Why Students Share Misinformation on Social Media: Motivation, Gender, and Study-level Differences

ACALIB-01639; No. of pages: 10; 4C: The Journal of Academic Librarianship xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

The Journal of Academic Librarianship

Why Students Share Misinformation on Social Media: Motivation, Gender, and Study-level Differences Xinran Chen ⁎, Sei-Ching Joanna Sin, Yin-Leng Theng, Chei Sian Lee Wee Kim Wee School of Communication and Information, Nanyang Technological University, 31 Nanyang Link, 637718, Singapore

a r t i c l e

i n f o

Article history: Received 4 May 2015 Accepted 6 July 2015 Available online xxxx Keywords: Misinformation sharing Social media Motivation Gender differences Study-level differences Characteristics of information

a b s t r a c t The increasing use of social media for information sharing has elevated the need for information literacy (IL) education to prepare students to be effective information creators and communicators. One concern is that students sometimes indiscriminately forward misinformation. Understanding the reasons behind misinformation sharing would help the development of IL intervention strategies. Guided by the Uses and Gratifications approach and rumor research, undergraduate and graduate students in Singapore were surveyed on why they share misinformation on social media. Gender and study-level differences were investigated. Over 60% of respondents had shared misinformation. The top reasons were related to the information's perceived characteristics, as well as self-expression and socializing. Accuracy and authoritativeness did not rank highly. Women had a higher prevalence of sharing and intention to share misinformation. Undergraduate and graduate students differed in their reasons for sharing misinformation. The former share (and intend to share) more misinformation than the latter, but the difference was not statistically significant. Because many of the reasons cited were social in nature, IL training should address the social motivations propelling such behavior. Social media systems may also develop features that encourage users to flag debunked postings and allow a correction to be displayed alongside the misinformation. © 2015 Elsevier Inc. All rights reserved.

INTRODUCTION The rise of social media has not only changed how people stay connected, but also brought about considerable opportunities and challenges in students' information behavior. The changing information horizon and shifting information behavior patterns have implications for information literacy (IL) education. College students are particularly active users of various social media platforms (Duggan, Ellison, Lampe, Lenhart, & Madden, 2015). They use social media for both academic and everyday life information seeking (Head & Eisenberg, 2011; Kim, Sin, & Yoo-Lee, 2014; Shao, 2009). While library and information science (LIS) professionals recognize the collaborative information seeking potential of social media, they are also cognizant of the varying quality of social media information. Given the ease with which information is posted and shared, misinformation—defined as information that has been shown to be inaccurate (Karlova & Fisher, 2013)—can circulate on social media quickly and widely (Mintz, 2012b). Misinformation can cause suspicion and fear among the public. It can also have harmful effects on individuals' well-being (Ferrara, 2015). There is, therefore, a pressing need to prepare students to be proficient social media users

⁎ Corresponding author. E-mail addresses: [email protected] (X. Chen), [email protected] (S.-C.J. Sin), [email protected] (Y.-L. Theng), [email protected] (C.S. Lee).

who are careful and responsible when sharing information on social media. The efforts to develop an IL program suitable for the new information environment are multi-pronged. These include reexamining the scope and focus of IL (e.g., critical IL, IL 2.0, and meta-literacy), developing standards and best practices, and conducting empirical investigations on students' social media information behavior. In terms of the last category, most studies have focused on perception and use of social media (e.g., Kim, Sin, & Tsai, 2014; Lim, 2009; Morris, Teevan, & Panovich, 2010; Zhang, 2012), as well as on the criteria and strategies used in evaluating the credibility of social media information (e.g., Kim & Oh, 2009; Kim & Sin, 2014a; Lim, 2013; Walsh, 2010). Extant studies are invaluable in shedding light on students as consumers of social media information. There is, however, a dearth of research on students as information sharers. While there are malevolent misinformation-spreaders on social media, misinformation would not have gone so viral without the participation of regular social media users (i.e., those who do not have malicious intent). Many regular users unwittingly propel the spread of misinformation when they undiscerningly forward misinformation to their own social networks (Ratkiewicz et al., 2010). Some of this misinformation sharing could be prevented. Different from rumor, which is defined as information that is unverifiable at the moment (DiFonzo & Bordia, 2007), misinformation is inaccurate information that has already been refuted. Thus, users could conceivably take steps to discover the information to be inaccurate. Currently, the extent to which students

http://dx.doi.org/10.1016/j.acalib.2015.07.003 0099-1333/© 2015 Elsevier Inc. All rights reserved.

Please cite this article as: Chen, X., et al., Why Students Share Misinformation on Social Media: Motivation, Gender, and Study-level Differences, The Journal of Academic Librarianship (2015), http://dx.doi.org/10.1016/j.acalib.2015.07.003

2

X. Chen et al. / The Journal of Academic Librarianship xxx (2015) xxx–xxx

share misinformation with their online friends is unclear; if they do share misinformation, what motivates them to do so is also unclear. RESEARCH QUESTIONS AND IMPLICATIONS In light of the research gap, the main objectives of this exploratory study are to understand the characteristics of students who share misinformation on social media and their reasons for doing so. The goals are to provide insights for the development of IL intervention strategies on reducing misinformation sharing among students, as well as to provide a basis for the development of further large-scale research on this topic. The study explores two main research questions (RQ). • RQ1a: What are students' perceptions of, and experiences with, misinformation sharing on social media? • RQ1b: Are there differences in their perceptions and experiences by (i) gender and (ii) different levels of higher education (hereafter, study-level differences)? • RQ2a: What are the reasons behind their misinformation sharing on social media? • RQ2b: Are there differences in their reasons by (i) gender and (ii) study-level?

By examining the reasons behind students' social media misinformation sharing, a hitherto unexplored area, this study's findings will contribute to IL education, and to IL and information behavior literature, in three ways. First, the study goes beyond analyzing students' use of traditional scholarly and web resources to examine their use of social media—an increasingly popular source of information. Second, the study investigates students as information sharers rather than as mere information consumers. Lastly, the study moves beyond cognitive factors to include affective reasons for students' information behavior. With greater understanding of the motivations behind students' misinformation sharing and the potential individual factors affecting such behavior, IL educators can develop pertinent strategies to prepare students to be responsible users of social media. LITERATURE REVIEW SOCIAL MEDIA AND INFORMATION LITERACY EDUCATION Social media is “a group of Internet-based applications that build on the ideological and technological foundations of Web 2.0, and that allow the creation and exchange of User Generated Content” (Kaplan & Haenlein, 2010, p. 60). These social media applications include collaborative projects, blogs, content communities, social networking sites, virtual game worlds, and virtual social worlds (Kaplan & Haenlein, 2010). A key aspect of social media is that it lets users communicate, share information, and collaborate (Anttiroiko & Savolainen, 2011). The potential for social media to play a role in collaborative information seeking and social change is notable (Bertot, Jaeger, & Grimes, 2010; Shah, 2012). At the same time, inaccurate information is rampant on social media (Mintz, 2012b). Given that students are active social media users (Duggan et al., 2015), how they use this medium (where information is of varying quality) is an area of concern. Research often shows that people in general, and students in particular, are not particularly diligent in their information seeking (Connaway, Dickey, & Radford, 2011; Kim & Sin, 2011). To illustrate, in a survey on students' evaluation of social media information, almost 60% of respondents reported that they never verified if sources were cited properly in the social networking site (SNS) messages they received (Kim et al., 2014b). Even in situations where students do conduct a credibility assessment, their evaluative strategies are faulty at times. For example, students sometimes use ineffective peripheral cues (e.g., a site's design or the nicknames of a Wikipedia article's editors) to evaluate the credibility

of related social media information (Kim & Sin, 2014b; Lim & Simon, 2011; Yaari, Baruchson-Arbib, & Bar-Ilan, 2011). Cognizant of the changing information landscape, there has been some discussion of broadening the scope of IL to cover non-academic information seeking and the use of social media (Farkas, 2011; Spiranec & Zorica, 2010). For instance, IL educators are preparing students to engage effectively with information in a collaborative environment. What is more, some libraries have started to develop guidelines and resources for the evaluation and use of social media information (Bridges, 2012; Mitrano, 2011; Witek & Grettano, 2012). Problematic actions such as students' sharing of misinformation on social media are an area that IL training has the potential to help tackle. REASONS BEHIND MISINFORMATION SHARING ON SOCIAL MEDIA MISINFORMATION SHARING ON SOCIAL MEDIA There are notable incidents of crisis-related misinformation sharing on social media. Examples include misinformation surrounding Ebola (Anagnostopoulos et al., 2014; Oyeyemi, Gabarron, & Wynn, 2014) and the 2011 riots in the United Kingdom (Guardian Interactive team, Procter, Vis, & Voss, 2011). Other misinformation may take the form of daily life advice that appears on social media sites repeatedly over a long period of time (Frost, 2002). Notably, inaccurate messages often continue to go viral even after being debunked, whereas the correct information does not receive as much attention (Friggeri, Adamic, Eckles, & Cheng, 2014; Oyeyemi et al., 2014). Indeed, catchiness—rather than truthfulness—often drives information (and misinformation) diffusion on social media (Ratkiewicz et al., 2010). Several characteristics of social media also may exacerbate the spread of misinformation. First, unlike traditional media, social media lacks rigorous quality control mechanisms. Furthermore, social media applications make it easy to disseminate information, including misinformation. One can forward messages to many receivers quite effortlessly; it is often as simple as a mouse click. Misinformation on social media can thus quickly reach many individuals, which can cause confusion and unnecessary anxiety among the public (Budak, Agrawal, & Abbadi, 2011). To help curb the spread of misinformation, it is important to understand the motivations driving the sharing of misinformation on social media. MOTIVATION AND SOCIAL MEDIA USE: THE USES AND GRATIFICATIONS APPROACH It remains unclear what motivates regular users to share misinformation on social media. However, factors that motivate social media usage in general have been investigated in a number of studies. While these usage studies have a different focus than the current research, they can provide insights on potential motivational factors for misinformation sharing on social media. Previous studies have investigated the motivations for using SNS and photo sharing sites (Dunne, Lawlor, & Rowley, 2010; Kim, Kim, & Nam, 2010; Nov, Naaman, & Ye, 2009). The reasons behind news sharing on social media (Lee & Ma, 2012) and information sharing on mobile gaming sites (Lee, Goh, Chua, & Ang, 2010) were also studied. Oh and Syn (2015) investigated and compared the factors motivating users to share information on different social media platforms (Facebook, Twitter, Delicious, YouTube, and Flickr). Many of these studies on individual motivations are based on the wellrecognized Uses and Gratifications (U&G) approach (Lee & Ma, 2012, p. 332; Park, Kee, & Valenzuela, 2009; Ruggiero, 2000; Shao, 2009), and individual motivations are often found to be salient. For example, a study about Facebook use found that an individual's motivation is significantly related to the amount of time spent on Facebook, while personality is not shown to have as strong an impact (Ross et al., 2009). Four main motivation categories are identified in social media literature informed by the U&G approach (Lee & Ma, 2012; Park et al., 2009). They are: (1) entertainment, which is about using social media for personal enjoyment; (2) socializing, which refers to relationship

Please cite this article as: Chen, X., et al., Why Students Share Misinformation on Social Media: Motivation, Gender, and Study-level Differences, The Journal of Academic Librarianship (2015), http://dx.doi.org/10.1016/j.acalib.2015.07.003

X. Chen et al. / The Journal of Academic Librarianship xxx (2015) xxx–xxx

development and maintenance with one's network on social media; (3) information seeking, which focuses on meeting informational needs with social media applications; and (4) self-expression and status seeking, which refers to using social media to express oneself and gain reputation. Because these four motivations have been found significant in general social media usage, this study seeks to investigate their pertinence in the context of students' misinformation sharing on social media. PERCEIVED CHARACTERISTICS OF INFORMATION Beyond investigating individual motivations as guided by the U&G approach, this study also investigates another factor, the characteristics of information, which may also influence students' decisions to forward postings on social media. For example, some individuals might tend to forward postings that they find entertaining, while others might prefer to forward breaking news. IL training often highlights assessment criteria such as accuracy and the authority of the information source. It is thus of interest to investigate whether these characteristics also factor into people's social media information sharing decisions. It has been suggested that when it comes to information sharing on social media, users do not always feel that the authority of the information is a key concern. People tend to consider information that “a friend told them” to be reliable enough to share on social media (Mintz, 2012a). Studies on rumor (a phenomenon related to misinformation sharing) also suggest that information characteristics are worth examining. These studies indicate that people are more likely to spread rumors when the rumors are consistent with their beliefs or are threatening (Allport & Postman, 1947; DiFonzo & Bordia, 2007; Rosnow, 1991). On the other hand, some individuals consider rumor to be a form of information that merely serves as a conversational topic in daily life (Guerin & Miyazaki, 2006). Drawing from this body of literature, the current study examines the extent to which certain information characteristics contribute to misinformation sharing on social media. INDIVIDUAL DIFFERENCES GENDER DIFFERENCES Findings on gender differences in the broader information seeking and use literature (e.g., studies on library anxiety and gender) have been inconclusive (Onwuegbuzie, Jiao, & Bostick, 2004). Urquhart and Yeoman's (2010) meta-synthesis showed the varying findings on gender differences produced by different studies. While gender differences concerning access to computer and the Internet have become less significant over the years (Broos, 2005; Weiser, 2000), gender differences may still exist in various online activities (Lim & Kwon, 2010). For example, there were mixed results on the influences of gender on individuals' perceptions toward information seeking on the web (Zhou, 2014). Interestingly, gender differences seem to be more notable in social media usage, especially in SNS use. Women are more likely to use SNS in general, and particularly in terms of information seeking (Duggan et al., 2015; Kim & Sin, 2014b; Sin & Kim, 2013). They are more likely to frequently update their status as well as comment on others' postings (Hampton, Goulet, Rainie, & Purcell, 2011). Despite being more active users of SNS, female students were found to be less active on wikis, blogs, and Internet forums compared to their male counterparts (Kim & Sin, 2014b). Women also exercise more caution concerning the quality of information on online platforms such as Wikipedia (Lim & Kwon, 2010). Thus, it is interesting to investigate the nature and extent of gender differences in social media misinformation sharing. STUDY-LEVEL DIFFERENCES When compared to studies on gender differences, there are fewer studies investigating study-level differences (undergraduate vs. graduate students) when it comes to social media usage. Nevertheless, the research that has investigated study-level differences suggests that

3

graduate students are regular social media users, although they are not as active as undergraduates in terms of social media use (Lampe, Ellison, & Steinfield, 2006; Park, 2010; Sin & Kim, 2013). It is possible that the graduate students' additional years of education and experience with information evaluation and use may help them reduce problematic online behaviors such as misinformation sharing on social media, when compared to undergraduates. Nevertheless, because of the dearth of empirical research on this topic, it is unclear whether such a proposition is accurate. More empirical studies are needed to investigate potential study-level differences. METHOD RESEARCH METHOD AND INSTRUMENT The study used data collected through a survey questionnaire, which is a frequently used method suitable for collecting self-reported data about attitudes and behaviors (Neuman, 2011). The questionnaire comprised mainly closed-ended questions that were answered via a Likert-type scale. A few open-ended questions were also included so that respondents could express their opinions freely. In order to give respondents a clear understanding of misinformation (i.e., information that is commonly known to be false or is scientifically proven to be incorrect), concrete examples of misinformation were provided in the questionnaire. These examples showed realistic scenarios of misinformation that were currently being communicated on social media. There were six misinformation examples, all of which were selected from the Internet and have been scientifically proven to be untrue (Agranoff, Davis, & Brink, 1966; Batellier, Couty, Picard, & Brillard, 2008; Emery, 2012; Goldenberg, 2006; Saner, 2011; Saquete, 2010; The University of New Hampshire, 2000). To represent the various forms of misinformation typically seen on social media, the six examples represented a combination of news, factoids, photos, and daily life advice. To answer RQ1, the questionnaire asked respondents about their perceptions of, and behavior toward, misinformation sharing on social media. On a 7-point scale, respondents rated the frequency of misinformation sharing on social media by themselves and by others. For RQ2, respondents answered questions about their reasons for sharing misinformation on social media. Twenty-nine items were used in total. Individual motivations were measured with a list of 16 motivation items (see Table 1), representing the four main motivational categories of the U&G approach (i.e., entertainment, socializing, information seeking, and self-expression and status seeking) (Kim et al., 2010; Lee & Ma, 2012; Park et al., 2009). For information characteristics, a list of 13 characteristics was developed, drawing from the literature on information quality and rumor research, discussed above (DiFonzo & Bordia, 2007; Mintz, 2012a; Rieh, 2002; Rosnow, 1991). Demographic information, including gender and study-level data, was also collected to address RQ1b and RQ2b. The questionnaire was pilot-tested and revised before distribution. It was available in both print and online versions, and it took about 10 min to complete. SAMPLING METHOD The study population was comprised of students from two public universities in Singapore who were social media users aged 18 to 29 years old. This age range was selected as it is comparable to other social media studies that use college student samples (Correa, Hinsley, & de Zúñiga, 2010; Wyllie, Zhang, & Casswell, 1998). Similar to many developed countries, social media usage is high in Singapore. When compared to 14 other nations including the United States and the United Kingdom, Singapore respondents reported spending the most time on social media (Firefly Millward Brown, 2010). The social media usage behavior of Singapore college students was found to be very similar to that of U.S. students in a comparative study (Kim & Sin, 2014a).

Please cite this article as: Chen, X., et al., Why Students Share Misinformation on Social Media: Motivation, Gender, and Study-level Differences, The Journal of Academic Librarianship (2015), http://dx.doi.org/10.1016/j.acalib.2015.07.003

4

X. Chen et al. / The Journal of Academic Librarianship xxx (2015) xxx–xxx

Table 1 Students' perceptions and behaviors related to misinformation sharing Mean

SD

Categories

Frequency

Percentage

Extent to which the friends of respondents share misinformation on social media

3.97

1.38

Frequency in which respondents themselves share misinformation on social media

2.70

1.63

Extent to which respondents intend to share misinformation on social media in the future

3.26

1.62

1: No one 2: Very few people 3: Some people 4: Neutral 5: Many people 6: Nearly everyone 7: Everyone 1: Never shared 2: Hardly share 3: Sometimes share 4: Neutral 5: Regularly share 6: Frequently share 7: Always share 1: Surely will not share 2: Probably will not share 3: Maybe will not share 4: Neutral 5: Maybe will share 6: Probably will share 7: Surely will share

9 13 51 16 69 9 4 55 32 41 9 23 9 2 26 47 21 26 37 13 1

5.26% 7.60% 29.8% 9.36% 40.4% 5.26% 2.34% 32.2% 18.7% 24.0% 5.26% 13.45% 5.26% 1.17% 15.2% 27.5% 12.3% 15.2% 21.6% 7.60% 0.58%

As a beginning study on a seldom-investigated topic, this exploratory study aimed to propose and investigate possible reasons behind misinformation sharing and analyzed potential differences between groups among the sample. The study was not intended for population generalization. The questionnaires were disseminated to students around the campuses of the two universities using a convenience sampling approach, which is an effective approach when population generalization is not a focus (Neuman, 2011). Because data were collected through a non-probability sampling method, the sample cannot be considered a representation of all students. The findings should be interpreted with this caveat in mind.

undesirable aspects of their behavior are being observed, they might alter their natural behavior to look as good as possible to observers (Hertwig & Ortmann, 2008). As sharing inaccurate information might not be perceived as a socially desirable behavior, it is possible that some respondents underreported the degree to which they shared misinformation. We thus posit that the findings are a conservative estimation of the prevalence of misinformation sharing. A non-reactive method (e.g., content analysis) could be employed in future studies to facilitate method triangulation with survey findings.

FINDINGS DATA ANALYSIS METHOD RESPONDENT CHARACTERISTICS Descriptive statistics and inferential testing were conducted with SPSS. Descriptive statistics were used for all research questions. For RQ1b and RQ2b, which involve inferential testing of gender and study-level differences, independent sample t-tests were used (Field, 2009). An independent sample t-test is a commonly used statistical technique for comparisons between two unrelated groups (Field, 2009), which fit the characteristics of the study (i.e., women vs. men; undergraduate vs. graduate student). Assumptions diagnostics such as Levene's tests for equality of variances were conducted to confirm that the data were suitable for t-test analyses.

Two hundred questionnaires were distributed, from which 171 complete responses were received and analyzed. This resulted in a response rate of 85.5%. Of all respondents, 57.3% were women (n = 98), more than men (n = 73, 42.7%). A majority of the respondents were between 21 and 25 years of age (71.3%); the average age was 24 (SD = 2.1). There was an almost equal share of undergraduates (n = 85, 49.7%) and graduate students (n = 86, 49.7%). All respondents (n = 171) were social media users, and 81.3% of the respondents used social media every day.

LIMITATIONS As this exploratory study used a non-probability sampling technique, the resultant sample is not representative of all university students. Future studies could test these relationships again with larger sample sizes using probability samples, which could buttress the external validity of the findings. Multiple testing of the same population using the same instrument would help verify the stability reliability of the instrument, as is the case with testing different populations to gauge the instrument's representative reliability. In terms of internal validity, the survey instrument was developed based on previous studies and was pilot-tested. This provided support for face validity. Nevertheless, research on the current topic is at its nascent, formative stage. There is room to further improve the validity and reliability of the instrument. Another limitation is that the study used self-reported data. Similar to all studies that involve human participants, respondents' subjectivity and reactivity are potential methodological limitations (Neuman, 2011). When participants are aware that some socially

RQ1A: PERCEPTIONS AND BEHAVIORS RELATED TO MISINFORMATION SHARING Nearly all respondents (n = 162, 94.7%) reported having seen their friends share misinformation on social media (Table 1); on a 7-point scale (with 1 indicating “no one” and 7 indicating “everyone”), the mean was 3.97 (SD = 1.38). About two-thirds of the respondents (n = 116, 67.8%) reported that they themselves had shared misinformation on social media. The frequency with which they shared such misinformation was 2.7 on a 7-point scale (SD = 1.63), with 1 indicating they never shared misinformation and 7 indicating they always share it. Respondents also indicated their intention to share misinformation in the future; the sample mean was 3.26 on a 7-point scale, with 1 indicating “will surely not share” and 7 indicating “will surely share” misinformation in the future (SD = 1.62). Notably, only 15.2% of respondents (n = 26) reported that they would surely not share misinformation in the future.

Please cite this article as: Chen, X., et al., Why Students Share Misinformation on Social Media: Motivation, Gender, and Study-level Differences, The Journal of Academic Librarianship (2015), http://dx.doi.org/10.1016/j.acalib.2015.07.003

X. Chen et al. / The Journal of Academic Librarianship xxx (2015) xxx–xxx

5

RQ1B: GENDER AND STUDY-LEVEL DIFFERENCES Female students on average scored higher on all three misinformation perception and behavior variables than their male counterparts (Fig. 1). The independent sample t-tests show that gender differences are statistically significant for two of the three variables: the frequency in which respondents themselves share misinformation on social media, t(169), p = .014 (MMen = 2.34, SD = 1.44; MWomen = 2.96, SD = 1.72) and the extent to which respondents intend to share misinformation on social media in the future, t(169), p = .029 (MMen = 2.95, SD = 1.58; MWomen = 3.49, SD = 1.62). In terms of study-level differences, undergraduates scored higher on all three variables than graduate students (Fig. 2). However, of the three variables, only one of them—the extent to which their friends share misinformation on social media—is shown to be statistically significant in the independent sample t-tests, t(169), p = .012 (MUndergrad = 4.24, SD = 1.32; MGrad = 3.71, SD = 1.40). RQ2A: REASONS BEHIND MISINFORMATION SHARING On a 7-point scale (1 as “completely disagree” to 7 as “completely agree”), respondents indicated to what extent they agreed that each of the 29 items matched their reasons for sharing misinformation on social media (Fig. 3). The mean scores of all items are presented in Table 2 in descending order. The top three reasons were: “The information can be a good topic of conversation,” “The information is interesting,” and “The information is new and eye-catching,” all of which were related to information characteristics. Some individual motivations also appeared as top reasons for sharing misinformation. For example, the ranked fourth was “Sharing helps me get other people's opinions regarding the information/event,” which was a motivation based on the U&G information seeking category; additionally, the reason ranked fifth was a motivation in the U&G self-expression category: “I can express my opinion by sharing that information.” On the other hand, accuracy ranked 24th out of the 29 reasons. While accuracy is an important factor when evaluating the credibility of information, and is often taught in IL training, it was not prominent in respondents' sharing behaviors. The authority of information sources, another important factor for assessing information credibility, ranked even lower at 26th out of 29 reasons. RQ2B: GENDER AND STUDY-LEVEL DIFFERENCES IN THE REASONS BEHIND MISINFORMATION SHARING GENDER DIFFERENCES The descriptive data suggest some gender differences in the reasons behind misinformation sharing. Table 3 shows the top reasons by gender. Both genders gave the reasons “The information can be a good

Fig. 1. Gender differences in perceptions and behaviors related to misinformation sharing. Note: answer scales of the questions are as follows: extent friends share misinformation (left image): 1 (no one) to 7 (everyone). Frequency respondents themselves share misinformation (middle image): 1 (never) to 7 (always). Extent respondents intend to share misinformation (right image): 1 (will surely not share) to 7 (will surely share).

Fig. 2. Study-level differences in perceptions and behaviors related to misinformation sharing. Note: answer scales of the questions are as follows: extent friends share misinformation (left image): 1 (no one) to 7 (everyone). Frequency respondents themselves share misinformation (middle image): 1 (never) to 7 (always). Extent respondents intend to share misinformation (right image): 1 (will surely not share) to 7 (will surely share).

topic for conversation” and “The information is interesting” as the first- and second-ranked reasons, respectively. The third-ranked reason differed by gender. For men, “Sharing helps me get other people's opinions regarding the information/event” was the third-ranked reason. For women, it was “The information is new and eye-catching.” Gender differences in the reasons for misinformation sharing on social media were investigated with independent sample t-tests (Table 4). Significant gender differences were found in 10 of the 29 items, in which women gave higher ratings to all 10 significant items in comparison to men. “Sharing helps me bookmark useful information” had the largest gender differences. This is followed by: “Sharing helps me enhance interpersonal relations”; “Sharing helps me keep updated on the latest happenings”; “Sharing helps me keep in touch with friends”; “Sharing helps me get other related information”; “Sharing is a good way to relax”; “The information is fun”; “Sharing is a culture and I share like others do”; “I feel enjoyment while sharing”; and “Sharing helps me interact with people.” Nine of the 10 significant reasons were related to motivations, and only one reason belonged to the information characteristics category.

STUDY-LEVEL DIFFERENCES For undergraduates, the top three reasons for misinformation sharing were a mix of motivational factors and information characteristics (Table 5). They were: (1) “The information can be a good topic for conversation”; (2) “Sharing helps me get other people's opinions regarding the information/event”; and (3) “I can express my opinion by sharing that information”. For graduate students, the top three reasons were all related to the characteristics of information. They were: (1) “The information is interesting”; (2) “The information is new and eye-catching”; and (3) “The information can be a good topic for conversation”. The descriptive differences between undergraduate and graduate students can be found in Table 6 and Fig. 3. The independent sample t-tests found eight statistically significant study-level differences (Table 6). Six of the significant differences were related to motivations; two were related to information characteristics. The significant reasons (in descending order of mean study-level difference) were: (1) “Sharing helps me bookmark useful information”; (2) “Sharing helps me get other people's opinions regarding the information/event”; (3) “I can express my opinion by sharing that information”; (4) “The information can be a good topic for conversation”; (5) “Sharing helps me keep updated on the latest happenings”; (6) “The information is consistent with my belief/assumption”; (7) “Sharing helps me get other related information”; and (8) “Sharing makes me feel influential”. For all eight reasons, undergraduates had a higher mean score than graduate students.

Please cite this article as: Chen, X., et al., Why Students Share Misinformation on Social Media: Motivation, Gender, and Study-level Differences, The Journal of Academic Librarianship (2015), http://dx.doi.org/10.1016/j.acalib.2015.07.003

6

X. Chen et al. / The Journal of Academic Librarianship xxx (2015) xxx–xxx

Fig. 3. Reasons behind misinformation sharing on social media, by gender and study-level.

DISCUSSION Sharing misinformation on social media was found to be relatively common among the sample; indeed, a large majority of respondents (94.7%) have seen their friends doing so. Because the sharing of Table 2 Reasons behind misinformation sharing on social media Rank Reasons

Mean SD

1 2 3 4

5.25 5.13 5.03 5.02

1.34 1.31 1.38 1.46

4.95 4.89 4.87 4.83 4.75 4.75 4.74

1.44 1.37 1.46 1.61 1.53 1.57 1.29

4.74 4.71 4.69 4.50 4.39 4.25 4.25 4.24 4.23 4.12 4.09 4.09 4.01 3.74 3.58 3.48 3.37 3.22

1.46 1.42 1.73 1.67 1.51 1.62 1.54 1.61 1.74 1.57 1.60 1.47 1.60 1.57 1.62 1.58 1.78 1.47

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

The information can be a good topic for conversation. The information is interesting. The information is new and eye-catching. Sharing helps me get other people's opinions regarding the information/event. I can express my opinion by sharing that information. Sharing helps me interact with people. The information is fun. Sharing helps me keep updated on the latest happenings. Sharing helps me keep in touch with friends. Sharing helps me get other related information. The information provides understanding of a particular event/situation. The information is current. The information seems useful. Sharing helps me bookmark useful information. Sharing is good for keeping boredom away. I feel enjoyment while sharing. Sharing is a good way to relax. The information seems important. Sharing is a culture and I share like others do. Sharing is a good way of killing time. Sharing helps me enhance interpersonal relations. The information is consistent with my belief/assumption. The information comes from my close friends/family. The information seems accurate. Sharing makes me feel influential. The information comes from authoritative sources. Sharing makes me look good to others. I want to be the first one among others to share. The information looks frightening.

misinformation is not a socially desirable action, and based on the research method literature, we posit that respondents might have underreported their own misinformation sharing. It is thus noteworthy that—potential social desirability issues notwithstanding—a majority of respondents (67.8%) still admitted to having shared misinformation themselves. An even larger proportion (84.8%) indicated that they might share misinformation in the future. This suggests that IL professionals need to proactively address students' tendency toward misinformation sharing. For example, IL educators may dedicate a segment of IL sessions to ask students to reflect on their motivations in sharing misinformation, and to encourage them to critically revaluate and discuss the negative consequence of misinformation sharing. The reasons given for misinformation sharing shed light on possible areas of intervention. Not many of the reasons ranked near the top were informational; among the top five reasons, only one—ranked fourth—was related to information seeking. The other reasons reflected respondents' tendency to share information that they perceived fun and interesting, and being motivated by reasons that were social in nature. On the other hand, factors that are often discussed in IL education as Table 3 Top five reasons of misinformation sharing on social media by gender Rank Men

Women

1

The information can be a good topic for conversation.(M = 5.18)

2

The information is interesting. (M = 5.04) Sharing helps me get other people's opinions regarding the information/event. (M = 4.95) The information is new and eye-catching. (M = 4.92)

The information can be a good topic for conversation. (M = 5.31) The information is interesting. (M = 5.19) The information is new and eye-catching. (M = 5.11) Sharing helps me interact with people. (M = 5.10) I can express my opinion by sharing that information. (M = 5.09)

3

4

5

The information is current. (M = 4.77)

Please cite this article as: Chen, X., et al., Why Students Share Misinformation on Social Media: Motivation, Gender, and Study-level Differences, The Journal of Academic Librarianship (2015), http://dx.doi.org/10.1016/j.acalib.2015.07.003

X. Chen et al. / The Journal of Academic Librarianship xxx (2015) xxx–xxx Table 4 Results of t-tests on gender differences in reasons behind misinformation sharing Reasons

1. The information can be a good topic for conversation. 2. The information is interesting. 3. The information is new and eye-catching. 4. Sharing helps me get other people's opinions regarding the information/event. 5. I can express my opinion by sharing that information. 6. Sharing helps me interact with people. 7. The information is fun. 8. Sharing helps me keep updated on the latest happenings. 9. Sharing helps me keep in touch with friends. 10. Sharing helps me get other related information. 11. The information provides understanding of a particular event/situation. 12. The information is current. 13. The information seems useful. 14. Sharing helps me bookmark useful information. 15. Sharing is good for keeping boredom away. 16. I feel enjoyment while sharing. 17. Sharing is a good way to relax. 18. The information seems important. 19. Sharing is a culture and I share like others do. 20. Sharing is a good way of killing time. 21. Sharing helps me enhance interpersonal relations. 22. The information is consistent with my belief/assumption. 23. The information comes from my close friends/family. 24. The information seems accurate. 25. Sharing makes me feel influential. 26. The information comes from authoritative sources. 27. Sharing makes me look good to others. 28. I want to be the first one among others to share. 29. The information looks frightening.

Mean

T-tests

Men Women Mean diff.

t

p

5.18

5.31

−0.13

0.62 .537

5.04 4.92

5.19 5.11

−0.15 −0.19

0.72 .471 0.87 .384

4.95

5.08

−0.14

0.58 .565

4.77

5.09

−0.32

1.41 .160

4.62

5.10

−0.49

2.18 .031⁎

4.56 4.49

5.09 5.08

−0.53 −0.59

2.27 .025⁎ 2.34 .021⁎

4.44

4.99

−0.55

2.26 .026⁎

4.44

4.99

−0.55

2.22 .028⁎

4.60

4.85

−0.24

1.18 .242

4.77 4.64 4.26

4.71 4.77 5.01

0.05 −0.23 .815 −0.12 0.55 .582 −0.75 2.71 .008⁎⁎

4.22

4.71

−0.50

4.11 3.95 4.38 3.95

4.60 4.48 4.14 4.46

−0.49 2.11 .037⁎ −0.53 2.16 .032⁎ 0.24 −0.98 .327 −0.51 2.09 .038⁎

3.93

4.45

−0.52

1.86 .064

3.78

4.38

−0.60

2.42 .017⁎

4.22

4.00

3.86

4.27

−0.40

4.14 3.60 3.68

3.91 3.85 3.51

0.23 −0.92 .357 −0.24 1.01 .316 0.17 −0.67 .501

3.21

3.68

−0.48

3.40

3.36

3.19

3.24

1.88 .062

0.22 −0.86 .389 1.74 .085

1.91 .058

0.04 −0.15 .885 −0.05

0.23 .817

well as in information retrieval literature ranked rather low: accuracy (24th); authority of sources (26th); perceived usefulness (13th), and importance of information (19th). All in all, these findings suggest a significant difference between the considerations behind social media misinformation sharing and those behind other information seeking and evaluation behaviors. In the broader information behavior literature, the importance of non-cognitive factors such as affect and motivation are increasingly being recognized (Nahl & Bilal, 2007). In regards to misinformation sharing, this study suggests that notable numbers of non-cognitive reasons (e.g., self-expression and socialization) are also at play. Consequently, IL education should not be limited to teaching the criteria and strategies of credibility assessment. Rather, non-informational motivations should also be addressed. What is more, the negative consequences of misinformation sharing should be included as topics in IL training. IL educators can tailor their messages based on students' reasons for misinformation sharing. For example, this study has found

7

Table 5 Top five reasons of misinformation sharing on social media by study-level Rank Undergraduates

Graduate students

1

The information is interesting. (M = 5.09) The information is new and eye-catching. (M = 5.00) The information can be a good topic for conversation. (M = 4.98) The information is fun. (M = 4.81)

2

3

4

5

The information can be a good topic for conversation.(M = 5.53) Sharing helps me get other people's opinions regarding the information/event. (M = 5.40) I can express my opinion by sharing that information. (M = 5.28) Sharing helps me bookmark useful information. (M = 5.20) The information is interesting. (M = 5.16)

Sharing helps me interact with people. (M = 4.78)

that respondents were motivated by reasons related to self-expression and social interaction. Thus, to deter misinformation sharing, IL educators could highlight the fact that misinformation sharing might hurt a student's reputation (e.g., friends may view the misinformation sharer as being undiscerning and untrustworthy). This study has found that more women than men share (and intend to share) misinformation. This is a surprising and interesting finding, as prior studies suggest that women are more cautious online and are more critical of the quality of online information (Lim & Kwon, 2010). Research on the selective model of information processing also suggests that women tend to use a more holistic approach, and will evaluate more cues, when processing information (Darley & Smith, 1995; Meyers-Levy & Maheswaran, 1991). Thus, one would not have anticipated that women would be found to share misinformation more frequently. Further research is needed to investigate whether the aforementioned gender difference is a recurring pattern, and also to identify the reasons behind the finding. We present a few possible explanations here for further examination. First, as noted earlier, women tend to post and share more messages on some platforms such as SNS (Hampton et al., 2011). It is therefore possible that increased misinformation sharing is in part due to their higher rate of social media sharing in general. This explanation is not entirely satisfactory, however, as it does not address the point concerning women being more cautious in terms of information evaluation. A second explanation may have stronger implications for IL education. It is plausible that, when compared to men, women might find the social aspects of social media sharing to be more salient than its informational aspects. Women were reported to have stronger social and communicative intentions than men in the area of Internet use (Rodgers & Harris, 2003). The current findings provide some support for this view. There were four social reasons in the list of 29 items (i.e., “Sharing helps me interact with people”; “Sharing helps me keep in touch with friends”; “Sharing is a culture and I share like others do”; and “Sharing helps me enhance interpersonal relations”). Gender differences were significant for all four social reasons, with women citing them more often than men. Weber, Blais, and Betz (2002) found that women are more risk-averse then men in various domains (finance, health/safety, recreational, and ethical), but not in the social domain. Along this line, we posited that women may not be fully perceiving their information/misinformation sharing as an informational activity, during which they are generally found to be more cautious (Lim & Kwon, 2010). They might instead be perceiving their social media information sharing as primarily a social activity, which is an area where they might not exercise the same level of caution. This tentative hypothesis requires more testing. In the meantime, IL education could highlight that social media users' everyday online social communication still has informational consequences (e.g., misinformation, if shared, might mislead or even harm their friends). Students should thus be encouraged to be vigilant

Please cite this article as: Chen, X., et al., Why Students Share Misinformation on Social Media: Motivation, Gender, and Study-level Differences, The Journal of Academic Librarianship (2015), http://dx.doi.org/10.1016/j.acalib.2015.07.003

8

X. Chen et al. / The Journal of Academic Librarianship xxx (2015) xxx–xxx

Table 6 Results of t-tests on study-level differences in reasons behind misinformation sharing Reasons

1. The information can be a good topic for conversation. 2. The information is interesting. 3. The information is new and eye-catching. 4. Sharing helps me get other people's opinions regarding the information/event. 5. I can express my opinion by sharing that information. 6. Sharing helps me interact with people. 7. The information is fun. 8. Sharing helps me keep updated on the latest happenings. 9. Sharing helps me keep in touch with friends. 10. Sharing helps me get other related information. 11. The information provides understanding of a particular event/situation. 12. The information is current. 13. The information seems useful. 14. Sharing helps me bookmark useful information. 15. Sharing is good for keeping boredom away. 16. I feel enjoyment while sharing. 17. Sharing is a good way to relax. 18. The information seems important. 19. Sharing is a culture and I share like others do. 20. Sharing is a good way of killing time. 21. Sharing helps me enhance interpersonal relations. 22. The information is consistent with my belief/assumption. 23. The information comes from my close friends/family. 24. The information seems accurate. 25. Sharing makes me feel influential. 26. The information comes from authoritative sources. 27. Sharing makes me look good to others. 28. I want to be the first one among others to share. 29. The information looks frightening.

Mean

T-tests

Undergrad.

Grad.

Mean diff.

t

p

5.53 5.16 5.06 5.40 5.28 5.01 4.92 5.08 4.74 5.00 4.87 4.84 4.89 5.20 4.73 4.35 4.08 4.34 4.29 4.14 4.00 4.34 4.22 4.14 3.99 3.74 3.54 3.51 3.27

4.98 5.09 5.00 4.65 4.63 4.78 4.81 4.58 4.77 4.51 4.62 4.64 4.53 4.19 4.28 4.43 4.42 4.15 4.19 4.31 4.24 3.85 3.97 3.87 3.50 3.43 3.42 3.24 3.17

0.55 0.07 0.06 0.75 0.65 0.23 0.10 0.50 −0.03 0.49 0.25 0.20 0.36 1.01 0.45 −0.08 −0.34 0.19 0.11 −0.17 −0.24 0.49 0.26 0.27 0.49 0.31 0.12 0.26 0.10

2.75 0.36 0.28 3.46 3.04 1.11 0.46 2.06 −0.11 2.05 1.29 0.88 1.66 3.99 1.77 −0.37 −1.36 0.81 0.44 −0.65 −1.02 2.04 1.15 1.10 2.05 1.24 0.50 0.96 0.42

.007⁎⁎ .722 .782 .001⁎⁎ .003⁎⁎ .269 .644 .041⁎ .911 .042⁎ .200 .382 .098 .000⁎⁎ .078 .715 .175 .421 .662 .519 .311 .043⁎ .251 .275 .042⁎ .218 .614 .339 .672

⁎ p b 0.05. ⁎⁎ p b 0.01.

and cautious in assessing information quality, both when they are conducting academic tasks as well as when they are connecting with friends on social media. Descriptive statistics show that undergraduate students share (and intend to share) misinformation more frequently than graduate students. This may be attributed in part to undergraduates' higher social media usage. It is also possible that graduate students are more circumspect in their social media behavior because of their additional years of education and experience. It is thus worth noting that, while the study found statistically significant study-level differences in the reasons behind misinformation sharing (RQ2b), the prevalence of misinformation sharing was statistically similar among undergraduates and graduate students (RQ1b). This indicates that IL professionals should also pay attention to graduate students' social media use. Table 3 shows that there were five reasons given by graduate students that show higher descriptive scores as compared to undergraduates. Three of these reasons relate to the U&G approach's entertainment category (i.e., “Sharing is a good way to relax”; “I feel enjoyment while sharing”; and “Sharing is a good way of killing time”). The other two reasons relate to the U&G's socializing category (i.e., “Sharing helps me keep in touch with friends”; and “Sharing helps me enhance interpersonal relations”). When developing IL training in social media for graduate students, then, the influence of these two U&G motivation categories may warrant more attention. The study also identifies some positive findings. Reasons related to the U&G approach's information seeking category, such as “Sharing helps me to obtain other people's opinions regarding the information/ event” and “Sharing helps me get other related information” ranked relatively high (4 and 10, respectively). This suggests that respondents were interested in gaining a better understanding of the information at hand. Fortunately, students can gain knowledge through further information gathering and discussion with other social media users. Indeed, if used properly, discussion on social media can serve as a

question-and-answer negotiation process; such interactions can help users to collectively make sense of happenings (Heverin & Zach, 2012). Further research could investigate questions such as the extent to which students engage in such critical discussions on social media, the ways in which they go about seeking information and clarification, and whether their actions indeed help to debunk inaccurate information. Therefore, IL training should encourage students to assess social media information critically and collaboratively, as well as to rebut false information on social media when they encounter it. Preferably, students should conduct such collaborative information assessment without forwarding the original misinformation to their networks. This is because such actions add to statistical counts, including “number of times shared”. Since undiscerning users sometimes do not read the comments related to a posting (rather, they inaccurately interpret a high number of shares as an indication of good information credibility), it is preferable not to further broadcast the original misinformation. In the same vein, the study's findings also have implications for the design of social media systems and interfaces. For example, a system with features that encourage individual users to flag officially debunked postings would be helpful, as would features that allow correction information to be displayed alongside the misinformation. Similarly, because some respondents share misinformation as a way to bookmark a posting (the reason ranked 14th), features that allow users to curate and store postings in a private collection could be adopted. CONCLUSION The study found that college students sometimes share misinformation on social media, often for non-informational reasons such as to share eye-catching messages or to interact with friends. The study also indicates significant gender and study-level differences in the reasons cited. The findings suggest that coordinated efforts are needed in IL education and in the design of social media applications to curb the

Please cite this article as: Chen, X., et al., Why Students Share Misinformation on Social Media: Motivation, Gender, and Study-level Differences, The Journal of Academic Librarianship (2015), http://dx.doi.org/10.1016/j.acalib.2015.07.003

X. Chen et al. / The Journal of Academic Librarianship xxx (2015) xxx–xxx

spread of misinformation. For the former, the goal should be to not only equip social media users with the ability to recognize misinformation, but also to cultivate an aspiration to stop the forwarding of such messages. Better yet, students should be encouraged to challenge misinformation when they encounter it. From a system design perspective, providing features that afford the flagging and rebuttal of misinformation would be helpful. Although it may not be possible to eradicate misinformation on social media completely, proactive IL intervention can help combat uncritical misinformation sharing. In light of the popularity and reach of social media, as well as its collaborative information seeking potential, it would be worthwhile to devote efforts to help students reduce their misinformation sharing.

References Agranoff, B.W., Davis, R.E., & Brink, J.J. (1966). Chemical studies on memory fixation in goldfish. Brain Research, 1(3), 303–309. http://dx.doi.org/10.1016/0006-8993(66)90095-3. Allport, G.W., & Postman, L. (1947). The psychology of rumor. New York: Holt, Rnehart & Winston. Anagnostopoulos, A., Bessi, A., Caldarelli, G., Del Vicario, M., Petroni, F., Scala, A., et al. (2014). Viral misinformation: The role of homophily and polarization. Retrieved from http://arxiv.org/pdf/1411.2893v1.pdf (arXiv preprint arXiv:1411.2893). Anttiroiko, A. -V., & Savolainen, R. (2011). Towards library 2.0: The adoption of Web 2.0 technologies in public libraries. Libri, 61(2), 87–99. http://dx.doi.org/10.1515/libr.2011.008. Batellier, F., Couty, I., Picard, D., & Brillard, J.P. (2008). Effects of exposing chicken eggs to a cell phone in “call” position over the entire incubation period. Theriogenology, 69(6), 737–745. http://dx.doi.org/10.1016/j.theriogenology.2007.12.006. Bertot, J.C., Jaeger, P.T., & Grimes, J.M. (2010). Using ICTs to create a culture of transparency: E-government and social media as openness and anti-corruption tools for societies. Government Information Quarterly, 27(3), 264–271. http://dx.doi.org/10. 1016/j.giq.2010.03.001. Bridges, L.M. (2012). Librarian as professor of social media literacy. Journal of Library Innovation, 3(1), 48–65. Broos, A. (2005). Gender and information and communication technologies (ICT) anxiety: Male self-assurance and female hesitation. CyberPsychology & Behavior, 8(1), 21–31. http://dx.doi.org/10.1089/cpb.2005.8.21. Brown, Firefly Millward (2010). The language of love in social media — New rules for brand engagement. Retrieved from http://www.fireflymb.com/Libraries/Social_ Media_Study/2010_SocialMediaStudy_SummaryReport.sflb.ashx. Budak, C., Agrawal, D., & Abbadi, A.E. (2011). Limiting the spread of misinformation in social networks. Proceedings of the 20th International Conference on World Wide Web (WWW '11) (pp. 665–674)http://dx.doi.org/10.1145/1963405.1963499. Connaway, L.S., Dickey, T.J., & Radford, M.L. (2011). “If it is too inconvenient I'm not going after it:” Convenience as a critical factor in information-seeking behaviors. Library & Information Science Research, 33(3), 179–190. http://dx.doi.org/10.1016/j.lisr.2010. 12.002. Correa, T., Hinsley, A.W., & de Zúñiga, H.G. (2010). Who interacts on the Web?: The intersection of users' personality and social media use. Computers in Human Behavior, 26(2), 247–253. http://dx.doi.org/10.1016/j.chb.2009.09.003. Darley, W.K., & Smith, R.E. (1995). Gender differences in information processing strategies: An empirical test of the selectivity model in advertising response. Journal of Advertising, 24(1), 41–56. http://dx.doi.org/10.2307/4188961. DiFonzo, N., & Bordia, P. (2007). Rumor psychology: Social and organizational approaches. Washington, DC: American Psychological Association. Duggan, M., Ellison, N.B., Lampe, C., Lenhart, A., & Madden, M. (2015). Social media update 2014. Retrieved from http://www.pewinternet.org/files/2015/01/PI_ SocialMediaUpdate20144.pdf. Dunne, Á., Lawlor, M. -A., & Rowley, J. (2010). Young people's use of online social networking sites — A uses and gratifications perspective. Journal of Research in Interactive Marketing, 4(1), 46–58. http://dx.doi.org/10.1108/17505931011033551. Emery, D. (2012, May 22). Maggots on the brain/sushi worms. Retrieved from http://urbanlegends.about.com/library/bl-brainworms.htm. Farkas, M. (2011). Information literacy 2.0. American Libraries. Retrieved from http://americanlibrariesmagazine.org/2011/11/01/information-literacy-2-0 Ferrara, E. (2015). Manipulation and abuse on social media. SIGWEB Newsletter. http://dx. doi.org/10.1145/2749279.2749283 (Retrieved from http://arxiv.org/abs/1503. 03752). Field, A. (2009). Discovering statistics using SPSS (3rd ed.). Los Angeles: Sage. Friggeri, A., Adamic, L.A., Eckles, D., & Cheng, J. (2014). Rumor cascades. Proceedings of the Eighth International AAAI Conference on Weblogs and Social Media (pp. 101–110) (Retrieved from http://www.aaai.org/ocs/index.php/ICWSM/ICWSM14/paper/download/8122/8110). Frost, C. (2002). Source credibility: Do we really believe everything we're told? Aslib Proceedings, 54(4), 222–228. http://dx.doi.org/10.1108/00012530210443320. Goldenberg, D. (2006). How to cook an egg (and create a viral sensation). Retrieved from http://www.gelfmagazine.com/archives/how_to_cook_an_egg_and_create_a_viral_ sensation.php. Guardian Interactive team, Procter, R., Vis, F., & Voss, A. (2011). How riot rumours spread on Twitter. Retrieved from http://www.guardian.co.uk/uk/interactive/2011/dec/07/ london-riots-twitter.

9

Guerin, B., & Miyazaki, Y. (2006). Analyzing rumors, gossip, and urban legends through their conversational properties. Psychological Record, 56(1), 23. Hampton, K., Goulet, L.S., Rainie, L., & Purcell, K. (2011). Social networking sites and our lives. Retrieved from http://pewinternet.org/~/media//Files/Reports/2011/PIP%20-% 20Social%20networking%20sites%20and%20our%20lives.pdf. Head, A.J., & Eisenberg, M.B. (2011). How college students use the web to conduct everyday life research. First Monday, 16(4). http://dx.doi.org/10.5210/fm.v16i4.3484. Hertwig, R., & Ortmann, A. (2008). Deception in experiments: Revisiting the arguments in its defense. Ethics & Behavior, 18(1), 59–92. Heverin, T., & Zach, L. (2012). Use of microblogging for collective sense-making during violent crises: A study of three campus shootings. Journal of the American Society for Information Science and Technology, 63(1), 34–47. http://dx.doi.org/10.1002/asi.21685. Kaplan, A.M., & Haenlein, M. (2010). Users of the world, unite! The challenges and opportunities of social media. Business Horizons, 53(1), 59–68. http://dx.doi.org/10.1016/j. bushor.2009.09.003. Karlova, N.A., & Fisher, K.E. (2013). A social diffusion model of misinformation and disinformation for understanding human information behaviour. Information Research, 18(1) (Retrieved from http://informationr.net/ir/18-1/paper573.html). Kim, J.H., Kim, M. -S., & Nam, Y. (2010). An analysis of self-construals, motivations, Facebook use, and user satisfaction. International Journal of Human Computer Interaction, 26(11– 12), 1077–1099. http://dx.doi.org/10.1080/10447318.2010.516726. Kim, S., & Oh, S. (2009). Users' relevance criteria for evaluating answers in a social Q&A site. Journal of the American Society for Information Science and Technology, 60(4), 716–727. http://dx.doi.org/10.1002/asi.21026. Kim, K. -S., & Sin, S. -C.J. (2011). Selecting quality sources: Bridging the gap between the perception and use of information sources. Journal of Information Science, 37(2), 178–188. http://dx.doi.org/10.1177/0165551511400958. Kim, K. -S., & Sin, S. -C.J. (2014a). Social media as information sources: Use and evaluation of information from social media. 2013 OCLC/ALISE research grant report. Retrieved from http://www.oclc.org/content/dam/research/grants/reports/2013/kim2013.pdf. Kim, K. -S., & Sin, S. -C.J. (2014b). Use of social media in different contexts of information seeking: Effects of sex and problem-solving style. Proceedings of the Information Behaviour Conference (ISIC) 2014 (Retrieved from http://www.informationr.net/ir/201/isic2/isic24.html). Kim, K. -S., Sin, S. -C.J., & Tsai, T.I. (2014a). Individual differences in social media use for information seeking. The Journal of Academic Librarianship, 40(2), 171–178. Kim, K. -S., Sin, S. -C.J., & Yoo-Lee, E.Y. (2014b). Undergraduates' use of social media as information sources. College & Research Libraries, 75(4), 442–457. http://dx.doi.org/10. 5860/crl.75.4.442. Lampe, C., Ellison, N., & Steinfield, C. (2006). A face(book) in the crowd: Social searching vs. social browsing. Proceedings of the 20th Anniversary Conference on Computer Supported Cooperative Work (CSCW '06) (pp. 167–170). http://dx.doi.org/10.1145/1963405. 1963499. Lee, C.S., Goh, D.H. -L., Chua, A.Y.K., & Ang, R.P. (2010). Indagator: Investigating perceived gratifications of an application that blends mobile content sharing with gameplay. Journal of the American Society for Information Science and Technology, 61(6), 1244–1257. http://dx.doi.org/10.1002/asi.21305. Lee, C.S., & Ma, L. (2012). News sharing in social media: The effect of gratifications and prior experience. Computers in Human Behavior, 28(2), 331–339. http://dx.doi.org/ 10.1016/j.chb.2011.10.002. Lim, S. (2009). How and why do college students use wikipedia? Journal of the American Society for Information Science and Technology, 60(11), 2189–2202. Lim, S. (2013). College students' credibility judgments and heuristics concerning Wikipedia. Information Processing and Management, 49(2), 405–419. Lim, S., & Kwon, N. (2010). Gender differences in information behavior concerning Wikipedia, an unorthodox information source? Library & Information Science Research, 32(3), 212–220. http://dx.doi.org/10.1016/j.lisr.2010.01.003. Lim, S., & Simon, C. (2011). Credibility judgment and verification behavior of college students concerning Wikipedia. First Monday, 16(4). http://dx.doi.org/10.5210/fm. v16i4.3263. Meyers-Levy, J., & Maheswaran, D. (1991). Exploring differences in males' and females' processing strategies. Journal of Consumer Research, 18(1), 63–70. Mintz, A.P. (2012a). If it's on the Internet, it must be true. In A.P. Mintz (Ed.), Web of deceit: Misinformation and manipulation in the age of social media. Medford, N. J.: CyberAge Books, Information Today, Inc. Mintz, A.P. (Ed.). (2012). Web of deceit: Misinformation and manipulation in the age of social media. Medford, NJ: Information Today. Mitrano, T. (2011). “It takes only a generation:” Information literacy in the digital age. Retrieved from http://archive-edu.com/page/4840981/2014-10-30/http://www.it. cornell.edu/policies/publications/generation.cfm. Morris, M.R., Teevan, J., & Panovich, K. (2010). What do people ask their social networks, and why? A survey study of status message Q&A behavior. Proceedings of the 28th SIGCHI Conference on Human Factors in Computing Systems. 3. (pp. 1739–1748). Nahl, D., & Bilal, D. (2007). Information and emotion: The emergent affective paradigm in information behavior research and theory. Medford: Information Today. Neuman, W.L. (2011). Social research methods: Qualitative and quantitative approaches (7th. ed. International ed.). Boston: Pearson Education. Nov, O., Naaman, M., & Ye, C. (2009). Analysis of participation in an online photo-sharing community: A multidimensional perspective. Journal of the American Society for Information Science and Technology, 61(3), 555–566. http://dx.doi.org/10.1002/asi. 21278. Oh, S., & Syn, S. Y. (2015). Motivations for sharing information and social support in social media: A comparative analysis of Facebook, Twitter, Delicious, Youtube, and Flickr. Journal of the Association for Information Science and Technology. http://dx.doi.org/ 10.1002/asi.23320.

Please cite this article as: Chen, X., et al., Why Students Share Misinformation on Social Media: Motivation, Gender, and Study-level Differences, The Journal of Academic Librarianship (2015), http://dx.doi.org/10.1016/j.acalib.2015.07.003

10

X. Chen et al. / The Journal of Academic Librarianship xxx (2015) xxx–xxx

Onwuegbuzie, A.J., Jiao, Q.G., & Bostick, S.L. (2004). Library anxiety : Theory, research, and applications. Lanham, Md.: Scarecrow Press. Oyeyemi, S.O., Gabarron, E., & Wynn, R. (2014). Ebola, Twitter, and misinformation: A dangerous combination? Thes BMJ, 349. http://dx.doi.org/10.1136/bmj.g6178. Park, J. -H. (2010). Differences among university students and faculties in social networking site perception and use: Implications for academic library services. The Electronic Library, 28(3), 417–431. http://dx.doi.org/10.1108/02640471011051990. Park, N., Kee, K.F., & Valenzuela, S. (2009). Being immersed in social networking environment: Facebook groups, uses and gratifications, and social outcomes. CyberPsychology & Behavior, 12(6), 729–733. http://dx.doi.org/10.1089/cpb.2009.0003. Ratkiewicz, J., Conover, M., Meiss, M., Gonçalves, B., Patil, S., Flammini, A., et al. (2010). Detecting and tracking the spread of astroturf memes in microblog streams. arXiv preprint arXiv:1011.3768. Retrieved from http://arxiv.org/abs/1011.3768 Rieh, S.Y. (2002). Judgment of information quality and cognitive authority in the Web. Journal of the American Society for Information Science and Technology, 53(2), 145–161. http://dx.doi.org/10.1002/asi.10017. Rodgers, S., & Harris, M.A. (2003). Gender and e-commerce: An exploratory study. Journal of Advertising Research, 43(03), 322–329. http://dx.doi.org/10.1017/S0021849903030307. Rosnow, R.L. (1991). Inside rumor — A personal journey. American Psychologist, 46(5), 484–496. http://dx.doi.org/10.1037/0003-066x.46.5.484. Ross, C., Orr, E.S., Sisic, M., Arseneault, J.M., Simmering, M.G., & Orr, R.R. (2009). Personality and motivations associated with Facebook use. Computers in Human Behavior, 25(2), 578–586. http://dx.doi.org/10.1016/j.chb.2008.12.024. Ruggiero, T.E. (2000). Uses and gratifications theory in the 21st century. Mass Communication and Society, 3(1), 3–37. http://dx.doi.org/10.1207/s15327825mcs0301_02. Saner, E. (2011). How to pamper a panda. Retrieved from http://www.guardian.co.uk/ world/2011/dec/05/how-to-pamper-a-panda. Saquete, R. (2010). The myth of the antiradiation cactus. Retrieved from http://www.computableminds.com/post/cactus/monitor/radiation/antiradiation/ ultraviolet. Shah, C. (2012). Collaborative information seeking: The art and science of making the whole greater than the sum of all. Berlin: Springer. Shao, G. (2009). Understanding the appeal of user-generated media: A uses and gratification perspective. Internet Research, 19(1), 7–25. http://dx.doi.org/10.1108/ 10662240910927795.

Sin, S. -C.J., & Kim, K. -S. (2013). International students' everyday life information seeking: The informational value of social networking sites. Library & Information Science Research, 35(2), 107–116. http://dx.doi.org/10.1016/j.lisr.2012.11.006. Spiranec, S., & Zorica, M.B. (2010). Information literacy 2.0: Hype or discourse refinement? Journal of Documentation, 66(1), 140–153. http://dx.doi.org/10.1108/ 00220411011016407. The University of New Hampshire (2000). Kentucky fried chicken hoax. Retrieved from http://www.unh.edu/BoilerPlate/kfc.html. Urquhart, C., & Yeoman, A. (2010). Information behaviour of women: Theoretical perspectives on gender. Journal of Documentation, 66(1), 113–139. Walsh, J. (2010). Librarians and controlling disinformation: Is multi‐literacy instruction the answer? Library Review, 59(7), 498–511. http://dx.doi.org/10.1108/ 00242531011065091. Weber, E.U., Blais, A. -R., & Betz, N.E. (2002). A domain-specific risk-attitude scale: Measuring risk perceptions and risk behaviors. Journal of Behavioral Decision Making, 15(4), 263–290. http://dx.doi.org/10.1002/bdm.414. Weiser, E.B. (2000). Gender differences in Internet use patterns and Internet application preferences: A two-sample comparison. CyberPsychology and Behavior, 3(2), 167–178. http://dx.doi.org/10.1089/109493100316012. Witek, D., & Grettano, T. (2012). Information literacy on Facebook: An analysis. Reference Services Review, 40(2), 242–257. http://dx.doi.org/10.1108/00907321211228309. Wyllie, A., Zhang, J.F., & Casswell, S. (1998). Positive responses to televised beer advertisements associated with drinking and problems reported by 18 to 29-year-olds. Addiction, 93, 749–760. Yaari, E., Baruchson-Arbib, S., & Bar-Ilan, J. (2011). Information quality assessment of community generated content: A user study of Wikipedia. Journal of Information Science, 37(5), 487–498. http://dx.doi.org/10.1177/0165551511416065. Zhang, Y. (2012). College students' uses and perceptions of social networking sites for health and wellness information. Information Research, 17(3), 523 (Retrieved from http://www.informationr.net/ir/17-3/paper523.html). Zhou, M. (2014). Gender difference in web search perceptions and behavior: Does it vary by task performance? Computers & Education, 78, 174–184.

Please cite this article as: Chen, X., et al., Why Students Share Misinformation on Social Media: Motivation, Gender, and Study-level Differences, The Journal of Academic Librarianship (2015), http://dx.doi.org/10.1016/j.acalib.2015.07.003