CPLUG RESEARCH REPORT No. 4

THE PERFORMANCE OF CAMDEN PUBLIC LIBRARIES

Prepared by A.Templeton

29^th January 2004

1. Introduction

There has been a long period between the issue of the last CPLUG research report in June 1999 and the drafting of this document. In the intervening time, central government has invested a considerable amount of money in library facilities throughout the country. This investment was concentrated in the provision of public access IT equipment and was intended to improve computer literacy in the general population. The new “People’s Network” facilities have proved extremely popular and have had a marked, positive effect on the number of visits made to public libraries within Camden. Presumably, they have also succeeded in satisfying the government’s intensions in Camden.

The rapidly rising number of people visiting each of Camden’s libraries has been carefully recorded by CPLUG within its webpages ¹. Although acknowledging that there has been an improvement in one narrow measure of library performance, CPLUG has been reluctant to join in the general rejoicing expected by some within the Council. The reason for that reluctance is that there has been no corresponding general improvement in the number of books issued from the libraries. Indeed, the most recent figures indicate a falling trend in many of Camden’s libraries. This report attempts to isolate some of the reason(s) for this worrying development.

2. Factors Affecting the Performance of Public Libraries

In common with all activities involving the general public, there is a wide range of factors that determine the performance of a public library. Some of these are under the simple control of library managers (such as marketing) and some are not (such as location).

There is also a group of factors, which are only partially controlled by professional library managers. These are generally those budget items which are of high cost and high visibility. Inevitably, Councillors set, or heavily influence, the cost limits and professional library managers have to do their best within the imposed constraints. It is not CPLUG’s view that this is inherently a bad situation (its called the democratic process). However, it is noticeable that the “robbing Peter to pay Paul” atmosphere of Camden budget setting usual casts the library budget in the role of Peter.

The major part of the money spent by Local Authorities on services within their boroughs comes from central government. The government imposes a duty on Councils to spend almost all of this money in the manner specified by the government. However, money provided by the government for provision of library services is one of the few exceptions. Thus, when this is combined with a predisposition of some Councillors to seek the easy, expedient way out of any difficulty, the library budget usually loses.

Modern public libraries have many tasks, but the one activity that the public considers to be pre-eminent is that of book lending. This is the rock on which libraries base their traditional education function and it is as relevant today as 150 years ago.

More than 700,000 Londoners have no qualification and 23% of London’s population have low levels of numeracy and literacy (ALG’s document “Keyfacts 2004”). This is well above average UK levels. People within this group desperately need the chance to quietly improve their skills in order to give them the opportunity to compete in today’s jobs market (the London unemployment rate is 7.1%, again well above the national rate – there are more unemployed people on London than in Scotland and Wales combined). Due to the high concentration in Camden of refugees and other deprived groups, the situation here must be worse than most other boroughs.

The current concentration on computer literacy can generate great benefits for the average Londoner but it has no relevance for those that cannot read. Access to appropriate books combined with some sympathetic tuition is the medicine required to cure this social sickness.

For those that can read, write and do their everyday sums, the government has set the challenge of life-long learning to enhance their work-skills and to live a fuller life. The simplest, cheapest, most flexible way of undertaking that task is with books borrowed from a public library.

3. Measuring Performance

As book lending is considered by the public to be the core activity of public libraries, it is not unreasonable to judge the performance of a library service on this aspect. If a library service performs well in other ways but fails in this, the public perception is one of failure. Library managers may well argue that they are set many performance targets and to concentrate on just one is unfair. This is a valid point, but the fact remains that the public judges on grounds that it considers to be important and is unimpressed by other criteria.

At the creation of Camden, in 1965, the borough inherited a strong, well-supported library service, with 17 libraries (each open for 60 hours per week) and, for a number of years thereafter, it remained so. However, in the early 1970s a change of attitude occurred within Camden Council and the previous commitment to the provision of excellent library services became far more muted. The number of libraries was reduced to 14, but for a decade Camden’s library service managed to survive reasonably well by means of good management. This period of stability came to an end in the mid 1980s when Camden decided that library services in the borough should be sacrificed to bolster other, more favoured, activities. This set the pattern for the council’s attitude for the next two decades. The effect of this history of waning commitment can be seen in the following charts. In order to avoid distorting factors caused by strikes, library refurbishment closures and the general turbulence of the most recent years, these charts cover the years up to 1997 only. Although the loan bookstock data is almost complete, that for other parameters is less so for the early years of Camden’s existence. The data analysis is therefore restricted to the years 1975 – 1997. Twenty-three years of data is sufficient to establish, with confidence, any trend within the data.

The above chart of the variation of Camden libraries bookstock with time presents one of the obvious variables affecting the number of issues from the borough’s libraries. The comparison of the variation of bookissues and loan bookstock with time, shown below, confirms that there is a relationship between the two parameters. The measure of the closeness of that relationship is the correlation coefficient, where complete matching is represented by a value of 1. In this case, the value of the calculated correlation coefficient is 0.9380, i.e. the match is very good, but not quite perfect – as indicated by visual inspection.

It is not surprising that the match between bookstock and bookissue variations (above chart), whilst being good, was less than perfect. The number of books issued from a borough’s public libraries must be dependent on several factors. The second obvious variable affecting the number of book issues is the number of library opening hours available to the public to access that stock (chart below).

A visual inspection of the above chart, showing the variation of the number of book issues and opening hours with time, leads to the conclusion that there is also a close relationship between these two parameters. This is confirmed by the correlation coefficient having a value of 0.9498 – very similar to that obtained for the coefficient for issues versus loan bookstock.

The correlation between loan bookstock / opening hours and bookissues is impressively good in both cases. However, it is unlikely that two completely independent parameters could each have such a high, similar value coefficient. Therefore, is there another variable which controls bookissues and which is related to those parameters already investigated?

Both the number of books on the shelves and the opening hours of the libraries are factors which determine the availability of library books to the public. Availability can be defined as the number of book hours available to library users i.e. book stock multiplied by opening hours. The chart below compares the variation of this parameter (calculated as the sum of individual library availabilities, due to opening hours being library dependent) with the variation of bookissues.

The correlation coefficient of the new book availability factor versus bookissues has improved to 0.9746, a significant step towards the perfect match. This composite factor, therefore, represents a powerful tool for predicting the effect that changes to the level of bookstock or opening hours will have on library performance. However, the ultimate driving force behind performance variations is frequently expenditure. The chart above compares the variation of library expenditure with bookissues and shows that there is not as close a correlation between the two as for that obtained for book availability versus issues. Presumably, the rather large difference between the two correlation coefficients (0.7515 & 0.9746 respectively) is, at least partially, due to the library service attempting to meet many performance targets.

4. Conclusions

It is possible to suggest that this investigation has only found what was perfectly obvious already i.e. that the number of books issued from a borough’s public libraries is very dependent on the opening hours of the libraries and the size of their bookstocks. Yet it is these “facts” that are often denied and contradicted by Local Authorities desperate to save money by closing libraries and/or reducing opening hours and/or reducing book funds.

It is usually suggested by Councillors supporting library closure policies that, if a library is closed, borrowers will transfer en masse to neighbouring libraries. This is in spite of research by Sheffield University² showing that a significant proportion of borrowers simply stop using public libraries when their local branch is closed. There may be a unique Local Authority area in some corner of the UK where this behaviour pattern does/will not occur, but this report shows that it is certainly not Camden.

Those Councils unwilling to fund a good library service but wishing to avoid the risk of a public backlash usually adopt a policy of “death by a thousand cuts” - gradually reducing bookstock and opening hours. It is clear from this investigation that such a policy is indeed a means of bringing slow death to a library service. Of course, this approach has the additional attraction for Councils that the resultant reduction in library usage can then be cited as the main reason for a subsequent closure programme – presented as being the fault of the users for not using the libraries.

The disadvantage of the slow death approach to a library funding issue is that it needs very careful management. A public backlash is always possible, especially when applying the final coup de grâce.

The extremely good correlation between the book issues and book availability parameters suggests that it may be possible to calculate the level of book issues expected from a borough’s public libraries, if the book availability is known. A linear regression was carried out on the data used in this investigation and resulted in the equation:

Book Issues = 679982 + 0.03354 x Book Availability

The level of agreement of this equation with the actual data can be gauged from the following chart.

The level of agreement is quite good. However, it would be very optimistic to expect the equation to accurately predict the number of books issued when this is far from the range covered by this investigation. Errors introduced by extrapolation increase as extrapolation increases. This would be especially so for low levels of issues. In this circumstance, many unpredictable variations have a large effect.

The data used in this investigation does not include that for years after 1997. Apart from the conflict which marked some of those missing years, the major event in this period was the introduction of ICT services into Camden’s public libraries. Internet access via library computers has proved immensely popular and has changed the way many people regard their local library. It has certainly caused an increase in the number of visits made to public libraries. However, the commonly held belief that the number of books issued from a library is closely tied to the number of visits made to those libraries, has not been borne out by recent Camden statistics.

The most likely explanation for this is that the historic dependencies still hold in the main. Nothing has been changed fundamentally except that there is now an additional reason for a visit to a library. Only if the Library Service can convert those extra, computer related visits to dual purpose, computer/book visits will there be a change in the book issue trends. This is something that can, with advantage, be the subject of further investigation.

¹ http://librarylondon.org/localgroups/camden/cplug.htm

² “Assessing What Libraries Mean to Users” by Richard Proctor, pp 10–17, CPLUG Seminar Oct 99, “Do Libraries Have a Future?” Available at: http://www.librarylondon.org/localgroups/camden/seminar.htm