Managing and Assessing Internet Service Needs and Use of Under-Served Populations at Public

Managing and Assessing Internet Service Needs and Use of Under-served Populations at Public Libraries

Yunkeum Kim

University of Wisconsin-Madison, 600 N. Park St.Madison, WI53706Email:

As the range of public library services with new technology has diversified, defining, measuring, assessing and managing the quality of these electronic services has become a critical issue. Internet connections at public libraries have increased dramatically since 1990 due to the availability of key external funding sources. The major goal of those funding sources was to expand public access to computers and digital information, especially for the public of low-income or under-served populations; thus,there has been continuing increase inpublic library Internet services at the library, for example, the services at Internet workstations. Although there have been attempts to measure the performance of public library network services by library and information science researchers using network statisticsbased on “technical-driven” performances, there has been little research dealing with service quality and satisfaction of Internet service users at the public libraries based on “user-driven” performance evaluation. This paper describes the process of survey research using 386 Internet service users at four public libraries and discusses how public libraries need to approach those under-served populations’ Information needs and use at Internet workstations which can expand the possibility of new direction for future public library Internet services.

Introduction

Over the past ten years, public libraries in the United States have made significant changes, providing networked library services such as online library catalogs, online reference services, online databases, new electronic methods of document delivery, access and resource- sharing across libraries, digital libraries, etc. According to a national survey of public library outlet Internet connectivity statistics (Bertot and McClure, 1998) sponsored by the National Commission on Library and Information Science (NCLIS), public library Internet connections have increased dramatically from 20.9% of public libraries in 1994 to 72.3% in 1997. Based on a sample of 2,500 of the nation’s 15,718 public library outlets, public library Internet connections have continued to increase from 83.6% in 1998 to 95.7% in 2000 (Bertot and McClure, 2000). The updated statistics from the 2002 study (Bertot and McClure, 2002) showed that 98.7% of public library outlets have an Internet connection and especially public libraries outlets with more than 40% poverty have reached 100% connectivity in 2002.

McClure, Ryan, and Bertot (2002) pointed out that many innovative public library network services could be developed due to the availability of key external funding sources such as the Library Services and Technology Act (LSTA), E-rate, the Bill and Melinda Gates Foundation, and other private, federal, state and local sources. The major purpose of these external funding sources is to provide access to technology through libraries for the public, especially for those who have low income, less education, and/or fewer abilities to access networked information services and resources at home. As one of the methods of providing equal access to technology-based services, public libraries have offered public access Internet workstations. However, as shown in public library network statistics (McClure, Bertot, & Zweizig, 1994; Bertot, McClure, & Zweizig, 1996; Bertot, McClure, & Fletcher, 1997; Bertot and McClure, 1998, 2000; Bertot, McClure, & Ryan, 2001, Bertot and McClure, 2002), the measure of effectiveness of public access Internet workstation has focused on physical (technical) Internet access at public libraries such as the number, usage, or maximum speed of public access Internet workstations.

However, since the statistical reports have focused on “technical quality” of public library Internet services (i.e. speed, number of workstations) from a library’s perspective rather than on “service needs” in terms of judgment attributes from a users’ perspective (user-based approach), they have not functioned as tools to measure the quality of public library Internet services from the users’ perspective. Consequently, there is a need for a study that seeks to provide a systematic framework for assessing public library Internet services based on users’ Internet service perceptions at public libraries.

This paper presents the findings of 386 survey results at four public libraries in Midwest region exploring Internet service needs and use of under-served populations.

Theoretical Background

There have been many studies that measured and evaluated “quality” of library services throughout the decades. Hernon and Dugan (2002) indicated that “quality” might be viewed from two different perspectives: “technical quality” (i.e. more mechanical and procedural aspects that ensure that services function effectively and efficiently) and “customer quality” (i.e., aspects related to user perceptions of service delivery). Traditionally, public library services were measured by focusing more on the “technical quality” of Hernon and Dugan’s taxonomy. For example, public libraries have several statistical tools to evaluate their service performances. The two basic tools most widely adopted by public libraries, Output Measures for Public Libraries (Van House and others, 1987) and Planning and Role Setting for Public Libraries (McClure et al., 1987) are reflected in two national public library statistical systems, the Federal-State Cooperative System (FSCS) and Public Library Data Services Statistical Report (PLDS). The Federal-State Cooperative System (FSCS), established in 1986, consists of nationwide public library statistics collected annually through the NationalCenter for Education Statistics (NCES) in U.S. Department of Education. The Public Library Data Services Statistical Report (PLDS), first published in 1987, also consists of public library statistics, but these are collected and disseminated annually through Public Library Association, a division of the American Library Association (Liu, 1999). However, since these public library statistical reports do not include electronic library services, there have been efforts by researchers to develop manuals to measure and evaluate public library networked services (Bertot, McClure and Ryan, 1999, 2000, 2001; Bertot and McClure, 1998; Bertot, McClure and Fletcher, 1997).

Bertot et al.’s manual and statistical reports (1997, 1999, 2000 and 2001) have contributed in many ways to give guidelines for public library directors to measure the performance of their networked services (i.e., guidelines that define and measure the concept of acceptable public access to the Internet via public libraries). Using the manuals by Bertot and others, libraries could use a core standard of national statistics and measures that describe public libraries’ performance on the Internet and Web-based services and resources. Even though what is being evaluated has changed from the traditional services, the manual focused on physical or functional outcomes of public library Internet services such as number of public library Internet workstations, number of public access Internet workstation uses, the maximum speed of public access Internet workstations, the technology infrastructure to support public Internet access workstations, number of virtual reference transactions, number of virtual visits to networked library resources, etc. However, the number of Internet workstations or maximum speed of Internet workstations is not always indicative of library users’ service experience. The manuals by Bertot and others include a basic user assessment of the library’s electronic resources and services using a questionnaire and focus group interviews. Bertot (2001) pointed out the need for the new service quality measures for libraries in the networked environment: “The quality measurement (in the manual) is limited to predominantly technology-generated logs, counts of equipment and/or equipment usage, and transaction data (e.g., reference). . . . Adopting a user-based approach to library network service quality assessment is critical.” Also Fitzsimmons and Fitzsimmons (p. 64, 1998) noted, “The success of technological innovations, particularly for the front office, depends on customer acceptance. The impact on customers is not always limited to a loss of personal attention. Customers also may need to learn new skills. The contribution of customers as active participants or co- producers in the service process must be considered when marketing changes in the service delivery system.”

However, until the early 1990s, the measurement of “service quality” in the LIS field was relatively new. For example, since the late 1970s, LIS researchers started to renew their attention to library users, such as how to translate users’ information needs and information seeking and library use behavior into managing library services. However, most of the research was narrowly focused on examining specific user groups’ information seeking or needs; the information search process results of both trained or untrained searchers (Cochrane & Markey, 1983; Saracevic and Kantor, 1988) to trying to identify the model of users’ information search process form cognitive perspectives (i.e., Kuhlthau’s Information Search Process, 1988; Belkin’s Anomalous States of Knowledge (ASK), 1980; Dervin’s Sense-making model 1977). Despite their contributions to users’ motivation, perceptions and attitudes studies in LIS field, the seeking and search behavior research did not deal with user-based criteria for the measurement of library service quality and the analytical tool to apply such criteria to the evaluation of a specific library or its specific services (Nitecki, 1995).

In a recent research study, The Impacts of the Internet on Public Library Use (D’Elia and Roger, 2000), sponsored by the Urban Library Council, a random telephone survey of 3,097 adults conducted during the spring of 2000 showed that 75.2% of Internet users also used the library and 60.3% of library users also used the Internet. Also they found that the use of the library and use of the Internet were inversely related to age. Library users were significantly younger than library nonusers and Internet users were significantly younger than Internet nonusers. They reported that use of the library was not significantly related to race, but the use of the Internet was (Hispanic (n = 350), 32.1%; Black (n = 357), 43.7%; American Indian (n = 21), 54.5%; White (n = 2,133), 58%; Asian (n = 112), 66.7%). However, as shown in the results, the data reported in the study has limitations due to its sample distribution of race population. Another interesting finding from this research is the respondents’ rating on Internet service characteristics. The Internet received higher rating for: ease of getting there, time to get there, availability (hours of access), range of resources, expectation of finding what is sought, ability to act immediately on the information obtained, up-to-datedness of information, fun, enjoyability of browsing, and ability to work alone (compared to working around other people in the library).

Bertot and McClure (2000) collected public library outlet Internet connectivity data by outlet metropolitan status and poverty level. The national survey netted over 200 electronic responses and 1,888 mailed responses, and showed that 83.6% of public library outlets have some type of Internet connection; 91.0% of urban, 88.1% of suburban and 78.4% of rural public libraries were connected to the Internet; 84.1% of library outlets serving populations of less than 20% poverty have an Internet connection, 80.9% of library outlets of 20-40% poverty and 79.9% of library outlets of greater than 40% poverty have an Internet connection. In their follow-up research Bertot and McClure (2001) reported that Internet connections had increased to 95.4% of libraries, and of those outlets that do provide public access Internet services, each has an average of 8.3 graphical workstations (rural libraries with 4.9 workstations as compared to an average of 17.3 in urban libraries).

However, relatively little empirical research has been conducted to date to systematically and extensively uncover the underlying relationship between demographic characteristics of Internet service users and past experience of Internet use and the level of expertise and users’ service satisfaction or the intention to refer the library’s Internet service to others.

Methodology

To identify the relationships between users’ demographic variables (gender, education, race, income, occupation status) and Internet use variables (Internet use frequency, past use experience, Internet use Instructional class taken, level of Internet use expertise) and Internet user satisfaction or Internet service referral intention, following hypothesis were tested:

Hypothesis 1. There will be positive relationships between demographic characteristics of Internet service users such as Gender, Age, Race, Education level, Income, and Occupation status (independent variables) and users’ Internet service satisfaction (dependent variable).

Hypothesis 2. There will be positive relationships between Internet use variables such as Years of Use, Class Taken, and Level of Expertise (independent variables) and users’Internet service satisfaction (dependent variable).

This study was conducted at four public libraries in upper Midwest. The selection of the sample sites was based on the following considerations: 1) library size and number of Internet workstations in adult section, 2)service area metropolitan status, and 3) service area ethnic diversity. Libraries were selected from medium to large public libraries. Next, public libraries a minimum of six Internet workstations were selected. Two urban libraries located in metropolitan downtown areas and two suburban public libraries were selected for the study.

The data were collected for a week at each library site for totaling 40 hour per week per library. The users of public library Internet workstations from four participating public libraries were approached upon finishing their use of the Internet workstation by asking if they would participate in filling out a survey questionnaire. In order to avoid a bias of selecting any particular set of users, all users at Internet workstations were approached. During a month long period of survey collection, 389 survey responses were collected. Of 389 survey participants who responded to this study, 386 completed all parts of the survey and appropriately answered the survey questionnaire. From observation from each site, approximately ¼ of Internet users at the library were regular daily users. Among 386 survey responses, 115 responses were collected from Library A, 121 from Library B, 73 from Library C, and 77 from library D. All the data collected from the four libraries were coded in MS Excel file and later converted to SPSS/PC 11.0 data file for further statistical analysis.

Data Analysis

Demographic Characteristics of Survey Participants

Using descriptive statistics, survey participants’ demographic characteristics of gender, education, age, race, total annual household income, and occupational status were analyzed and are summarized in Table 1. In terms of gender, there were more male survey respondents (66.6%) than female (32.1%), although the actual male and female users were about 50/50 from a month-long observation during the survey period at the four research sites. In education level, 61.6% of respondents were either college graduates (31.3%) or had some college (30.3%), followed by graduate school (13.7%) and high school (13.0%), etc. In age, about half of the respondents were at the age level of 25-44 with 25-34 (28.0%) and 35-44 (22.3%), followed

Table 1. Demographic Characteristics of Survey Participants (Aggregate Data)

by 18-24 (18.7%) and 45-54 (18.4%), etc. In terms of race, the majority (72.3%) of the survey participants were White/Caucasian, followed by Black/African American (14.0%). About two thirds (65.3%) of the respondents had annual household income of less than $40,000, and more than 60% of the participants were employed either full time (44.0%) or part-time (17.1%), while 20.5% were unemployed.

Since the four research sites represent different communities, further comparisons of the four public library participants’ demographic characteristics were made, summarized in Table 2. In all of the four library sites, more male Internet users (63.3 – 73.0%) participated in the survey than female participants (26.1 – 35.5%). Among the four libraries, the suburban Library C reported higher education levels (67.1% with college graduate or higher degrees) of Internet service users than those at other libraries. For the age level of Internet service users, Library A showed a lower percentage of Internet service users who were 55 and over (5.2%) than the other three libraries, while more than half of the users at Library C and Library D were between 18 and 34. Library A showed a higher percentage of Black/African American Internet users (38.4%), while the other three libraries (Library B, C, and D) showed the range of 78.1% - 88.4% White/Caucasian Internet users.

Even though there was a big Hispanic population for Library B and it was expected to get more survey responses from this group (from the observation about 20% of Internet users were Hispanic), only three Hispanic users

(2.5%) agreed to participate in the survey mainly due to their language barrier.

The average Internet users’ household income for city central public libraries, Library A and Library D, was much less than the suburban public libraries, Library B or Library C. For example, approximately 40% of Library A Internet service users and 50% of Library D users’ annual income level was less than $20,000. Library D showed the highest percentage of unemployed Internet users (approximately 30%). From many of the survey participants who marked on the “other (please specify)” for the occupational status question, “self employed” was the most frequently specified answer.

Past Internet User Experience and Level of Expertise

The descriptive statistics in Table 3 show that 57.5% of Internet users used Internet workstations more than once a week. Many survey participants also wrote in the survey or mentioned to the researcher during the survey that they were daily users of the library’s Internet workstations. Over 90% of Internet users had the experience of using the Internet for more than a year, of which 45.3% had 1-5 years of experience and 36.5% 6-10 years.

About three quarters of Internet users answered they have Internet access outside of the library. Among the Internet users, 72% did not take any instructional courses related to Internet use. Regarding the levels of Internet use expertise, only 11.9% answered that they belong to the beginner level, while 63.5% of them considered their

Table 2. Comparison of Demographic Characteristics of Four Research Site Survey Participants