79 Facebook as a Library Tool: Perceived vs. Actual Use Terra B. Jacobson Terra B. Jacobson is Public Services Manager in the Library at Prairie State College; e-mail: Terra.jacobson@ gmail.com. The author would like to thank Dr. Noriko Hara for all of her assistance and guidance with this paper. © Terra B. Jacobson As Facebook has come to dominate the social networking site arena, more libraries have created their own library pages on Facebook to cre- ate library awareness and to function as a marketing tool. This paper examines reported versus actual use of Facebook in libraries to identify discrepancies between intended goals and actual use. The results of a 2009 study about the use of Facebook in libraries are used as a guide to gauge the perceived and actual uses for Facebook in this study. Results of the test reveal that the two ranks are not statistically different, but that there is a noticeable difference when looking at the perceived and actual rankings qualitatively. n the past few years, libraries have begun to examine the possibilities available to them through social networking sites like Twitter and Facebook as a tool for library awareness and marketing. As Facebook has come to dominate the so- cial networking site arena, more libraries have created their own library pages on Facebook to create library awareness and to function as a marketing tool. This has spurred a large number of how-to articles about the uses for Facebook in libraries as well as research about how librarians and libraries use Facebook. This paper exam- ines reported versus actual use of Face- book in libraries to identify discrepancies between intended goals and actual use. The results of the 2009 study by Hendrix, Chiarella, Hasman, Murphy and Zafron, about the use of Facebook in libraries, is used as a guide to gauge the perceived and actual uses for Facebook in this study.1 Literature Review Although Facebook’s existence is hardly new, its use in the library setting is just taking hold. Library literature relating specifically to Facebook has a very lim- ited range at this time and is not widely published or is found in smaller or digital- only journals. A review of the existing lit- erature shows that most research relating to Facebook in the library setting fits into one of five categories, with some of these categories consisting of only one paper. This review examines each of these five categories to determine what is missing in library literature on Facebook and to explain what this study has to contribute to library literature. Currently, the most popular type of literature is “How-To” based. Included in this category are also articles relating to best practices as well as ideas for the use of Facebook. This area includes a study by Charnigo and Barnett-Ellis, in crl-88r1 80  College & Research Libraries  January 2011 which they survey academic librarians to understand their knowledge of Facebook use on campus as well as the potential it has to be used as an academic resource. Their study found that librarians were wary about the actual academic purpose of Facebook with “54 percent of those surveyed” stating that it did not serve an academic purpose.2 The lack of academic purpose found in Facebook may be what led to the many how-to and library pro- motion articles that followed this 2007 study. This includes a wide range of purposes for Facebook, from Cooper ’s article addressing how to create custom or use-existing library-related applications to articles stating the marketing benefits of Facebook for libraries.3 With an onslaught of Facebook sug- gestions, the analysis of library-centered case studies soon followed suit in dis- cussing individual library experiences using Facebook in a variety of outreach and marketing levels. Glazer describes the attempt to use Facebook to market the Rutgers library, sending out mass messages through Facebook as well as using it to create closer ties with other campus organizations.4 Yet this study does not provide any measurable results, just a qualitative discussion about their personal experience. This type of article is reflected in much of the case studies found in the literature; similar discus- sions occur in both Adamich’s study, “The ‘Facebook’ Effect on FDLP Communica- tion,” and in Powers, Schmidt and Hill’s “Why Can’t we be Friends? The MSU Libraries Find Friends on Facebook.”5 Both of these studies look at how their library has used Facebook, but none can really evaluate its effectiveness; a crux in the study of Facebook as it is very difficult to measure. The previous studies, as well as others, have often included an element of the sur- vey method. The next category of litera- ture falls under the area of student-based research. Two articles that conducted studies relevant to the use of Facebook in libraries both examine the effectiveness of Facebook as an outreach tool through student perspectives. Mathews’ article de- scribes the author’s experience in contact- ing students via Facebook and the results of his “friend requests.” The author found that it helped in “promoting the library” and prompted some questions about library resources.6 The author suggests Facebook as an outreach tool to connect with students and encourages “colleagues to do the same.”7 Another student-based study, by Chu and Meulemans, investi- gates the willingness of students to com- municate with professors via Facebook and offers “practical integration” ideas for library use of Facebook.8 They suggest the use of Facebook for “content delivery, reference assistance, outreach, and infor- mation literacy instruction” and suggest that students may be willing to commu- nicate with organizations via Facebook, although their research showed that students were reluctant to communicate with faculty this way.9 Another type of study concerning the use of Facebook in libraries is the service- provided analysis. This looks at a particu- lar service that the library is attempting to provide. The study by Mack, Behler, Roberts, and Rimland follows the imple- mentation and promotion of reference services via Facebook. The study found that from September through the end of the semester in December, 126 reference questions were received via Facebook out of the 441 total reference questions tracked. This was the highest of all areas, prompting the authors to recommend that libraries create a Facebook page to connect with their patrons.10 The final subset of literature is the perceived-use study. This is currently a small subset of literature containing an article by Hendrix, Chiarella, Hasman, Murphy, and Zafron. This study uses the survey method to question librarians in the health sciences about their libraries’ use of Facebook. This survey provided a listing of what librarians believe their library uses Facebook for. The results provided illustrated that most librarians Facebook as a Library Tool: Perceived vs. Actual Use  81 say they use their Facebook page to “push out announcements,” “post photos,” and “to be where users are.”11 The authors concluded that it was not possible to determine “whether or not Facebook is a useful and effective application.”12 This is one of the very few studies to address what Facebook is used for in the library setting and provides useful guidelines for perceived use for the analysis presented in this paper. This current study attempts to add to the literature on library Facebook use, as there is a lack of research studies in this area. What none of the previous literature addresses are actual library Facebook pages and their content. This study compares the perceived use of library Facebook pages as viewed by librarians with actual library Facebook pages. This examination will shed light on what libraries are actually using Facebook for and help in establishing a set of actual uses for libraries. The study will add to the literature and future research by providing librarians with a list of actual uses they can implement and also provide researchers with an outline of services that can potentially be evaluated for ef- fectiveness in further research. Methodology This study used the results from the Hendrix et al. study to set a framework for the comparison of actual library Face- book pages, groups, and profiles to what librarians are stating they are used for (in other words, perceived use).13 The top responses from their survey will be used to compare with actual use. The current analysis uses two data col- lection steps to obtain the broadest range of data. The investigation began with a selection process, determining which library Facebook pages would be used in the evaluation. Since many libraries do not update their pages regularly, shown by the Hendrix et al. study with some libraries only performing “0 to 20 minutes of maintenance a week,” library Facebook pages, for this study, had to be carefully selected to gain insight on Facebook page performance.14 To avoid performing an analysis on Facebook pages that were not updated at all in the past month, a listserv of college librarians were asked for their input on frequently updated (at least once a week) library pages. The Facebook Web site was also searched for frequently updated library pages to include in the study. This frequency of updates is important since, if the pages are not active, there would have been nothing to observe. The selection process resulted in 12 library Facebook pages for analysis. To be sure to provide an even analysis of libraries, 2 libraries from 6 categories of libraries were selected to be observed. These categories were as fol- lows: Small Colleges/Universities (under 3,000 students); Midsize Colleges/Univer- sities (3,000–10,000 students); Large Col- leges/Universities (10,000 students and above); Branch Libraries (part of a larger university system); Community Colleges (offering up to associates degrees); Special Libraries (subject-specific collections). Data collection for this study was per- formed on the same day for all individual library Facebook pages (April 14, 2009). A form was created to collect data from the pages as objectively as possible (see Appendix 1). Each library was assigned a random number from 1 through 12 to distance the results from other known information about each institution. The areas selected for data collection on the Site Observation Form are all possible in- dicators for the uses indicated in the Hen- drix et al. study as well as other relevant data that would be useful for possible correlations.15 Information collected in the Site Observation Form (see Appendix 1) included the date of Facebook page cre- ation for each page, the number of fans, the number of tracked page changes and the number of days the page was in exis- tence. The form also collected information about the use of specific applications such as photos, events, and reference help. It also collected information on the total number of wall posts and the number of 82  College & Research Libraries  January 2011 posts from fans as well as other optional applications that were added to the Face- book pages. The data collected from this form was entered into two tables for data analysis. The first table (see table 1) contains data relating to the maintenance and use of the Facebook page. This includes infor- mation about how many fans, wall mes- sages, fan messages, photos, events, and boxes applications could be found on the page. It also contains information about how many days each library Facebook page has existed and the average days between update (calculated by the days in existence divided by the total updates). The second data set (see table 2) is a col- lection of yes/no data where 1=yes and 0=no. These data calculate how many add-on applications are used on each of the library Facebook pages and how many times each application is used between the 12 library pages. Additionally, to prepare for the analy- sis, the survey results from Hendrix et al. were categorized into 6 tiers, rank- ing from highest to lowest number of responses. The items with the highest number of responses were ranked as tier 1, the second highest as tier 2, and so on through 5 tiers of “use.” Any data from the survey that did not generate any re- sponse was stated to have “no use.”16 This ranking system is also used in the analysis to compare the perceived use from the Hendrix study to the data retrieved from this study. These uses were developed by Hendrix in their 2009 study as possible uses for Facebook. The types of use and their rank from the Hendrix data are as follows: Tier 1: Announcements/Marketing, Photos, To be where users are Tier 2: Reference services Tier 3: Forum for Users Tier 4: RSVP to events, OPAC search, Database Search Tier 5: Employee Announcements, Employee Communication No Use: Administer a course, Tutorials, Post Audio, Post Video, Podcasts17 Results Data collected from the Site Observation Form was input into their corresponding Excel spreadsheets totaling the number of fans, wall messages, fan messages, photos, events, “boxes” applications, “add-ons,” updates, and days in existence (quantity of use). Additional data were also gathered about how many library pages used particular applications such as reference services, video, discussion boards, and the like (yes/no responses). Table 1 Facebook Page Maintenance and Use * Data format for this item only 1=yes, 0=no Library Number Fans Wall Messages Fan Messages Photos Info Tab Used* Events Boxes Total Updates Days in Existence Avg. Day Update 1 62 10 1 6 1 5 1 15 81 5.40 2 20 28 0 10 1 2 4 55 449 8.16 3 103 69 0 3 1 32 6 82 201 2.45 4 147 140 0 25 1 45 3 45 340 7.50 5 217 19 10 0 1 3 6 21 45 2.14 6 534 153 7 70 1 0 4 179 312 1.74 7 57 16 1 1 1 2 1 24 56 2.30 8 132 17 4 0 1 0 0 19 489 25.74 9 369 111 7 2 1 0 4 115 439 3.82 10 228 37 9 252 1 7 8 45 348 7.73 11 100 10 2 7 1 2 0 16 73 4.56 12 804 41 5 53 1 11 4 47 239 5.09 Total 2773 651 46 429 12 109 41 663 3072 76.63 Mean 231.08 54.25 3.83 35.75 1.00 9.08 3.42 55.25 256.00 6.39 Facebook as a Library Tool: Perceived vs. Actual Use  83 Therefore, two different types of data were collected and analyzed: data on how many libraries used a particular applica- tion and the quantity of use. Averages were collected for information pertain- ing to the quantity of use, and ratios for updates were also determined. These data are useful in understanding library Facebook usage as a whole, whereas the yes/no data for how many libraries use particular applications were very useful in determining the tier ranking for specific applications. Both were used to determine the actual use to compare to the Hendrix et al. survey data as a ranking tool, which will be discussed in the analysis section.18 The data sets will not be discussed at length in this paper, as these data were used primarily for the ranking proce- dure discussed in both the analysis and discussion sections of this paper; the collected data are available in table 1 and table 2. These data did allow for a Pearson Correlation to be conducted to determine if any of the data showed any correlations, meaning if any one aspect of Facebook use was at all related to other types of use. Pearson Correlations were run using SPSS to determine relation- ships between how many fans a Face- book page has and how many events, add-on applications, and updates a Face- book page has. It was found that there is no relationship between how many events are posted on a Facebook page and how many fans a page may have, with a negative correlation of .096. There were moderate relationships found in both the number of add-on applications and fans, a positive correlation of .554, as well as the relationship concerning how many updates had occurred on the Facebook page and the number of fans the page had, a positive correlation of .478. This does not mean that these add- ons or number of updates increases the number of fans a page may have, but that there is some relationship between these items (although none of them are statistically significant at the 0.05 level). Also, it must be noted that the sample size in this analysis is very small (n=12). It may be possible that frequent updates and an increased number of add-on applications may help in developing a larger Facebook fan base, and it would be worth consideration in library Facebook page development. Table 2 Number of add-On applications on Facebook Pages Note. 1=yes, 0=no for this data set. Library Number Ref/Ask Help Notes Links IM Video Fav. Pages Extended Info Search Works Flickr Discussion Books Contact Us Rss/Blog Total Add ons 1 1 0 0 0 0 0 0 0 0 1 0 0 0 2 2 0 0 0 0 0 1 0 0 0 1 0 0 0 2 3 0 0 1 0 0 1 0 0 0 0 0 0 0 2 4 0 0 0 0 1 1 0 0 0 0 0 0 0 2 5 1 1 1 0 0 1 0 0 0 0 0 0 0 4 6 1 0 0 1 0 0 0 1 0 1 0 0 1 5 7 0 1 1 0 0 1 0 0 0 0 1 1 0 5 8 0 0 0 0 0 1 0 0 0 0 0 1 0 2 9 1 1 1 1 0 0 0 0 0 0 0 0 0 4 10 0 0 1 0 0 0 0 0 0 1 0 0 1 3 11 0 0 0 0 0 0 0 0 0 0 0 0 1 1 12 0 1 0 0 1 0 1 0 0 1 0 0 0 4 Total 4 4 5 2 2 6 1 1 0 5 1 2 3 36 Mean 0.33 0.33 0.42 0.17 0.17 0.50 0.08 0.08 0.00 0.42 0.08 0.17 0.25 3.00 84  College & Research Libraries  January 2011 From the data collected with the Site Observation Form, it was then deter- mined whether the library Facebook pages were performing the expected uses presented in the Hendrix study. The data were gathered in table 1 and table 2 and were then used to determine the ranking of each Facebook “use” that was previously determined by Hendrix et al. For each time a library Facebook page had an application on its page, the task was counted as a “use” of that item. This could result in a total of up to 12 out of 12 uses for any one application. Each ap- plication was associated with a “use” that was defined by Hendrix et al. that was determined by each use (such as Photos, Reference Services, and the like) being matched up with plausible Facebook ap- plications, options, or sections of the page that would indicate that a particular use was occurring in library Facebook pages. For example, the use “Forum for Users to Communicate” would be observed through the use of the Facebook discus- sion tool and by the number of messages left on the wall that were communication from “fans.” If these were used, then this would confirm the use “Forum for Users to Communicate,” and based on how many of the twelve pages confirmed this use, that particular use would be ranked (according to the same tiers laid out in the methodology section). The ranking requirements are indicated as follows: Tier 1: at least 10 libraries perform this use Tier 2: at least 8 libraries perform this use Tier 3: at least 6 libraries perform this use Tier 4: at least 4 libraries perform this use Tier 5: at least 1 library performs this use No Use: indicated by 0 libraries per- forming this use These rankings were created to mirror the information provided by Hendrix et al. to best recreate a similar and equally placed difference between the tiers. In this tier system, zero indicates “no use” since any actual use of that particular “use” would have been determined by the data collected. Analysis To understand how each use was classi- fied into a specific tier for ranking of its particular presence on Facebook pages, each “use” ranking will be explained, outlining which data supports the place- ment of that “use.” The use “Announce- ments/Marketing” was determined by the information provided by the Facebook Wall and the number of fans a page had as well as the events posted. In this case, 11 out of 12 libraries (all numbers concerning how many libraries will be out of a total of 12 libraries) posted information about events and 8 libraries announced research services and sources. The average update time for library pages was every 6.4 days with an average of 9 events posted per library site. These results ranked this use in tier 1. The use “Photos” used Flickr applications and photos posted in the photo section of the page to determine the ranking of this use. The average number of photos per page was 35.75, and 10 out of 12 libraries had photos on the page. However, 6 of these pages only had pro- file photos posted. Therefore, this is not the primary use of that Facebook site and the use was ranked accordingly, at tier 3. The next use to be examined is “to be where users are,” which used users’ posts and fans to determine its ranking. The data show an average of about 231 fans per site with all but three pages having over 100 fans, but there are only 3.8 fan messages per library on average. Only 6 have more than 2 posts individually and 4 library pages have 0 posts. This is not what one would consider user par- ticipation, and therefore this use was also ranked at tier 3. The next use, “reference services,” is an outreach tool that has been discussed in past literature, but only 4 libraries provided an option to contact the library (determined by reference contact information or ask a librarian services), resulting in a ranking at tier 4. The use “forum for users to communicate” was determined by the discussion options as well as wall posts. Out of 12 libraries, 5 had a discussion section, which shows ef- Facebook as a Library Tool: Perceived vs. Actual Use  85 fort, but most of these discussions had no past or current participants. Along with this finding, only 3.8 out of 54.25 average wall posts were from fans, which is not a high percentage of fan input. This study is not discussing effectiveness of these tools, but the actual use on the library’s end, resulting in this use being ranked in tier 4. Along the same vein of fan com- munication, the use “RSVP to events” is evaluated by the use of the events tab. The data collected showed an average of 9 events posted per library with 9 out of the total 12 libraries posting events. Since it is not possible to see the actual replies to events from viewing the profile alone, this was ranked based on the number of libraries posting events, giving it a high rank of tier 2. The next items examined were library search tools. In the first of these uses, “OPAC Search,” evaluation was deter- mined by library pages that included a search tool. A total of 10 libraries supplied an OPAC or catalog search on their site, making it highly ranked as tier 1. The other search tool, “Database search,” only had 2 out of 12 libraries using this tool, which was evaluated by the presence of a database search utility on the pages; this resulted in it being ranked at tier 5. Originally in the Hendrix et al. study, the use “post video” was determined to have no use.19 This study found that “post video” was actually ranked at tier 5, with 2 out of 12 libraries using Facebook to post video (determined by the actual posting of video on the page). The last items listed with use in the Hendrix et al. survey, “employee announcements” and “employee communication,” were determined by listings on the RSS feed as well as the events tool.20 Both of these uses were not mentioned in either the RSS feed or the events section. This resulted as these items having no use. A difficulty in this determination of use was that this research does not allow for examination of private communication through Face- book; therefore, this finding of no use could be incorrect. Other items resulting in no use in the Hendrix et al. study—administer a course, tutorials, post audio and pod- casts—also were found to have no use on the library Facebook pages.21 These were determined by the posting of course materials or links, tutorials, and audio posts (both general and podcasts). None of the pages examined used these tools, resulting in no use for these areas. Items found to have “no use” do not necessar- ily mean that these are not used in other library Facebook pages, but this does tell us that the use of these tools is very low in general, since this sample covered various library types. Discussion of Perceived Use and Actual Use To illustrate the change in rank order for the “uses” evaluated in this study, figure 1 shows the comparison between the Hendrix et al. study and the results of this analysis or the “Actual Facebook Con- tents.”22 Each item is listed from highest to lowest ranking on the predetermined tier system (Tier 1–5 and No Use). The number to the left of each use is its deter- mined tier, and items that were not found on the pages or were listed as having no use were classified in the category “no use.” The results of this analysis show that many of the same “uses” had no use; added to the category of no use is “em- ployee communication” and “employee announcements,” and the use “video” was moved to a tier 5 category since some Facebook pages (2) used video applica- tions. Other major differences noticeable in figure 1 is the higher ranking of OPAC Searches, RSVP to events, and the drop in ranking from the Hendrix et al. survey of the photo, to be where users are, forum for users, and database search uses. The surveys both seem to be on par with the use, announcements/marketing, perhaps confirming that this may be a good use of this tool. To fully examine the change between the perceived and actual use of Facebook by librarians, a Wilcoxon Signed-Rank 86  College & Research Libraries  January 2011 Test was conducted that assesses the null hypothesis for the data sets. Each use was ranked against itself as shown in figure 1, and all items that had a rank of “no use” were ranked as a value of 6 to be able to conduct the Wilcoxon Signed-Rank Test with SPSS version 16. The null hypothesis in this test is that the two samples do not differ. Results of this test accepted the null hypothesis. The results of the Wilcoxon Signed-Rank Test shown in tables 3 and 4 illustrate that the p value of .467 is not statistically significant, leaving the results to be questionable when relating to the differences in rank between the perceived and actual use. What these data present is that the two rankings are not statistically signifi- cantly different. Qualitatively, though, the Hendrix et al. survey ranked many of these items slightly higher, which could indicate that librarians perceive their use of Facebook to be much more active than it actually is even though the Wilcoxon Signed-Rank Test accepts the null hypothesis that their differences are not statistically significant. This could be a result from the homogenous nature of the librarians surveyed (all academic health science librarians) in the Hendrix et al. study resulting from beliefs de- veloped in their particular community of practice.23 This result could also be influenced by the attempt to classify these items into a particular ranking system with the addition of a rank of 6 to calculate the Wilcoxon Signed-Rank Test. Also, many nonparametric tests were examined for use in this study (Mann- Whitney test, Spearman Rank Order Correlation Coefficient, and others), but none of them were correct for the type FigUre 1 Comparison in raking between Hendrix et al. (2009) and actual Facebook Contents Hendrix et al. (2009) Survey actual Facebook Contents 1. Announcements/Marketing 1. Announcements/Marketing (-) 1. Photos 1. OPAC Search (↑) 1. To be where users are 2. RSVP to events (↑) 2. Reference services 3. Photos (↓) 3. Forum for Users 3. To be where users are (↓) 4. RSVP to events 4. Reference Services (↓) 4. OPAC search 4. Forum for Users (↓) 4. Database Search 5. Database Search (↓) 5. Employee Announcements 5. Video (↑) 5. Employee Communication No Use No Use Administer a course Administer a course (-) Tutorials Tutorials (-) Post Audio Post Audio (-) Video Employee Communication (↓) Podcasts Employee Announcements (↓) Note. Each tier can contain more than one use. There is no limit to how many uses may fall into a specific tier classification. (-) = No change in rank; (↓) = decrease in rank; (↑) = increase in rank. Facebook as a Library Tool: Perceived vs. Actual Use  87 of ranking data used in this comparison; it is still difficult to say if the Wilcoxon Signed-Rank Test can appropriately match these types of data, but it was the best fit out of all the nonparametric tests. Although the results of the test reveal that the two ranks are not statistically dif- ferent, I observe that figure 1 illustrates a noticeable difference when looking at the types of use in both the perceived and actual rankings qualitatively. There may not be a large difference in rank, but it seems that there is a difference of one or two tiers in many of the uses for Facebook presented here. Issues/Recommendations This study is not without issues that influence the results presented here. Because not many library pages are still active (not updated within the past 6 months), it was very difficult to select candidates for this study. Also, inactive pages are not represented in this study, which would greatly influence the results of this evaluation. However, since this study is looking at use, it was determined by the author that pages must be in use to evaluate the current uses. Along with this issue of use, the study is also based on the researcher’s perception of what Facebook programs and application are predictors of particular uses, and the evaluation in this study is only performed by one researcher. Perhaps further evaluations would consider conducting multiple evaluations of each site by various evalu- ators. It is also difficult to determine or predict use since, although an applica- tion may be provided (such as a discussion board), that does not necessarily mean that it is being used by “fans.” This was not accounted for in this study since this ob- servation does not attempt to predict use of the Facebook pages by patrons or “fans” but how librarians and libraries use Facebook as a library tool. Although this study confirms many of the same uses as the Hendrix et al. study, it does illustrate that there are specific ways in which libraries and librarians can best use Facebook. The results presented here show that many libraries are using Facebook primarily as a marketing tool, and it may be valid to assert that this is currently the best use in the library realm. However, uses for communication from patrons or “fans,” communicating library needs, and as a forum/discussion space for users may not be an ideal use. Most pages observed in this study did not have any “fan” posts or user discussion even though the space was provided for them. Librarians state that many of them are doing more with their Facebook pages than what is actually showing up in their page content. Librarians need to be aware of how much they are actually doing with their Facebook pages and look at why they created or want to create a Facebook page. This study shows that we, as librar- ians, may be more ambitious in our hopes for our Facebook presence than we can actually materialize. Table 3 Wilcoxon Signed-rank Test: ranks N Mean rank Sum of ranks Perceived - Actual Negative Ranks 7a 4.93 34.50 Positive Ranks 3b 6.83 20.50 Ties 5c Total 15 a. Perceived < Actual | b. Perceived > Actual | c. Perceived = Actual Table 4 Wilcoxon Signed-rank Test: Test Statisticsb Perceived - Actual Z -.728a Asymp. Sig. (2-tailed) .467 a. Based on positive ranks. b. Wilcoxon Signed-Ranks Test 88  College & Research Libraries  January 2011 Conclusion Librarians must be aware that using Facebook as a tool requires a lot of atten- tion, and, if a library cannot commit to updating its Facebook page at least once a week or more, it may not be a very useful or successful tool. From this study, it also was found that Facebook would be a bet- ter tool for “active libraries,” or libraries that host a lot of events, exhibits, work- shop, and other activities, as its top use is for announcements and marketing. Also, librarians should not get too attached to Facebook, as there is always the next tool or social networking site that people are using. Web 2.0 applications move quickly, and the Internet is constantly changing; librarians should be prepared to leave their hard work behind to jump to the next tool. Notes 1. D. Hendrix and others, “Use of Facebook in Academic Health Sciences Libraries,” Journal of the Medical Library Association 97, no. 1 (2009): 44–47. 2. Laurie Charnigo and Paula Barnett-Ellis, “Checking Out Facebook.com: The Impact of a Digital Trend on Academic Libraries,” Information Technology and Libraries 26, no. 1 (2007): 30. 3. Jason D. Cooper, “Facebook Applications for the Library Community,” The Alabama Li- brarian 58, no. 1 (2008): 8–11; Susan Jennings and Jamie Price, “Be My Friend: Using Facebook in Libraries,” Tennessee Libraries 58, no. 2 (2008), available online at www.tnla.org/displaycommon. cfm?an=1&subarticlenbr=213 (accessed 12 March 2009). 4. Harry Glazer, “Clever Outreach or Costly Diversion? An Academic Library Evaluates its Facebook Experience,” College and Research Libraries News 70, no. 1 (2009): 11–19. 5. Tom Adamich, “The Faceboook Effect on FDLP Communication” Documents to the People 36, no. 2 (2008): 29–32; Amanda Clay Powers, June Schmidt, and Clay Hill, “Why Can’t We Be Friends? The MSU Libraries Find Friends on Facebook,” Mississippi Libraries 72, no. 1 (2008): 3–5. 6. Brian S. Mathews, “Do You Facebook? Networking with Students Online,” College and Research Libraries News 67, no. 5 (2006): 307. 7. Ibid., 307. 8. Melanie Chu and Yvonne Nalani Meulemans, “The Problems and Potential of MySpace and Facebook Usage in Academic Libraries.,” Internet Reference Services Quarterly 13, no. 1 (2007): 80–84. 9. Ibid., 83. 10. D. Mack and others, “Reaching Students with Facebook: Data and Best Practices.” Electronic Journal of Academic and Special Librarianship 8, no 2 (2007): http://southernlibrarianship.icaap.org/ indexv8.html (accessed March 12, 2009). 11. Hendrix, “Use of Facebook in Academic Health Sciences Libraries,” 44–47. 12. Ibid. 13. Ibid. 14. Ibid. 15. Ibid. 16. Ibid. 17. Ibid. 18. Ibid. 19. Ibid. 20. Ibid. 21. Ibid. 22. Ibid. 23. Ibid. Facebook as a Library Tool: Perceived vs. Actual Use  89 Appendix 1: Site Observation Form: Page 1 90  College & Research Libraries  January 2011 Site Observation Form: Page 2