Microsoft Word - ES_1781 Evidence Based Library and Information Practice 2008, 3:3 68 Evidence Based Library and Information Practice Evidence Summary Half of Vermont Academic and Public Libraries Lack Written Confidentiality Policies Despite Directors’ Support for Principles of Patron Confidentiality A Review of: Magi, Trina J., “The Gap between Theory and Practice: A Study of the Prevalence and Strength of Patron Confidentiality Policies in Public and Academic Libraries.” Library & Information Science Research 29 (2007): 455-70. Reviewed by: Stephanie Walker Acting Chief Librarian Brooklyn College – City University of New York Brooklyn, NY, United States of America Email: swalker@brooklyn.cuny.edu Received: 11 December 2006 Accepted: 11 January 2007 © 2008 Walker. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Abstract Objectives – To examine the prevalence and strength of patron confidentiality policies and practices at academic and public libraries in Vermont; to examine variances in policies by type and size of library and by qualifications of library directors; and to examine the level of support among Vermont library directors for strengthening state law to clarify that personally identifiable information about library patrons is confidential and should only be released with a court order. Design – Survey (non-randomized) of 213 library directors. Setting – Academic and public libraries in Vermont (USA). Subjects – The initial audience was defined to be the directors of the 213 academic and public libraries in Vermont (as the persons to whom the survey was directed). Ultimately, results from 149 returned surveys were used as a basis for analysis. In addition, a written report of the information gathered was provided to the governing bodies of the Vermont Department of Libraries and the Vermont Library Evidence Based Library and Information Practice 2008, 3:3 69 Association. The author also anticipates that the results of the survey will be of general interest to academic and public librarians. Methods – A literature review of previous research on library policies and practices regarding patron confidentiality was completed. As well, the author examined the various state and federal laws and library association codes of ethics which address issues of the confidentiality of library patron data. A paper questionnaire was mailed on January 11, 2006, to directors of 188 municipal and incorporated public libraries (including 11 combined public and school libraries) and 25 college and university libraries, for a total of 213 libraries targeted. Cover letters explaining the purpose of the study and assuring respondents of the confidentiality of their responses were included in the mailing. The contact information came from a list supplied by the Vermont Department of Libraries. The questionnaire had previously been tested by five librarians from the aforementioned list, and slight changes had been made to clarify questions and instructions. Survey respondents were given an initial deadline of February 1, 2006, but responses were received and included through March 13, 2006. Various tactics were used to boost response rates, including electronic discussion list announcements, usage of pre-addressed stamped return envelopes, usage of personalized address information and a personal signature by the researcher on the cover letter, reminder postcards, and a coupon for free ice cream. One hundred and fifty-one questionnaires were returned. Two questionnaires were returned incomplete, and were removed from both the target population and the responses, leaving a total of 149 out of 211 possible responses and a response rate of 71%. For this response rate, all calculations were determined to have a margin of error of +/-4.4% with 95% confidence. The survey was constructed to address five research questions, as detailed below: 1) How many requests for personally identifiable patron information have libraries received in the past year, and from whom did those inquiries come? 2) How many libraries have a written confidentiality policy, and what are the characteristics of libraries and library directors that have policies? 3) How current are those policies? 4) How many policies prohibit release of patron information without a court order or other binding legal document (e.g., subpoena or search warrant)? How many policies call for attorney review of any such court orders or other binding legal documents? 5) Do librarians support the idea of strengthening the state law relating to confidentiality of library records? The questions were closed-ended in order to allow responses to be coded and easily analyzed, although space was provided for respondents to offer additional comments. Though a deadline of February 1, 2006 was set, responses were received as late as March 13, 2006. Person’s chi-square and Fisher’s exact test were used to measure statistical significance of differences among various respondent groups. Main Results – The survey respondents estimated that they received a minimum of 1,228 requests for patron information within the past year, and 46% of libraries had received at least one such request. Academic libraries, which made up 13% of the sample, received 11% of requests. These queries included requests for personally identifiable information about patron reading interests, use of library materials and use of services. It was not possible to accurately quantify requests received because many respondents indicated that their library had Evidence Based Library and Information Practice 2008, 3:3 70 received numerous requests, and did not supply specific numbers; in such cases, there was generally no indication that the library kept track of such requests. (and perhaps did not themselves track such requests). Of interest was the finding that because of gag order provisions in laws such as the PATRIOT Act, which prohibit librarians from reporting that such requests for information have even been made, it is likely that requests made by law enforcement agencies are under-reported. However, according to the responses to the survey, 90% of total inquiries came not from law enforcement officials, but from parents, guardians, spouses, or partners of library patrons or from teachers or professors of library patrons. In academic libraries, 60% of requests were received from teachers or professors of patrons. Only 1% of total requests were reported as coming from law enforcement officials. Additional groups making requests included college or university administrators (1% of total inquiries, exclusively in academic libraries); persons representing other government agencies but not law enforcement (0.3% of total inquiries, exclusively in public libraries); persons representing companies or non-governmental organizations (0.08%, one inquiry only, in a public library); and ‘others’ (6% of total inquiries, all but one of which were at public libraries). ‘Others’ included the president of a friends group, caregivers of elderly or disabled or homebound patrons, friends or acquaintances of patrons, book group participants, and people picking up reserved materials. Despite heightened awareness of confidentiality issues (due to the PATRIOT Act and related heavily publicized legal actions), only 48% of survey respondents indicated that their library had written policies or procedures detailing how requests for personally identifiable information should be addressed. Fifty-six per cent of academic libraries had such policies, as compared to 47% of public libraries. Thirty-five per cent of libraries without such policies indicated that they were working on developing these policies or planned to create them within a year. As well, 64% of libraries whose directors hold MLS degrees or the Vermont Department of Libraries’ Certificate of Public Librarianship had confidentiality policies, as compared to 38% of libraries where the director held neither qualification. Data were also analyzed according to the number of years of experience that the library director had, but results obtained from this comparative analysis were not statistically significant. Thirty-six per cent of libraries with two or fewer paid personnel had written confidentiality policies, as compared to 41% of libraries with 3-5 paid personnel, 59% of libraries with 6-10 paid personnel, and 95% of libraries with 11 or more paid personnel. The policies tend to be up-to-date: of libraries with confidentiality policies, 46% had reviewed or updated their policy within the past year, and 86% had done so within the past three years. Of libraries with confidentiality policies, 86% require a court order or other binding legal document, such as a subpoena or search warrant, before they will release patron information. There were no statistically significant differences in this matter when data were analyzed according to library type, number of paid personnel, or years of experience of the library director or the number of paid library personnel, but there were differences when data were analyzed according to whether the library director held an MLS degree or Vermont Department of Libraries’ Certificate of Public Librarianship, or neither qualification. In libraries where the director held one of the abovementioned qualifications, 93% required presentation of a binding legal document before permitting release of patron information; in libraries Evidence Based Library and Information Practice 2008, 3:3 71 where the director held neither qualification, this fell to 58%. Fifty-two per cent of libraries with confidentiality policies required that an attorney review the binding legal documents requesting patron information before such information could be released. Libraries which were more likely to include such a provision in their policies included libraries in which the director held either an MLS degree or the Vermont Department of Libraries’ Certificate of Public Librarianship (62%, versus only 8% of libraries where the director holds neither qualification), and libraries with larger numbers of paid personnel (20% of libraries with 2 or fewer personnel had provisions requiring attorney review of binding legal documents, versus 79% of libraries with 3-5 paid personnel, 50% of libraries with 6-10 paid personnel, and 61% of libraries with 11 or more paid personnel). Finally, 74% of respondents supported strengthening Vermont law to enhance safeguards to confidentiality of library records. Twenty-five per cent stated that they would need more information before making a decision, and 1% stated that they would not support such changes. There was a statistically significant difference by library type: 83% of academic library directors supported strengthening the law as described above, as compared to 73% of public library directors. The article also suggested that education and support should be directed at those libraries which have no confidentiality policy. Conclusion – In 2005, nearly half of Vermont academic and public library directors received at least one request for personally identifiable information about patrons’ use of library resources or services, and in total, at least 1,228 such requests were received. It is difficult to be sure about either the number or prevalence of such requests, as some laws, such as the PATRIOT Act, prohibit libraries from revealing that such inquiries have been made. However, only 1% of reported requests for such information were made by law enforcement agencies; it is far more common for requests to come from parents, guardians, spouses, partners, teachers, or professors of library patrons. Despite the number and prevalence of requests received, and despite heightened awareness of patron confidentiality issues due to recent publicity on the topic of legal action taken by libraries as a result of requests for information made under the PATRIOT Act, fewer than half of the libraries studied have written policies or procedures addressing how such requests should be handled. Of libraries that do have such policies, a large majority (86%) require presentation of a binding legal document before they will release patron information, and over half (52%) require attorney review of the binding legal document before they will comply with the request. A large majority (74%) of Vermont library directors also support strengthening state law to protect confidentiality of library records. Commentary The survey response rate was quite strong, at 71%. The author notes that “a high response rate is important in minimizing non-response bias in the survey results. Non-response bias occurs when the opinions of the people who respond are significantly different from the opinions of those who do not. … [Several] research methods textbooks suggest that researchers should strive to achieve response rates of at least 50%, 60%, or 75%.” (462). By these criteria, the author has rather successfully dealt with non-response bias; the response rate is under 75% but exceeds other stated acceptable rates. The sample size, however, is small and localized, and it would be very useful to repeat the survey in other states or Evidence Based Library and Information Practice 2008, 3:3 72 even nationally, and then to compare results. Another possible area for further research is how requests for patron information are handled in libraries where no confidentiality policy exists. It would also be interesting to further explore the correlation that was found between the likelihood of a library having a written confidentiality policy and the library director holding an MLS degree. The author speculates that this correlation could have something to do with the fact that MLS programs may have included information about the importance of confidentiality, or alternatively that libraries which have gone to the trouble of codifying their practices in this area may be more inclined to hire directors with MLS degrees. The survey did not explore this, and it could present an additional avenue for further research. Likely understatement of the total number of requests for patron information due to the usage by respondents of words like “many” or “countless times” rather than specific numbers is of some concern. It would be preferable, of course, to have precise numbers. It is unclear whether library directors were asked if they formally keep track of exact numbers of requests for patron information. There is also possible inherent (and unavoidable) inaccuracy in numbers of requests reported as coming from law enforcement agencies; under the federal PATRIOT Act, such requests cannot be reported. Nonetheless, the research presented in this article is likely to be valuable to a number of audiences. The Vermont Department of Libraries and the Vermont Library Association can use the information gathered in designing programs to better serve their constituents. Other library associations and state or federal library organizations can use this information in a similar fashion. The information presented (including both the overview of existing research and relevant federal and state legislation, and the summary of this research) is also likely to be useful to library directors or other librarians charged with examining patron confidentiality practices or developing patron confidentiality policies at their libraries. It should be noted that while this study is a US-based study, and perhaps has specific resonance for US-based libraries due to the PATRIOT Act, the article will be of interest to libraries in other regions. Libraries around the world deal with confidentiality issues in one way or another.