Transcript Initial Readings of Contemporary Data on Buddhist Monasteries

Initial Readings of the Data About
Contemporary Chinese Buddhist
Monasteries
Jiang Wu 吴疆
Department of East Asian Studies
Daoqin Tong 童道琴
School of Geography & Development
The University of Arizona
Introduction to the Data
BGIS
 ECAI: Atlas of Chinese Religion
 China Data Center

Assumptions of Chinese Monasteries
Buddhist monasteries are fundamentally
independent and local institutions.
 They are one of the types of local
institution which has been allowed to
grow in China.
 Temple building activities are largely
spontaneous endeavors undertaken by
local communities
 Thus, temple building can be retreated as
an index to social and cultural
development.

Purpose of this Study





Changing the paradigm in the study of
Buddhism
From sectarian-based model to monastery-or
place-based study
Identify various social, cultural, economic factors
(viables) and their relationships to temple
building
Identify patterns in the growth of Buddhism
through history
Understand the transformation of Chinese
society
Methods
Data sampling
 Exploratory Spatial Data Analysis (ESDA)
 Regression analysis
 Historical approach
 Quantitative and qualitative research

Exploratory Spatial Data Analysis
(ESDA)
Allow users to describe and visualize
spatial distributions, discover patters of
association, clusters, etc.
 Explore the properties of datasets
without the need for formal model
building
 We believe that the temple distribution is
not random
 Spatial autocorrelation

Spatial Autocorrelation
Refers to the coincidence of attribute similarity
and locational similarity (Anselin 1988)
 Moran’s I (Anselin 1995)
◦ Provides the degree of linear association
between values observed at different
locations
◦ Positive vs. negative

n
n

 w (k ) x x
ij
I 
i
i 1 j 1
n
n
 x x
i 1 j 1
i
j
j
China Temple Distribution
Tianjin
Hebei
Shanxi
Neimenggu
Liaoning
Jilin
Heilongjian
Shanghai
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Xinjiang
Ningxia
Qinghai
Gansu
Shaanxi
Xizang
Yunnan
Guizhou
Sichuan
Chongqing
Hainan
Guangxi
Guangdong
Hunan
Hubei
Henan
Shandong
Jiangxi
Fujian
Anhui
Zhejiang
Jiangsu
Beijing
Temple Distribution by Provinces and Types
2500
2000
1500
TotalTemples
Association
cave
Chapel
1000
Hall
House
Monastery
Temple
500
Unknow
Others
0
Moran’s I
I=0.3951
 P-value =0.0015 with 9999 random
permutation
 Positive spatial autocorrelation

LH
HH
LL
HL
Local Indicator of Spatial
Autocorrelation (LISA)
Capture local spatial clustering (Anselin
1995)
 Provinces that are statistically significant

Factors to Explain Variability in
Temple Distribution

Linear regression
Y    1 X 1   2 X 2  ..   p X p
◦ Dependent/Response variable (Y)
 number of temples in a province
◦ Independent/Explanatory variables (X’s)







Population
Income
Rural/urban
media (TV, newspaper, internet users)
Ethnicity
Education
Transportation
Regression Results

R-square 0.69
Coefficient
t-statistic
p-value
-780.995
-2.085
0.048
Population
0.022
0.458
0.651
Income
0.146
4.193
0.000+
Roads (km)
0.001
0.420
0.678
Museum
11.345
4.064
0.000+
Internet users
-1.129
-2.241
0.035
HS_above
-33.920
-4.350
0.000+
Constant
Population (10,000)
Income (yuan)
Internet users (10,000)
HS_above (%)
Statistically Insignificant
Population: Population does not
contribute significantly to the variality of
temple distribution.
 Roads (km): transportation does not have
correlations with temple distribution.
 Interpretation: Chinese population is huge
and transportation has been welldeveloped. Thus they have minimum
impact.

Positive Correlation




Income: Higher income level tends to boost
the number of temples.
Museum: the existence of museum indicates
the existence of more temples
Interpretation: Economic growth stimulates
the growth of Buddhist institutions.
As cultural indicators, museums and
monasteries have similar role in local society
as they require local investment. (Note:
some temples might have been appropriated
as museums.
Negative Correlations



Internet users: The area where the number
of internet users increases may have negative
impact on the distribution of Buddhist
institutions.
HS_above: people with above high-school
education may have negative impact on the
building of Buddhist institutions.
Interpretation: Higher education may
discourage the development of Buddhist
institutions. (Not necessarily Buddhism as a
whole.)
Future works
Narrow the scales to country level
 Seeking continuities with data in Tang and
Qing
 Incorporating William Skinner’s Macroregion theory more effectively
 Conducting residual analysis to identify
the defects in the original data collection
