THE PERFORMANCE OF HIGH RESOLUTION NEUTRON POWDER DIFFRACTOMETE

Introduction

Corporate social responsibility refers to a commitment to improving the welfare of people who involved, general public through responsiveness to business practices, and the contribution of company resources. CSR entails raising funds for charity organizations and aligning company growth objectives with ethics in order to foster a sense of support and community that benefits all parties involved as well as society as a whole (Ahmed, Zehou, Raza, Qureshi, & Yousufi, 2020).

Annual reports, sustainability reports, and company websites can all show how businesses are responding to CSR. However, there is still a lack of interest from the academic community in conducting CSR research, particularly in developing countries (Gunawan, 2019). Because Indonesia faces several fundamental challenges in achieving the Sustainable Development Goals (SDGs) that must be met by2030, the growth of CSR is critical (Gunawan, 2019). Among the challenges are poverty reduction, human rights, environmental health and safety issues, pollution and waste disposal, socio-political insecurity, and the need for foreign investment (Raynard & Forstater, 2002; Goyal, Van Der Leij, & Moraga-Gonzï¿½lez, 2006; Trisnawati, Sembel, Gunawan, & Waluyo, 2017). Corporate social responsibility represents a vital development to contribute social and environmental protection. Furthermore, according to stakeholder theory, corporations that adopt socially responsible practices will benefit from reciprocity in the form of more profitable investments and incentives for enterprises to engage more in CSR initiatives. As a result, the purpose of this study is to see if Corporate Social Responsibility (CSR) is linked to financial success and can be expressed as a function of four social variables: GRI participation, KEHATI index, CG compliance, and Global Compact awardees (GC). Second, the Social Behavior Index was utilized to determine the negative relationship between financial performance and corporate social responsibility, which includes four social variables as equally weighted components. Three economic variables or ratios are used to indicate financial performance in both cases: ROA (return on assets), ROE (return on equity), and Tobin's Q.

Stakeholder theory in this study claims the company's resources in implementing the right company appropriately. Where companies that do not carry out social responsibility costs can be significant and represents a financial cost that reduces profits. Some investments are more profitable and create incentives for firms to boost investment in CSR initiatives if the company releases a socially responsible policy, Margolis and Walsh (2003) came to the same conclusion. Regardless of whether CSR is an independent variable (109 studies) or the dependent variable, the results reveal the strongest positive associations (18 studies).

The majority of studies in the CSR and FP fields have been completed, but the results are still limited. The reasons for this are that (1) there is no standard approach that can be used as a benchmark for comparing studies, and (2) there is no rigorous method for measuring CSR returns. (Karagiorgos, 2010). (Rodriguez-Fernandez, 2016)ï¿½ compares these issues in the context of developed countries, and the data he collects is for a year. While this research is in developing countries, data from the past three years with a larger sample size was used. Thus, hopefully this research can provide more in-depth information.ï¿½ Based on the researchers acknowledgement, there is no research has conducted a two-way relationship regardless of CSR and FP independent or dependent variables as was done by (Gunawan, 2019).

As a result, the goal of this study is to fill in the gaps in the Indonesian literature and emphasize the fundamental linkages between CSR and FP. This study is likely to persuade business boards that social policy should be a key component of their overall strategy. This research should be useful to all interest groups claiming or threatening the company's well-being, according to stakeholder perspectives. Additional research is needed to apply various research approaches to provide a better understanding and description of CSR studies in Indonesia, such as CSR studies on corruption, due to the country's uniqueness and significant number of problems (Joseph et al., 2016) (Gunawan, 2019). As a result, this research proposes two research questions:

RQ1: Does Financial Performance Influence Corporate Social Responsibility?

RQ2: Does Corporate Social Responsibility affect financial performance?

The major goal of this study is to answer these two concerns by doing statistical testing of the association between CSR and FP. The results obtained by other authors (McGuire, Sundgren, & Schneeweis, 1988) (Molina & Clemente, 2010); (Choi, Kwak, & Choe, 2010); (Karagiorgos, 2010); (Jo & Harjoto, 2011); (Rodriguez-Fernandez, 2016) have concluded that the two research questions which are expected to have positive signs have been achieved. This means that all are related to one another, regardless of the independent or dependent CSR and FP variables.

This article is divided into five sections: section 1 is a general introduction, section 2 is a theoretical background and literature review, section 3 is the research hypotheses, and section 4 is the methodology. Section 2 introduces CG, CSR, and FP as the three main pillars that underpin our conceptual model. This section, under "Proposed models and hypothesis formulation," lays forth two main assumptions about whether CSR explains FP and vice versa.

Method

Research design

This study formulated six multivariate regression models for this empirical analysis three for Model 1 and three for model 2. To analyze the statistics (SPSS) widely used in empirical research.

Sample

The sample consists of Indonesian companies listed on the Indonesia Stock Exchange in 2017, 2016, 2015 in the form of random data from those listed on the Indonesia Stock Exchange. The final sample total from table 3 is 855 over three years. This information is obtained from the Indonesia Stock Exchange website (https://www.idx.co.id), financial summary, and audited consolidated financial statements because different accounting systems will cause a lack of homogeneity in the calculation of financial ratios. Jackling and Johl (2009) conducted a previous study on the financial data of the same company. Financial variable data was obtained from the SABI6 database and checked with the AMADEUS7 database.

Table 1

Industry observation percentage

Industry	Observation	Percentage
Manufacture	339	50%
Services	351	50%
Mining	165	100%

Measurement

To assess the social dimension of the company, the following aspects have been taken into account:

GRI: based on the GRI index assessment, the numerical ratings are: A+: 1, A: 0.9, B+: 0.8, B: 0.7, C+: 0.6, C: 0.5 . A + if the company meets the criteria G4, A for the company still using GRI G3.1, B + for GRI G3, B for GRI G2, C + for GRI G1, and C for the GRI standard. Information is collected from the GRI website: https://database.globalreporting.org.

KEHATI: Value 1 if the company belongs to KEHATI, and 0 otherwise.

COMPL RECOM: calculated by dividing the number of satisfied recommendations by the total number implemented. The researchers recommend assigning a score of 1 to the recommendation that was met, 0.5 to the part that was met, and 0 points for what was described but not met. The data has been obtained from the corporate governance table (level of follow-up to the AGM) of the company's Annual Report. GC: Take a value of 1 if the company has signed the Global Compact, and 0 otherwise. Data is processed from the Global Compact Network webpage: https://www.unglobalcompact.org

The INDEX used in Model 2 is calculated using the same weighted sum of the four variables as (Belu & Manescu, 2013) were all given the same weighted average of the four CSR variables used to assess CSR.

Results And Discussion

Statistical analysis performed on the initial sample did not yield valid results. The correlations show shallow values and neither support the model that offers an agreed explanation. The researchers decided to separate each sample group from the sector of each company.

Table 2 shows the descriptive statistics for the chosen sample. The percentage of people that followed the CG guidelines ranged from 100% to 75%, with a mean of 91.9 percent. KEHATI is used by 5.3 percent of companies, and 1.5 percent have signed a GC. In terms of the GRI report, it equates to 0.52. Finally, the index variable ranges from 37.5 percent to 93.8 percent, with an average of 39.4 percent. One company signed the GC and GRI to create the KEHATI sustainability report, with a maximum value of 93.8 percent. The maximum and minimum ROE values for the financial variable are 11.04 and 0. There are 1.43 and 0 for ROA, respectively. For Tobin's Q are 5.66 and 0.69. The mean values were 0.29, 0.08 and 1.71.

Normality test refers to see whether the residual value is normally distributed or not (Ghozali, 2013). In Table 3, it is directed directly to the manufacturing sector in model 1 and model 2 the results show the Asymp The Sig test values for each equation ROA, ROE, Qtobin, and INDEX are above the specified alpha (0.05). It can be concluded that H0 is accepted, which means data

normally distributed in both models. There are no fine results for the services and mining sectors.

The autocorrelation test determines whether there is a link between the error in period t and the error in period t-1 in a linear regression model (previous). Table 3 in the Durbin-Watson column shows autocorrelation in Model 1 ROE 1.66. Based on the autocorrelation test criteria, this result does not meet the criteria du <d < 4 - du where du table is 1.8199. The equation concludes that H0 is rejected for the ROE variable, which means there is a positive or negative autocorrelation in the data. Meanwhile, in model 2 du table 1.8094 the results meet the criteria 1.8094 < 2.102 <4 - 1.8094 which means that H0 is accepted because there is no positive or negative autocorrelation in model 2.

The heteroscedasticity test examines if the variance of one observation's residue and the variance of another observation's residue in the regression model is unequal. A suitable regression model is homoscedasticity, or the absence of heteroscedasticity (Ghozali, 2013). In the sig column of each variable in Table 4, model 1 and model 2 have a significance value greater than 0.05. It can be concluded that H0 is acceptable, implying that the data is homoscedastic.

Table 5 demonstrates the F statistical test, which determines if all of the independent variables in the model have a combined effect on the dependent variable (Ghozali, 2013). The significance level is 0.05. Based on the F test criteria, the regression model can be used for model 1:

GRI, KEHATI, COMPLRECOM, and GC simultaneously affect ROA.

GRI, KEHATI, COMPLRECOM, and GC have no effect on ROE.

GRI, KEHATI, COMPLRECOM and GC simultaneously affect QTOBIN.

As for model 2: ROE, ROA, and QTOBIN simultaneously affect INDEX.

The statistical test t-value (Asymp Sig) 0.05 demonstrates the extent to which one explanatory/independent variable can explain the variation of the dependent variable on its own (Ghozali, 2013). Table 6 shows that the results of multiple linear regression using the regression equation model 1:

GRI partially have no effect on ROA.

KEHATI partially has no effect on ROA.

COMPLRECOM partially affects ROA.

GC partially has no effect on ROA.

GRI partially does not affect ROE.

KEHATI partially does not affect ROE.

COMPLRECOM partially does not affect ROE.

GC partially does not affect ROE.

GRI partially did not affect QTOBIN.

KEHATI partially did not affect QTOBIN.

COMPLRECOM partially affects QTOBIN.

GC partially affects QTOBIN.

Model 2:

ROE partially does not affect INDEX.

ROA partially does not affect INDEX.

QTOBIN partially affects INDEX.

The coefficient of determination (R2) is a statistic for measuring how effectively a model can explain independent variable fluctuations. The coefficient of determination is a number that ranges from 0 to 1. The ability of the independent variable to explain the dependent variable is limited if the R2 value is low. A value around one suggests that the independent variable has almost all of the information needed to predict changes in the dependent variable (Ghozali, 2013). Table 6 shows that the value of column R in our model 1 takes into account the ROA and QTOBIN variables. The correlation values are 0.196 and 0.274, which means that the relationship between GRI, KEHATI, COMPLRECOM, GC on ROA and QTOBIN is weak but definite. For model 2 the value is 0.217 which means that the relationship between ROE, ROA, and QTOBIN on INDEX is weak but definite.

Results And Discussion

normally distributed in both models. There are no fine results for the services and mining sectors.

ï¿½The heteroscedasticity test examines if the variance of one observation's residue and the variance of another observation's residue in the regression model is unequal. A suitable regression model is homoscedasticity, or the absence of heteroscedasticity (Ghozali, 2013). In the sig column of each variable in Table 4, model 1 and model 2 have a significance value greater than 0.05. It can be concluded that H0 is acceptable, implying that the data is homoscedastic.

GRI, KEHATI, COMPLRECOM, and GC simultaneously affect ROA.

GRI, KEHATI, COMPLRECOM, and GC have no effect on ROE.

GRI, KEHATI, COMPLRECOM and GC simultaneously affect QTOBIN.

As for model 2: ROE, ROA, and QTOBIN simultaneously affect INDEX.

GRI partially have no effect on ROA.

KEHATI partially has no effect on ROA.

COMPLRECOM partially affects ROA.

GC partially has no effect on ROA.

GRI partially does not affect ROE.

KEHATI partially does not affect ROE.

COMPLRECOM partially does not affect ROE.

GC partially does not affect ROE.

GRI partially did not affect QTOBIN.

KEHATI partially did not affect QTOBIN.

COMPLRECOM partially affects QTOBIN.

GC partially affects QTOBIN.

Model 2:

ROE partially does not affect INDEX.

ROA partially does not affect INDEX.

QTOBIN partially affects INDEX.

Table 2

Descriptive statistic

	N	Minimum	Maximum	Mean	Std. Deviation
GRI	339	0,5	1	0,5176991	0,09252867
KEHATI	339	0	1	0,0530973	0,22455905
COMPLRECOM	339	0,75	1	0,9188295	0,11434894
GC	339	0	1	0,0147493	0,12072578
INDEX	339	0,375	0,9375	0,3939227	0,07584116
LNASSET	339	0	34,5693	25,8595257	5,12990277
ROE	339	0	11,0404	0,2901239	1,01028733
ROA	339	0	1,4327	0,080469	0,13007758
QTOBIN	339	0,6925	5,6625	1,7060265	1,3188018
Valid N (listwise)	339

Table 3

Normality Test

(MODEL 1) ROA		all companies	manufacture	services	mining
N		843	339	351	165
Normal Parameters^a,b	Mean	0	0	0	0
	Std. Deviation	0,394782	0,12754379	0,54067309	0,36908528
Most Extreme Differences	Absolute	0,349	0,218	0,402	0,296
	Positive	0,349	0,218	0,402	0,296
	Negative	-0,339	-0,204	-0,394	-0,238
Test Statistic		0,349	0,218	0,402	0,296
Asymp. Sig. (2-tailed)		,000^c	,200^c	,000^c	,200^c
ROE
N		843	339	351	165
Normal Parameters^a,b	Mean	0	0	0	0
	Std. Deviation	3,80243607	1,0053733	0,64623164	8,34596514
Most Extreme Differences	Absolute	0,411	0,378	0,382	0,4
	Positive	0,411	0,378	0,358	0,4
	Negative	-0,402	-0,34	-0,382	-0,335
Test Statistic		0,411	0,378	0,382	0,4
Asymp. Sig. (2-tailed)		,061^c	,062^c	,000^c	,000^c
QTOBIN
N		843	339	351	165
Normal Parameters^a,b	Mean	0	0	0	0
	Std. Deviation	1,36052248	1,26845027	1,36465047	1,39224876
Most Extreme Differences	Absolute	0,206	0,191	0,19	0,191
	Positive	0,206	0,191	0,19	0,191
	Negative	-0,125	-0,18	-0,101	-0,104
Test Statistic		0,206	0,191	0,19	0,191
Asymp. Sig. (2-tailed)		,145^c	,071^c	,182^c	,121^c
ï¿½ MODEL 2
N		843	339	351	165
Normal Parameters^a,b	Mean	0	0	0	0
	Std. Deviation	0,07606069	0,07403285	0,07741491	0,0734273
Most Extreme Differences	Absolute	0,415	0,408	0,461	0,473
	Positive	0,415	0,408	0,461	0,473
	Negative	-0,262	-0,203	-0,303	-0,334
Test Statistic		0,415	0,408	0,461	0,473
Asymp. Sig. (2tailed)		,120^c	,200^c	,000^c	,000^c

Table 4

Autocorelation test

(MODEL 1)
	R	R Square	Adjusted R Square	Std. Error of the Estimate	Durbin-Watson
ROA	,196^a	0,039	0,024	0,12849776	1,909
ROE	,099^a	0,01	-0,005	1,01289302	1,66
QTOBIN	,274^a	0,075	0,061	1,27793769	2,045
(MODEL 2)
INDEX	,217^a	0,047	0,036	0,07447484	2,102

Table 5

Heteroscedastic Test

(MODEL 1)		Unstandardized Coefficients		Standardized Coefficients	t	Sig.
		B	Std. Error	Beta
ROA (1)	(Constant)	0,263	0,068		3,876	0
	GRI	-0,009	0,071	-0,008	-0,131	0,896
	KEHATI	-0,023	0,03	-0,048	-0,785	0,433
	COMPLRECOM	-0,191	0,052	-0,201	-0,689	0,402
	GC	0,095	0,055	0,105	1,725	0,085
	LNASSET	-0,001	0,001	-0,03	-0,559	0,577
ROE (1)	(Constant)	1,403	0,598		2,346	0,02
	GRI	-0,195	0,623	-0,019	-0,313	0,754
	KEHATI	-0,133	0,261	-0,032	-0,511	0,61
	COMPLRECOM	-1,062	0,457	-0,128	-0,324	0,453
	GC	0,326	0,484	0,042	0,674	0,501
	LNASSET	0,001	0,01	0,004	0,068	0,946
QTOBIN (1)	(Constant)	0,231	0,524		0,442	0,659
	GRI	-1,274	0,545	-0,139	-336	0,789
	KEHATI	0,464	0,228	0,123	0,232	0,435
	COMPLRECOM	0,938	0,4	0,127	0,346	0,563
	GC	1,076	0,423	0,153	0,54	0,789
	LNASSET	0,018	0,009	0,11	0,038	0,877
(MODEL 2)
INDEX (1)	(Constant)	-0,022	0,017		-1,26	0,209
	LNASSET	0,001	0,001	0,066	1,304	0,193
	ROE	-0,002	0,004	-0,026	-0,434	0,665
	ROA	0,102	0,03	0,203	0,427	0,602
	QTOBIN	0,016	ï¿½ï¿½ 0,003	0,315	0,189	0,897

Table 6

Statistic Test F

(MODEL 1)
	Sum of Squares		df	Mean Square	F	Sig.
ROA (1)	Regression	0,221	5	0,044	2,672	,022^b
	Residual	5,498	333	0,017
	Total	5,719	338
ROE (1)	Regression	3,348	5	0,67	0,653	,660^b
	Residual	341,642	333	1,026
	Total	344,99	338
QTOBIN (1)	Regression	44,032	5	8,806	5,392	,000^b
	Residual	543,831	333	1,633
	Total	587,863	338
(MODEL 2)
INDEX (1)	Regression	0,092	4	0,023	4,129	,003^b
	Residual	1,853	334	0,006
	Total	1,944	338

Table 7

Statistic Test T


(MODEL 1)		Unstandardized Coefficients		Standardized Coefficients	t	Sig.
		B	Std. Error	Beta
ROA (1)	(Constant)	0,182	0,081		2,237	0,026
	GRI	0,046	0,085	0,033	0,542	0,588
	KEHATI	0,015	0,036	0,026	0,426	0,671
	COMPLRECOM	-0,143	0,062	-0,125	-2,29	0,023
	GC	0,121	0,066	0,113	1,843	0,066
	LNASSET	0	0,001	0,004	0,079	0,937
ROE (1)	(Constant)	0,832	0,642		1,296	0,196
	GRI	0,047	0,669	0,004	0,07	0,944
	KEHATI	-0,031	0,28	-0,007	-0,112	0,911
	COMPLRECOM	-0,657	0,491	-0,074	-1,339	0,182
	GC	0,499	0,52	0,06	0,96	0,338
	LNASSET	0,001	0,011	0,006	0,111	0,911
QTOBIN (1)	(Constant)	0,379	0,81		0,467	0,641
	GRI	-0,691	0,844	-0,048	-0,819	0,413
	KEHATI	0,652	0,353	0,111	1,845	0,066
	COMPLRECOM	1,533	0,619	0,133	2,475	0,014
	GC	2,246	0,655	0,206	3,426	0,001
	LNASSET	0,008	0,014	0,031	0,587	0,557
(MODEL 2)
INDEX (1)	(Constant)	0,357	0,021		16,639	0
	LNASSET	0,001	0,001	0,039	0,725	0,469
	ROE	-0,001	0,005	-0,013	-0,208	0,836
	ROA	0,063	0,037	0,108	1,711	0,088
	QTOBIN	0,01	0,003	0,175	3,248	0,001

After conducting statistical research on manufacturing companies, the theory that was originally proposed can be confirmed. In the mining and services industry, no excellent results were identified. Hypotheses H1.3 and H1.4 for ROA and QTOBIN are valid based on Model 1. According to Model 2, the second hypothesis for the QTOBIN variable can be stated that organizations that get the greatest outcomes in the QTOBIN variable have a higher Social Behavior Index value and, as a result, are differentiated by adopting stronger overall CSR practices. We can state as other authors (McGuire et al., 1988; Charlo Molina and Moya Clemente, 2010; Choi et al., 2010; Karagiorgos, 2010; Harjoto and Jo, 2011; (Rodriguez-Fernandez, 2016) came to the conclusion that the analyzed connection showed the predicted positive indicators. Despite these encouraging findings, we acknowledge that our research has certain limitations. To begin with, the geographic and chronological scope is restricted. Second, a two-way relationship between the social and financial factors is created as a result of the feedback, and it is unclear which occurred first.

Conclusion

This bidirectional link in CSR-FP proved to be good in both directions, despite its weakness. As a result, we advise businesses to raise their CSR investment, which will lead to an increase in FP and beyond. Companies with more financial stability have a higher Social Behavior Index. This gives positive feedback, encouraging corporations to (1) invest financial resources in CSR initiatives. (2) Check to see how their CSR investments are benefiting their bottom line.

References

Ahmed, Mansoora, Zehou, Sun, Raza, Syed Ali, Qureshi, Muhammad Asif, & Yousufi, Sara Qamar. (2020). Impact of CSR and environmental triggers on employee green behavior: The mediating effect of employee well‐being. Corporate Social Responsibility and Environmental Management, 27(5), 2225ï¿½2239. Google Scholar

Belu, Constantin, & Manescu, Cristiana. (2013). Strategic corporate social responsibility and economic performance. Applied Economics, 45(19), 2751ï¿½2764. Google Scholar

Choi, Jong Seo, Kwak, Young Min, & Choe, Chongwoo. (2010). Corporate social responsibility and corporate financial performance: Evidence from Korea. Australian Journal of Management, 35(3), 291ï¿½311. Google Scholar

Ghozali, Imam. (2013). Ghozali, Imam. 2013. Aplikasi Analisis Multivariate dengan Program SPSS. Semarang: Badan Penerbit UNDIP. Analisis Multivariate Dengan Program SPSS. Google Scholar

Goyal, Sanjeev, Van Der Leij, Marco J., & Moraga-Gonzï¿½lez, Josï¿½ Luis. (2006). Economics: An emerging small world. Journal of Political Economy, 114(2), 403ï¿½412. Google Scholar

Gunawan, Juniati. (2019). The development of corporate social responsibility in accounting research: evidence from Indonesia. Social Responsibility Journal. Google Scholar

Jackling, Beverley, & Johl, Shireenjit. (2009). Board structure and firm performance: Evidence from Indiaï¿½s top companies. Corporate Governance: An International Review, 17(4), 492ï¿½509. Google Scholar

Jo, Hoje, & Harjoto, Maretno A. (2011). Corporate governance and firm value: The impact of corporate social responsibility. Journal of Business Ethics, 103(3), 351ï¿½383. Google Scholar

Joseph, Corina, Gunawan, Juniati, Sawani, Yussri, Rahmat, Mariam, Noyem, Josephine Avelind, & Darus, Faizah. (2016). A comparative study of anti-corruption practice disclosure among Malaysian and Indonesian Corporate Social Responsibility (CSR) best practice companies. Journal of Cleaner Production, 112, 2896ï¿½2906. Google Scholar

Karagiorgos, Theofanis. (2010). Corporate social responsibility and financial performance: An empirical analysis on Greek companies. Google Scholar

Margolis, Joshua D., & Walsh, James P. (2003). Misery loves companies: Rethinking social initiatives by business. Administrative Science Quarterly, 48(2), 268ï¿½305. Google Scholar

McGuire, Jean B., Sundgren, Alison, & Schneeweis, Thomas. (1988). Corporate social responsibility and firm financial performance. Academy of Management Journal, 31(4), 854ï¿½872. Google Scholar

Molina, M. J. Charlo, & Clemente, I. Moya. (2010). El comportamiento financiero de las empresas socialmente responsables. Investigaciones Europeas de Direcciï¿½n y Economï¿½a de La Empresa, 16(2), 15ï¿½25. Google Scholar

Raynard, Peter, & Forstater, Maya. (2002). Corporate social responsibility: Implications for small and medium enterprises in developing countries. Google Scholar

Rodriguez-Fernandez, Mercedes. (2016). Social responsibility and financial performance: The role of good corporate governance. BRQ Business Research Quarterly, 19(2), 137ï¿½151. Google Scholar

Trisnawati, Estralita, Sembel, Roy, Gunawan, Juniati, & Waluyo, Waluyo. (2017). Pengaruh Kualitas Manajer Pajak Terhadap Penghindaran Pajak Dengan Etika Machiavellian Sebagai Pemediasi. Jurnal Ekonomi, 22(3), 393ï¿½420. Google Scholar

(MODEL 1)
	R	R Square	Adjusted R Square	Std. Error of the Estimate
ROA (1)	,196^a	0,039	0,024	0,12849776
ROE (1)	,099^a	0,01	-0,005	1,01289302
QTOBIN (1)	,274^a	0,075	0,061	1,27793769
(MODEL 2)
INDEX (1)	,217^a	0,047	0,036	0,07447484