Electronic databases in public libraries: issues of organisational adoption.

Based on research which evaluated content, usability, and use of electronic databases in public library settings, user adoptions are examined. This is done in terms of Everett Rogers' analysis of the diffusion of innovations. In this case, it was an organisation, the State Library of Victoria, which decided to adopt an innovation on behalf of the public libraries of Victoria. Nevertheless, Rogers' framework is useful for understanding why staff and library users were slower than might be expected to accept or adopt electronic databases for their information seeking. Training is the main way in which this situation could be changed

**********

Librarians in all sectors are playing a key role in the diffusion of electronic technologies. It is through libraries that many people use technologies such as cdroms, online databases or the internet for the first time. The attitudes of librarians to new technologies are therefore very important. For example, in the early stages of the introduction of a set of online databases into a public library, the extent to which librarians understand and promote the databases, and assist users, will be crucial to their acceptance. As Holland says, it is through persuading patrons to adopt information/networked technology that librarians act as change agents in the wider diffusion process. `The concepts of diffusion, change agency, and critical mass provide ... a structure and criteria to analyse how effective [librarians] are, and reasons why some projects succeed and others fail'. (1)

In relation to this concept of the librarian as change agent in the diffusion of innovation this article discusses a project which was undertaken in Victoria in 2000. The project evaluated the content, usability and use of four sets of electronic databases--World magazine bank (Ebsco Host); two Gale products, Health reference library (now Health and wellness resources center) and Custom database; and Electric library (at that stage, an Infosentials product). These electronic databases were provided on a trial basis in the public libraries of Victoria by the State Library of Victoria (SLV), through the Gulliver program. (2) Members of Enterprise Information Resources Group (EIRG)/ Information and Telecommunications Needs Research Group (ITNR) at Monash University were involved in the evaluation of the pilot program at four Victorian public libraries (Bayside and Port Phillip in metropolitan areas; Corangamite, a regional library service and Casey-Cardinia in a semi rural area). The chief investigators, Associate Professor Frada Burstein and Dr Kirsty Williamson received an ARC Spirt grant in partnership with SLV/Vicnet, to undertake the research. (3) SLV/Vicnet also contributed funding to the project. As well as trialing the databases with individual users, focus groups were conducted with staff in each of the public libraries. With both groups of participants, interesting and rich data emerged from the fieldwork. Although an important focus of the project was the evaluation of the databases from a user perspective, this paper is largely concerned with user adoption of the databases for finding information in a library context.

Diffusion of innovations

Everett Rogers is a famous name in the literature about the diffusion of innovations. In 1962 he published a landmark study Diffusion of innovation which is now in its fourth edition. (4) In the latest edition he analyses and critiques the 4,000 publications available at that time, compared with the 405 publications which provided the supporting evidence for the first edition in 1962. The theoretical framework he began to develop in the original edition has been revised and updated in each subsequent edition. Rogers describes five attributes of innovation

* `Relative advantage is the degree to which an innovation is perceived as better than the idea it supersedes. This might include economic profitability or social prestige, for example.' (5) As Callahan says, in terms of computer applications in libraries, the `relative advantages' of speed and access are readily apparent (6)

* `Compatibility is the degree to which an innovation is perceived as consistent with the existing values, past experiences, and needs of potential adopters.' This might include compatibility with sociocultural values and beliefs, with previously introduced ideas, or with client needs for innovation (7)

* `Complexity is the degree to which an innovation is perceived as relatively difficult to understand and use.' Rogers generalises that `The complexity of an innovation, as perceived by members of a social system, is negatively related to its adoption'. (8) He postulates that the perceived complexity of the home computer in the early 1980s `... was an important negative force in their rate of adoption'

* `Trialability is the degree to which an innovation may be experimented with on a limited basis.' (10) The idea is that if people have a chance to try out an innovation without serious financial commitment then they will be more likely to adopt. This particularly applies to early adopters. Later adopters are surrounded by peers who provide their experience and encouragement

* `Observability is the degree to which the results of an innovation are visible to others.' (11) The idea here is that some innovations are more observable and can more easily be described to others

According to Rogers, the way in which these attributes are perceived by those who are the beneficiaries of the innovation determines both the speed of adaptation and the degree to which a specific innovation is adopted, modified or ignored. In diffusion of innovation theory, as he expounds it, people making a decision to adopt an innovation first gain knowledge about the existence of the innovation and its attributes. They form a favourable impression, are persuaded to go further (the persuasion stage) and eventually decide to adopt and implement the innovation

What happens, however, when an institution or organisation makes a decision to adopt an innovation at government or administrative level, rather than at the level of the individual adopter? Haythornwaite postulates that `When adoption decisions are made at the organisational level ... the adoption process for the individual is cast into disarray, with the persuasion stage following rather than preceding the decision phase'. (12) Although Haythornwaite was referring to the impact of the adoption of computers in the classroom on teacher librarians, the adoption of electronic databases by the administration of a state library on behalf of all the public libraries in the state (undoubtedly in consultation with at least some of the chief librarians) is a comparable situation. As Haythornwaite said of the teacher librarians in her study, public librarians in position `... are more than facilitators, responsible for removing the barriers to adoption--they are also individual adopters, faced with the challenges and stresses of an adoption decision'. (13) They can embrace the innovation with enthusiasm, reinvent it to match their desires, (14) use it spasmodically, or not use it at all.

Haythornwaite's findings were that the teacher librarians interviewed saw their role in computing as `... occurring by default, fate, chance, or necessity'. (15) In terms of the Rogers' five attributes of innovations, they formed a very favourable impression of the innovation with regard to relative advantage; were concerned about the complexity and the resultant learning required of them; reported both positive and negative observable results; saw use of computers as compatible with their basic jobs and skills; and felt that they had had the opportunity to try out pieces of computer applications (trialability) and that this had led to a positive attitude to the adoption of the corresponding full system. (16)

Following a description of the method used for the Monash University/State Library of Victoria evaluation of the use of electronic databases in public libraries in Victoria there will be a discussion of the study in relation to the issues of adoption of innovations, outlined above.

Method: evaluation of use of electronic databases in Victorian public libraries

Collection of qualitative data took place in four Victorian public library services in the first half of 2000. The research in each case began with a focus group of library staff about their perceptions of content and usability of each of two databases, as well as general access issues. A representative of SLV was included in each of the focus groups. These representatives were from VISionN, a department of the State Library which answers reference queries from the Victorian public library service. Attempts were made to include library staff in varying roles in each of the focus groups eg a library manager and a library technician.

The focus groups were followed by the trialing of the databases with ten library users who varied in age and gender, from each of the library services. The databases were evaluated with ten further library users at SLV in the first half of 2001. In the case of the SLV component of the research an attempt was made to locate users who were already using one of the four databases. The difference between these users and those recruited in the four public libraries is that none of the latter group was discovered using a database and none had used one before. This reflects the difference between the public library clientele and that of SLV where there were 33 terminals available at the one time, from which users could access the databases.

Each library service was allocated two of the four electronic databases offered through Gulliver, for the purposes of the research. As was pointed out in one group, it was a pity that staff had not been able to compare all four Gulliver databases. However, using two databases at a time was a matter of practicality, as it would have been too time consuming for each librarian to test all four, and impossible to discuss them adequately in a hundred minute focus group session. The setting of homework, requiring staff to follow a topic of their choice in their two databases and to answer questions related to content and usability prior to their involvement in their focus group, resulted in data of high quality from these focus groups.

In discussing the findings which throw light on the extent to which the electronic databases had been adopted by librarians and library users, it is mainly data from the focus groups which will be used. The findings should not be seen as reflecting the present situation in Victoria, as the data were collected some time ago. Rather, the paper presents a snap shot in time involving four library services, plus a small sample of users from SLV.

Findings: relative advantage

There was a number of indications from the focus groups that librarians saw the electronic databases as providing advantages, compared with some of the other forms of information available at present. They saw the databases as giving them access to a large amount of information in searchable form. It was clear that one of the big pluses of the databases is that librarians feel confident about the authority of the information which is found on them. The items on the databases are well cited and they have confidence in the authority of the sources.

It is not surprising that the authority of the sources was an important issue for library staff. For a number of them, it was a feature that marked the databases off from the internet; particularly for health related queries, which are common, and a sphere in which authority can have practical significance for users. At the same time, the fact that many articles retrieved from the databases were from professional or academic publications led some focus group participants to question their utility for the broader public. Against this, others pointed to the ability of at least one of the databases to sort articles by reading level and experience.

Difficulties in always finding relevant content, which may as much be a consequence of user inexperience as problems with database content, prompted some participants to value the internet by comparison, as at least `something' could always be found. Even then, it was suggested it might be worth persisting with the Gulliver databases `because if you get the hit then you get the full text. It's a quick transaction.' On the other hand, a common observation was that many users have a very instrumentalist approach to information retrieval, wanting broad information rather than journal articles--`something they can take home'.

It was noted that the topic of the search is crucial, both in relation to individual databases and in general. Some topics eg historical subjects, are not well suited to the Gulliver databases. Electric library was seen as very valuable because it indexes current newspapers. The availability of the Australian and Herald-Sun was seen as `fantastic'..

A common complaint across all the group was that Australian content in the databases was less than had been expected. The view expressed in one group by an SLV staff member, that at present Custom database `is really not very good for Australian material', went unchallenged. Elsewhere it was noted the SLV had done all it could on this front. In another focus group it was argued that World magazine bank and Custom database were limited in terms of Australian content. Some of the discussion around particular searches carried out by staff members offered concrete examples to back up such concerns. For instance, problems in retrieving relevant sources on the controversy in Australia about tampons and the goods and services tax. On the other hand, while some focus group members considered Electric library and Custom database to be `very American', at least one noted that results in Electric library could be limited to Australasian sources. And while one participant thought that Custom database's content was too `heavy going' and `academic' for `normal public library users', a number of staff users valued its access to full text.

The availability of full text was a very important reason why the Gulliver databases were seen as having the advantage over other electronic resources. As one focus group participant commented `People would like a citation which they can find'. Focus groups considered that the large amount of full text on the Gulliver databases was a big plus. In one session, Gulliver was compared favourably with Austrom and Austguide both of which provide only citations. On the other hand some librarians felt that more full text or all full text would improve the databases.

In summary, while there were reservations expressed about aspects of the Gulliver databases, on the whole they were seen as having relative advantage, in keeping with Rogers' first attribute of innovations. (17) Given that the only library users who had used the databases prior to the trials were from SLV, there is no reliable data from the perspective of users about the issue of relative advantage.

Findings: compatibility

There is indirect support from the project data for the proposition that electronic databases are compatible `... with the existing values, past experience, and needs of potential adopters'. It was clear that librarians in the focus groups saw the databases as fitting into the public library setting, and as providing a valuable tool for finding information for themselves and users. The fact that the internet was being increasingly used within the library meant that many users now having experience with electronic searching. On the other hand, a common point made in the focus groups was that the primary use of library public pcs was for internet access which was not for reference. Users were using it `for recreation eg email or looking for sites'. For such users, icons on the pc desktop or library home page flagging the Gulliver databases were simply an impediment, `just something that is in the way' as they navigate towards a familiar website like Yahoo. At the same time, some participants felt that many users lacked broader contextual knowledge about internet and pc use. In one case it was claimed that many users relied on the previous user's setup to access their email, as they did not know how to type in urls. One idea was for users to be encouraged to close down the web browser when finished, obliging the next user to start from scratch, and thus encounter the library home page or desktop with its links to Gulliver and elsewhere. Users, it was argued in one focus group, also continue to display unrealistic expectations as to what it involved in information retrieval, whether on the internet or a database. `A lot of them aren't prepared for the searching you have to do to find information.'

The implications of these points about users is that there is some mismatch between their experiences with electronic sources and those that the databases might offer. This is confirmed by the fact that during the fieldwork, the researchers discovered not one library user in any of the four libraries who had used the databases. During the additional fieldwork at SLV, where there are 33 terminals for public use, it took ten hours to find ten users using one of the Gulliver databases. It seems clear that, despite the seeming compatibility with the existing values and needs of library users, the electronic databases had not been widely adopted with enthusiasm in the libraries.

Findings: complexity

Perceived complexity, Rogers' third attribute, was a very important reason why adoption of the databases had not been as rapid as had been hoped. There was strong evidence for this in the librarians' focus groups. For example, the importance of feeling confident in the use of various search strategies was a common theme. The researchers concluded that the databases were not being used to an optimal level, even by experienced library staff. For example, the use of different kinds of searches as were then available eg keyword, natural language, relevance, resulted in different search results, depending on the database being used. Staff in the focus groups could not reach a consensus about the effectiveness or otherwise of different search strategies. Some thought natural language searches worked well on certain Gulliver databases and less so on others, and there was lively discussion about which search options worked best with Health reference library. Once again, training for both library staff and users is needed for best use. As one librarian said `Staff need training on all the different aspects of these databases because they are not always apparent from the first screen'. The individual user trials attested that ignorance of search strategies in general and in terms of particular databases and queries, was a significant reason for many of the problems encountered.

Discussions in the focus groups about the variations of interfaces between the databases, and the sorts of confusions that these might entail, also indicated that librarians considered the databases to be complex. Some of these variations are more a matter of `chrome' ie superficial appearance, perhaps in part the consequence of a proprietary impulse to differentiate a product from that of its near competitors. Others, such as key commands, screen layouts and search options, could be more significant. As one library staffer put it `If I was a member of the public and I was confronted with all these databases--everyone has a different interface, everyone does something different--it would be an absolute nightmare'. Again the emphasis on training in the focus groups indicated that librarians saw the databases as requiring skills for effective use. The idea of training for the databases for staff, as well as for users was a popular one. Training was needed so that staff could practice with the databases.

Findings: trialability

Rogers' fourth attribute concerns the degree to which an innovation may be experimented with on a limited basis. (19) At first glance this attribute appears to be easily met, as staff and users alike have ready access to the databases without any financial commitment at all. There are other barriers to trialability, however. In terms of staff, a major barrier to individual experimentation with the databases was the lack of time. A recurrent theme of the focus groups was that `time is always pressing' and `a big issue' in a public library context. In terms of users their lack of awareness of existence of the databases was clearly an impediment. In the focus groups, concern was expressed among some staff about the level of promotion for the Gulliver databases. According to one member, in their library at least, `if people don't ask for assistance, they many never know that they're there'. Individual users may also be diffident about experimenting with the database without support from a librarian. They may fear embarrassment if they strike problems. Librarians seemed to think that, where users did use the databases, it was not uncommon for staff to provide close support. As was said in one group, users would probably very interested in Gulliver `if you offered to do the work for them'. While some users were adventurous, many continued to expect that staff would find materials for them, on the shelves or online.

Another issue, in terms of trialability, was access to the databases. In most libraries the solid demand for user access terminals raised issues for patron access to databases if, and when, this was required. One library had designated some pcs as `15 minutes only'. Other libraries were considering the possibility of setting aside one or more pcs specifically for public reference use. It was argued in one group that this was a luxury that could not be afforded, as it would mean `a computer sitting there not doing much'. Furthermore, it was claimed, `the shortage of terminals is probably one of the reasons we don't introduce people to the databases'.

A number of libraries reported problems in accessing the databases. Whether due to configuration of proxy servers, individual IP addresses for terminals, passwords or some other reason, each group had some stories on this front. Teething problems of this nature are not unusual of course, and the situation may well have changed since the focus groups met. At least one group, however, reported that such difficulties had led a number of users to abandon their attempts to access Gulliver.

Findings: observability

With regard to the degree to which the results of an innovation are visible to others, (20) the data indicate that the databases provided through the Gulliver program were not highly visible. This was certainly the case with users, as has been indicated above. In some cases, even staff found that the databases did not spring readily to mind. One focus group member suggested that staff were not always aware of the Gulliver databases, her colleague adding `we forget they're there and we get very busy'. There was a widespread view that the databases could be promoted better. Indeed it was suggested that current low levels of database usage could be reversed with better promotion--not simply among users, but also among staff.

Conclusion

The findings of this project were surprising to the researchers. The major focus of the research was to be the evaluation of the content and usability of the electronic databases. While most librarians in the focus groups saw content as of paramount importance, including the extent to which full text was provided, they felt confident about the authority of the information found on all the databases. They were, on the whole, unconcerned about usability issues, feeling that all the databases were of a reasonable standard.

The issues which were seen to be of greatest importance were access, time pressures, and training. It was these three issues which were seen to be responsible for the fact that, at the stage when the research was undertaken, adoption of the databases had been slower than expected. First, there were major issues of access to the databases--whether the proxy server or passwords were being used. Problems with internet connections, including slow response times, often exacerbated more specific difficulties. A second significant finding concerned time. Innovations never occur in a vacuum. They are always introduced into particular social structures and have varying implications for different subjects within those structures. In this case, talk by focus group participants of time pressures as possible obstacles to adoption hint at an array of further questions about the work regimes in public libraries today. Together with lack of training these problems were seen to deter staff from using the databases and affected perceptions of their value. In terms of Rogers' key attributes, the databases did not present favourably in terms of complexity, trialability and observability.

Training was seen as the principal way in which greater use the databases could be encouraged. It was considered that the first step was to train library staff, before users could be encouraged and trained to use the databases. Even some librarians who had been trained by SLV staff appeared not to have used the databases extensively. The researchers concluded that training should involve hands on tasks which familiarise staff with databases. The homework, which required library staff to use the databases and answer questions prior to the focus groups, was seen as most beneficial. Many staff commented that they felt much more confident about using the databases as a result of the research. A new project seeks to remedy this situation more broadly by developing the competency of public librarians in using online databases. It began in September 2001, in NSW and ACT and is funded by the State Library of NSW and the National Library of Australia. Future articles will report the findings of this project.

The study which is the focus of this paper was undertaken at the early stage of a major Victorian project the aim of which is to promote online services in public libraries. The fact that nearly all library services in Victoria decided to subscribe to the databases in 2001 is an indication of the value placed on them and that the adoption process is underway.

References

(1) Holland, M Diffusion of innovation theories and their relevance to understanding the role of librarians when introducing users to networked information The electric library 15(5) 1997 p394

(2) For a description of the Gulliver program see Whitehead, D and Toohey, R Gulliver: head, shoulders (and knees) above the rest? Vala conference Melbourne 2000

(3) ARC Spirt grants (now called ARC Linkage) are awarded for projects which involve industry/academic research partnerships. One of the options available under the Spirt (Linkage) program is to include one or more PhD or Masters students in the project for which funding is sought. The Monash/SLV project involved a Masters student, Nettie Handley, who received an Australian postgraduate award (industry) to take part in the project

(4) Rogers, E The diffusion of innovation New York, The Free Press 1995

(5) ibid p212

(6) Callahan, D The librarian as change agent in the diffusion of technological innovation The electric library 9(1) 1991

(7) Rogers op cit p224

(8) ibid p242

(9) ibid p243

(10) ibid

(11) ibid p244

(12) Haythornwaite, C Exploration of the impact of `Computers in the classroom' on school librarians p6 http://alexia.lis.uiuc.edu/~haythorn/ school_di.htm

(13) ibid p8

(14) Markus, M Power, politics and MIS Communicating of the ACM 26, 1983 p430-444, cited by Haythornwaite op cit p8

(15) Haythornwaite op cit p24

(16) ibid p 16-23

(17) Rogers op cit p212

(18) ibid p224

(19) ibid p243

(20) ibid p244

All of the authors are associated with Enterprise Information Research Group (EIRG), in the School of Information Management and Systems at Monash University. Dr Kirsty Williamson is the Director of Information and Telecommunication Needs Research (ITNR), a research group which is affiliated with EIRG and also with School of Information Studies, Charles Sturt University, Wagga Wagga. Professor Don Schauder is the chair of ITNR and Dr Steve Wright and Louise Stockfeld are Research Fellow and Research Associate, respectively. Associate Professor Frada Burstein is an EIRG researcher and Nettie Handley is a Masters student who was awarded an Australian post graduate award (industry) to participate in the project. The paper reflects the philosophy of ITNR which has a specifically user focus in relation to information and telecommunications needs of community groups, as well as EIRG's broader aims which are to improve enterprise and interpersonal information flow and recorded memory. Address: School of Information Management and Systems Level 7, 26 Sir John Monash Drive Monash University Caulfield East Vic 3145 Tel[phone] fax[phone] [e-mail] www.sims.monash.edu.au/research/eirg/index.html and www.infotech.monash.edu.au/itnr

Kirsty Williamson, Don Schauder, Steve Wright, Louise Stockfeld and Nettie Handley School of Information Management and Systems Monash University Victoria Refereed paper