4.1 Library Profile

4.1.1 Geographical Distribution and Type of Depository Library

The geographical distribution of questionnaire respondents compared to the geographical distribution of existing depository libraries is revealed in Figure 1. The similarity between the two distributions shows that the 450 questionnaires analyzed in t his report represent an unbiased geographical sample of the total number of depository libraries. Of the existing 949 DSP libraries 84.5% are Canadian; 91.1% of questionnaire respondents are Canadian. The breakdown of Canadian libraries by type is as f ollows: academic, 31.3%; public, 51.5%; government, 7.6%, other, 9.6%. This breakdown is not applicable to foreign libraries. Canadian questionnaire respondents are distributed as follows: academic, 34.9%; public, 55.3%; government, 9 %; other, 1.7%. Again, the questionnaire sample reflects the DSP proportionately.

Fig. 1. Geographic Distribution of Questionnaire Respondents

Of those that responded to the questionnaire, public libraries are the most common (50.8%), followed by academic (38.1%) and government (9.6%) libraries. Other types make up only 1.5%. A healthy percentage (74.2) have a designated person in charge of their government documents collections.

Depository libraries serve heterogeneous populations, ranging from a minimum of 85 people to a maximum of over 45 million – a typical library serves a population of 15,000 (median figure). This wide disparity in size is reflected in the number of item s (including print materials, CD-ROMs, microforms, etc.) held in collections: as few as 100 at the low end to a maximum of 12 million.

Findings allow a picture of the typical depository library to emerge:

Only 107 (23.8%) libraries reported having a data librarian. A typical library has no data librarian (median is 0).

Government publications from all jurisdictions make up a significant proportion (more than 10%) of library collections in 149 of the 450 responding depositories: 33 (7.6%) report that over 30% of their materials consists of official documents, 30 (6.9%) indicate holdings of between 21% and 30%, and 86 (19.7%) say that 11% to 20% of their materials fall into this category. Holdings of less than 10% are reported by 288 libraries (65.9%). Thirteen respondents did not answer this question.

A more striking finding shows the prominence of Canadian federal publications in collections of official documents. Only 145 libraries (33%) report holdings of less than 40% in this category. The rest (67%) state that materials from Canadian federal sources constitute more than 40% of government document holdings. Answers were missing in 10 cases. Table 1 provides a breakdown.

Depositories were asked to provide information about the organization and shelf arrangement of their document collections, and the availability of special equipment and materials for visually-disabled patrons.

Separate collections are the practice in 28.8% of cases while documents are integrated with the main collection and the reference collection respectively in 10% and 4.7% of reporting libraries as shown in Figure 2. Since 1986 70 depositories (16.1%) have moved from separate groupings to integrated collections, a finding that may reflect a growing tendency to mainstream government publications. The majority (55%) provide a mixed arrangement, that is, some materials are shelved with the main collectio ns, some in a separate area. Other systems were reported by 1.3% of libraries, a figure that includes those who chose more than one category of organization.

Little consistency in shelf arrangement is evident from the findings of this study. (Figure 3). A few libraries, 21 (6.1%) and 24 (7%) respectively, employ the CODOC system and the Canadian classification scheme, while 65 (18.9%) use Statistics Canada catalogue numbers. Alphabetical arrangements by agency prevail in 68 cases (19.8%); the largest number, 166 (48.3%), employ some other shelf arrangement. Among those other arrangements imposed on documents collections (as revealed in comments by respon dents) are: the Dewey Decimal Classification, various adaptations of CODOC, alphabetical arrangements by subject, call numbers designed in-house, accession number order, and arrangement in vertical files.

Specialized products and equipment for use by visually disabled patrons are provided as follows:

Table 2 Special Equipment

4.1.3 Importance of Government Publications

Respondents were asked in Question 45 to rate the importance of government publications in their collections.

The pie chart in Figure 4 shows that government publications are seen to be "very important" or "essential" in about 68% of depositories. They are considered to be "somewhat important" in a little more than 29% of lib raries; only 3.4% of respondents said they were "not important." A chi-square test showed that public libraries are less likely to consider government publications to be important than are academic or government libraries (p<0.001). Results are displayed in Figure 5. Not surprisingly, official documents are most important to government depositories of which 69.77% said they were "essential" and none reported they were "not important." While public and academic libra ries were fairly close in their "not important" and "very important" responses, a much larger percentage of academic libraries said that these materials were "essential" to their organizations (40.48% vs. 20.26% for public libraries) and a larger percentage of public libraries stated that they were only "somewhat important" (37.89% vs. 22.02% for academic). Given that academic libraries are research oriented, have more users requiring access to statistical and o ther materials issued by governments, and that they often serve patrons who are referred to them by public libraries, it is perhaps not surprising that official publications are deemed by them to be more important.

It has been suggested that the significance of the government documents collection as perceived by the library is related to the population that the library serves. A Spearman Correlation Coefficient test was conducted between the population served an d the perceived significance of government documents. The test result showed that is no relationship between these two variables

4.2 Facilities for Electronic Access

4.2.1 Hardware Facilities

Online Public Access Catalogues (OPACs)

A large majority of depository libraries, 386 (87.1%), are equipped with online catalogues and of these 51.8% provide dial in access. Library catalogues are accessible from other libraries in 40.9% of cases and from the Internet in 35.1%.

Question nine elicited information about bibliographic control of government publications in the depositories, specifically the types of catalogue used (summarized in Table 3). The majority (257 or 62%) list their documents in the library’s main OPAC. While only 3 (0.72%) maintain a separate online catalogue and 2 (0.48%) use microfiche, larger numbers make use of card catalogues. Eighteen (4.3%) and fifteen (3.6%) use the main card catalogue or a separate card catalogue respectively to provide acce ss to their document collections. The most interesting finding here shows that as many as 103 (almost 25%) employ a mixed arrangement, that is, documents are, for the most part, listed in two places. the most prevalent combination consists of the main O PAC and a separate card catalogue. The next most common mix is made up of the main OPAC and the main card catalogue.

An attempt was made to discover how many personal computers are available for public use in the depositories compared to the number reserved for use by staff. This was a difficult question to answer. Libraries were asked to record the number of PC s in use "system wide" and while there seemed to be some confusion about this term results do provide information on how PCs are distributed. There is wide variability among respondents and the standard deviation is high. The typical library provides se ven PCs for exclusive staff use and six for use by library patrons. So, on average, the number of PCs allotted for staff members exceeds the number accessible to the public.

By dividing the number of PCs by the population served it can be seen that the typical library has .4 public service PCs per 1000 people served. Academic libraries have 2 PCs per 1000 population, government libraries have 1.7 and public libraries hav e only .15 PCs for every 1000 people they are mandated to serve. While these figures do not show how many PCs are needed per 1000 people served they should be kept in mind when one considers that, according to a recent Statistics Canada survey, only 7.4% of Canadian households access the Internet from home. The Statistics Canada data, part of the Household Facilities and Equipment Survey and collected for the first time in 1996, show that while 1.8 million households have a computer wit h a modem which links consumers to a wide range of services including the World Wide Web, only half of those households with modem-equipped computers have used the Internet. Deterrents to use may vary, says Statistics Canada; among them, the cost of Inte rnet services, long distance charges, and the mysteries of cyberspace. The survey refers to home access only and excludes access at work, school and libraries, etc. It is reasonable to surmise that as the amount of government information available on th e Internet increases greater demands will be made on depository libraries for access to it.

Libraries were asked to report the number of public service PCs available with various CPUs including 286, 386, 486, Pentium, and Macintosh . For purposes of data analysis, when no count was given for one or more categories (i.e. the line was left bl ank) it was assumed that the libraries had zero computers of that type. It is possible, however, that libraries do possess PCs of the types omitted but that the respondents do not know the exact number. Figures provided here, therefore, may be an underes timate and this should be kept in mind when the results are being interpreted. The distribution of these different types of CPU is shown in Figure 6. It is very encouraging to know that the majority of these computers are 486 and Pentium (69.4%) which are ideal for Internet access. Figure 7 shows the distribution of CPU types across the depositories.

The total number of public service PCs has been calculated for each library and the frequency distribution of this variable is presented in Fig. 8. This figure reveals a strikingly uneven distribution of computers throughout the depositories. The maj ority of libraries have very few computers while a very small percent of libraries have a large number of computers. The average number of public service PCs a library has is 17.6 (mean) while the number of public service PCs in a typical library is 6 (m edian). The minimum number is 0 while the maximum number is 300 with a standard deviation of 35.6 representing a huge variability among libraries. This fact presents a significant implication for the electronic access to official publications. If gover nment information is primarily in electronic form, the majority of libraries located at the low end of Figure 7 may not have a sufficient number of public service PCs to satisfy patron needs. Figure 9 shows the same pattern of distribution for the three types of libraries: academic, public, and government.

Fig. 9 Frequency Distribution of Number of Public Service PCs
for Different Types of Libraries

The amount of random access memory (RAM) in a computer is one indicator of the type of software it is capable of running. RAM size is shown in Figure 10. In the libraries surveyed 89% of public service computers had a least 4MB of RAM, adequate f or running Windows 3.1. Only 48%, however, are equipped with over 8MB of RAM, the practical minimum for Windows95. This shows that while it is safe to develop most, if not all, government information for the Windows environment, both Windows 3.1 and Win dows95 should be supported and Windows95 products will need to be carefully designed and tested to accommodate minimum installations of this version of Windows running in 8MB or less of RAM. The 11% that have less than 4MB of RAM will be restricted to DO S or extremely stripped down installations of Windows 3.1. Accommodating these systems will not be easy although use of text-based Internet browsers and DOS interfaces on text-only products may allow at least some information to be made available on them . As for Windows NT, only 5% of systems have more than 16MB of RAM, the minimum for NT. So far at least, development specifically for NT is probably not worthwhile but Windows 3.1 and Windows95 products should be tested for compatibility with NT.

Figure 11 shows that academic libraries enjoy a higher percentage of public access PCs with RAM in excess of eight megabytes.

Next, the libraries were queried about the number of public service CD-ROM drives mounted on stand alone PCs, and the number attached to a network (see Questions 18 and 19). Of the 450 respondents 394 answered this question. Once again there is wid e variability with the typical library reporting two CD-ROMs available on stand alone PCs, and only one accessible through a network. There are a total of 1803 standalone CD-ROM drives reported and 3307 networked CD-ROM drives reported. Therefore, about two-thirds (64.7%) of CD-ROMs in all these libraries are networked.

Figure 12 shows the percentage of CD-ROMs available by type of drive. The type of CD-ROM drive reflects the average speed of data access compared to the original CD-ROM drives introduced in the 1980s (shown in Fig. 12 as 1x drives). That is, a 2x dri ve is double the speed of the original, 4x is quadruple, and so on.

Clearly, 4x is the most common CD-ROM drive in the responding libraries. These drives were standard at the time of the survey and offered a good price/performance ratio, making them popular for use in network servers, which often have seven or more dr ives. While the standard is now 10x and increasing fast (16x and 20x drives are now available), 4x drives will remain common because libraries often cannot afford frequent upgrades. The 4x drives are sufficient for most textual products and for still gr aphics. It would seem reasonable that this is the type of content best suited to depository libraries. Full motion video with or without audio requires faster drives in order to be effective, especially when served on a network, and will need to be intr oduced more slowly with due notice to libraries which may or may not be able to support it.

Figure 13 shows that there is no significant discrepancy among the three types of libraries in following the pattern of having 2x and 4x drives dominant in the library.

Printers

The availability of printers is crucial in providing adequate access to electronic government publications as opposed to materials in print form. Therefore the number of public service printers in different categories (Dot matrix, Ink jet, Laser, Colo ur) has also been investigated. Figure 14 presents the results of this investigation. It should be noted that the most common type of public service printer is the dot matrix (representing about 41% of the total number of printers). The slow speed of dot matrix printers in producing graphics and Windows fonts should be kept in mind when making decisions regarding electronic access to government information.

Figure 15 looks at the printer types with figures broken down by type of library. In general, the pattern holds for all libraries, however it is interesting to note that public libraries are much closer to an even distribution of types of printer whil e government libraries have a large percentage of inkjet printers in service.

Libraries were asked to indicate the number of public service PCs equipped with various operating systems including MS-DOS, Windows 3.1, Windows NT, Windows 95, and others. Data are summarized in Fig. 16. It should be pointed out that the 20% of public service PCs equipped with MS-DOS only will have difficulty accessing information on the World Wide Web via a graphical interface, which means that sites using images, frames and multimedia active content will be inaccessible or difficult for them.

Figure 17 shows that there is a higher percentage of Windows 3.x systems in government libraries, but there is otherwise no significant discrepancy among the three types of libraries.

Ninety-eight percent of the depository libraries replied to the question asking whether or not they have a local area network (LAN) or a wide area network (WAN). Seventy-three percent of libraries reported "yes" and 27% reported "n o". For those libraries which do have networks, it is not clear whether they have a LAN or a WAN because the survey question did not ask for this breakdown. Of those who are connected to a network 71.7% allow public access to their LAN/WAN while in 28.3% of libraries access is restricted to library staff.

For those libraries not yet networked question 26 inquired about their LAN/WAN plans: 32.8% plan to install a network within one year, 17.2% within two years, 9.8% reported a plan later than 2 years, and 40.2% reported no plan.

The vast majority of depository libraries (89%) reported having access to the Internet. Those that do not yet have Internet access were asked if they have a plan for this. They responded as follows: 71.1% are planning for access within one yea r, 17.8% within two years, 8.9% have a plan of later than 2 years, and only 2.2% reported having no plan for Internet access.

For the most part Internet service is provided by a Commercial ISP (47.6%) or by a College/University (43.9%) (See Figure 18.). Freenet is the source for 5.1% of depositories (from Figure 19, it can be seen that these are all public libraries) while 3 .4% use some other means of access.

In examining comments of the respondents it was noted that many libraries gave "SOLS" or "SOLScape" as their Internet provider. SOLS is the Southern Ontario Library Service, an agency of the Ontario government responsible for coord inating many aspects of public library service in the province. One of SOLS’ current programs is to provide free dial-up Internet accounts to staff at public libraries. This program is the reason for the number of respondents giving SOLS as their provid er, but it is not a stable provider since funding is provided year-by-year. Alternative providers indicated through the comments include municipal governments, local consortia (public libraries partnering with municipal governments, boards of education, and other agencies), and nonprofit agencies.

There is a wide disparity among depositories in the number of public service PCs that provide Internet access; the typical library has two. Of the depositories that offer Internet access a large proportion (71.8%) offer Internet service to their patr ons; 28.2% allow staff use only.

The distribution of the percentage of PCs using the various types of Internet connection is shown in Figure 20.

Fig. 20 Type of Internet Connection

At first glance it appears that high bandwidth (i.e. fast) T1 and higher connections are quite common compared to the lower bandwidth (i.e. slower) ISDN (digital phone lines) and SLIP (modem over ordinary phone lines) connections, with 74% of computers having access to the faster type of connection. To look at this more closely the data were broken down by type of library and the distributions for each type of library were compared. The following graph (Figure 21) shows the results.

When the figures for PCs with T1 or higher connections are examined it is obvious that the largest number of PCs with access to such connections are found in academic libraries. A chi-square test was conducted to determine the statistical significance of this contrast. The result showed that there is a highly significant relationship (p<.001) between the type of library and the type of Internet connection. Public libraries are more likely to be using lower bandwidth connections such as ISDN or SL IP than academic libraries. The primary reason for this is that T1, T3, and other high bandwidth connections are expensive, putting them beyond the reach of all but the largest and wealthiest public libraries. By contrast, most universities now use the high bandwidth connections because of research needs and the availability of funding for universities to act as regional nodes on the Internet.

Public libraries make up the largest single sector of libraries among depositories (50.8% as opposed to 38.1% of academic libraries) and they serve larger populations (a median of 28,913) than academic libraries do (a median of 5596). Public libraries , however, generally have lower bandwidth connections to the Internet. In order to reach this important user community government information on the Internet will need to be manageable over these lower bandwidth connections. Academic libraries, while th ey do have access to the higher bandwidth connections of their institutions, must share this bandwidth with hundreds or even thousands of other users across campus, so that the available bandwidth at any given moment may not be more than what is available to public libraries. Therefore, managing the content to be accessible to lower bandwidth will benefit them as well.

There are a number of measures that can be taken in formatting content that will benefit users with low bandwidth connections. Text and still graphics should be preferred over audio and full motion video data which require a large amount of bandwidth and perform poorly even on low bandwidth ISDN connections. Active content (i.e. programs which run on the Internet) should be avoided or use server-based approaches (CGI scripts, Microsoft's Active Server Pages technology) instead of browser-based appr oaches (ActiveX and Java programs).

Table 4 displays collected data about Web browsers, while Figure 22 breaks this data down by type of library.

Findings show that the dominant Web browser among all depositories regardless of type is Netscape Navigator (84.8%), a fact which reflects the computer industry as a whole where Netscape Navigator is the dominant browser. Netscape's dominance does car ry some implications for publishing on the Internet. Netscape and Microsoft's Internet Explorer both have their own proprietary modifications to basic HTML. Avoidance of such proprietary extensions where possible will allow access to the maximum number of users. Where use of proprietary extensions is necessary, Netscape's are more suitable since Navigator is most widespread in depository libraries.

When active content is incorporated into government information Java, which is common to both Netscape and Microsoft, is a likely preference rather than ActiveX which is unique to Microsoft. Where possible, server side approaches to active content (su ch as CGI) may be employed as they are browser independent. Finally, since almost 10% of depository browsers are text based, government information on the Internet should be tested in both text and graphical browsers to ensure that these users are not de nied access to vital information. Change is rapid in this area and developments need to be watched closely; the balance between Microsoft's Internet Explorer and the older, more popular Netscape Navigator may shift over the next year or so.

In the matter of fees for service, a crucial factor in providing citizens with full access to information about their federal government's activities and publications, the study found that the vast majority of depositories equipped with an Internet connection (89.2%) do not charge their patrons for access. Printing services, however, which patrons are likely to want in combination with Internet use, are another matter. As table 5 shows, a substantial proportion (67.2%) impose charges on th eir patrons for printing. As reflected in Figure 23 it is clear that public and academic libraries are much more likely to charge for printing, a fact that will no doubt result in higher costs for patrons who wish to access information through the Inter net.

4.3 Support and Preparedness for Electronic Access

4.3.1 Current Use

Use of Canadian federal government documents in electronic vs. printed form is investigated by the following two questions: "What is the frequency of use of printed government publications in your library/resource centre?", and &q; uot;What is the frequency of use of electronic government publications in your library/resource centre?" Five categories of answers were provided: very low, low, moderate, high, and very high. The words "printed" and "electron ic" were bolded in the original questionnaire to highlight the difference between the two questions. The response rates were 98.2% to the first question and 82% to the second. The results are summarized in Figure 25.

A clear pattern of use of the two formats of government publication emerges in Figure 25. Overall, electronic formats are used much less frequently compared with print. A total of 45.3% of libraries reported very low use of electronic government in formation while only 7% libraries reported very low use of printed publications. The typical frequency of printed format use is "moderate" while the typical frequency of electronic format use is "very low". Remarks were solicited at the end of both questions by asking "Any comments on your response?" Many respondents provided observations; some were so eager to express their opinions that they extended their answers to the next page of the questionnaire. An examination o f the comments helped us understand the reasons for the relative low use of electronic documents. A recurrent theme is the lack of computer equipment and staff time to assist users. Absence of public awareness of electronic sources and lack of staff time to promote them were also cited as reasons for their low use. Some respondents expect an increase in the use of digital materials as staff and patrons becomes more computer literate. Many also note that they have just begun to acquire government inform ation in electronic form or that current collections are limited resulting in very low use compared to the use of traditional print materials.

To investigate the difficulties library users have in using electronic formats of government information, questions were asked about the type of help sought by patrons. Three types are listed: help with hardware operation, help with software, and help with understanding the content of government publications. Respondents were asked to rank these from the least frequent to the most frequent. Just under 50% of respondents answered this question. The ranking results are presented in Figure 26.

The figure shows the overall pattern of assistance sought: help with software is most frequently requested, followed by help with understanding the content of the documents themselves. Help with hardware is least frequently asked for. The written c omments received corroborate the conclusion that software help is most needed. Many respondents complained about the variety of different search engines available and difficulties in using them, the lack of user friendly interfaces, the unfamiliarity of users with the software, or the general lack of computer literacy among users. It is understandable that help with hardware is relatively infrequently sought because many libraries have designated workstations for searching government information and u sers generally need not deal with hardware problems.

The pattern of staff preparedness is reflected in Figure 27 and shows, first of all, that depositories seem to be better able to assist patrons in using CD-ROMs than they are in providing help with Internet. This is not surprising given the fact t hat the Internet is a relative newcomer, especially in public libraries; it does indicate that more work in training and documentation is needed in this area. Libraries will be helped considerably if Internet and CD-ROM products are designed to provide a common "look and feel" so that staff and patrons can move easily from one to the other.

It is also worth noting, as Figure 27 shows that the mean number of staff able to conduct training sessions is much smaller than the mean number who can provide direct assistance to patrons. This will have a significant impact on the ability of deposi tories to offer in-house training for both staff and patrons. Finally, the numbers show that the mean number of staff requiring training is about the same as the mean number of staff currently able to assist patrons with electronic resources. This mean s that many depositories are short of the trained personnel needed to assist with electronic access and competent delivery of electronic government publications will only be possible with increased training assistance. Figure 28 shows that public librari es require more staff training, a finding that is confirmed by comments made by respondents. Many lament the absence of funding, the dearth of training programs, and the lack of time available for acquiring and passing on expertise in dealing with electr onic sources of government information.

As for the availability of suitable training in electronic resources (including financial means, number and quality of available courses, time for training, etc.) it was rated by a majority of respondents (51.7%) as "poor" or "very poor" and confirmed in the written comments. Available services were rated "satisfactory," "very good," or "excellent" by 34.8%, 11.8% and 1.6% respectively. See Figure 29.

When these data were broken down by type of library, a chi-square test showed a significant difference between public, academic, and government libraries (>0.05), as the graph in Figure 30 illustrates. About 62% of public libraries rated availabil ity of training as "poor" or "very poor," as opposed to around 41% of academic libraries and 35% of government libraries. Probable reasons for these discrepancies are the facts that academic libraries may have access to training opportunities on campus a nd governments may tend to have better resources available for employee education and training in a field which, after all, is their own. Since academic and government libraries were earlier in adopting electronic access technologies it is reasonable to expect them to have better training available; they will have had more time to develop in-house resources and partnerships with training agencies.

When asked if their governing bodies had set goals for the introduction of electronic resources, 60% of respondents answered "no." See Figure 31. Thus, a majority of depository libraries are introducing services in this area without a clear set of goals. In some cases this may mean that goals simply have not been formalized by the governing body (e.g. public library board), while in others it may mean that projects in electronic access are proceeding on an ad hoc basis.

A significant difference in the establishment of goals by governing bodies among library types is revealed when data are broken down and a chi-square test performed (p<0.05). Upwards of 50% of public libraries have established goals - the highest p ercentage, as can be seen in Figure 32 . This may be due to the fact that they are most directly accountable to taxpayers through formal boards that include elected public officials. Academic libraries, while they are accountable to their university o r college administrations, are not necessarily directly accountable to the institutions’ funding agencies. Similarly, government libraries are normally part of a larger department and are accountable to senior management rather than the department's gove rning body.

A striking absence of formal policies for the management of electronic materials is evident from the results displayed in Table 6. Libraries were asked about the existence of such guidelines in a range of areas including collection development, chil dren's access, acquisition/receiving, reference service, cataloguing, archiving, and mirroring.

This shortcoming may be due to some extent to the relative novelty of such materials and a lack of defined professional standards to which policy makers can refer (for example, MARC cataloguing standards for data files have been developed and are still evolving). Collection development fares best in this area with 22% of libraries having a written policy. Archiving and mirroring are the weakest at 2.4% and 0.9% respectively. This is probably because most libraries do not yet see these activities an important for them in the electronic age. Should electronic access to government information via the Internet become the norm libraries will need to be engaged in locally mirroring commonly accessed materials on their own networks. It is likely, however, that DSP will need to be involved in any broader mirroring of government information as this will have to be arranged with universities or other providers with sufficient disk space and bandwidth. Local partnerships such as the Halinet project in Halton Region, Ontario, could also provide mirroring and archiving of government materials for their members. In any case, the lack of formal written policies in these matters will certainly retard development of local archiving and mirroring facilities by lea ving development of such facilities on a purely ad hoc basis.

Where the availability of physical facilities (space, lighting, ease of use, etc.) for the operation of computers is concerned, Figure 33 shows that most respondents (about 75%) rate them as "satisfactory" or better. But again, a chi-square test done to compare different library types showed a highly significant difference (p<0.001) between public libraries' rating of physical facilities and those of academic and government libraries, Figure 34. Public libraries were more likely to rate facilitie s as being "poor" or worse (about 36%), while in academic and government libraries the percentage of ratings below "satisfactory," (i.e., "poor" or "very poor") is below 20%. The percentage of public libraries reporting "very good" facilities is around 1 5%, about half of either of the other categories.

This contrast is no doubt due to funding discrepancies and the nature of the institutions. Academic libraries are important to the functioning of colleges and universities and are therefore better supported and may also benefit from private beneficiar ies and foundation grants. Government libraries are usually not stand alone facilities but are part of larger civil service office facilities and benefit from general improvements to the larger facilities and government buying plans for capital items suc h as furnishings. Unionized and regulated government offices may also tend to have standards in place for ergonomics, access for the disabled, etc. Public libraries, by contrast, have to rely primarily on municipal and provincial tax bases for funding, putting them in competition for funds with such essential services as policing, fire fighting, and various kinds of infrastructure. This is a competition that is difficult for public libraries to win since they are often viewed more as recreational facil ities rather than essential services. Lack of funds will inevitably lead to inferior facilities needed to support the new technologies.

These findings and conclusions were confirmed by the answers to the question "How would you rate the adequacy of financial support for electronic access available to your library/resource centre?" Results are displayed in Figures 35 and 36. Most d epositories (about 60%) rated their funding "satisfactory" or better. As with adequacy of physical facilities, however, there were important differences among types of libraries. A chi-square test showed a highly significant result (p<0.001) - public libraries (50% in fact) are more likely to rate their funding "poor" or "very poor." The graph in Figure 36 shows almost a sliding scale with about 36% of academic libraries rating funding as "poor" or "very poor" and about 17% of government libraries p roviding these ratings. Because of the clientele served by academic and government libraries, and the need to provide up-to-date information to researchers, there has been a stronger incentive for these libraries to pursue electronic access, hence the gr eater availability of funding. In addition, these libraries can frequently capitalize on broader institutional developments such as campus and government resources to reduce their own funding requirements.

Despite these discouraging findings, at least in regard to public libraries, it is heartening to note that depositories report an interest on the part of their governing bodies in increasing access to electronic information. Figure 37 shows that 90% o f respondents recorded an interest level of at least "moderate" and at least 61% indicated an interest level of at least "high." In this case a chi-square test found no significant difference in interest levels among the various library types. While thi s news does not automatically translate into support for the necessary capital and operating budgets, it does show that there is a base of interest which can be built upon.

Depositories that have not developed detailed plans for future acquisitions and installation of the new technologies were asked what they perceived as the reasons for this. The graph in Figure 38 makes clear that the two main reasons are the cost of h ardware and lack of staff time to develop in-house expertise, both of which rated over 20%. Since a good basic business computer generally costs about $2,000.00 to $2,500.00 at any time, in addition to the costs of network hardware such as interface card s, hubs, and routers, the former is a significant issue. In a large public or university library system with multiple branches or departments costs can multiply rapidly. New technologies such as network computers may help stabilize or even slightly redu ce these costs, but in many cases the savings on desktop computers will have to be sunk into better servers and higher bandwidth networks to make systems function effectively. It should also be noted that libraries can ill afford the cost of network admi nistrators. And some respondents report reductions in their materials budgets to support the purchase of hardware and software.

In regard to staff time available for development, libraries of all types have been subjected to restructuring (and not always efficient restructuring) over the past decade or so. In some cases staff members have had to take on duties that were former ly assigned to two or more persons. Until library staff are able to find time to better cope with new and more demanding tasks it will continue to be difficult to allow for attendance at courses, workshops, etc. Even organizations that enjoy the luxury of dedicated systems personnel may find their staff members spending much of their time maintaining and supporting existing systems with little time left for learning how to deal with new developments and working up plans for inevitable technological chan ge.

As noted above, the Canadian federal government is moving ahead in the provision of its publications in electronic form. What will the impact of this intended shift be on citizens’ access to government information? Will the change promote or impe de accessibility? To investigate this issue, respondents were asked whether use would increase, decrease, or remain unchanged if federal documents were available primarily in electronic form. This question calls for some speculation; there is re ally no objective way to predict the future on this issue. To reach a better understanding of the respondents’ answers comments were again solicited at the end of the question "What are your reasons for the perceived change?".

About eighty-nine percent of the depositories answered this question. Interestingly, among those who responded, answers were evenly divided among the three categories: 34.3% respondents think the use would increase, 34.3% said it would decrease, and 31.3% believed that it would remain unchanged. The main reasons for the perceived change can be summarized as follows:

The main reasons provided are: users will be able to access more documents than they have been able to in the past; electronic information is more current than printed materials; increased off site use (e.g. in branch libraries); availability of 2 4 hours access; easy keyword searching; increased access without the necessity of having more shelf space (this would alleviate the current lack of space for printed documents). Some respondents also pointed out that "increase" is contingent up on the availability of computers.

The main reasons provided are: insufficient computer hardware and software; increased reliance by users on staff for assistance; no staff available evenings and weekends; patron preferences for printed format (especially older users); alienation of those who do not have the time or inclination to learn how to use the software before accessing the content of electronic sources; difficulty in printing and/or downloading ; difficulty in browsing.

The main reasons provided are: different types of media are suitable for different types of material; use may go up as users become more computer literate; users will accept whatever format is made available to them; change depends on budget and cost of electronic service.

Because public, academic, and government libraries serve very different populations representing a large segment of society, comparisons among them were made to determine if they differ in this perceived change. Table 7 provides the cross tabulation o f the two variables with each cell showing the number of respondents and the percentage in that category. Although the three types of library have a somewhat different split in the perceived changes, a chi-square test revealed no significant difference ( p=0.09) among the three types of library on this issue.

At present printed documents are much more frequently used in the depository libraries than are their electronic counterparts. Print materials are seen to be user friendly and are used for research purposes and popular reading alike; improved bibl iographic control would further increase their use.

Relatively low use of electronic sources is attributed to insufficient and/or outdated computer equipment and inadequate staff support in providing assistance to patrons and in promoting the new technologies. Librarians themselves recognize the necess ity of raising the levels of staff expertise if users are to be assured continued access to official publications as the transition to a more electronic DSP proceeds.

The most frequent requests for help in using electronic formats of government information are those dealing with software. Electronic formats present formidable barriers to patrons who are unfamiliar with computers; a plethora of formats and access protocols contributes to user confusion and reluctance to access digital sources. Library staff must spend more time in coaching, training and encouraging their patrons.

Respondents to the study are divided (with an almost even split among the three categories) on their estimate of use of government information should it be provided primarily in electronic form, a direction in which the Canadian federal government se ems certain to move. There is no relationship between the type of library and the perceived change.

Public libraries and academic libraries are proportionally represented in all the three groups of perceived change. Some believe that electronic use will increase because electronic searching is easier and faster and allows more access points, better content coverage, and off site searching. Others predict a decrease mainly because of lack of equipment, shortage of staff time, and patrons’ preference for printed formats. Those who perceive use to remain unchanged express uncertainty on the issue and point to the complexities of the matter; many factors are involved such as availability of funding, cost of service etc.

Respondents were asked about their expectations regarding electronic resources. "By the year 2000 what do you expect will be your library/resource centre’s most common medium for access to government information?" They were requested to ra nk four possible formats: CD-ROM/Disk, Internet, Microform, and Paper. Results are displayed in Figure 39. The Internet had the highest percentage of first rankings, perhaps because of extensive media coverage and its growing popularity. Libraria ns are generally aware of the advantages of the Internet as a publishing medium and are interesting in using it. Paper was a strong second in percentage of number one and number three rankings and third in number two rankings. Some libraries, especiall y the smaller ones, are moving slowly in developing electronic services and may not anticipate completion of their automation and networking projects by the turn of the century which, after all, is less than three years away. CD-ROM and disk had a relati vely low percentage of number one rankings but had the highest percentage of second and third rankings showing that these formats are still expected to have a place, though not as significant a place as the Internet in the delivery of documents. Micro form fared poorly, having the highest percentage of number four rankings and the lowest percentage of number one, two, and three rankings.

The division on these issues represents a significant degree of uncertainty among depositories on the future of government information use when it is primarily in electronic form. These findings are similar to the argument made by Bernadine Hoduski th at government information in multiple formats will best serve the needs of diverse users and provide equal opportunity for access. (Abbott-Hoduski, p. 252) It is therefore recommended that the government proceed with caution on the intended move into t he electronic format of its publications. Both positive and negative effects of such a major move should be fully investigated before full implementation takes place.

Go to the following section

Return to the top of this page

Return to the table of contents