Featured Articles

Characteristics and Trends in the Public Library Data Service Report

Access to timely and relevant data is an essential component for public libraries. The Public Library Data Service (PLDS) survey has been providing such data to the library community since 1988. PLDS is an annual survey conducted on behalf of the Public Library Association (PLA) where public libraries from the United States and Canada have provided information on finances, library resources, annual use figures, technology, and additional yearly special categories. The results of this survey are published by PLA and available as both an online subscription database and yearly print report.¹ Results available directly in that report are not duplicated in the analysis available here. Rather, this report presents trends across the data and across population groups.

Summary

• Per $1,000 of expenditures, libraries tended to experience slightly less output between the 2008 and 2009 reports.


• Libraries serving a smaller population tended to have higher per capita measures in terms of library visits, holdings, and circulation; however, libraries serving a larger population tended to have higher per capita measures in collection turnover. No specific pattern was seen for registrations as a percentage of population or in-library use of materials.


• A small drop was seen in the prevalence of Web content such as basic library information, online catalogs, community links, and so on, but some wireless features and virtual reference services increased.


• Seven-year trends showed statistically significant increases in income and expenditures per capita as well as per capita measures including library visits, holdings, circulation, and collection turnover.

Research Method

This study used data from the 2003–09 PLDS reports. In 2009, 852 of 1,645 libraries responded to an invitation to participate in the PLDS survey, a response rate of 51.8 percent. Of the responding libraries, 829 completed more than 20 percent of the survey, were validated following data checking, and were included in the final report. For this article, all valid answers were
analyzed, including libraries that did not answer more than 20 percent of the survey. For each of the continuous variables, the following statistics were obtained: number of libraries reporting, minimum, maximum, mean, and standard deviation. For each
of the categorical variables, frequencies of responses and percentage of libraries responding were calculated. Most statistical calculations were performed using Microsoft Excel. Statistics were verified and additional statistical tests were conducted using
SPSS v.172 to examine longitudinal data and study the effect of population group on financial and output measures.

Overall Service

The responding libraries in the 2009 PLDS Report ranged in size from those serving 1,150 to 4,045,873 people in their legal  service areas. Figure 1 presents the distribution of participating libraries by population of legal service areas represented. In 2008, 703 PLDS libraries served 88,268,921 registered patrons from a population of 176,322,718 in the United States and Canada. Library holdings ranged from 7,800 to 16,141,095 items among 783 reporting libraries. PLDS libraries on average also
circulated items 1,857,097 times (n=773), performed 277,991 reference transactions (n=731), had 1,123,752 library visits (n=734), provided programs to 55,664 patrons (n=764), and received 18,880 interlibrary loan materials from other libraries (n=764) while providing 18,698 materials to other libraries (n=754). Table 1 provides a statistical summary of selected
descriptive characteristics of participating PLDS 2009 public libraries.

Operating Finances

Income and expenditure measures are a primaryconcern of many libraries. Averaged over all respondents, ibrary income was $48.61 per capita in the egal service area while library expenses were $45.65 er capita in the legal service area. With the exception f the spike seen for the population range of 25,000–49,999 in the 2008 survey, income and expenditure patterns by population group for the 2009 survey followed the same pattern as in the 2008 survey (see figures 2 and 3). Income and expenditures per person were largest for the libraries serving the smallest population and smallest for libraries serving the largest population with the second largest income and expenditures for the population group of 25,000–49,999. Fluctuations by population were relatively identical between income and expenditure. Income per capita (F_(8,380)=86.43,p<0.001) and
expenditure per capita (F_(8,380)=81.344,p<0.001) varied significantly by population group.

Library Outputs in Relation to Expenditure

Often there is a need to validate that expenditure has yielded a desired result within a public library. Table 2 summarizes various library outputs as a function of expenditure. In 2008 per $1,000 spent, libraries on average received 149 library visits, circulated 235 materials, saw 9 patrons attend programs, fielded 29 reference transactions, measured 48 uses of materials within the library, and registered 17 new patrons. Compared to 2008 survey values, all numbers were down slightly on average, meaning that per $1,000 of expenditures, libraries were experiencing slightly less output measures on average,  although the differences tested were not statistically significant when looking at paired sample t-tests between the 2008 and 2009 surveys. This reduction continues a trend from the previous year however.

Selected Library Services by Population

The population of legal service area of individual libraries influences many relevant output measures. Several of these measures are visualized in figures 4–9.

For most of the trends analyzed, a clear spike can be seen in the visualizations. For example, when viewing library visits per capita (figure 4), libraries serving a population under 5,000 see more than twice as many (2.7 times) visits per capita as any other library group, while libraries serving the largest populations see the smallest average visits per capita. Furthermore, libraries in the three groups serving populations between 5,000 and 49,999 saw nearly 1.75 times as many visits per capita as the libraries serving the largest population group. Visits per capita varied significantly by population group (F_(8,366) = 97.991, p<0.001). All library group averages and the overall average with the single exception of the 5,000–9,999 population group were increased from last year. The overall average library visits per capita for all reporting libraries was 6.17 (n=734).

Although not as skewed as library visits, in-library use of materials per capita also saw the largest number for libraries serving a
population under 5,000; however, the smallest numbers were now seen for libraries in the middle ranges of 50,000–99,999 and 25,000–49,999 (figure 5). Overall, this number was statistically unchanged from last year and tended to go up for larger libraries and down for smaller libraries, but not always. The average in-library use of materials per capita for all libraries reporting was 1.94 (n=313).

The only trend showing an almost linear aspect is a comparison of the average percent of population registered with the library to the population of the legal service area (figure 6). Like many of the trends though, libraries serving a population under 5,000
showed the largest value while libraries serving populations of 1,000,000 and above showed the smallest value with a generally decreasing value as the population increased. If outliers are not excluded (as in the figure and the previous analysis so as to allow comparison to last year when they were also not excluded), all group averages except the 5,000–9,999 population group were increased from last year. The overall average for all reporting libraries was 63.68 percent (n=703).

Average holdings per capita (figure 7) shows the same pattern as library visits per capita. Libraries serving populations under 5,000 had average holdings more than 3.25 times those of other libraries. Generally, as the population served grew, the average holdings per capita decreased, with the largest population group having the smallest value. Libraries serving a population of 5,000–9,999 had the second largest value of average holding per capita at 5.62, and were 2.75 times as high a value as the lowest value of 2.04 for libraries serving populations over 1,000,000. Holdings varied significantly by population
group (F_(8,366) = 37.166, p<0.001). The average holdings per capita for all reporting libraries was 3.74 items (n=782).

Showing a similar but not as skewed pattern, average circulation per capita saw the largest value for the smallest population group and the smallest value for the largest population group (figure 8). No clear trend existed for other population groups; however, the values varied significantly overall by population group (F_(8,366) =76.069, p<0.001). The overall average for all reporting libraries was 10.01 items circulated per capita (n=733).

Collection turnover allows a library to see how often items within the collection are being used by dividing the total circulation by the total holdings. Unfortunately, the current PLDS survey does not distinguish types of circulation, so this is not an entirely accurate portrayal, but it can give a general picture of collection usage. When analyzing collection turnover (figure 9), the graph
skews in the opposite direction as other figures. Larger population libraries have a larger average collection turnover than smaller population libraries, with the largest collection turnover for libraries serving a population of 500,000 to 999,999 and the smallest collection turnover for libraries serving a population under 5,000 (a 3.5 times difference). The overall average for all reporting libraries was 2.98 items circulated per item held (n=748).

Technology

Technology is a major component in the modern public library that has become almost indespensible in numerous aspects. The majority of PLDS libraries make use of numerous technology tools and services, as shown in table 3.

Surprisingly, the number of libraries that reported having a website dropped since last year’s report, from 98.17 to 92.43 percent. There was also a slight drop in those reporting “yes” to the presence of particular features on their websites including online catalogues, online reference services, online databases, and personalized patron accounts, although all were present the majority of the time. The most common website content included programming information/events calendars and basic library information such as hours of operation and location. Other website content such as children/young adult pages, community links, library staff created content, and library Friends pages were also present the majority of the time.

When Internet access was filtered, it was generally performed by the library staff. For 57.81 percent of reporting libraries, Internet access was filtered by staff, while only 10.34 percent of libraries allowed patron control of Internet filtering. These numbers are not exclusive in that a library may allow filter control by both staff and patrons.

Libraries continue to recognize the importance of expanding technology services in some areas. While wireless Internet access remained about the same at 88.22 percent, libraries with wireless Internet access extending outside the library increased to
46.03 percent. To meet the needs of an increasingly online society, libraries continue to increase reference services in virtual modalities. While chat services remained about the same (31.37 percent), instant messaging (19.47 percent) and e-mail/Web
forms (62.14 percent) both saw increase in usage over last year for providing virtual reference services to patrons.

Access to locally produced digitized content remained essentially the same as last year at 43.63 percent.

Seven-Year Trends

For the 2010 report, the PLDS survey will be using a different research pool, encompassing all U.S. and many Canadian public libraries. The database restructure will make longitudinal comparison more difficult. Therefore, several longitudinal trends were
analyzed this year for the data currently available for libraries that have consistently completed the PLDS survey. Although more than 800 libraries completed the survey for any given year, 403 libraries completed the PLDS survey each and every year from
2003–09, allowing comparison of data by paired sample t-tests.

First, because income and expense per capita are important measures for most libraries, these were compared (figure 10). A statistically significant increase in both income per capita and expenditures per capita is seen almost every year with a probability of at most 0.01. The only non-significant finding was between FY2006 and FY2007 for income per capita (t₍₄₀₂₎=0.560,p = 0.576). Between FY2007 and FY2008, both income per capita (t₍₃₉₃₎=3.394,p<0.001) and expenditures per capita
(t₍₃₉₁₎ = 7.008, p<0.001) showed a statistically significant increase as for all other years except as previously noted.

Over the course of the seven years studied, income per capita changed by 31.45 percent. During the same period, expenditures
per capita increased by a slightly less 30.44 percent. Both of these values were higher than the calculated inflation rate of 22.93 percent for June 2001 through June 2008.³

Looking at a few key output trends, one also sees increases over the years (figure 11). Although small, holdings per capita increased over the years. Statistically, the differences between adjacent years has not been significant for holdings per capita, but the change between years 2002 and 2008 overall is statistically significant (t₍₃₉₃₎ = 3.397, p<0.001).

Changes in library visits per capita were more significant, showing that each year more people are visiting public libraries. Except for a somewhat linear trend from FY2005 to FY2007, other years saw increases in average library visits per capita,
although not statistically significant on a per year basis. For FY2007 compared to FY2008, the increase was 0.15 visitors on average per capita, the only year showing a statistically significant change from the previous year (t(374)=3.280,p<0.001). The overall change from 2002 to 2008 is also significant though (t₍₃₄₄₎ = 6.025, p<0.001).

Circulation per capita shows the most dramatic changes on average per year during this time period. The average circulation per capita rose 0.61 items between FY2004 and FY2005 and then dropped 0.38 items between FY2005 and FY2006; however, neither of these changes was significant by a paired samples t-test. Still, the changes between FY2007 and FY2008 (t₍₃₉₁₎ = 7.241, p<0.001) as well as the overall change between years 2002 and 2008 (t₍₃₉₁₎ = 7.544,p<0.001) were both significant showing that circulation per capita is increasing on average for public libraries during this time frame.

Collection turnover combines the circulation and holdings measures and results in a curve shaped somewhere in between the other two. This measure has seen a statistically significant increase both between the years 2007 to 2008 (t₍₃₈₄₎ = 4.33, p<0.001) and for the overall change from 2002 to 2008 (t₍₃₈₄₎ = 7.088,p<0.001). The average collection turnover has risen 0.42 items circulated per item held during this six-year time period.

References

1. Available from www.ala.org/ala/mgrps/divs/pla/plapublications/pldsstatreport/index.cfm (accessed Apr. 21, 2010).


2. For more information, see www.spss.com (accessed Apr. 21, 2010).


3. Inflation Data, http://inflationdata.com, and Bureau of Labor Statistics, www.bls.gov/CPI (both accessed Apr. 21, 2010).