The link between the use of self-service holds and patron privacy issues has been identified by several authors. Bowers has squarely addressed the privacy concerns raised by self-service holds, arguing that the manner in which self-service holds have been implemented in many libraries violates both the privacy provisions of the American Library Association Code of Ethics as well as many state statutes governing the privacy of library circulation records. Further, Bowers urges librarians to advocate for library policies that would protect user privacy in this area.¹⁸

Chmara echoes Bowers’ concerns about materials placed on self-service hold shelves, stating:

The library must ensure that its hold policy does not encroach upon the privacy of patrons. For example, if the library places hold materials on an open shelf for patrons to peruse the materials to find their own hold requests, the library should consider using a wrapper that would cover the title of the work. Otherwise, other patrons will be able to discover what materials are being requested by particular patrons.”¹⁹

Researchers have identified the direct connection between the display of hold material on publicly accessible hold shelves and invasion of patron privacy, yet this practice still exists.

Numerous studies have dealt with aspects of library patron privacy, although none have dealt specifically with privacy issues in relation to material placed on self-service hold shelves. These studies include several that survey the historical, legal, and policy aspects of the privacy of patron records.²⁰ Several authors have also written comprehensive guides to privacy and confidentiality issues that affect libraries and their patrons.²¹ In addition, Magi reviewed scholarship from a variety of disciplines to identify common themes about why privacy is important to both individuals and to society.²²

In a further study, Magi surveyed public and academic libraries in Vermont to determine whether the libraries had policies regarding confidentiality of library records and to assess the attitudes of library directors regarding strengthening state laws governing library privacy. Only 48% of the library directors who responded had formal privacy policies at their libraries, although 35% of libraries without privacy policies had plans to develop them in the near future. The study found, however, that the policies that had been adopted, in most cases, provided a high level of protection of patron privacy. In addition, although only 48% of libraries had formal policies, 74% of library directors supported the idea of amending state statutes to provide a greater degree of protection of patron privacy.²³

Studies of libraries outside the United States have also revealed a lack of attention to policies addressing patron privacy. Schepman, Koren, Horvat, Kurtovic, and Grgic found that 75% of the libraries in the Netherlands and 72% of the libraries in Croatia did not have written privacy policies. They also found, that in many cases, librarians were not aware of the privacy implications of new technologies that they were considering for adoption.²⁴ A study of academic and special libraries in the United Kingdom revealed that an even lower percentage of these libraries (14%) had written privacy policies.²⁵ In another study examining library confidentiality and privacy in a global context, Shachaf surveyed codes of ethics of library associations in multiple countries and found coverage of these topics in most codes (27 out of 31 studied).²⁶

Sturges et al. surveyed attitudes of library users towards confidentiality of information held by libraries. They found a high level of confidence among users regarding the security of data held by libraries, but that, “there exists a genuine gap between the privacy protection that the users believe they can expect from the library, and the preparedness of librarians for providing this protection.”²⁷ Similarly, Sutlieff and Chelin found a high level of confidence among university students in the United Kingdom in their libraries’ protection of their personal data which was not necessarily justified by actual library practices.²⁸

Magi surveyed directors of public and academic libraries to identify the policies and practices that were in place at their libraries to protect patron privacy. She found a high awareness of privacy issues among directors, although a minority of libraries still engaged in practices, such as contacting patrons by postcard or leaving phone messages that could have led to a disclosure of patron circulation information to others besides the patron. She also found that library directors had a high level of confidence in their own ability to observe their libraries’ confidentiality policies, but a lower level of confidence in the ability of their staff to do so.²⁹ Magi’s study is one of the few that looks at actual library practices that affect patron privacy.

This review of the literature shows that the issues of patron privacy and confidentiality of library records are of great concern to librarians and researchers. The question remains as to how this concern translates into actual library practices. A second question concerns the strength of librarians’ commitment to patron privacy when pitted against other values, such as service to patrons. This study explores both of these questions and examines actual library practices regarding self-service hold shelves.

The target population includes every public and academic library in the State of Michigan. Data was collected using a Google™ survey emailed in April 2011 to the directors of every public and academic library in the state. (See the Appendix A for the survey instrument.) Library directors were identified through the use of the Interactive Library Directory of the Library of Michigan.³⁰ The survey was distributed in electronic form by sending a link to the survey to the e-mail accounts of the directors of the target population as listed in the directory. In libraries with multiple branches, directors of every branch were surveyed. There are 756 libraries in the target population; 609 are public libraries or branches of public libraries, and 146 are academic libraries.

The target population excluded special libraries because the researchers determined that special libraries are sufficiently different from public and academic libraries that they should not be included in this study. The target population was limited to libraries in the state of Michigan because it was determined that a broader geographic study was not feasible due to time and other constraints. In addition, the researchers determined that there were a sufficient number of libraries in the population selected to obtain meaningful results.

In an effort to boost response rate, a follow-up email with a link to the survey was sent to all non-respondents on May 3, 2011. This email was sent from the personal email account of one of the researchers and it encouraged non-respondents to participate in the survey.

Survey responses were downloaded into a Google Docs™ spreadsheet and also exported to the software program SPSS for statistical analysis. Each of two researchers independently categorized the responses to every open-ended question that was recorded and an inter-rater reliability analysis using the Kappa statistic was performed to determine consistency among the two raters.

Of the target population of 756 library directors, 70 emails were returned as undeliverable and one survey was inadvertently sent to a special library. These numbers were subtracted from the total making the number of surveys actually sent and received 685. Of these, 559 were sent to directors of public libraries and 126 to directors of academic libraries. Two hundred and eighty-four of the 685 library directors who received surveys responded, for an overall response rate of 41.4%. With this response rate, estimated percentages reported below for the total sample have a margin of error of +/- 5.93% with 95% confidence.

Table 1 describes the respondents by type and size of library. Out of the 284 respondents, 232 (81.7%) are public libraries and 52 (18.3%) are academic libraries. These percentages correspond to the percentages of public and academic libraries in the total population of Michigan libraries surveyed, 81.6% and 18.4%, respectively.

Table 1 shows further that 60 (21.1%) of the 284 directors of Michigan libraries who responded to the survey report that their libraries use self-service holds. Of the 60 libraries that use self-service holds, 56 (93.3%) are public libraries and four (6.7%) are academic libraries. Based on this sample, between 15% and 27% of Michigan libraries use self-service holds +/- 5.93% with 95% confidence.

As noted in table 6, eleven libraries (3.9% of total respondents) are considering implementing self-service holds in the future. Nine of these are public libraries and two are academic libraries.

Among public libraries that use self-service holds, 27 (48.2%) have been doing so for four or more years; 15 (26.8%) for two to three years; 10 (17.9%) for one to two years; and four (7.1%) for less than one year. Three (75%) academic libraries have been using self-service holds for four or more years and one (25%) academic library for less than a year.

Two-hundred and two (71.1%) of the librarians who responded to the survey have either a master’s or doctorate degree in library and information science. In public libraries, the decision to implement self-service holds was made by library employees including directors and librarians with input from nonlibrarian staff. Seventeen public libraries (30.4%) report that their boards were also involved in the decision.

The survey asked libraries with self-service holds to rank each of four options: space issues at the circulation desk; freeing up staff time for other work; cost savings (labor or other costs); and patron convenience as “very important,” “moderately important,” or “not important” to their decisions to implement self-service holds. Libraries were also given the option of answering “no opinion” to each of the four options. Table 2 summarizes libraries’ responses to this question, with results displayed in order of the importance attached to each item.

Respondents also had the option of providing other reasons for why they implemented self-service holds. Only five respondents identified reasons other than those listed in table 2. Four reported that their purchase of a new integrated library management system was a reason that they decided to use self-service holds. One reported that, once an addition to the library had been completed, the library had space to accommodate bookshelves for self-service holds and decided to implement them.

As shown in table 3, libraries use various methods of identifying patrons on items being held on publicly accessible hold shelves, the most common being by full first and last names of patrons. Twenty-seven libraries (45%) use this method. Twenty-four (40%) identify the patron by full last name and first initial(s). Seven (11.7%) use either a partial last name or a first name only. In total, 97.7% of libraries identify patrons with all or part of the patrons’ names. Only two libraries (3.3%) identify the patron without a full or partial name; these libraries use library account numbers only. None of the libraries report using a nickname or alias chosen by the patron.

Table 4 shows the methods libraries use to conceal the titles of the items on self-service hold shelves. Twenty-two (37.3%) wrap the item to cover the title. Sixteen (27.1%) place the item, either spine down or facing back on the shelf. Eleven libraries (18.6%) use a combination of wrapping and placing the spine down or facing back on the shelf. Nine (15.3%) do nothing to cover the title and one library (1.7%) places hold items in cloth bags.

Table 5 shows the method used to conceal the title of the item placed on hold, if any, for libraries that identify patrons by all or part of the patrons’ names. Out of the fifty-seven libraries that identify patrons by all or a portion of their names, 50 (87.7%) use some method of concealing the titles of items that are placed on self-service hold. Seven libraries (12.3%) that identify the patron by full first and last names or by full last name and one or two initials do not use any method of concealing the title of hold items. The two libraries that identify the patron only by library account number do not take any steps to conceal the title of the book and are not included in this table.

Libraries that do not use self-service holds were asked whether they had ever considered using self-service holds. Forty-eight libraries reported that they had considered using self-service holds and 172 libraries reported that they had not considered using self-service holds. Eighty-four of these 220 libraries responded to a follow up question asking them to explain why they had decided against using self-service holds. This suggests that some of the 172 libraries that had answered “no” to the first part of the question had, in fact, considered self-service holds. From the 84 responses, the researchers identified 103 separate reasons that libraries did not implement self-service holds. These reasons were evaluated and assigned to the categories reflected in table 6. The inter-rater reliability was found to be Kappa = 0.89. According to Landis and Koch this is almost perfect agreement.³¹

The most common reason given for not implementing self-service holds was lack of space, followed by concerns for privacy, and lack of self-service checkout facilities.

Space concerns included both allocation of space within the library as well as overall lack of space. Some libraries reported they would have to give up valuable shelf space for circulating materials. Several said they simply did not have enough room. One respondent said the library would “have to gut a bathroom” to make room for them.

Libraries that do not use self-service holds expressed concerns about privacy that were tied to specific privacy laws or library confidentiality policies, while other libraries identified specific ways in which the use of self-service hold shelves could lead to a breach of confidentiality. One respondent said “it [self-service holds] is a violation of the Library Privacy Act” while another said the library was located in a “very small town [and] everyone knows just about everything about everyone. If they want to know what someone is reading they would just as soon look on the hold shelf as ask them, and [they] occasionally do.” Another small public library director noted that “items … could still be perused by other library patrons.” A director of a university library stated that “placing a hold book with a person’s name attached in a public area would violate confidentiality protection.” A director of a public library with a service population of between 25,000 and 99,000 indicated that the library actually discontinued the use of self-service holds because “people picked up the wrong items that belonged to other patrons and tried to see who was requesting what.” The library stated, “We considered it a privacy issue and decided to go back to having holds requested at the Circulation Desk.”

At least one public library indicated that it did not implement self-service holds due to its board’s concerns about privacy. The respondent stated “In the past self-service holds were not implemented due to patron privacy. We have had complaints about privacy in the past and the Board expressed concern”

The connection between self-service holds and self-service checkout was also noted by a number of respondents. “Even if we did have space, without self checkout it didn’t seem a good option… . It only seems to make sense if your library has self-check—otherwise what point is there in getting an item off a public hold shelf, only to have to stand in line at the circ desk to check it out.”

Security concerns included both concerns regarding theft of library materials and concerns about patrons taking items that were on hold for a different patron. One library commented, “Many holds items come from other library’s (sic) and we are not comfortable letting them get too far from our sight while we’re waiting for the patron who requested them to come pick them up.” Another cited the “potential for somebody to swipe somebody else’s high demand item.”

One of the themes that emerged from the 14 responses classified as “other” was the concern over losing personal contact with library patrons. A public library with a service population of fewer than 10,000 does not use self-service holds because “our patrons really like the one on one contact with staff.” Another small public library said “We are never too busy to wait on patrons…this is a good opportunity to give them personal attention.” Three libraries said they did not have the staff to service such a system. Other reasons given for non-implementation included a collection that does not circulate and one that is entirely on-line. Two libraries answered that self-service holds would cause confusion for their patrons.

Libraries that use self-service holds were asked whether they had changed or modified any of the procedures they use to prepare materials for placement on the shelves after they began using self-service holds. Of 26 total responses to this question, 23 described general changes to processing the holds, including changes in how the library generates hold slips and where self-checkout stations are located. Only three responses described changes that were implemented to better protect patron privacy. One respondent said that the library “gradually went to covering more of the book to help with privacy concerns, and another said that the library was “more secure in hiding titles.” A third library said that it was placing items backwards on the shelf to protect privacy and discourage browsers.

The final survey question asked all respondents whether they had any additional comments about their library’s approach to and practices regarding self-service holds. Eleven of the 24 libraries that offered comments emphasized patron and staff satisfaction with self-service holds. According to one library, “[patrons] honestly don’t care a smidgen if there is a privacy law, they just want convenience.”

One respondent said that its check-out lines are shorter. Another said that, as a result of offering this service, the number of holds in the library has increased greatly and consequently so have the library’s circulation statistics. Three libraries mentioned increased theft of items from their hold shelves, particularly video games.

Nine libraries mentioned privacy issues. Several noted that they locate their hold shelves near the circulation desk to discourage browsing and to allow for monitoring by staff. Three described incidents where patron privacy was actually compromised by another patron.