1. INTRODUCTION:

Microfinance: Concept and Definition

Although often used interchangeably, microfinance is quite different from micro credit. Micro financing, according to Conroy (Conroy JD, 2003) is the delivery of financial services to poor and low income households with limited access to formal financial institutions. Rolando ( Rolando GT, 2010) opines that microfinance is an excellent way of supporting entrepreneurs. Microfinance products are often accompanied by training in financial literacy and business management, while the later refers to small loan facility provided to poor people to motivate them to become self-employed (Remenyi and Quinones, 2000).

In simple words, micro credit includes credit activities only, but microfinance includes credit as well as non-credit activities. Micro credit is thus one aspect of microfinance. Micro credit refers to small loan programmes operated by non-profit organizations, i.e., these programs have no profit motives. By contrast, microfinance refers to profit-making ventures (Ledge Wood, 1999) supporting that microfinance ventures do not meet the definitional requirement of NGO.

Purpose of Microfinance:

Microfinance scheme is offered to people below poverty line who lack collateral and cannot take bank’s assistance for banking and allied services. The purpose of microfinance is to raise the earnings of low-class people and let them have access to deposits and loans. Microfinance is thus defined as formal schemes designed to improve the well-being of the poor through better access to saving services and loans. While both informal finance¹ and microfinance serve poor people who do not have bank accounts, informal finance derives from the grassroots, bottom-up demand of the poor for appropriate financial services, whereas microfinance derives from donor-driven, top-down supply (Schreiner, 2001). The task force on supportive policy and regulatory framework for microfinance (NABARD, 1999) defined microfinance as “provision of thrift, credit and other financial services and products of very small amounts to raise income levels of clients, and improve their living standards.” International Labour Organization (ILO) described microfinance as an economic development approach that involves providing financial services through institutions to low income clients. MIX (Microfinance Information Exchange) describes microfinance as “Microfinance services-as opposed to financial services in general- are retail financial services that are relatively small in relation to the income of a typical individual. In short, microfinance is the supply of loans, saving and other basic financial services to the poor, (CGAP).

¹Informal finance is defined as contract or agreements conducted without reference or recourse to the legal system to exchange cash in the present for promises of cash in the future.

2. LITERATURE REVIEW:

Youssef L. and Tkiouat M. (2017) have chosen 10 experts from the three chief Moroccan MFIs. The experts comprises mainly of consultants, market research executives, and credit officers. The author has collected the required information through face-to-face discussion and online questionnaires. The major points covered in questionnaire is assigning the weights to different steps of microfinance lending process and to selected indicators. In this case study, the author found that the Moroccan MFIs lending process is not suitable and different steps require more improvements. The author has also suggested that the improvement in microfinance process can lead to better results and supports in achieving their objectives.

Datar, SM. et. al. (2008) have done research on the performance of microfinance institutions of Morocco regarding their lending process. They have found that it does not mean that proper loan management leads to generating profits. The clients are using loan fund for purchasing more inventory instead of business expansion. And just holing more inventory does not mean that they are able to sale more units at a profit. Though they are able to sell the goods at a profit does not mean that they are able to support their business reinvestment, household requirements and loan reimbursements.

Mahanta et.al. (2012) have describes three different aspects of microfinance, first is about the growth of microfinance in India and other countries of the world, second is the role of NABARD and other national banks in growth of SHGs and Grameen banks the role of government in regarding policy framing for micro borrowers. In the study they have observed that if the program is pushed with capacity building to achieve poverty alleviation. The fund is used for personal consumption and social purpose only if required skill training is not provided to them hence it becomes important to provide skill based training like weaving, handicrafts, carpentry etc. including agriculture and non-agriculture. The clients should be provided technical and professional aid after loan sanctioning.

Badatya, K. C. (2006) has selected multi stage random sampling method for data collection. At first level, there were selection of districts from three regions namely Telengana, tribal belts of Telengana and Rayalseema. The total sample selected constituted 56 SHGs and 310 members selected from agencies like RRBs, commercial banks and cooperatives as also from different NGOs.The beneficiaries have reported change in their social lives after joining of SHGs. From total respondents, 75 percent of respondents were agree that there is improvement in their self-confidence, better role in decision-making and also increase teamwork among group members. The matured /older groups (87% for SHGs of 7 or more years, 69 per cent for SHGs of 4-6 years old) opined that they have experienced is improvement in their self-confidence, better role in decision-making and also increase teamwork among group members. In case of newly formed groups (53% for SHGs of 3 years old), 82 per cent of SHGs reported better literacy and education.

Malleswari (2010) defined empowerment as a process of change by which individuals or groups gain power and ability to take control over their lives. It involves well-being, access to resources, increased self-confidence, self-esteem, respect, participation in decision making, bargaining power, control over benefits and their own life. It can be viewed as a means of creating a social environment in 231 which one can take decisions and make choice either individually or collectively for social transformation.

3. RESEARCH METHODOLOGY:

OBJECTIVES:

i. Identify and examine the elements influencing the women empowerment

ii. Classification of factors affecting overall women empowerment

Research Framework:

The overall women empowerment was checked on 5 parameters namely 1) Loan Procedure- this factor demonstrates the different loan procedural stages and the issues faced by the respondents, 2) saving related matters – this includes the issues faced by the respondents at the time of maturity amount withdrawal, 3) use of loan amount- it indicates constrains in using of loan amount and its repayment matters, 4) autonomy – it indicates different matters related to women’s autonomy. 5) Change in women’s life – it describes different aspects of changes in women’s life. The response were scored on 5 point likert scale (1= strongly agree to 5= strongly disagree)

4. TESTS AND RESULTS:

RELIABILITY:

Mainly there are three types of Reliability. (A) Test-Retest (B) Split-half (C) Construct-reliability. Here, researcher try to understand whether construct/statements are frame are reliable or not? The construct reliability can be derived with Cronbach’s alpha, the tools is applied to understand the reliability aspects, if the value of the Cronbach alpha is more than 0.4, then it is acceptable. It is more than 0.6, then it is moderate and if it is more than 0.7, it is desirable. In this case, In this case, the coefficients range from 0.842 to 0.711 for all 5 factors which indicates the items have relatively high internal consistency and all statements are reliable.

Table 1.1 Cronbach’s alpha value

Cronbach’s alpha (α) value
Factor	Cronbach's Alpha	Cronbach's Alpha Based on Standardized Items	N of Items
LOAN PROCEDURE	0.842	0.804	7
SAVING	0.806	0.784	6
USE OF LOAN AMOUNT	0.721	0.598	9
AUTONOMY	0.767	0.746	8
EMPOWERMENT	0.711	0.709	11

DISCRIMINANT ANALYSIS:

Researcher wants to understand there are various independent factors (loan procedure, saving related matters, usage of loan amount, women’s autonomy and change in women’s life) having significance impact on women empowerment. Such type of analysis derived with the tools of multiple regression model but unfortunately this tools cannot classified all the factors into two parts like which factors and with what value creates high level of influence and which factors creates low level of influence. To address this issue, researcher has applied the techniques called the discriminant analysis. Here, researcher has taken the opinion of 627 respondents, first table of analysis case processing summary indicates this value. None of them are missing out. Therefore, all the observations researcher has received are valid responses.

Discriminate analysis divides the respondents into major two groups from which one group is responding likely towards the overall women empowerment while another groups responding unlikely towards the women empowerment. Here, the whole data of 627 respondents are is divided into two groups according to their responses.

Table 1.2 Group statistics

Group Statistics
CHANGES		Mean	Std. Deviation	Valid N (list wise)
CHANGES		Mean	Std. Deviation	Unweighted	Weighted
LOW	CLP1	16.04	3.170	168	168.000
	CSR1	15.83	2.984	168	168.000
	CUL1	21.54	2.543	168	168.000
	CAU1	18.37	2.821	168	168.000
	CWL1	23.21	2.021	168	168.000
HIGH	CLP1	20.41	4.558	459	459.000
	CSR1	19.91	4.158	459	459.000
	CUL1	26.00	4.604	459	459.000
	CAU1	24.60	4.533	459	459.000
	CWL1	32.11	5.327	459	459.000
Total	CLP1	19.24	4.651	627	627.000
	CSR1	18.82	4.277	627	627.000
	CUL1	24.80	4.598	627	627.000
	CAU1	22.93	4.978	627	627.000
	CWL1	29.72	6.115	627	627.000

The above table provides statistical evidence regarding significant differences between means of high women empowerment and low women empowerment for all independent variables. There are 459 respondents whose responses are likely towards women empowerment while there are 168 respondents whose responses are unlikely towards women empowerment.

1. H0: There is no significance difference in the discriminant score with respect to loan procedure of high and low women empowerment.

H1: There is significance difference in the discriminant score with respect to loan procedure of high and low women empowerment.

2. H0: There is no significance difference in the discriminant score with respect to saving related matters of high and low women empowerment.

H1: There is significance difference in the discriminant score with respect to saving related matters of high and low women empowerment.

3. H0: There is no significance difference in the discriminant score with respect to use of loan amount of high and low women empowerment.

H1: There is significance difference in the discriminant score with respect to use of loan amount of high and low women empowerment.

4. H0: There is no significance difference in the discriminant score with respect to women’s autonomy of high and low women empowerment.

H1: There is significance difference in the discriminate score with respect to women’s autonomy of high and low women empowerment.

5. H0: There is no significance difference in the discriminate score with respect to change in women’s life of high and low women empowerment.

H1: There is significance difference in the discriminate score with respect to change in women’s life of high and low women empowerment.

CLP denotes cumulative loan procedure statements. There are 07 statements under this category with 5 point scale for each statements makes total score of 35. The mean value is 19.24 indicates that majority of opinion are towards agree and strongly agree with std. deviation of 4.651. CSR denotes cumulative saving related statements. There are 06 statements under this category with 5 point scale for each statements makes total score of 30. The mean value is 18.82 indicates that majority of opinion are towards agree with std. deviation of 4.277. CUL denotes cumulative use of loan amount related statements. There are 09 statements under this category with 5 point scale for each statements makes total score of 45. The mean value is 24.80 indicates that majority of opinion are towards agree and strongly agree with std. deviation of 4.598. CAU denotes cumulative autonomy related statements. There are 08 statements under this category with 5 point scale for each statements makes total score of 40. The mean value is 22.93 indicates that majority of opinion are towards agree and strongly agree with std. deviation of 4.978. CWL denotes cumulative empowerment statements. There are 11 statements under this category with 5 point scale for each statements makes total score of 55. The mean value is 29.72 indicates that majority of opinion are towards agree and strongly agree with std. deviation of 6.115. All the hypothesis are failed to accept and it indicates no significant difference in the discriminant score with respect to change in all factors of high and low women empowerment.

Table 1.3 Tests of equality of Group Means

Tests of Equality of Group Means
	Wilks' Lambda	F	df1	df2	Sig.	Results
CLP1	0.826	131.251	1	625	.000	Null Hypothesis is failed to Accept
CSR1	0.821	136.005	1	625	.000	Null Hypothesis is failed to Accept
CUL1	0.815	141.88	1	625	.000	Null Hypothesis is failed to Accept
CAU1	0.692	277.807	1	625	.000	Null Hypothesis is failed to Accept
CWL1	0.584	444.315	1	625	.000	Null Hypothesis is failed to Accept

In the above table for CLP (Cumulative loan procedures statements) Wilks’ Lambda is 0.826 at 0.000 significance level which is less than 0.05, it means the null hypothesis is failed to accept. Same like this, for all independent variables, considering Wilks’ Lambda value and significance level in all cases, null hypothesis is failed to accept or alternate hypothesis is accepted that is there is significance difference in the discriminant score with respect to each independent variables of high and low women empowerment. For those factors/ independent variables, its wilks’ lambda is lowest is most influential factor. Here, Wilks’ lambda is lowest for change in women’s life which is 0.584 it means it is the most influential factor.

Table 1.4 Box's Test of Equality of Covariance Matrices

Log Determinants			Test Results
CHANGES	Rank	Log Determinant	Box's M		256.263
LOW	5	9.211	F	Approx	16.884
HIGH	5	13.137		df1	15
Pooled within-groups	5	12.498		df2	413135
The ranks and natural logarithms of determinants printed are those of the group covariance matrices.				Sig.	0.12
			Tests null hypothesis of equal population covariance matrices.

The test results indicates significance value of 0.120 and it indicates that the data do not differ significantly from multivariate normal. It indicates that the researcher can proceed with the further analysis.

H0: There is no significance difference in discriminant score with respect to intra-relationship amongst the all independent variables.

H1: There is significance difference in discriminant score with respect to intra-relationship amongst the all independent variables.

Here, null hypothesis is accepted. Here, all the independent variables having significant relationship, it indicates that all the independent variables are closely related with one another. Therefore, while applying discriminant score, deliberately researcher considered as a separate groups therefore their intra-relationship will not create any further issue and researcher did not violate the assumption of multi-collinerily.

Summary of Canonical Discriminant Functions:

Canonial correlation is the correlation between the dependent variables, it shows very strong relationships amongst all independent variables.

H0: There is no significance difference between observed and expected independent variable impact on discriminant score of all five independent variables namely loan procedure, saving related matters, use of loan amount, women’s autonomy and change in women’s life.

H1: There is significance difference between observed and expected independent variable impact on discriminant score of all five independent variables namely loan procedure, saving related matters, use of loan amount, women’s autonomy and change in women’s life.

Table 1.5 Eigenvalues

Eigenvalues
Function	Eigenvalue	% of Variance	Cumulative %	Canonical Correlation
1	1.12^a	100.0	100.0	.653

a. First 1 canonical discriminant functions were used in the analysis.

The Eigen value suggests the ratio of variance explained. The larger Eigenvalue indicates a strong function. The canonical relation indicates a correlation among the discriminant scores and the levels of these dependent variables (loan procedure, saving related matters, usage of loan amount, women’s autonomy and change in women’s life). The greater the correlations value shows better the function that discriminates the values. The value of 1 is considered as perfect. Here the value of canonical correlation is 0.653 and it suggests very strong relationship. As Eigen value is 1.12, it is highly acceptable.

Table 1.6 Wilks' Lambda

Test of Function(s)	Wilks' Lambda	Chi-square	df	Sig.
1	.432	345.998	5	.000

As the table indicates Chi-square value of 345.998 with .00 significance level which is less than 0.05, null hypothesis is failed to accept and alternate hypothesis is accepted. The value of Wilks’ lambda is 0.432 which indicates 43.2% unexplained variability which is fairly occurred good value.

Checking for relative importance of each independent variable

By comparing the standardised coefficient, it is likely to recognise which independent variable is more discriminating than the other variables. The greater the discriminating powers the greater the standardised discriminant coefficient. The Standardised Canonical discriminant function coefficient is shown in the below table. The change in women’s life has the highest discriminating power with highest discriminant coefficient of .885 542 followed by saving related matters (0.108), usage of loan amount (0.102), loan procedure (0.100) while women’s autonomy has negative coefficient of -0.043. It indicates that change in women’s life is the best predicator amongst all of whether women empowerment us high performer or low performer.

Table 1.6 Structure Matrix

Standardized Canonical Discriminant Function Coefficients		Structure Matrix		Canonical Discriminant Function Coefficients
	Function		Function		Function
	1		1		1
CLP1	0.1	CWL1	0.978	CLP1	0.024
CSR1	0.108	CAU1	0.773	CSR1	0.028
CUL1	0.102	CUL1	0.553	CUL1	0.025
CAU1	-0.043	CSR1	0.541	CAU1	-0.01
CWL1	0.885	CLP1	0.532	CWL1	0.189
		Pooled within-groups correlations between discriminating variables and standardized canonical discriminant functions		(Constant)	-6.972
		Variables ordered by absolute size of correlation within function.		Unstandardized coefficients

Formulating the Discriminant Function:

The normal form of the Discriminant Function is

Z = a + b1x1 + b2x2 + b3x3 + b4x4+b5x5

Where,

Z = Dependent variable, a = constant term

b1, b2, b3, b4 and b5 are the corresponding unstandardized discriminant function coefficient

x1, x2, x3 x4 and x5are the independent variables

D= constant + (.024*Loan procedures) + (.028*saving related matters) + (.025*use of loan amount) - (.010*women’s autonomy) + (.189*changes in women’s life) – 6.972

Table 1.7 Functions at Group Centroids

Functions at Group Centroids
CHANGES	Function
CHANGES	1
NO	-1.423
YES	.521

Unstandardized canonical discriminant functions evaluated at group means

The Function of the Group Centriod provides middling discriminant score of the two groups. These two scores are equivalent in total values but have opposite sign discriminating the score. The centroids are the extreme point for formulating the decision rule and are symbolised below:

-1.423 0 0.521

Low responds towards women empowerment has mean of -1.423 while high responds towards women empowerment has mean of .521.

Table 1.8 Prior Probabilities for Groups

CHANGES	Prior	Cases Used in Analysis
CHANGES	Prior	Unweighted	Weighted
NO	.268	168	168.000
YES	.732	459	459.000
Total	1.000	627	627.000

The prior probabilities give us the number of opinions used in the analysis and the dispersal of the opinions into groups used as an initial point in the analysis. It gives the weighted value, which is further used in the scheming of the centriod value.

Table 1.9 Classification Function Coefficients

	CHANGES
	NO	YES
CLP1	.151	.197
CSR1	.309	.363
CUL1	.757	.805
CAU1	.161	.141
CWL1	.504	.872
(Constant)	-20.458	-32.127

Developing and Analysing the Confusion Matrix:

For validating the projecting capacity of the discriminant Function, the equation is subjected to the data collected on the five independent variables. The values from the actual data collected is replaced in the unstandardized discriminant function and the decision rule is used to analyse the performance.

Table 1.10 Classification Results

Classification Results^a,c
		CHANGES	Predicted Group Membership		Total
		CHANGES	NO	YES	Total
Original	Count	NO	158	10	168
	Count	YES	29	430	459
	%	NO	94.0	6.0	100.0
	%	YES	6.3	93.7	100.0
Cross-validated^b	Count	NO	153	15	168
	Count	YES	33	426	459
	%	NO	91.1	8.9	100.0
	%	YES	7.2	92.8	100.0

a. 93.8% of original grouped cases correctly classified.

b. Cross validation is done only for those cases in the analysis. In cross validation, each case is classified by the functions derived from all cases other than that case.

c. 92.3% of cross-validated grouped cases correctly classified.

As seen in above table, 93.8% of respondents were classified correctly into high or low response towards women empowerment. A discriminant analysis was piloted to forecast whether a respondents was high or low towards women empowerment or not. The predictor variables were loan procedure, saving related matters, use of loan amount, women’s autonomy and changes in women’s life. Significant mean difference was perceived for all the independent variables on the dependent variable that is women empowerment. While the log determinants were quite similar, the Box’s M indicated that the assumption of equality of covariance matrices was violated. However, given the large sample, this problem is not regarded as serious. The discriminate function revealed a significant association between group and the predictors, accounting for 43.2 % of between group variability. Closer Analysis of structure matrix revealed significant independent variables namely change in women’s life (0.978), women’s autonomy (0.773), use of loan amount (0.553), saving related matters (0.541) and loan procedure (0.532) and it indicates all independent variables are strong. The cross validated classification showed that overall 93.8% were correctly classified.

REFERENCES:

1. Badatya, K. C. (2006). Microfinance for Microenterprises: An Impact Evaluation Study of Self Help Groups (Evaluation Study Series No. 13). . Andhra Pradesh Regional Office, Hyderabad. NABARD.

2. Brana S. (2013), “Microcredit: An Answer to the Gender Problem in Funding”, Small Business Economics, Vol. 40, pp. 87-100.

3. Conroy JD. (2003). The Challenges of Micro financing in Southeast Asia. Financing Southeast Asia's Economic Development.

4. Datar, S. E. (2008). In microfinance, clients must come first. . Stanford Soc Innov Rev , 6(1): 38–45.

5. Dichter T. 1999, Non-governmental Organisations (NGOs) in Microfinance: Past, Present and Future. Available from: www.esd.worldbank.org/html/esd/agr/sbp/end/ngo.htm (Accessed on: 05.06.18)

6. Ledgerwood, Joanna (1999); Microfinance Handbook; Sustainable Banking with the poor: An institutional and financial Perspective; World Bank, Washington D.C.

7. Mahanta, P. S. (November 2012). Status of Microfinance in India - A Review. International. Journal of Marketing, Financial Services & Management Research 2012, 1(11).

8. Malleswari, B. (2010). Micro-Finance and Women Empowerment. New Delhi: Serials Publications. .

9. Remenyi and Quinones (2000); Microfinance and Poverty Alleviation: Case studies from Asia and the Pacific. New York. 79. p.131-134,253-263.

10. Rolando GT. (2010). Government’s role in promoting social entrepreneurship. Institute for Social Entrepreneurship in Asia.

11. Schreiner, Mark. (2000); “Informal Finance and the Design of Microfinance”, Development in Practice, Vol. 11, No.5, pp.637-640.

12. Verma R., M. R. (2012). Micro–Finance – A Critical Analysis Of Rural India. international seminar, (p. 23).

13. Wright G.A.N. 1999, Examining the Impact of Microfinance Services – Increasing Income or Reducing Poverty. Small Enterprise Development, Vol. 10, No. 1, pp. 39-47.

Received on 37.07.2018 Modified on 10.08.2018

Int. J. Rev. and Res. Social Sci. 2018; 6(3):273-278.

DOI: 10.5958/2454-2687.2018.00027.8