CC BY
9
УДК 338.49
МЕТОД ФАКТОРНОГО АНАЛИЗА ДЛЯ ОЦЕНКИ КАЧЕСТВА ЖИЗНИ НАСЕЛЕНИЯ КИТАЙСКИХ ПРОВИНЦИЙ
FACTOR ANALYSIS METHOD OF CHINESE PROVINCES POPULATION LIFE
QUALITY EVALUATION
Сян Сяо Ган / Xiang Xiao Gang Уханьский текстильный университет
В статье обосновано применение метода статистического многофакторного анализа и определена система индикаторов для оценки качества жизни населения в различных провинциях Китая. Экспериментальные расчеты выполнены на основе национальных статистических данных КНР по 31 региону за 2013 год. Сравнительный анализ показывает, что есть очевидные региональные различия в качестве жизни населения Центральных, Восточных и Западных провинций. Рассчитанный Индекс Муниципалитетов и его локальные компоненты могут быть использованы для ранжирования регионов и обоснования направлений социально-экономической политики.
Ключевые слова: многофакторный анализ, качество жизни, валовой внутренний продукт, индекс муниципалитетов, ранжирование регионов.
1. Introduction
Under the policy of reform and opening up our national economy has achieved a long-term and rapid development process. At the same time, the standard of the living quality of domestic residents has been improving continuously. The life of the residents gets a huge change, especially in recent years. According to the statistical data [1] in 2013 the GDP of our country was 9.24 trillion dollar, American GDP was 16.8 trillion dollar, Japan's GDP was 4.9 trillion dollar and Russia's GDP was 2.1 trillion dollar. From the total economy in the world the United States was on the first place, China ranked the second. Japan was the third, Russia was the seventh. Despite having a very large population, China's GDP per capita was only 6807 dollar. At the same time the US GDP per capita was 53143 dollar, Japan's GDP per capita was 38492 dollar, Russia's GDP per capita was 14612 dollar. Our country GDP per capita is around 90th place in the world and it is far behind comparing with the developed countries.
Nowadays, there are some uncertainties in the international environment, such as Ukraine crisis, Greece problem, the South
China Sea issue etc. Residents' life in China has basically entered into a well-off society, but the living level still is not comprehensive, not balanced and not high. There are many deficiencies and shortcomings. There are still most obvious problems that the life level of the bottom people is very low, the gap between the rich and the poor is gradually expanding, and the problem of corruption is very sensitive. So we are still developing, and it is necessary to go a long way to achieve the four-level modernization of the country.
American economist Rostow believes that the economic growth consists of six levels. They are: the traditional society, preparation before takeoff, taking off, maturing, mass consumption and life quality [2]. The developed countries have already completed the first modernization stage from the agricultural society to the industrial society. At present, they are on the second stage of the transformation from the industrial society to the knowledge society. On the first stage economic growth is the primary goal. In the second stage the quality of life is the primary goal. Whether it is in foreign or domestic the research about the quality of life has a long his-
tory. So, there are a lot of relevant theoretical perspectives [3-12]. It is generally believed that "the quality of life is the guarantee of the society to the personal life and the satisfaction of the individual to the social life". This ind i-cates that the level of economic and civilization development of a country or region is reflected in the life quality of the residents. Quality of life is the ultimate goal of the social development. We are in the critical period of economic development. At this time the research on quality of life and related issues has a very practical significance. It can give advice for the government to formulate related policy of improving the residents' quality of life and to fully accelerate the speed of construction of a well-off society in an all-round way.
2. Index system of the urban residents' life quality evaluation
Life quality evaluation index system refers to the form a life quality index system by selecting a certain number of indicators, obtain a comprehensive index by using various mathematical models and processing method and analyze of the life quality by using the comprehensive index.
The result received by the European Statistical Laboratory (ESL) shows that the effectiveness of the index number to the results of the study is marginally declined. If the number of the indicators in the model is less than 20, the effectiveness will be low. If the number of indicators in the model is between 20 and 30, the validity of the model will be about 75%-90%. If the index number reaches 40 or more, the effectiveness will be above 90% [13]. Considering the principle of the selecting indicators, such as objective existence, availability, applicability, etc., this article defines 26 following indicators of the residents' life quality.
X1 = Urban residents' food cash consumption expenditure per capita, unit of measure: Yuan.
X2 = Urban residents' clothing cash consumption expenditure per capita, unit of measure: Yuan.
X3 = Urban residents' residence cash con-
sumption expenditure per capita, unit of measure: Yuan.
X 4 = Urban residents' household equipment cash consumption expenditure per capita, unit of measure: Yuan.
X5 = Urban residents' transportation and communication cash consumption expenditure per capita, unit of measure: Yuan. X6 = Urban residents' number of cars for household use per 100 people, unit of measure : Unit.
X 7 = Urban residents' cultural services, education and recreation cash consumption expenditure per capita, unit of measure: Yuan. X8 = Urban residents' medicine and med ical service cash consumption expenditure per capita, unit of measure: Yuan. X9 = Urban residents' insurance consumption expenditure per capita, unit of measure: Yuan.
X10 = Residents' high-grade apartments and villas consumption per capita, unit of measure: Yuan.
X11 = Urban residents' disposable income per capita, unit of measure: Yuan. X12 = Marriage rate = 1-divorce rate, unit of measure: %.
X13 = Urban residents' registered employment rate, unit of measure: %. X14 = Urban residents' coverage rate with access to tap water, unit of measure: %. X15 = Urban residents' coverage rate with access to gas, unit of measure: %. X16 = Urban residents' area of paved roads per capita, unit of measure: s.q.m. X17 = Urban residents' number of public transportation vehicles per 10000 people, unit of measure: Unit.
X18 = Urban residents' number of doctors per 1000 people, unit of measure: Unit. X19 = Number of enrolled students in higher education per 1000 people, unit of measure: Person.
X20 = Collections of public libraries owned per capita, unit of measure: Copy. X21 = Per capita public green areas, unit of measure: s.q.m.
X22 = Engel's coefficient of urban house-
holds' positive index = 1- Engel's coefficient of urban households, unit of measure: %. X23 = Infant survival rate, unit of measure: %.
X24 = Life expectancy (estimated value of 2010), unit of measure: Age. X 25 = Adult literacy rate = 1-adult illiteracy rate, unit of measure: %. X26 = Urban employee basic pension insurance rate, unit of measure: %.
Among them, the X1-X5 variables are the inner basic family life quality consumption indicators, including food, clothing, live, equipment, traffic and communication. The X6-X10 variables are the inner additional family life quality consumption indicators, includ-
ing cars, entertainment, health care, insurance and luxury. The X11-X13 variables are the indicators of the inner family life quality in concept, including wealth, marriage and work. The X14-X17 variables are the indicators of the external basic social life quality environment, including water, gas, road and bus. The X18-X21 variables are the indicators of the external additional social life quality environment, including medicine, education, knowledge and environment. The X22-X26 variables are the indicators of the external social life quality environment in concept, including equality, development and civilization. They are shown in the following table 1.
The decomposition of the index structure
Table 1
First level index Second level index Third level index Fourth level index Variable
Index of life quality Index of inner family life quality Index of inner basic family life quality Food X1
Clothing X2
Live X3
Equipment X4
Traffic and communication X5
Index of inner additional family life quality Cars X6
Entertainment X7
Health care X8
Insurance X9
Mansions X10
Index of inner family life quality in concept Wealth X11
Marriage X12
Work X13
Index of external social life quality environment Index of external basic social life quality environment Water X14
Gas X15
Road X16
Bus X17
Index of external additional social life quality environment Medicine X18
Education X19
Knowledge X20
Environment X21
Index of external social life quality environment in concept Equality X22, X26
Development X23, X24
Civilization X25
3. Factor analysis model
It is assumed that X = {xi,...,Xp)is a p-dimensional random column vector, F = {F,.., F ) is an ^-dimensional random column vector, A = (Aj )pxm is a
pxm -matrix, and s = {s,...,sp) is a p-dimensional random column vector.
X1 = a11F1 + a12 F2 + ••• + a1mFm + s1 X 2 = a21F1 + a22 F2 + ••• + a2mFm + s2
X p = ap1F1 + ap2 F2 + ••• + apmFm +sp
It can be written as: X = AF + s.
In actual situation, X is the variable that can be observed, but F is the non-observed factor variable. A is the factor loaded matrix, and s is a special factor, which is the equivalent to the residual error in multiple regression analysis. It is assumed that p > m, COV{F,s) = 0 .That means that F and s is not relevant. It is assumed that F,.., F are not relevant and their variances are 1. Moreover, s1,...,spare not related and
their variances are different [14].
4. Data input and run
Through the access to the 2013 and 2014 China Statistical Yearbook we get the major national statistical data [1, 15]. In the course of the input, the life expectancy X24
variable data is adopted with the 2010 national survey data. The divorce rate X12 variable and the X22 Engel coefficient are the inverse index, which are forward processed. And some other variable data is preliminary arranged and calculated. The results of the research are experimentally confirmed. Then we use the Analysis Factor command process in the mathematical statistics software SPSS, select the corresponding option, run and get the following results.
5. Data result
Table 2
KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .540
Bartlett's Test of Sphericity Approx Chi-Square 921.622
df 325
Sig. .000
The first factor analysis test is the feasibility test. KMO test result is 0.54, which is more than 0.5. And the significance level probability value of the Bartlett's Test of Sphericity is 0.000, which is less than 0.05. So it is feasible to make factor analysis for the sample.
Table 3
Total Variance Explained
Initial Eigen values Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings
Co mpo nent Total % of Varianc e Cumulati ve % Total % of Varianc e Cumulati ve % Total % of Varianc e Cumulati ve %
1 2 3 4 5 6 7 8 9 10
1 11.18 43.033 43.033 11.18 43.033 43.033 8.134 31.285 31.285
2 3.331 12.812 55.845 3.331 12.812 55.845 4.026 15.486 46.770
3 2.406 9.255 65.100 2.406 9.255 65.100 2.597 9.990 56.761
able 3
1 2 3 4 5 6 7 8 9 10
4 1.989 7.648 72.748 1.989 7.648 72.748 2.559 9.841 66.602
5 1.315 5.057 77.805 1.315 5.057 77.805 1.780 6.846 73.447
6 1.046 4.023 81.829 1.046 4.023 81.829 1.755 6.750 80.197
7 1.023 3.935 85.763 1.023 3.935 85.763 1.447 5.566 85.763
8 .789 3.035 88.798
9 .626 2.406 91.204
10 .488 1.876 93.080
11 .356 1.368 94.448
12 .316 1.216 95.663
13 .253 .972 96.636
14 .193 .741 97.376
15 .177 .681 98.058
16 .138 .532 98.590
17 .107 .412 99.002
18 .088 .338 99.340
19 .055 .212 99.552
20 .039 .148 99.700
21 .029 .111 99.811
22 .021 .082 99.893
23 .013 .050 99.944
24 .011 .043 99.987
25 .003 .010 99.997
26 .001 .003 100.000
According to the table 3, it can be known that there are totally 7 factor variables, which characteristic value is more than 1, can be extracted. And their variance contribution rate is respectively defined as \ = 43.033%, ¿2= 12.812%, \ = 9.255% , A4 = 7.648%, = 5.057% , A6 = 4.023%, ^ = 3.935%. The top seven factor variables' cumulative variance contribution rate
7
value lis equal to= 85.763%, which
i=1
value is more than 85%. It is sufficient to describe the comprehensive level of the life quality.
Table 4 is the factor loaded matrix and table 5 is the rotated factor loaded matrix. Table 5 is more suitable to explain the meaning of the 7 F13F2,...,F7 variable factors than table 4. It is noticed that there are some blank in the table 4 and table 5 and he blank means the value is less than 0.1.
Table 4
Table 5
Component Matrix"
Rotated Component Matrix"
Component
1 2 3 4 5 6 7
X11 .919 -.238 -.111 .118
X9 .903 -.153 .162 -.107 .171
X5 .893 -.150 .212 .144
X7 .872 -.310 -.111 .127 .116
X24 .860 .256 -.249
X4 .859 -.145 -.135 .171 .297
X26 .857
X3 .827 -.121 -.171 -.131 .303
X1 .744 -.360 -.301 .206 -.212 -.273
X6 .708 .578 .213
X20 .702 -.353 -.245 -.349 .186 -.219
X8 .683 .399 .305 -.311 -.171 -.140 -.103
X17 .670 .321 .160 -.278 -.325
X15 .653 .529 -.280 .159 .228 .156 -.131
X10 .650 -.471 -.375 -.105
X19 .600 .320 .120 -.438 -.355
X2 .579 .297 .427 -.118 -.228 -.236 .241
X14 .513 -.158 .395 .376 -.173 -.456
X25 .506 .667 -.313 .141 .311
X12 -.278 -.599 .491 .209 .347
X22 .398 .504 .305 -.344 .422 .367 .103
X18 .161 -.441 .728 -.236 .242
X23 .447 -.273 .496 -.126 .205 -.193 .107
X16 -.151 .447 .665 .284 -.294 -.114
X21 .487 .154 .519 -.244 .511
X13 .229 -.398 .352 .477 -.284 .387 .159
Component
1 2 3 4 5 6 7
X11 .892 .273 .133 .143 .130
X1 .880 .235 -.298
X7 .867 .210 .142 .139 .245
X5 .860 .170 .279 .119 .145 .131
X10 .847 -.114 -.267
X3 .822 .336 .183
X20 .817 .222 -.282 -.289
X4 .792 .318 .244 .151 .259 -.163
X9 .688 .481 .230 .290 -.110 .239
X26 .686 .470 .183 .141
X24 .675 .497 .339 -.192 .140
X19 .276 .782 -.263 .162
X8 .263 .750 .303 .273 -.264
X17 .278 .731 .110 .164 .267 .231
X2 .245 .621 .117 .359 .367 -.105 -.228
X6 .297 .566 .220 .497 .195 .301 .253
X22 .179 .884 .330 -.114
X25 .263 .305 .766 -.371 .107 -.151
X15 .421 .324 .627 -.293 .183 .317
X18 .893 -.218 .152
X23 .325 .124 .706
X21 .121 .918
X16 -.282 -.143 .628 .598
X13 .152 .141 -.145 .179 .846
X12 -.555 -.201 -.120 .640 .265
X14 .397 .139 .219 .774
According to table 5, we can explain more the F factor loaded coefficient value of on the X11, X1, X7, X5 X10, X3 and X20 variables are greater than 0.817, and on the variables X4, X9, X26 and X24 are greater than 0.675. F can be interpreted as the economic level factor of the life quality.
The factor loaded coefficient value of F on the X19, X8, X17, X2 and X6 varia-
bles are greater than 0.566. F2 can be interpreted as the education development level factor of the life quality.
The factor loaded coefficient value of F3 on the X22, X25 and X15 variables are
greater than 0.627. F3 can be interpreted as the
social equality and civilization factor of the life quality. The factor loaded coefficient value of F4 on the X18 and X23 variables are great-
er than 0.706. F4 can be interpreted as the medical level factor of the life quality.
The factor loaded coefficient value of F5 on the X21 and X16 variables are greater than 0.628. F5 can be interpreted as the public
facilities factor of the life quality.
The factor loaded coefficient value of F6 on the X13 and X12 variables are greater than 0.640. F6 can be interpreted as the con-
Table 6
Component Score Coefficient Matrix
Component
1 2 3 4 5 6 7
X1 .176 .038 -.291 -.086 .125 -.056 -.027
X2 -.003 .167 -.125 .137 .243 -.099 -.226
X3 .139 -.171 .130 .068 .015 -.079 .019
X4 .121 -.036 .068 -.081 .114 .185 -.244
X5 .144 -.138 .073 .028 .114 -.014 .010
X6 -.074 .134 .035 .158 .035 .134 .128
X7 .123 -.063 .045 -.008 -.013 .127 -.085
X8 -.062 .248 -.032 .086 -.033 -.165 -.041
X9 .027 .081 .063 .040 -.096 .142 -.076
X10 .163 -.069 -.103 -.036 -.099 -.092 .048
X11 .130 -.036 -.041 .004 .054 .024 .015
X12 .025 -.218 .116 -.026 -.060 .377 .146
X13 -.029 .053 .018 -.023 .034 .524 -.146
X14 -.005 -.003 -.040 .084 -.075 -.082 .577
X15 .008 -.024 .249 -.135 -.020 .008 .189
X16 -.044 -.021 -.066 .003 .306 -.094 .401
X17 -.109 .310 -.048 -.024 -.165 .145 .154
X18 -.063 -.029 .047 .373 -.107 .009 .086
X19 -.085 .374 -.110 -.202 -.178 .050 .118
X20 .152 -.087 -.067 .086 -.100 -.261 .063
X21 .065 -.114 -.044 .036 .604 .001 -.131
X22 -.069 -.137 .496 .163 -.046 .052 -.069
X23 .037 -.061 -.046 .307 .115 -.098 .004
X24 .053 .084 .043 -.132 -.017 -.005 .048
X25 -.013 -.025 .354 -.167 -.062 .049 -.009
X26 .072 .068 -.042 .017 .056 -.051 -.054
cept of marriage and employment factor of the life quality.
The factor loaded coefficient value of F7 on the X14 variable is greater than 0.774. F can be interpreted as the pub lic natural resource factor of the life quality.
According to Table 6 we can get 7 fac- as follows.
tor score functions and they can be written
F = 0.176X1 -0.003X2 + 0.139X3 +......+ 0.053X24 - 0.013X25 + 0.072X26
F = 0.038X + 0.167X2 - 0.171X +......+ 0.084X24 - 0.025X25 + 0.068X26
F = -0.027X1 -0.226X2 + 0.019X3 +......+ 0.048X24 - 0.009X25 - 0.054X26
SPSS will automatically compute the contribution rate of the 7 factor variables as
score of the 31 samples on the 7 factor vari- weight, we can get the comprehensive index
ables, and the score could be written as vari- and the corresponding order. ables fac1_1 - fac1_7. Using the variance
index = (fac1_1 * A + fac1_2 * A2 + fac1_3 * A3 + fac1_4 * A4 + fac1_5 * As +
fac1_6 * A6 + fac1_7 * A)/ A '
The result is in the following table 7. five top data font is bold and shading, and
It is noticed that, from the second col- the after five data font is italic and shading.
umn to the tenth column in the table 7, the
Table 7
Factor variable score, comprehensive index and the order
facl 1 fac1 2 fac1 3 fac1 4 fac1 5 fac1 6 fac1 7 index order
Beijing 1.24463 2.82751 .56452 1.68128 -.40320 2.82671 -.84415 1.328 2
Tianjin .20103 2.36429 -.42804 -.45003 -.63492 -.42306 1.95087 .400 6
Hebei -.91740 -.30341 1.14315 .00437 .06008 -.06429 1.51074 -.312 24
Shanxi -.84147 -.56207 2.06963 .71422 -1.04914 .88975 .38017 -.222 17
Inner Mongolia .02410 -.30520 .32852 1.34423 2.34332 -.99833 -.82717 .175 8
Liaoning .13157 .55613 .65033 -.16145 -.39677 -.65644 -.16920 .143 10
Jilin -.65772 .62251 1.07114 1.33682 -.31672 -1.55138 -.97672 -.138 14
Heilongjia ng -.77880 .94617 -.11347 -.20696 -.32850 -1.95220 -.48250 -.413 26
Shanghai 3.91069 -.83205 -.25111 .35376 -2.06727 -1.84552 -.31544 1.619 1
Jiangsu .56448 .39387 .03700 -.71639 .94247 .55082 .99886 .409 5
Zhejiang 1.26806 .12832 .06301 .66476 .96657 .24707 .72239 .823 4
Anhui -.14799 -.42454 -.44621 -.88051 .53662 -.32591 .92250 -.206 15
Fujian .76008 -.55707 .15273 -.33017 .05217 .18641 .51297 .321 7
Jiangxi -.33313 -.80265 .22429 -.97401 .30687 .37019 .29024 -.301 23
Shandong -.15146 -.02412 .45774 .13240 1.97847 .18191 1.25922 .165 9
Henan -.61090 -.12966 .70889 -.63922 -.58920 .86812 -2.00856 -.393 25
Hubei -.35126 .64303 -.66062 -.98189 -.41097 -.27216 .53720 -.251 19
Hunan -.31153 -.39032 .62650 -.82132 -1.02730 -.42540 .11072 -.296 22
Guangdong 1.74340 -1.10276 .24941 .47033 1.45773 1.26356 -.57312 .898 3
Guangxi .04066 -1.39968 .54600 -1.11376 -.21466 .43687 .13786 -.215 16
Hainan .50326 -.61375 -1.09637 -1.30342 -.24127 1.02498 1.37640 .023 11
Chongqing .18176 1.17182 -1.33240 -1.43113 1.67532 -.78442 -1.68808 -.021 12
Sichuan -.21835 .52622 -.56649 -1.13076 -.15837 -.38306 -1.43780 -.286 21
Guizhou -.41302 -1.21459 .16833 -.70658 -.43383 .64265 -1.80822 -.512 30
Yunnan -.49845 -.83446 -.39081 .33179 -.77747 -.17762 .17823 -.433 27
Tibet -1.01937 -.84906 -4.03986 2.06546 -.58461 .78432 -.00920 -.888 31
Shaanxi -.70463 1.43150 -.19598 -.85673 -.63032 .51246 .24177 -.239 18
Gansu -.68693 -.35681 -.06559 -.43439 -.10680 .99821 -1.14396 -.456 29
Qinghai -.86789 -.90348 .20872 1.58690 -1.23550 .21575 .68959 -.438 28
Ningxia -.27072 -.57428 .29662 1.15993 1.52321 -.92148 .28701 -.026 13
Xinjiang -. 79268 .56859 .02041 1.29246 -.23600 -1.21855 .17738 -.258 20
6. Analysis and conclusion
According to the score value of the 31 samples on the 7 individual score factor, we can separately explain them in detail.
First of all, we can rank the value of the score variable fac1_1 of the factor variable F and get the result. The top five provinces are Shanghai, Guangdong, Zhejiang, Beijing and Fujian. These are Municipality directly under the central government or the eastern coastal economically developed provinces. The after five provinces are Xinjiang, Shanxi, Qinghai, Hebei and Tibet. These are less developed provinces in the West and central provinces.
Secondly, we can rank the value of the score variable fac1_2 of the factor variable
F and get the result. The top five provinces are Beijing, Tianjin, Shaanxi, Chongqing and Heilongjiang. The development of education in these provinces is better. The after five provinces are Tibet, Qinghai, Guangdong, Guizhou and Guangxi. And the development of education in these provinces is relatively low.
Thirdly, we can rank the value of the score variable fac1_3of the factor variable
F and get the result. The top five provinces are Shanxi, Hebei, Jilin, Henan and Liao-ning. The social equality and civilization in these provinces is better. The after five provinces are Sichuan, Hubei, Hainan, Chongqing and Tibet. And the social equality and civilization in these provinces is relatively low.
Fourthly, we can rank the value of the score variable fac1_4 of the factor variable
F4 and get the result. The top five provinces are Tibet, Beijing, Qinghai, Inner Mongolia and Jilin. The medical condition in these provinces is better. The after five provinces are Hubei, Guangxi, Sichuan, Hainan and Chongqing. Medical condition in these provinces is poor.
Fifthly, we can rank the value of the score variable fac1_5of the factor variable
F and get the result. The top five provinces are Inner Mongolia, Shandong, Chongqing, Ningxia and Guangdong. Public facilities in these provinces are better. The after five provinces are Yunnan, Hunan, Shanxi, Qinghai and Shanghai. The public facilities in these provinces are poor.
Sixthly, we can rank the value of the score variable fac1_6 of the factor variable
F and get the result. The top five provinces are Beijing, Guangdong, Hainan, Gansu and Shanxi. These provinces have the better concept about the marriage and employment. The after five provinces are Inner Mongolia, Xinjiang, Jilin, Shanghai and Heilongjiang. These provinces have the lower concept about the marriage and employment.
Seventhly, we can rank the value of the score variable fac1_7 of the factor variable
F and get the result. The top five provinces were Tianjin, Hebei, Hainan, Shandong and Jiangsu. The public natural resource in these provinces is better. The after five provinces are Gansu, Sichuan, Chongqing, Guizhou and Henan. The public natural resource in these provinces is relatively poor.
According to the result of the comprehensive index, the top five regions are Shanghai City, Beijing City, Guangdong Province, Zhejiang Province and Jiangsu Province. Their comprehensive indexes separately are 1.619, 1.328, 0.898, 0.823 and 0.409. Among these areas, Shanghai is the most economically developed municipalities in China, Beijing is China political center, and Guangdong, Zhejiang and Jiangsu are China most economically developed eastern coastal provinces. The top five regions totally match the corresponding characteristics of these regions in China.
According to the result of the comprehensive index, the after five regions are Yunnan Province, Qinghai Province, Gansu Province, Guizhou Province and Tibet Province, province. Their comprehensive indexes separately are -0.433, -0.438, -0.456, -0.512 and
-0.888. These provinces are located in the southwest of China, or in the northwest of China. The economic level is low. Ranking and the actual situation is also very consistent for these areas. It should be pointed out that the comprehensive index is negative, which indicates that the life quality in these areas is lower than the average level of the whole China. (Comprehensive index of average level equal 0)
In addition, it is noticed that the position of the index of the Guangxi province is in the median among the 31 provinces. Its comprehensive index is -0.215. The value is less than 0, so its urban resident's life quality is lower than the average level of the whole China. The result means that the urban resident's life quality level of the whole country is not more than the average level and various resources are inclined to the Municipality directly under the central government and the eastern coastal provinces.
7. Suggestion and expectation
According to the above analysis, it can be seen that the life quality ranking of urban residents in different provinces of China in 2013 is basically in line with the area's characteristic and the national conditions of our country. Eastern area and central municipalities are on the top, Middle area and northeast areas are in the median, and the West area is the lowest. It is necessary for the country to further strengthen the economic investment and the policy contribution in the central and western regions from the national level. Otherwise, with the further development of the economy, the gap will be further expanded, which is not conducive to the progress and development of harmonious society.
In terms of individual indicators, the quality of life in some provinces may be high, but it is not ideal in some indicators. For example, the low level of education development in Guangdong does not accord with the quality of life of the whole. The indicators of public facilities and the concept of marriage and employment of Shanghai
are very low. It also does not conform to the life quality of the first comprehensive ranking.
The quality of life in some provinces may be a low overall, but some indicators are more prominent. For example, medical condition in Tibet and Qinghai is better than most other areas. The index of the concept of marriage and employment in Gansu is better than most other areas. For the provinces that the comprehensive index is not matched with the individual indicator, they can improve the quality of life by strengthening the weak link.
All in all, the country also needs to further deepen the reform and strengthen its weak link, so that the life quality of majority people can been improved.
Of course, this paper still has some deficiencies. The data indicators perhaps are not very comprehensive and objective. The interpretation of indicators may not be very clear and appropriate. The method of the Data analysis is relatively simple. The results are not completely satisfactory. All these need to be amended and added in the future research.
REFERENCES
1. National Bureau of Statistics of the People's Republic of China. China Statistical Yearbook -2014 [M]. Beijing: China Statistics Press.2014.
2. Walt Whitman Rostow. The Stages of Economic Growth [M]. Beijing: China Social Sciences Press, 2010.
3. Zha qi feng,Xu wen song. Comprehensive evaluation study of Jiangsu city residents' living qua lity [J]. Journal of Jiangsu University (social sciences). 2003(4), 106-109.
4. Zhang li hua,Du hong wei,Liu xue qin. The application of factor analysis in evaluating the living quality of urban residents[J]. Journal of Hebei Polytechnic University (social science edition). 2008(1), 28-32.
5. Liu xin hua. Necessity and software operation of positive management in factor analysis [J]. Journal of Chongqing Institute of Technology(Natural Sc i-ence).2009(9),152-155.
6. Zheng zong sheng,Wu su yao,He chuan qi. The comparison of the quality of life of the 120 countries
in the world[J]. Theory and Modernization. 2006(4),15-20.
7. Hu fang. An analysis of index of life quality of the Netherlands and its revelation to China [J]. Journal of Guangdong University of Technology (social sciences edition). 2007(1), 20-22.
8. Zhou chang cheng,Yuan hao. A global vision of weighted distribution in construction of comprehensive indices of living quality[J]. Jiang Hai Academic Journal. 2002 (1), 94-99.
9. Bao yue hua,Chen qiang. Indicator system of quality of urban development from perspective of urban functions[J]. Journal of Tongji University (Natural science). 2011(5), 778-784.
10. Feng xiao tian,Yi song guo. The quality of family life in urban areas: the indicators and the structure[J]. Sociological Studies. 2000 (4), 107-118.
11. Li sha.Germany living quality index system research and enlightenment [J]. The World of Survey and Research. 2011(1), 60-64.
12. Fan bo nai. The evaluation on quality of citizen's living of China [J]. Journal of Zhejiang University (Humanities and Social Sciences). 2006(4), 122-130.
13. European Statistical Laboratory [EB/OL].http ://esl.jrc.it.
14. Zhang wen tong. SPSS statistical classification advanced tutorial [M]. Beijing: Higher Education Press, 2004.
15. National Bureau of Statistics of the People's Republic of China. China Statistical Yearbook -2013[M]. Beijing: China Statistics Press.2013.
Рукопись поступила в редакцию 3.08.2015.
FACTOR ANALYSIS METHOD OF CHINESE PROVINCES POPULATION LIFE QUALITY
EVALUATION
Xiang Xiao Gang
In the article the statistical multifactor analysis method application is substantiated and the indicator system for the China different provinces population life quality evaluation is determined. The experimental calculations are carried out on the basis of the PRC's 31 regions national statistical data for the period of 2013. Compar ative analysis shows that there are the obvious regional differences in the quality of life of Central, Eastern and Western provinces population. The Municipalities Index calculated and its local components can be used in region ranking and in social and economic policy directions justification.
Keywords: multifactor analysis, life quality, gross domestic product, municipalities index, the region ranking.
