used to measure both“injunctive social norms” (INJ) and “descriptive social norms” (DES), with three items for each construct. Additionally, “habit” (HAB) was measured with three items by Limayem, Hirt, & Cheung (2007) and Verplanken and Orbell (2003).
Both the dependent and the independent variables were measured on a 7-point Likert scale (with 7 representing “strongly agree” and 1 “strongly disagree”). All items (see the Appendix section) were translated from English to Dutch for the convenience of the respondents.
4. Results 4.1. Measures
To address the questions on which factors influence photo sharing continuance intention and on whether the predictors of photo sharing intention on Facebook differ between male and female Facebook users aged between 18 and 25 years, a multigroup structural equation modeling was performed. However, before testing the structural model, requirements of instrument reliability and validity must be fulfilled. Maximum likelihood estimate procedures require the fulfill-ment of the normality assumption (Curran, West, & Finch, 1996). Based on the guidelines for normality assessment (skewness < 3; kurtosis < 10) Kline (2005) prescribes, the values for the complete dataset and for the male and the female datasets could be considered adequately normal, and hence, the empirical test for multivariate normality was satisfactory.
Confirmatory factor analysis using AMOS 22.0 was per-formed to determine the scales’ convergent validity. According to Hair, Black, Babin, and Anderson (2006), an item is significant if its factor loading (FL) is higher than .50. At the construct level, Hair et al. (2006) recommend calculat-ing composite reliability (CR) instead of Cronbach’s alpha when using structural equation modeling (SEM). The average variance extracted (AVE) for the different constructs was also calculated, as it measures the amount of variance captured by the construct in relation to the amount of variance that is attributed to the measurement error (Fornell & Larcker, 1981).
Both CR and AVE provide a good indication of the con-structs’ convergent validity. AVE values are adequate for all the factors (>.5; Fornell & Larcker,1981), while CR values are all higher than the recommended value of .6 (Bagozzi & Yi,
1988).Table 1presents the factor loadings for all the measures and the CR and AVE values for all the constructs.
Additionally, discriminant validity of the various constructs was also determined by comparing the AVE of a specific construct to the values of the squared correlation between that construct and other constructs. Fornell and Larcker (1981) propose that to establish discriminant validity, AVE must be higher than the squared correlation coefficient between factors. This requirement was met across all pairs of constructs for the overall sample (n = 473) and for the male (n = 179) and the female (n = 294) samples. Table 1 shows the convergent and the discriminant validity of the research instruments.
Table 2presents the fit indices for the measurement model. Several indices were used to assess the fit of both measurement and structural models. Since X2is sensitive to sample size, the ratio of X2to the degrees of freedom was considered. A value smaller than 5 for X2signifies an acceptable fit between the hypothesized model and the sample data (Wheaton, Muthen, Alwin, & Summers,1977).
Additionally, three fit indices were considered to draw comparisons to the baseline model. Hu and Bentler (1999) and Schreiber, Stage, King, Nora, and Barlow (2006)
recommend using the root-mean-square error of
approximation (RMSEA) as a measure of absolute fit and the comparative fit index (CFI) and Tucker–Lewis index (TLI) to determine the model’s incremental fit. An accep-table model fit is achieved if the CFI and the TLI values are higher than .90 and the RMSEA value is lower than .08 (Hair et al., 2006). Based on these criteria, the measure-ment model tested with data from 473 respondents can be considered to have a good fit [X2 = 707.70, df = 389, X2/ df = 1.82, TLI = .97, CFI = .97, RMSEA = .04].
To further see whether there are hidden relationships among the independent variables prior to the test of the structural model, a correlation analysis was also conducted. Values on Table 3 clearly indicate that strong correlations among the independent variables do not exist. Table 3 also shows the mean and the standard deviation values for both the independent variables and the dependent variable.
4.2. Multigroup Analysis
Structural equation modeling was performed to test the research model. More importantly, to determine whether or not the factors influencing the intention to continue sharing photos on Facebook differ between male and female
Table 1.Convergent and discriminant validity of the research constructs.
Overall sample (n = 473) Male sample (n = 179) Female sample (n = 294)
FL CR AVE FL CR AVE FL CR AVE
Entertainment (ENT) ENT1 .805 .884 .718 .820 .871 .693 .796 .891 .732
ENT2 .897 .903 .794
ENT3 .838 .769 .769
Self-presentation (SPR) SPR1 .876 .879 .784 .886 .885 .793 .864 .876 .779
SPR2 .895 .895 .901
Communication (COM) COM1 .812 .851 .656 .830 .864 .679 .793 .838 .632
COM 2 .828 .832 .823
COM3 .789 .810 .769
Competence-based trust (TCO) TCO1 .850 .817 .690 .884 .770 .630 .820 .851 .741
TCO2 .811 .692 .900
Character-based trust (TCH) TCH1 .832 .923 .799 .782 .912 .777 .867 .932 .820
TCH2 .943 .977 .921
TCH3 .904 .874 .927
General trust in FB contacts (TFC) TFC1 .884 .933 .823 .853 .931 .819 .904 .933 .823
TFC2 .964 .961 .975
TFC3 .871 .897 .837
Injunctive social norms (INJ) INJ1 .947 .957 .880 .912 .940 .831 .967 .969 .913
INJ2 .992 1.012 .983
INJ3 .872 .799 .915
Descriptive social norms (DES) DES1 .820 .914 .780 .813 .917 .788 .811 .905 .761
DES2 .946 .970 .927
DES3 .880 .873 .876
Habit (HAB) HAB1 .759 .862 .677 .758 .852 .659 .758 .865 .682
HAB2 .933 .917 .936
HAB3 .765 .750 .772
Intention to continue sharing photos (INT) INT1 .795 .914 .641 .786 .915 .643 .795 .913 .638
INT2 .842 .810 .858
INT3 .863 .847 .872
INT4 .705 .731 .681
INT5 .791 .791 .801
INT6 .796 .840 .769
Table 2.Fit indices of the invariance test for the measurement model. X2
df X2
/df RMSEA CFI TLI CFIΔ TLIΔ
Configural invariance model 1236.94 778 1.590 .04 .96 .95
Metric invariance model 1267.97 799 1.587 .04 .96 .95 .00 −
respondents, multigroup analysis was conducted. Test of the initial model reveals that its fit is almost acceptable, as indi-cated by the figures inTable 4.
However, after a close inspection of the modification indices, it was decided to correlate two benefit constructs (impression management and updating) and the two trust in Facebook constructs (competence-based trust and character-based trust). This slight modification in the model resulted in the reduction in the values for the X2/df and RMSEA and an increase in the CFI and TLI values, which means that the proposed research model reasonably fits the data. Table 5 presents the fit indices for the revised structural model.
A meaningful gender-specific comparison, however, requires that the instrument measuring the theoretical con-struct of interest must possess adequate cross-gender equiva-lence. To meet this requirement, configural, metric, and scalar invariances need to be established to compare the findings for the two groups of respondents (Hair et al.,2006; Steenkamp & Baumgartner,1998; Vandenberg & Lance,2000).
Looking at the fit values (e.g., CFI, TLI) and the ΔCFI value (a value higher than .01 indicates a significant drop in fit; Cheung & Rensvold,2002) onTable 5, it can be concluded that the full model has configural invariance; hence, the test of metric invariance can proceed (Vandenberg & Lance, 2000). With metric invariance, factor loadings of items across groups are constrained to be equal (Vandenberg & Lance, 2000). Again, fit values (Table 5) indicate that metric invariance has been satisfied, which means that the pattern of the rela-tionship between indicators and the latent constructs they measure are similar across two groups (Frazier et al.,2014).
As the model is metrically invariant, assessment of scalar invariance was then performed. Fit indices again indicate that scalar invariance has been achieved, which suggests that group differences in the means of the observed items are caused by the variations in the means of the latent construct (Steenkamp & Baumgartner, 1998). As the model possesses configural, metric, and scale invariance, a multigroup analysis was deemed appropriate.
4.3. Hypothesis Testing Based on the Untrimmed Model Analyses of the structural model for both male and female respondents reveal that among male Facebook users, aged 18 to 25 years, INT is strongly attributable to SPR (β = .43, p < .001) and COM (β = .33, p < .001). Moreover, HAB (β = .232, p < .001) also plays a role in their inclination to continue sharing photos on the aforementioned platform. These findings support hypotheses 1b, 1c, and 5.
For female Facebook users, the effects of SPR (β = .39, p < .001), COM (β = .18, p. < .01), and HAB (β = .31, p < .001) on INT are also statistically significant. Hence, hypotheses 1b, 1c, and 5 are also supported. Interestingly, however, DES (β = .18, p < .001), but not INJ, has also been found to strongly predict female Facebook users’ INT, thereby support-ing hypothesis 4b.
In the slightly modified structural model, SPR and COM were correlated, just as correlation was established between TCO and TCH. Correlations between the first pair (male: r = .56, p < .001; female: r = .46, p < .001) and the second pair (male: r = .75, p < .001; female: r = .63, p < .001) are statistically significant.
4.4. Model Trimming
Since ENT, TFC, INJ, and TCH have no significant effects on INT for both male and female respondents, the model was eventually trimmed with the removal of those constructs in various phases. Stepwise removal of the constructs was based on their p values. INJ was removed first, then TFC, and then ENT, and finally TCH. The removal of INJ, TFC, and ENT did not change the results, as SPR, COM, HAB, and DES remained significant predictors of INT. However, the removal of TCH in the fourth stage resulted in TCO’s statistically significant effect on female respondents’ INT. Hence, despite the fact that TCO did not have a statistically significant effect on INT in the original model, TCO was not removed in the next phase, and thus, a fifth model was not tested.
Table 3.Mean scores, standard deviation values, and inter-correlations of the research constructs (N = 473).
M (SD) SPR ENT COM TCO TCH TFC INJ DES HAB INT
SPR 4.16 (1.45) 1 ENT 3.66 (1.54) .29** 1 COM 4.93 (1.17) .44** .29** 1 TCO 4.04 (1.30) .20** .12* .18** 1 TCH 4.24 (1.40) .17** .12* .20** .61** 1 TFC 4.98 (1.25) .12* .04 .12* .26** .25** 1 INJ 2.28 (1.29) .11* .19* −.02 .02 .00 .02 1 DES 5.27 (1.11) .18** .16* .21** .08 .11* .26** .00 1 HAB 3.18 (1.48) .22** .38** .26** .23** .14** .08 .26** .28** 1 INT 4.50 (1.32) .51** .28** .48** .28** .21** .16** .08 .30** .43** 1
**Correlation significant at 0.01; *correlation significant at 0.05; constructs measured on a 7-point Likert scale—7 strongly agree to 1 strongly disagree.
Table 4.Fit indices of the invariance test for the original structural model. X2
df X
2
/
df RMSEA CFI TLI CFIΔ TLIΔ Configural invariance
model
1882.42 848 2.22 .05 .91 .90 − −
Metric invariance model 1912.33 869 2.20 .05 .90 .90 .01 .00 Scalar invariance model 1924.63 878 2.19 .05 .90 .90 .00 .00
Table 5.Fit indices of the invariance test for the revised structural model. X2
df X
2/
df RMSEA CFI TLI CFIΔ TLIΔ Configural invariance
model
1580.04 844 1.87 .04 .93 .93 − −
Metric invariance model 1607.79 865 1.86 .04 .93 .93 .00 .00 Scalar invariance model 1620.75 874 1.85 .04 .93 .93 .00 .00
The trimmed model (4th model on Table 6 with five predictors: self-presentation- and communication-related benefits, descriptive social norms, competence-based trust in Facebook, and habit) of photo sharing continuance intention still possesses an acceptable fit. For male respondents, their intention to continue sharing photos on Facebook is predi-cated on three factors, namely self-presentation-related bene-fits (β = .40, p < .001), communication-related benebene-fits (β = .35, p < .001), and habit (β = .23, p < .001). Therefore, hypotheses 1b, 1c, and 5 are still supported.
In the case of female respondents, however, their Facebook photo sharing continuance intention is influenced by self-presentation-related benefits (β = .38, p < .001), habit (β = .30, p < .001), descriptive social norm (β = .18, p < .001), communication-related benefits (β = .17, p < .01), and competence-based trust in Facebook (β = .16, p < .01). These results support hypotheses 1b, 1c, 5, 4b, and 2a, respectively.
Presented in Table 6 are the fit indices of the invariance test for four alternative models (resulting from the stepwise removal of several predictors).
The significance of the overall difference in the effect of a predictor on the dependent variable across the two groups was also determined through a pairwise comparison based on a chi-square difference test. The X2value of 791.85 (df = 294) for the unconstrained model and the X2 value of 801.84 (df = 312) for the fully constrained model resulted in a difference of 9.98 (df = 18; p = .93), which suggests that the model for both male and female respondents is invariant, which further means that gender does not moderate the relationships between the different independent variables and INT.
A path analysis, by constraining one path at a time, was performed to fully verify this. For moderation to occur, the X2 value for a specific path must be greater than the X2values in at least one of the three levels of confidence (90% = 794.56; 95% = 795.69; 99% = 798.49). X2 values of the different constrained path onTable 7 clearly indicate that the predic-tors of INT do not differ between male and female respondents.
Additionally, pairwise comparison by looking at the critical ratio values (De Marco, Bonis, Vrignaud, Henry-Feugeas, & Peretti, 2006) also indicated that not one of the five paths is significant, which fully confirms the finding that gender does
not moderate the effects of the five predictors on photo sharing continuance intention. Results of the pairwise comparison using the critical ratio values are presented inTable 8.
5. Discussion of Results, Future Research Directions, and Practical Implications
5.1. Discussion of Results
Various factors are known to shape people’s decision to share various types of personal information on Facebook. While privacy concerns might reduce people’s disposition to share information, the urge to engage in self-presentation (SPR) and visual communication (COM), for instance, would suffice to nudge OSN site users to self-disclose. Results of the current study echo some of the findings reported in previous research. For both male and female Facebook users, self-presentation appears to be a very strong predictor of their intention to continue sharing photos on the site. The impact of impression management on information disclosure has not been solidly established in previous, except for one recent study (Ng,2014) that confirmed the positive relationship between the two constructs.
Nonetheless, the first finding of the current study is not entirely surprising since social networking sites primarily enable people to engage in self-presentation, considering the
Table 6.Fit indices of the invariance test for four alternative models.
X2 df X2/df RMSEA CFI TLI
Δ CFI
Δ TLI
Model 1 (INJ removed) Configural invariance model 1337.98 678 1.97 .05 .93 .92
Metric invariance model 1358.27 697 1.95 .05 .93 .92 .00 .00
Scalar invariance model 1370.27 705 1.94 .05 .93 .92 .00 .00
Model 2 (INJ & TFC removed) Configural invariance model 1079.41 530 2.04 .05 .93 .92
Metric invariance model 1096.51 547 2.01 .05 .93 .92 .00 .00
Scalar invariance model 1105.65 554 1.99 .05 .93 .92 .00 .00
Model 3 (INJ, TFC, & ENT removed) Configural invariance model 813.62 400 2.03 .05 .94 .93
Metric invariance model 826.89 415 1.99 .05 .94 .93 .00 .00
Scalar invariance model 834.90 421 1.98 .05 .94 .93 .00 .00
Model 4 (INJ, TFC, ENT,& TCH removed) Configural invariance model 643.15 290 2.22 .05 .94 .92
Metric invariance model 650.37 303 2.15 .05 .94 .93 .00 .01
Scalar invariance model 655.16 308 2.13 .05 .94 .93 .00 .00
Table 7.X2values for the various paths within the model.
Path X2 (df) SPR→ INT 792.02 (295) COM→ INT 793.19 (295) TCO→ INT 792.55 (295) DES→ INT 793.20 (295) HAB→ INT 791.86 (295)
Table 8.Standardized parameter estimates of the structural model.
Parameters Male (n = 179) Female (n = 294) Critical ratio
IMP→ INT 0.34 (.000)* 0.30 (.000)* −0.43
COM→ INT 0.34 (.000)* 0.19 (.007)** −1.36
TCO→ INT 0.07 (.421) 0.14 (.006)** 0.78
DES→ INT 0.10 (.139) 0.22 (.000)* 1.36
HAB→ INT 0.26 (.000)* 0.26 (.000)* 0.03
fact that one’s construction of a Facebook page emerges from an expectation “of a potential audience as much as it is a careful preening and production of the self” (Marshall,2010, p. 40). While the relationship between self-presentation and sharing personal information, in general, has remained ambivalent, previous studies (e.g., Hunt et al., 2014) have emphasized the important role of self-presentation in indu-cing photo sharing behavior. These findings could suggest that people prefer to engage in self-presentation by using visual cues (e.g., photos) instead of textual cues (e.g., daily updates), since the former might be perceived to be more effective in providing a more accurate and complete portrayal of the self than the latter.
The need to communicate with contacts has also been found to influence male and female users’ photo sharing continuance intention. Relationship maintenance, through regular communication via daily updates, is known to increase personal information sharing. Results of this study imply that people who regularly post personal photos on their Facebook page might have a strong need to visually commu-nicate with their contacts, considering the ability of photo-graphs’ cues to convey a more vivid notification of where and what those FB users have been up to.
Habit’s impact on photo sharing continuance intention should not also be discounted, as indicated by results of the current study. The fact that Facebook has been deeply embedded in the lives of young users signifies that whatever these users do on this site is highly ritualized and routinized (Debatin et al., 2009). This notion of the routinization of Facebook activities could be a sensible explanation for the strong effect of habit on both male and female users’ intention to continue posting photos on the site.
Considering the risks involved in sharing photos on Facebook (e.g., misappropriation by third parties, OSN site’s use of posted photos for commercial use; Andrews, 2012), users can only trust that the photos they share will not be abused. In the context of information sharing on OSN sites, the relevant trust targets primarily include the platform used for sharing (e.g., Facebook) and the intended recipient of the information, specifically members of a person’s online net-work (Beldad & Kusumadewi, 2015). In the current study, trust in Facebook and not in online contacts significantly contributes to female users’ willingness to continuously share photos on the platform.
Specifically, competence-based trust in the platform (the belief that the platform has the expertise and the technology to safeguard users’ information from unauthorized third-party access) can potentially increase photo sharing inten-tions. The fact that the effect of trust on photo sharing con-tinuance intention is significant among female users only somehow suggests that their decision to take certain risks (e.g., sharing personal photos online) is predicated on expec-tations that in the process they would be spared from experi-encing risky outcomes through certain safety measures (e.g., the platform’s use of security mechanisms). A possible expla-nation for the impact of competence-based trust in Facebook on female respondents’ photo sharing continuance intention could be the high level of privacy concerns they have when compared to men—as women are known to be more
concerned about online privacy violations (Cho et al., 2009; O’Neil,2001; Sheehan,1999; Youn & Hall,2008) and perceive more information sharing-related risks (Youn & Hall, 2008) than men do.
Interestingly, however, the impact of descriptive norms on photo sharing continuance intention is statistically significant for female users only and not for their male counterparts. Reviewing results of previous studies, Fischer and Arnold (1994) underscored that men and women consistently differ in their conformity and susceptibility to social influence, with the latter more receptive to social influence than the former. And this conclusion could partly explain the result of the current study.
It should be noted, however, that in this case, what matters most for female users’ decision to continue sharing photos is their awareness of what their contacts are doing (descriptive social norms) and not what those contacts expect them to do (injunctive social norms). In several studies, the differential impact of the two social influence dimensions on various forms of behavioral intention or behavior has been noted. While some studies highlight that behavioral intentions are functions of injunctive but not of descriptive social norms (Reno, Cialdini, & Kallgren,1993; Smith & Louis,2008), other studies report that the impact of descriptive social norms on behavioral intention is stronger than that of injunctive norms (Baumgartner, Valkenburg, & Peter, 2011; White, Smith, Terry, Greenslade, & McKimmie,2009).
In this study, the focal role of descriptive social norms on the behavioral intention studied could be attributed to the strikingly low level of respondents’ perception of how their peers expect them to act. On the contrary, research respon-dents, especially women, tend to highly agree that most of their peers actively post personal photos on their Facebook sites. The fact that descriptive social norms have a strong effect on female users’ intention to continue posting photos might be an indication of a compelling need among younger FB female users to engage in a behavior currently deemed fashionable.
5.2. Future Research Directions
While this study reveals some interesting insights and rela-tively surprising results, additional research into the theme of photo disclosure either on Facebook or on other OSN sites could still deepen our understanding of people’s photo shar-ing behavior. Before the advent of online social networkshar-ing, people used to archive their photographs using an album to be viewed by a selected number of people. Nowadays, photo-graphs, whether or not fit for public viewing, are somehow willingly posted online, sometimes for a much wider audience.
As previous students already found that younger users tend to share more information on Facebook than their much older counterparts (Christofides et al., 2011) and that the motivations and usage patterns of younger and older differ (Brandtzaeg & Skjetne, 2010), and considering that respon-dents of the current study were Facebook users aged 18 to 25, one wonders whether or not the factors influencing their photo sharing continuance intention would be similar to
users from different age clusters. Would habit and descriptive social norms still matter for the photo sharing intentions of baby boomers (born between 1945 and 1960) and generation Xers (born between 1961 and 1980)? Or would the role of trust be more salient for older users, considering the possibi-lity that they might be more risk-averse than their younger counterparts?
In this study, the object of the intention—personal photo-graphs—was broadly operationalized (e.g., photos that always include the person sharing). Hence, future research could advance a more nuanced view of personal photos by taking into account their various types and the degrees of risks associated with their disclosure and to see whether or not the factors influencing the disclosure of these photo types would substantially differ. The impact of habit on respon-dents’ photo sharing continuance intention could be due to the decision to not post highly compromising pictures of themselves. Probably, trust might be a stronger influencer if the decision involves the posting of intimate or private pictures.
The generalizability of the results of this study is also limited by the approach used for data collection (snowball sampling technique). As the respondents used in the study may not fully reflect the views, attitudes, and behaviors of a general popula-tion of Facebook users (even in the Netherlands), interpretapopula-tion of the results in relation to the Facebook photo sharing behavior of young adults must be done with caution. Data from a more representative sample could be used to test the model proposed in this research.
Furthermore, with the study’s focus on respondents from a single national group, one needs to be cautious in using the research findings to explain photo sharing intention or beha-vior of individuals from another cultural or national group. Variations in how people from different cultures value infor-mation privacy might have implications for the effects of certain factors, such as trust and benefits, on photo sharing disclosure intention. Hence, a study with a strong cross-cul-tural orientation would certainly provide in-depth insights into what people with different cultural backgrounds regard important when deciding to post photos online, as it is also known that people from various cultures have various moti-vations for using Facebook (Ji et al.,2010).
5.3. Practical Implications of the Results
Despite the relevant contributions of the current study to research into personal information disclosure on OSN sites, results of this research also have useful implications for OSN sites. One important point to seriously consider is how trust in an OSN site could be earned to eventually prompt users to fully savor the benefits of online connectivity through regular information sharing. The current study shows that people, specifically women, will not hesitate to continue posting their photos on an OSN site that can be trusted to provide the technical infrastructure to protect their users’ personal information. Although most OSN sites have privacy settings that people can use to define what and how much information to share to whom, those settings are often insufficiently uti-lized (Beldad,2016), either because users know nothing about
the existence of these settings or they simply lack the knowl-edge to use them (De Rosa,2016). OSN sites, therefore, have the obligation to ensure that those settings are easily accessible and learnable without so much effort.
Furthermore, the importance of social influence for female users’ decision to continue sharing their photos means that OSN sites, in general, should strengthen the use of cues that could give users an impression of the photo sharing behavior of their contacts (regardless of the type of photos shared). While platforms such as Facebook allow users to update their contacts of contents (e.g., photos, comments) that have been posted, other features could still be embedded into the current practice, for instance, by allowing users the possibility to indicate the quantity of photos they have shared and the regularity of their sharing.
Acknowledgment
The authors would like to express their gratitude to Mr. Mark Tempelman, pre-master research project course coordinator at the University of Twente, and all the students of the pre-master research project (1st semester, 2013–2014) for their substantial help in collecting the necessary data for the study.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
References
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179–211.
Ajzen, I. (2002). Residual effects of past on later behavior: Habituation and reasoned action perspectives. Personality and Social Psychology Review, 6 (2), 107–122.
Andrews, L. (2012). I Know Who You Are and I Saw What You Did: Social Networks and Death of Privacy. New York, NY: Free Press. Bagozzi, R., & Yi, Y. (1988). On the evaluation of structural equation
models. Journal of the Academy of Marketing Science, 16 (1), 74–94. Baek, K., Holton, A., Harp, D., & Yaschur, C. (2011). The links that bind:
Uncovering novel motivations for linking on Facebook. Computers in Human Behavior, 27, 2243–2248.
Baumgartner, S., Valkenburg, P., & Peter, J. (2011). The influence of descriptive and injunctive peer norms on adolescents’ risky sexual online behavior. Cyberpsychology, Behavior, & Social Networking, 14 (12), 753–758.
Beldad, A. (2016). Sealing one’s online wall off from outsiders: Factors influencing the use of Facebook’s privacy settings among young Dutch users. International Journal of Technology and Human Interaction, 12 (1), 21–34.
Beldad, A., De Jong, M., & Steehouder, M. (2011a). A comprehensive theoretical framework for personal information-related behaviors on the Internet. The Information Society, 27, 220–232.
Beldad, A., De Jong, M., & Steehouder, M. (2011b). I trust not therefore it must be risky: Determinants of the perceived risks of disclosing personal data for e-government transactions. Computers in Human Behavior, 27, 2233–2242.
Beldad, A., & Kusumadewi, M. (2015). Here’s my location, for your information: The impact of trust, benefits, and social influence on location sharing application use among Indonesian university stu-dents. Computers in Human Behavior, 49, 102–110.
Beldad, A., Van Der Geest, T., De Jong, M., & Steehouder, M. (2012). Shall I tell you where I live and who I am? Factors influencing the behavioral intention to disclose personal data for online government
transactions. International Journal of Human-Computer Interaction, 28 (3), 163–177.
Beldad, A. D. (2015). Sharing to be sociable, posting to be popular: Factors influencing non-static personal information disclosure on Facebook among young Dutch users. International Journal of Web Based Communities, 11 (3/4), 357–374.
Berendt, B., Gunther, O., & Spiekermann, S. (2005). Privacy in ecom-merce: Stated preferences vs. actual behavior. Communications of the ACM, 48 (4), 101–106.
Bhutta, C. B. (2012). Not by the book: Facebook as a sampling frame. Sociological Methods & Research, 41 (1), 57–88.
Brandtzaeg, P. B., & Skjetne, J. H. (2010). Too many Facebook‘friends’? Content sharing and sociability versus the need for privacy in social network sites. International Journal of Human-Computer Interaction, 26 (11–12), 1006–1030.
Braun, M. (2013). Obstacles to social networking website use among older adults. Computers in Human Behavior, 29 (3), 673–680. Burke, M., Kraut, R., & Marlow, C. (2011). Social capital on Facebook:
Differentiating uses and users. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp. 571–580), Vancouver, CA. doi:10.1145/1978942.1979023
Chang, C., & Chen, G. (2014). College students’ disclosure of location-related information on Facebook. Computers in Human Behavior, 35, 33–38. Chen, S., Yen, D., & Hwang, M. (2012). Factors influencing the
con-tinuance intention to the usage of Web 2.0: An empirical study. Computers in Human Behavior, 28 (3), 933–941.
Chen, Y. (2014). See you on Facebook: Exploring influences on Facebook continuous usage. Behaviour & Information Technology, 33 (11), 1208–1218.
Cheung, C., Chiu, P., & Lee, M. (2011). Online social networks: Why do students use Facebook? Computers in Human Behavior, 27, 1137– 1343.
Cheung, C., & Lee, M. (2010). A theoretical model of intentional social action in online social networks. Decision Support Systems, 49 (1), 24– 30.
Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 9 (2), 233–255.
Cho, H., Rivera-Sanchez, M., & Lim, S. (2009). A multinational study on online privacy: Global concerns and local responses. New Media & Society, 11 (3), 395−416.
Christofides, E., Muise, A., & Desmarais, S. (2011). Hey mom, what’s on your Facebook? Comparing Facebook disclosure and privacy in ado-lescents and adults. Social Psychological and Personality Science, 3 (1), 48–54.
Cialdini, R., & Goldstein, N. (2004). Social influence: Compliance and conformity. Annual Review of Psychology, 55, 591–621.
Culnan, M., & Bies, R. (2003). Consumer privacy: Balancing eco-nomic and justice considerations. Journal of Social Issues, 59 (2), 323–342.
Curran, P., West, S., & Finch, J. (1996). The robustness of test statistics to non-normality and specification error in confirmatory factor analysis. Psychological Methods, 1 (1), 16–29.
De Marco, G., Bonis, M., Vrignaud, P., Henry-Feugeas, M., & Peretti, I. (2006). Changes in effective connectivity during inci-dental and intentional perception of fearful faces. NeuroImage, 30, 1030–1037.
De Rosa, D. L. (2016, June 2). Kids, safety and social media. The Huffington Post. Retrieved from http://www.huffingtonpost.com/ denise-lisi-derosa/kids-safety-and-social-me_b_9176810.html
Debatin, B., Lovejoy, J. P., Horn, A. K., & Hughes, B. N. (2009). Facebook and online privacy: Attitudes, behaviors, and unintended consequences. Journal of Computer-Mediated Communication, 15, 83–108.
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 (1), epub.
Fischer, E., & Arnold, S. (1994). Sex, gender identity, gender role atti-tudes, and consumer behavior. Psychology & Marketing, 11 (2), 163– 182.
Fornell, C., & Larcker, D. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18, 39–50.
Frazier, T., Ratliff, K., Gruber, C., Zhang, Y., Law, P., & Constantino, J. (2014). Confirmatory factor analytic structure and measurement invariance of quantitative autistic traits measured by the Social Responsiveness Scale-2. Autism, 18 (1), 31–44.
Fusco, S., Michael, K., & Michael, M. (2010, June 7–9). Using a social informatics framework to study the effects of location-based social networking on relationships between people: A review of literature. International Conference on Mobile Business (pp. 151–171). Greece: IEEE.
Haferkamp, N., Eimler, S., Papadakis, A., & Kruck, J. (2012). Men are from Mars, women are from Venus? Examining gender differences in self-presentation on social networking sites. Cyberpsychology, Behavior, and Social Networking, 15 (2), 91–98.
Hair, J., Black, W., Babin, B., & Anderson, R. (2006). Multivariate data analysis (6th ed.). Upper Saddle River, New Jersey: Prentice Hall International.
Hollenbaugh, E., & Ferris, A. (2014). Facebook self-disclosure: Examining the role of traits, social cohesion, and motives. Computers in Human Behavior, 29 (3), 604–609.
Hollenbaugh, E., & Ferris, A. (2015). Predictors of honesty, intent, and valence of Facebook self- disclosure. Computers in Human Behavior, 50, 456–464.
Hu, L., & Bentler, P. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6 (1), 1–55. Huang, C., & Park, D. (2013). Cultural influences on Facebook
photo-graphs. International Journal of Psychology, 48 (3), 334–343. Hum, N., Chamberlin, P., Hambright, B., Portwood, A., & Schat, A.
(2011). A picture is worth a thousand words: A content analysis of Facebook profile photographs. Computers in Human Behavior, 27 (5), 1828–1833.
Hunt, D., Lin, C., & Atkin, D. (2014). Photo-messaging: Adopter attri-butes, technology factors and use motives. Computers in Human Behavior, 40, 171–179.
Ji, Y., Wang, G., & Zhu, Z. (2014). Online social networking behaviors among Chinese younger and older adolescent: The influences of age, gender, personality, and attachment styles. Computers in Human Behavior, 41, 393–402.
Ji, Y. G., Hwangbo, H., Yi, J. S., Rau, P. L. P., Fang, X., & Ling, C. (2010). The influence of cultural differences on the use of social network services and the formation of social capital. International Journal of Human–Computer Interaction, 26 (11–12), 1100–1121.
Kahneman, D. (2011). Thinking, fast and slow. New York, NY: Farrar, Straus, and Giroux.
Kisekka, V., Bagchi-Sen, S., & Rao, R. (2013). Extent of private informa-tion disclosure on online social networks: An explorainforma-tion of Facebook mobile phone users. Computers in Human Behavior, 29 (6), 2722– 2729.
Kline, R. (2005). Principles and practice of structural equation modeling (2nd ed.). New York, NY: Guilford.
Knibbs, K. (2014, May 28). 1.8 billion images are uploaded every day. The Daily Dot. Retrieved fromhttp://www.dailydot.com/technology/ mary-meeker-photo-report
Krasnova, H., Spiekermann, S., Koroleva, K., & Hildebrand, T. (2010). Online social networks: Why we disclose. Journal of Information Technology, 25 (2), 109–125.
Krasnova, H., Veltri, N., & Günther, O. (2012). Self-disclosure and privacy calculus on social networking sites: The role of culture (Intercultural dynamics of privacy calculus). Business & Information Systems Engineering, 4 (3), 127–135.
Lewis, J., & Weigert, A. (1985). Trust as a social reality. Social Forces, 63 (4), 967–985.
Limayem, M., Hirt, S. G., & Cheung, C. (2007). How habit limits the predictive power of intention: The case of information systems con-tinuance. MIS Quarterly, 31 (4), 705–737.
Litt, E., & Hargittai, E. (2014). Smile, snap, and share? A nuanced approach to privacy and online photo-sharing. Poetics, 42, 1–21.
Marshall, P. (2010). The promotion and presentation of the self: Celebrity as marker of presentational media. Celebrity Studies, 1 (1), 35–48.
Mayer, R., Davis, J., & Schoorman, F. (1995). An integrative model of organization trust. Academy of Management Review, 20 (3), 709–734. McLain, D., & Hackman, K. (1999). Trust, risk, and decision-making in organizational change. Public Administration Quarterly, 23 (2), 152– 176.
Mendelson, A. L., & Papacharissi, Z. (2011). Look at us: Collective narcissism in college student Facebook photo galleries. In Z. Papacharissi (Ed.), The networked self: Identity, community and cul-ture on social network sites (pp. 251–273). New York, NY: Routledge. Mesch, G. (2012). Is online trust and trust in social institutions asso-ciated with online disclosure of identifiable information online? Computers in Human Behavior, 28 (4), 1471–1477.
Ng, M. (2014). Consumer motivations to disclose information and par-ticipate in commercial activities on Facebook. Journal of Global Scholars of Marketing Science, 24 (4), 365–383.
Norberg, P. A., & Dholakia, R. (2004). Customization, information provision and choice: What are we willing to give up for personal service? Telematics and Informatics, 21 (2), 143–155.
O’Neil, D. (2001). Analysis of internet users’ level of online privacy concerns. Social Science Computer Review, 19 (1), 17−31.
Olivero, N., & Lunt, P. (2004). Privacy versus willingness to disclose in e-commerce exchanges: The effect of risk awareness on the relative role of trust and control. Journal of Economic Psychology, 25 (2), 243–262. Pew Research Center. (2015). The demographics of social media users.
Retrieved from http://www.pewinternet.org/2015/08/19/the-demo graphics-of-social-media-users
Qin, L., Kim, Y., Hsu, J., & Tan, X. (2011). The effects of social influence on user acceptance of online social networks. International Journal of Human-Computer Interaction, 27 (9), 885–899.
Reno, R., Cialdini, R., & Kallgren, C. (1993). The transsituational influence of social norms. Journal of Personality and Social Psychology, 64 (1), 104–112. Rui, J., & Stefanone, M. (2013). Strategic image management online.
Information, Communication & Society, 16 (8), 1286–1305.
Schreiber, J., Stage, F., King, J., Nora, A., & Barlow, E. (2006). Reporting structural equation modeling and confirmatory factor analysis results: A review. Journal of Educational Research, 99 (6), 323–338.
Sheehan, K. (1999). An investigation of gender differences in on-line privacy concerns and resultant behaviors. Journal of Interactive Marketing, 13 (4), 24−38.
Siibak, A. (2009). Constructing the self through the photo selection-visual impression management on social networking websites. Cyberpsychology: Journal of Psychosocial Research on Cyberspace, 3 (1), 1–9.
Smith, J., & Louis, W. (2008). Do as we say and as we do: The interplay of descriptive and injunctive group norms in the attitude–behavior relationship. British Journal of Social Psychology, 47 (4), 647–666. Smock, A. D., Ellison, N. B., Lampe, C., & Wohn, D. Y. (2011). Facebook
as a toolkit: A uses and gratification approach to unbundling feature use. Computers in Human Behavior, 27, 2322–2329.
Statistics Netherlands. (2013). Seven in ten internet users active on social media. Retrieved August 23, 2016, from https://www.cbs.nl/en-gb/ news/2013/40/seven-in-ten-internet-users-active-on-social-media
Statistics Netherlands. (2014). Statistical yearbook of the Netherlands 2014. The Hague/Heerleen, The Netherlands: Statistics Netherlands. Steenkamp, J., & Baumgartner, H. (1998). Assessing measurement
invar-iance in cross-national consumer research. Journal of Consumer Research, 25 (1), 78–107.
Strack, F., & Deutsch, R. (2004). Reflective and impulsive determinants of social behavior. Personality and Social Psychology Review, 8 (3), 220–247. Strack, F., & Werth, L. (2006). Reflective and impulsive determinants of consumer behavior. Journal of Consumer Psychology, 16 (3), 205–216. Taraszow, T., Aristodemou, E., Shitta, G., Laouris, Y., & Arsoy, A. (2010). Disclosure of personal and contact information by young people in social networking sites: An analysis using Facebook profiles as an example. International Journal of Media and Cultural Politics, 6 (1), 81–101.
Thelwall, M. (2008). Social networks, gender, and friending: An analysis of MySpace member profiles. Journal of the American Society for Information Science and Technology, 59 (8), 1321–1330.
Tifferet, S., & Vilnai-Yavetz, I. (2014). Gender differences in Facebook self-presentation: An international randomized study. Computers in Human Behavior, 35, 388–399.
Van Gool, E., Van Ouytsel, J., Ponnet, K., & Walrave, M. (2015). To share or not to share? Adolescents’ self-disclosure about peer relation-ships on Facebook: An application of the Prototype Willingness Model. Computers in Human Behavior, 44, 230–239.
Van House, N. (2007). Flickr and public image-sharing: Distant closeness and photo exhibition. Proceedings of CHI‘07 (Extended Abstracts on Human Factors in Computing Systems) (pp. 2717–2727), San Jose, CA.
Van House, N. A., Davis, M., Ames, M., Finn, M., & Viswanathan, V. (2005). The uses of personal networked digital imaging: An empirical study of cameraphone photos and sharing. Proceedings of CHI ’05 (Extended Abstracts on Human Factors in Computing Systems) (pp. 1853–1856), Portland, OR.
Vandenberg, R., & Lance, C. (2000). A review and synthesis of the measurement invariance literature: Suggestions, practices, and recom-mendations for organizational research. Organizational Research Methods, 3 (1), 4–70.
Varnali, K., & Toker, A. (2015). Self-disclosure on social networking sites. Social Behavior and Personality: An International Journal, 43 (1), 1–13.
Verplanken, B., & Aarts, H. (1999). Habit, attitude, and planned behaviour: Is habit an empty construct or an interesting case of goal-directed automa-ticity? European Review of Social Psychology, 10 (1), 101–134.
Verplanken, B. & Orbell, S. (2003). Reflections on past behavior: A self-report index of habit strength. Journal of Applied Social Psychology, 33 (6), 1313–1330.
Walther, J. B., Slovacek, C. L., & Tidwell, L. C. (2001). Is a picture worth a thousand words? Photographic images in long-term and short-term computer-mediated communication. Communication Research, 28, 105–134.
Wang, Y., Norcie, G., Komanduri, S., Acquisti, A., Leon, P. G., & Cranor, L. F. (2011). I regretted the minute I pressed share: A qualitative study of regrets on Facebook. Proceedings of the 7th Symposium on Usable Privacy and Security (pp. 1–14), Pittsburgh, PA. doi:10.1145/2078827.2078841
Wheaton, B., Muthen, B., Alwin, D. F., & Summers, G. F. (1977). Assessing reliability and stability in panel models. Sociological Methodology, 8, 84–136.
White, K., Smith, J., Terry, D., Greenslade, J., & McKimmie, B. (2009). Social influence in the theory of planned behaviour: The role of descriptive, injunctive, and in-group norms. British Journal of Social Psychology, 48 (1), 135–158.
Youn, S., & Hall, K. (2008). Gender and online privacy among teens: Risk perception, privacy concerns, and protection behaviors. Cyberpsychology & Behavior, 11 (6), 763–765.
Zhao, L., Lu, Y., & Gupta, S. (2012). Disclosure intention of location-related information in location-based social network services. International Journal of Electronic Commerce, 16 (4), 53–90.
About the Authors
Ardion D. Beldadis an assistant professor at the University of Twente, the Netherlands. His primary research interests include organizational trust creation and maintenance, determinants and consequences of online trust, information privacy and confidentiality in online and offline context, crisis communication, and socially responsible behaviors. Sabrina M. Hegneris a professor at the University of Applied Science in Bielefeld, Germany. She holds a PhD from the Universioty of Bremen in Germany. Her primary research interests include the creation of brand relationships, crisis communication, online and offline word-of-mouth communication, and socially responsible behaviors.
Appendix
Items used to measure the different constructs Intention to continue sharing photos (INT)
I am intending to continue posting photos of myself on Facebook. My intention to continue posting photos of me engaged in various activities (or present in different events) on Facebook is high.
I am very inclined to continue posting photos of me with my friends. I am very inclined to continue posting photos of me with my family. I am very sure I will continue posting photos of my visit to a fancy or interesting location.
I am highly inclined to continue posting photos of my trips or holidays. Entertainment-related benefits (ENT)
Posting photos is a way to pass the time. Posting photos provides distraction. Posting photos is a way to reduce boredom. Self-presentation-related benefits (SPR)
Posting my photos enable me to present myself to my social networks in a more favorable way.
I believe sharing my photos is a way to create a good impression about me in the eyes of my Facebook contacts.
Communication-related benefits (COM)
Posting my photos on Facebook is a way to inform my network members of my on-going activities.
With my photos on Facebook, my network members will know where I have been.
With my photos on Facebook, my network members will have an impression of my life.
Competence-based trust in Facebook (TCO)
Facebook uses appropriate technologies to protect Facebook users’ per-sonal information including photos.
Facebook possesses the expertise to ensure that Facebook users’ personal information including photos is safe from unauthorized access. Character-based trust in Facebook (TCH)
Facebook will not abuse photos that I post on the platform.
Facebook will not use my photos for other purposes without my consent. Facebook will not sell my photos to third parties.
General trust in Facebook contacts (TFC)
Members of my Facebook network can be trusted not to use my photos for unknown purposes.
Members of my Facebook network can be trusted not to abuse my photos.
I trust members of my Facebook network have the intention to respect my information privacy.
Injunctive Norm (INJ)
People who are influential to me think that I should post photos on Facebook.
People who are important to me think that I should post photos on Facebook.
My friends think I should post photos on Facebook. Descriptive Norm (DES)
Most people I am regularly in contact with post photos on Facebook. Most people in my Facebook network post photos on Facebook. Many people I know post photos on Facebook.
Habit (HAB)
Posting my photos on Facebook has become automatic to me. I consider it natural to post my photos on Facebook. Posting my photos on Facebook is something I do frequently.
