Losing Our Drawers

How well do these systems work? Overall, librarians told us they're satisfied with their automation systems, that they plan to continue purchasing major upgrades, and that the basic features offered by their systems are doing the job. In fact, most of those surveyed said they would purchase the same brand again.

What's interesting is that the ratings show that vendors, as a group, tend to do the same things well--or not so well. For example, all systems got fairly high grades for their technical support, while all were rated less highly for the quality of their technical manuals. (See chart below.)

RATING THE AUTOMATION SYSTEMS
These are the scores given to various automation systems by SLJ's survey respondents.  The top scores in each category are in bold.*
. 

*1=Highest Rating  5=Lowest Rating

Still, some vendors are clear leaders in the school-library market. Follett and Winnebago, for instance, have by far the largest installed base. (See chart, below.)

At  the same time, systems created with public and academic libraries in mind generally received lower scores. The exception was in the area of initial training, where some of these vendors scored substantially higher than their school-oriented brethren.

Most vendors got high marks for integrity--how trustworthy they've been in terms of living up to advertised claims--and for technical support. As one Winnebago user said: "We seldom have problems, and when we do, a toll-free call to technical support quickly resolves it."

Where many vendors fell down were in the areas of initial training and written documentation--those dense technical tomes that are supposed to help you operate your system. Also, few librarians rated their systems as "very easy to maintain," although most (69 percent) considered them manageable. This comment about automation systems from one respondent was typical: "We can't imagine reverting back to a card catalog, but be prepared to be overwhelmed with new tasks."

Following is a breakdown of the survey findings. Since the survey was conducted, there have been mergers in the automation market, notably Sagebrush and Winnebago. The survey reflects the state of the market prior to recent mergers. (See "About the Survey," below.)

Who's automated? Ninety percent of schools have automated circulation systems, and 85 percent have automated catalogs. More than half of those that have not yet automated plan to do so within two years.

As might be expected, more secondary schools are automated than elementary schools. Smaller schools with smaller collections and budgets are less likely to have automated circulation or catalog systems. About 30 percent of schools with collections of less than 5,000 items are not automated. And school librarians are either very loyal or still new to their systems: 84 percent said their current vendor is the only one they've used.

Purchasing decisions Fifty-five percent of school librarians spent six months to a year deciding which automation system to purchase. The average decision time was 10 months, with small schools spending less time than larger schools. Where do librarians get their information? The main source is word-of-mouth from colleagues. Other important sources are sales presentations (29 percent) and published reviews (12 percent). Eleven percent said automation decisions are made at the district level.

Most purchased systems Not surprisingly, two traditionally school-oriented companies, Follett and Winnebago, claim the largest share of the school automation market. (See chart, p. 70.) Follett dominates, with an impressive 44.2 percent of the market, while Winnebago has 22.8 percent. The numbers drop off precipitously from there, with Sagebrush the third highest at 7.3 percent. This vendor breakdown was similar for all types of schools: elementary, middle, and high schools; small schools and large; public and private. Of systems traditionally marketed to public or academic libraries, only DRA (2.5 percent) and Dynix (4.3 percent) have any significant penetration in the school-library market. These systems are slightly more common in schools with larger populations, budgets, or collections.

Vendor migration When asked what led them to switch to new vendors, 36 percent of our respondents said it was because all the schools in the district were moving to the same automation system. Other reasons included new features and services (30 percent), changing school technology (17 percent), or system discontinuation (17 percent). Several librarians commented that they consider new vendors when making major platform changes, such as moving from DOS to Windows, installing wide area networks, or migrating to new server platforms.

Web-based systems Twelve percent of respondents said they use a Web-based catalog. But only 1.5 percent said their systems were dependent on Internet access. That means most school libraries with Web-based catalogs must also maintain "proprietary clients," that is, catalog software installed on each computer.

Sticking to paper? Of those school librarians who don't have automated systems, 66 percent listed cost as the biggest obstacle. A few cited lack of administrative support or lack of technical skills. Finally, a small group, more for circulation (nine percent) than for catalogs (four percent), think that paper is simply easier and cheaper.

Are We Satisfied?Overall, librarians are pleased with their automation systems. Where there are weak points, they tend to exist in all systems. (See RATING THE AUTOMATION SYSTEMS chart above.)

User friendliness Let's face it: end user satisfaction is the chief goal of our having automated catalogs. In this area, Sagebrush Athena got the highest rating, with an average score of 1.47 on a scale of 1 to 5, with 1 being the highest. Fifty percent of respondents described this system as very friendly, which means, "Users find what they need with little outside assistance. The system teaches itself." Next came Follett, with an average score of 1.62, and Winnebago, with 1.74.

Efficiency and usability Here we asked librarians if their circulation system was efficient and user-friendly for staff. Most vendors received high ratings in this area, with Sagebrush Athena and Follett tied for the lead and Winnebago slightly behind. Fifty-seven percent of Sagebrush users and 55 percent of Follett users described these systems as "very efficient." This was defined as making it "easy to check [materials] in and out, print overdues, and check on the status of materials."

Vendor integrity In the crucial area of integrity, or trustworthiness, all vendors did quite well. Gateway received the highest rating, with 77 percent of users describing the company as "very trustworthy," meaning "all ads, documentation, and vendor reps portray the system capabilities honestly." Follett came in second, with Winnebago, and then Sagebrush Athena, close behind.

Technical support Ratings here were relatively strong as a whole. Gateway, Winnebago, and Sagebrush Athena are rated highest on this key dimension. Sixty-seven percent of Gateway users described the company's technical support as "excellent."

Training There's definitely room for vendor improvement in this area, where scores were lower overall (the average score was 2.9; only 15 percent of respondents gave their vendor a rating of excellent). Gateway was head-and-shoulders above the pack, with a score of 1.78, while school-library leaders like Follett and Winnebago received only 2.9 and 2.83 respectively. SIRS, DRA, and Dynix (Epixtech) also scored slightly better than others in this area. What we found is that while people at large schools usually receive initial training, 29 percent of small school staff said they had no training at all. Based on our own experience and respondents' comments, it's clear that many vendors require schools to pay an additional fee for training. A few respondents recommended that librarians make sure their schools set aside money for training, but clearly some schools do not budget for this expense. Said one respondent: "You can't learn it all from a manual, and besides, who has time to read a manual?"

Documentation This is another area where respondents were not enthusiastic. Only 21 percent described their system manuals as "comprehensive and easy to read." Sagebrush Athena and Follett scored highest here.

Maintenance Only 20 percent of users described their system as "very easy to maintain," meaning "even a technological novice could do it." The highest scores here went to Winnebago and Sagebrush Athena. About a third of the librarians who used these systems said they were very easy to maintain. Most librarians rated their system "manageable for someone who has basic computer skills." It was clear from comments that maintaining sophisticated automation systems stretches school staff, and that however good a vendor's technical support, there is also a need for district technology support. As one user put it, the vendor "can no more tell me if their program has somehow been corrupted by power outages or other incidents than they can fly me to Mars." Another person noted, "District technical support should be in on the ground floor when considering automation systems."

Downtime Systems in general appear to be quite reliable, with an average of only two to three days of downtime over the course of a year. Follett had the highest reliability rating, with more than 70 percent of its users saying the system is down one day or less per year. The systems that report more downtime are those more likely to run on wide area networks with multiple sites, such as DRA, Gateway, and Dynix.

Would you buy this system again? Most school librarians said yes. Only 14 percent said no. The most frequent reason given for not purchasing an automation system again was dissatisfaction with the system reports and with features of the catalog interface. The companies that had the most respondents saying they'd purchase their systems again were Follett (an average of 95.6 percent for all systems), Sagebrush Athena (92 percent), Alexandria (89 percent), and Winnebago (an average 88.5 percent for both systems). Those that had the fewest people saying they would purchase their systems again were Dynix (70 percent), DRA (67 percent), CASPR (52 percent), and Chancery (47 percent).

A Gateway note Gateway stood out in the ratings as the highest scorer on vendor integrity, training, and technical support. But this system, based on a server (AS400) typically used for district office functions, fell down badly in terms of user friendliness for staff and student. (See chart, p. 4.) This user comment was typical: "The software is 'green screen' and unfriendly. A graphical user interface is available at considerable additional cost."

Emerging Features
Like any technology, library automation systems are constantly evolving. So we asked survey respondents which newer features they use, how they like them, and, also, which they would look for in making their next automation purchase. (See feature chart, above) Spell check was clearly the most popular emerging feature. Some others we looked at include:

Authority control (Twenty-five percent use it; 31 percent of those who have it love it.) This is the ability of a system to check the consistency of subject headings. Fifty percent of respondents said they would look for authority control in judging a new system.

Links to Internet sites in catalog (Eighteen percent use it; 40 percent of those love it.) Though many respondents don't currently have this feature, some 53 percent said they would consider it in purchasing a new system.

13- to 15-digit barcodes (Seventeen percent use them; 39 percent of those love them.) Many librarians are familiar with the older barcodes, which were just a few digits long. But the newer barcodes--used by state and regional networks--are much longer, and provider greater information, such as a code for the location of a particular library. Being aware of these new barcodes is important, because if you link up with other libraries or systems, you don't want to have to re-barcode your materials.

Web-based catalogs (Twelve percent have them; 46 percent of those love them.) We were surprised that such a small number of libraries use Web-based catalogs, since many librarians consider them to be the hottest thing. But 52 percent of respondents said they would consider this feature in buying a new system.

Top Ten Tips

Our survey respondents had lots of advice for those in the market for a school-library automation system. Here are the most frequently repeated words of wisdom.

1.  Do your homework. Read the professional literature. Talk to colleagues. Look at a variety of systems before making a decision. One good resource is ALA's Library Technology Reports.
2.  Know what you need and want. Said one respondent: "Establish a set of criteria, look at a system, revise the criteria."
3.  Buy for the future. Try to predict the features you will need three years from now. Buying something that just meets minimum requirements will lead to frustration.
4. Be compatible. Try to buy a system that's compatible with other schools in the district and, if possible, with the local public library.
5.  Get training. This often requires extra fees or traveling. Include these costs in your automation budget.
6.  Get hands-on experience. Visit other schools that use the system and try it yourself. See vendor demonstrations. Try demo systems.
7.  Miscellaneous things to consider. Does vendor provide universal barcodes? Can MARC data can be imported and exported? Are reports produced useful to your library? Does the system offer district interoperability?
8.  Research tech support. Find out exactly what technical support services are offered. Also, include your local tech people in the decision-making process. You'll need them in the future.
9.  Weed your collection. Do this before you begin a conversion. As one librarian said: "Be ruthless."
10. Consider the end-user. Try to make his or her experience as easy as possible.

The crystal ball
What does the future hold for automation systems? In our opinion, increased networking is key. While schools continue to live in their own little worlds--with catalogs that don't go beyond their buildings--the rest of the library world is becoming a web of interconnecting systems. Both of our states, for instance--Minnesota and North Dakota--are developing statewide library information networks. These networks will enable users to search many if not all library systems in the state.

But participating in these networks means schools must join consortia, as some in Minnesota have done, or adopt union catalogs with Z39.50 servers, as schools in North Dakota have done. That's why we were disappointed to learn that only 30 percent of our colleagues would consider participating in union catalog projects beyond their districts. (See chart, p. 71.) Still, those involved in such arrangements were enthusiastic: "Consider joining an automation system that is shared with other schools in your region," said one respondent. "I have excellent technical support and interlibrary loan opportunities."

But many respondents were not even interested in integrating within their schools. Only 30 percent of respondents would, when shopping for a new system, consider compatibility with the new School Interoperability Framework. These are software infrastructure standards designed to help schools share data across systems, for instance, between library and student record-keeping software (SLJ, August 1999, p. 20).

In shopping for automation systems, it's crucial to learn as much as you can about emerging features. Things like Z39.50 and 15-digit barcodes may seem like frills now, but as networking expands, they will become even more important to know about and plan for. For instance, the increasing reliance on Z39.50 capabilities and interlibrary loan makes the integrity of your MARC records crucial, so that other systems can read your records and you can read theirs. And again, because networking is the wave of the future, several librarians surveyed recommended that you make sure the system you select is compatible with other schools and public libraries in your region.

Now is the time to tell vendors what we need to make sure our libraries keep up with changes that are afoot. In fact, many large-system vendors, such as DRA and Dynix, are looking seriously at the school market. School librarians should make their voices heard so that the features we need are added to systems purchased by consortia. It's possible that the lower survey scores for DRA and Dynix may reflect the frustrations of people who did not personally select a system for a school setting but who must use it as part of a consortium.

Overall, our survey shows that school librarians feel good about their automation systems. That's good news for vendors, and it's good news for librarians looking to purchase or upgrade systems. If you know your school's needs, can raise adequate funding for both a system and the training to run it, and can have access to adequate technical support, chances are you'll find an automation system that will not only ease your workload but will give your students better access to library resources. c

About the Survey
The survey instrument was designed by Jane Prestebak and Konnie Wightman. School Library Journal's research department mailed the survey, tabulated the data, and compiled the findings.

The survey was mailed in March to a random, representative sample of 4,000 School Library Journal subscribers. In order to reach smaller school libraries, an additional sample of 500 librarians was used from a random list from Market Data Retrieval.

We received a total of 1,572 responses, for an overall response rate of 35 percent. Ninety-three percent of those surveyed were building-level media specialists, and seven percent were district-level media supervisors. Ninety percent of the sample came from public school libraries and 10 percent from private schools.

Forty-eight percent served elementary grades, 46 percent served middle/junior high, and 35 percent served high schools. The total is more than 100 percent due to multiple grades served by some of the schools.

Analysis was done using a statistical package (SPSS). Measures of central tendency (means) were produced for all systems' ratings.