Can performance measurement analysis truly enhance microfinance service delivery? This study uncovers surprising insights from Gulu District, revealing how Balanced Scorecard principles are applied and highlighting critical gaps in innovation and growth that could reshape institutional effectiveness.
4.8.2 – Regression Analysis
The relationship between performance measurement and service delivery was further analysed using regression analysis. However, before the regression analysis was used, a graphical relationship using the scatter plot graph was obtained as shown in Figure 22.
Figure 22 – Scatter plot for Performance measurement & Service Delivery
[18_performance-measurement-analysis-in-microfinance-institutions_18]
Source: Research data August 2007
In the graph, performance measurement was taken as the independent variable and put in the X-axis, and service delivery was taken as the dependent variable and put in the Y-axis. The scatter plot graph was edited, and the best-fit straight line was produced. The fitted line is also called the linear regression of service delivery on the performance management.
To get more information about the line, a regression analysis was done and the results presented as follows: The first is the table of coefficients that was presented in Table 19.
Table 19 – Standardised & Unstandardised Coefficients
Variable | Unstandardised Coefficients | Standardised Coefficients | t | Sig. | |
B | Std. Error | Beta |
|
| |
(Constant) | 2.169 | 0.550 |
| 3.944 | 0.001 |
Performance measurement | 0.426 | 0.160 | 0.521 | 2.663 | 0.016 |
Source: Research data August 2007
In the table, the “Standardised coefficients” column contains the beta coefficient given by 0.521. The beta coefficient is equal to the Pearson’s coefficient between the two variables, and it tells how strongly the independent variable is associated with the dependent variable.
In the “Unstandardised Coefficients” column, two statistics are reported: B, which is the regression coefficient, and the standard error. There are also two statistics reported under B: one labelled as (Constant), the other labelled as (Performance measurement). The statistic labelled as “Performance measurement” is the regression coefficient whose value is 0.426, and is the slope of the line shown on the scatter plot, conventionally represented by lower case b. The one labelled as (Constant) is not actually a regression coefficient, but is the Y-intercept of a straight-line graph, represented by lower case a, whose value in the table is 2.169.
Since a straight-line graph is represented by formula:
Y = a + bX
Where, Y refers to the value of the dependent variable, a is the Y-intercept, b is the slope of the line which describes the relationship between the independent and dependent variables, and X is the value of the independent variable. When the values of a and b are substituted in the equation, we get an equation;
Y = 2.169 + 0.426X
Where X represents the performance measures and Y represents the service delivery. This linear relationship between X and Y (Performance measurement and Service delivery) is not perfect, since the correlation coefficient, found to be 0.521, was not 1 (or –1), and the scatter plot graph showed plenty of cases that did not fall directly on the line. Thus, it is clear to us that knowing the performance measurement level will not tell us exactly the service delivery levels. Furthermore, we are only analyzing a sample of cases and not the whole population to which we want to generalize our findings. It is clear that there is some error built into the findings.
The derived equation shows a positive relation between service delivery and the performance measurement. The interpretation of this result is that, using the balanced scorecard in the management of the microfinance institutions enhances performance and this proportionately enhances service delivery levels in the MFIs.
The t statistic is the regression coefficient divided by its standard error. The standard error is an estimate of the standard deviation of the coefficient. It can be thought of as a measure of the precision with which the regression coefficient is measured. If a coefficient is large compared to its standard error, then it is probably different from 0. In the given table, the t-test for Performance Measurement equals 2.663, and is statistically significant, meaning that the regression coefficient for Performance Measurement is significantly different from zero.
The p-value, which checks for the significance of the relationship, is given in the table by 0.016. Since this value is less than 0.05, we can reject the hypothesis that there is no relationship between performance measurement and service delivery, and confirm that some relationship exists between performance measurement and service delivery.
A more standardised statistic, which also gives a measure of the ‘goodness of fit’ of the estimated equation, is the model summary table, presented in Table 20.
Table 20 – Model Summary
R | R Square | Adjusted R Square | Std. Error of the Estimate |
0.521 | 0.271 | 0.233 | 0.38855 |
Source: Research data August 2007
This table displays R, R², adjusted R², and the standard error. R is the correlation between the observed and predicted values of the dependent variable. The values of R range from -1 to 1. The sign of R indicates the direction of the relationship (positive or negative). The absolute value of R indicates the strength, with larger absolute values indicating stronger relationships.
R² is the proportion of variation in the dependent variable explained by the regression model. The values of R² range from 0 to 1. Small values indicate that the model does not fit the data well. Adjusted R² attempts to correct R², to reflect more closely the goodness of fit of the model in the population.
R² is the Coefficient of Determination. The value of R² is calculated as follows; R² is the regression sum of squares (RegSS) divided by the total sum of squares (TotSS), which is RegSS/TotSS, which is the fraction of the variability in the response that is fitted by the model. The total sum of squares is the sum of the Regression, (RegSS) and Residual Sums of squares (ResSS) as shown in Table 20.
R² can be rewritten as follows:
If a model has perfect predictability, the residual sum of squares will be 0 and R²=1. If a model has no predictive capability, R²=0. In practice, R² is never observed to be exactly 0, the same way the difference between the means of two samples drawn from the same population is never exactly 0 or a sample correlation coefficient is never exactly 0. The value of R² in Table 20 and as indicated in the scatter plot graph of Figure 21 is, R²= 0.271 showing that there was about 27% shared variance between the service delivery and performance measures. Hence, the gain in prediction was minimal.
R is the Pearson correlation coefficient between the predicted and observed values, having a value of 0.521, and is the square root of R². The value of R=0.521 implies that the use of performance measures (the independent variable) in predicting the service delivery levels (the dependent variable) reduces the sum of squares of deviations about the prediction line (regression line), by only R² = 0.271, or 27% (Mendenhall and Reinmuth, 1978).
The other 73% of the variation in performance measurement is left unexplained. However, rectifying some of the negative factors found during the research which lowered the performance levels, such as inadequate funding, lack of staff training and lack of tools & equipment for the staff, to mention but a few, would reduce the high percentage of 73% and increase R².
The analysis of variance (ANOVA) table presented in Table 21 gives information on how well the model fits the data.
Table 21 – Analysis of Variance (ANOVA)
Source | Sum of Squares | df | Mean Square | F | Sig. |
Regression | 1.066 | 1 | 1.066 | 7.061 | 0.016 |
Residual | 2.869 | 19 | 0.151 | ||
Total | 3.935 | 20 |
Source: Research data August 2007
According to Table 21, the test statistics, F is given by 7.061, which is obtained by dividing the regression mean square (MSR) by the residual mean square (MSE). If we choose an α value of 0.05, then the critical F value from the F-distribution table, for a df of 1 in the numerator and 19 in the denominator is 4.381.
Since the test statistics is larger than the critical value, we reject the null hypothesis that there is no association between performance measurement and service delivery. Then, we conclude that there is an association between performance measurement and service delivery, or that our chosen model contributes information on performance management for the prediction of the service delivery levels of the selected MFIs.
In order to get an effective delivery of service, an MFI should be in a position to face the multiple factors, which affect the performance of the MFI. As shown in the Critical Microfinance triangle in Figure 4, internally, the MFI is faced with the institutional innovations such as technology, policies, organisational and management set-ups. Externally, the MFI is surrounded by human and social capital possessed by the poor, economic policies of the country and the quality of the financial infrastructure that supports the MFI. Improvements in the environment make it easier for the MFI to reach the three objectives; namely, Financial sustainability, Outreach to the poor and welfare or social impact on the poor (Meyer, 2002).
According to the results obtained from the research, the internal business processes of the selected MFIs were hindered due to lack of tools and equipment for the staff to perform effectively. The staff also lacked motivation and training to keep them abreast with the current trends in the MFI developments. According to the customers, the MFIs lacked funding which hindered the delivery of services to them, despite efforts of the staff to keep them in contact. There was also lack of communication, which limited the number of customers in the MFIs. More of the discussions on the findings are given in the subsequent Chapter.
Frequently Asked Questions
What is the relationship between performance measurement and service delivery in microfinance institutions?
The derived equation shows a positive relation between service delivery and performance measurement, indicating that using the balanced scorecard in the management of microfinance institutions enhances performance, which in turn enhances service delivery levels.
How was the performance measurement analysis conducted in the study?
The study employed both descriptive and cross-sectional approaches with quantitative methods, surveying 74 respondents from 8 microfinance institutions.
What were the findings regarding the Balanced Scorecard principles in microfinance institutions?
Findings indicate that microfinance institutions were indirectly using Balanced Scorecard principles, which proved effective for performance measurement through its blend of financial and non-financial metrics.