Page 125 - JOURNAL OF LIBRARY SCIENCE IN CHINA 2018 Vol. 44

124 Journal of Library Science in China, Vol.10, 2018

to the above-mentioned four major factors. On that basis, the number of employees (logcyry),
fiscal subsidy (logczbt), local permanent resident population by the end of that very year (logrk)
and GDP (loggdp) are used as independent variables and circulation person-time (logltrc) as a
dependent variable to carry out analysis of the four areas and discover inherent reasons affecting
the performance of county level libraries, by using the method of multiple linear regression.

2.1 Analysis of factors affecting circulation person-time of county level libraries

In general, circulation person-time of county level libraries from 2001 to 2015 increased along with
permanent resident population. It is also positively influenced by the number of employees and has
no direct positive correlation with economic growth factors such as GDP and fiscal subsidy. That
shows that there is no close links between the development of county level libraries and social and
economic development; institutions operate independently.
Calculation results indicate that growth of fiscal subsidy has minor positive impact on the
growth of circulation person-time of county level libraries in each area; the number of employees
has significant positive influence on the growth of circulation person-time of county level libraries
in the central and western areas; permanent resident population by the end of that very year has
remarkable impact on circulation person-time of county level libraries in the central area – when
population grows, circulation person-time grows as well; but GDP, an important indicator of social
and economic development has no direct positive impact on circulation person-time (see Table 5).

Table 5. Analysis of factors affecting circulation person-time of county level libraries.
Permanent resident Number of
GDP
Variable population by the end of (loggdp) Fiscal subsidy employees Constant R 2 Period
(logczbt)
that very year(logrk) (logcyry)
32.492*** -1.323*** 4.500*** -402.839**
Nationwide — 0.995
(2.973) (0.185) (0.569) (37.530)
0.533*** 0.196***
Eastern — — — 0.983
(0.031) (0.047)
13.662*** 0.157*** 4.929*** -180.983*** 2001-2015
Central — 0.953
(2.660) (0.041) (1.315) (30.456)
-0.858*** 0.725*** 3.422* -31.706**
Western — 0.890
(0.294) (0.260) (1.781) (15.058)
-0.589** 0.699*** 0.600***
Northeast — (0.232) (0.205) (0.047) — 0.788
Note:Figures within brackets indicate standard error. ***, ** and * respectively represent the significant levels of 1%,
5% and 10%. Data corresponding to permanent resident population by the end of that very year, GDP, fiscal subsidy and
number of employees in the table are regression coefficients when independent and dependent variables regress (see the
followings for the regression formula). Figures with brackets are P value of independent variables. In order to ensure co-
integration of the multiple regression formula, GDP data involving independent variables are the value by evaluating the
logarithm of GDP per capita, when carrying out multiple regression analysis of circulation person-time of libraries in the
western area. Other multiple regression formulas used logarithms of GDP.