Result and discussions 1. The development ROA, LDR, NPLs and OEOI PT. Bank Tabungan
462 Where,
Y = Return On Assets ROA a = A constant which is the value of the variable Y when the variable X is
0 zero β = Coefficient of the regression line
X
1
= Loans to Deposits Ratio LDR X
2
= Non Performing Loans NPLs X
3
= Operating Expense to Operating Income OEOI e = Residual
4. Result and discussions 4.1. The development ROA, LDR, NPLs and OEOI PT. Bank Tabungan
Negara Pesero Tbk. Based on data period 2010 – 2015 the development of ROA, NPLs and OEOI
were achieved by PT. Bank Tabungan Negara Pesero Tbk see Table 1 can be described as follows:
The average value of ROA achieved period 2010 – 2015 amounted 1.68 and until December 31, 2015 amounted to 1.61 lower than the national banks
2.32, and other groups such as state owned banks 3.31, foreign owned banks 1.68, foreign exchange banks 1.75, non-foreign
exchange banks 1.65; regional development bank 2.40, but higher than joint venture banks 1.01 Indonesian Banking Statistics, 2015. The
lowest value of ROA at 1.02 is achieved on III2014, while the highest value of ROA at 2.05 is achieved on IV2010.
The average value of LDR achieved period 2010 – 2015 amounted 107.96 and until December 31, 2015 amounted to 108.78 higher than the national
banks 92.11, and other groups such as state owned banks 88.58, foreign exchange banks 87.55, non-foreign exchange banks 85.95;
regional development bank 92.19, but lower than foreign owned banks 131.49 and ,joint venture banks 132.77 Indonesian Banking
Statistics, 2015. The lowest value of LDR BTN 98.19 is achieved on QI2014, while the highest value of ROA at 116.04 is achieved on QII2010.
The average value of NPLs achieved period 2010 – 2015 amounted 4.13 and until December 31, 2015 amounted to 3.42 higher than the national
463 banks 2.6, and other state owned banks, that is Bank BRI 2,02,
Bank BNI 2,70 and Bank Mandiri 2,29. Indonesian Banking Statistics, 2015. The lowest value of NPLs BTN 2.75 is achieved on
QIV2014 while the highest value of NPLs at 5.01 is achieved on QII2014. By looking at the average value of the NPL shows that the credit risk was faced
by BTN are at moderate albeit under tolerance required by Bank Indonesia at 3–5. But BTN must be able to control the risk of BTN lending so that BTN
has the ability to increase profits from lending.
The average value achieved OEOI period 2010 – 2015 amounted 84.38 and December 31, 2015 amounted to 84.83, higher than the national
banks 81.49; state owned banks 72.58; regional development bank 79.57 but lower than foreign exchange banks 94,98; non foreign
exchange banks 86.81 ; joint venture banks 87.55 and foreign owned banks 92.69. The highest value of OEOI BTN 89.91 is achieved on
QIII2014 and the lowest OEOI amounted to 80.26 achieved on QIII2012. By looking at the average value of OEOI is still below the tolerance required
by Bank Indonesia at 94 - 96. This means that, although BTN is able to control its operational efficiency so that the BTN still has the ability to
improve the operational efficiency.
Table 1 The Development of the LDR, NPLs, OEOI and ROA of Bank Tabungan
Negara, Tbk. Period 2010 - 2015 Percentage Tahun
LDR NPLs
OEOI ROA
I2010 II2010
III2010 IV2010
I2011 II2011
III2011 IV2011
I2012 II2012
113.97 116.04
114.3 108.42
110.33 110.85
112.27 102.57
102.77 108.3
4.00 4.12
4.22 3.26
4.04 4.35
4.18 2.75
3.22 3.46
84.85 84.69
84.24 83.28
83.80 84.92
85.05 81.75
81.18 80.54
1.94 1.92
1.93 2.05
1.93 1.85
1.70 2.03
1.99 1.98
464 III2012
IV2012 I2013
II2013 III2013
IV2013 I2014
II2014 III2014
IV2014 I2015
II2015 III2015
IV2015 Max
Min Average
110.44 100.9
98.19 110.58
109.4 104.42
100.53 105.17
108.54 108.86
109.71 109.94
105.71 108.78
116.04 98.19
107.96 3.68
4.09 4.77
4.63 4.88
4.05 4.74
5.01 4.85
4.01 4.78
4.70 4.50
3.42 5.01
2.75 4.13
80.26 80.74
83.31 83.31
83.29 82.19
86.55 89.17
89.91 88.97
85.53 85.40
85.84 84.83
89.91 80.26
84.38 2.01
1.94 1.60
1.58 1.63
1.79 1.39
1.11 1.02
1.14 1.53
1.55 1.50
1.61 2.05
1.02 1.68
Source: Published Financial Statements, period 2010 - 2015, www. Bank BTN co.id. downloaded November 10, 2015, processed
4.2. Multiple linear regression analysis Multiple linear regression analysis was used to determine basically
dependence of dependent variable with one or more independent variables, with the aim of estimating or predicting the average of
population data or average value of the dependent variable based on the value of the independent variable known Gujarati, 2003. By regression
analysis it can be seen whether there is influence between independent variables with the dependent variable. The results of multiple linear
regression analysis in this study can be seen in Table 2. Based on Table 2, the regression equation is as follows :
= 7.699 + 0.019 − 0.123
− 0.089 2
465 The equation above it can be explained as follows:
1. Constant value a of 7.699 which means a positive constant value. This shows if the LDR X
1
, NPLs X
2
and OEOI X
3
has a value of zero, then the ROA Y increase by 7.699
2. Regression coefficient for the variable LDR X
1
is 0.019, indicating a positive relationship between the LDR X
1
with ROA Y, meaning that if there is additional LDR X
1
per unit, assuming other variables remain the ROA Y was increased by 0.019 and vice versa.
3. Regression coefficient for the variable NPLs X
2
is -0.123, indicating a negative relationship between the NPLs X
1
with ROA Y, meaning that if there is additional NPLs X
1
per unit, assuming other variables remain the ROA Y was decreased by 0.123 and vice versa.
4. Regression coefficient for the variable OEOI X
3
is -0.089, indicating a negative the relationship between the OEOI X
3
with ROA Y, meaning that if the addition of OEOI X
3
for every one unit, assuming other variables constant, the ROA Y decrease by 0.089 and vice versa.
Table 2 Test Results of Multiple Linear Regression Coefficients
Model Unstandardized
coefficients Standardized
coefficients T
Sig. B
Std. error Beta
1 Constant 7.699
.998 7.716
.000 LDR
.019 .006
.280 3.309
.004 NPLs
-.123 .051
-.246 -2.422
.025 OEOI
-.089 .012
-.762 -7.401
.000 Dependent variable: ROA;
Source:Output SPSS 21.0. 4.3. Analysis of correlation coefficient and coefficient of determination
Correlation coefficient analysis was used to determine the direction and the strong relationship among the three independent variables. Those are
the variable LDR X
1
NPLs X
2
and OEOI X
3
with ROA as a dependent variable Y. see table 3.
Based on Table 3, it can be concluded that the variable the LDR X
1
, NPLs X
2
and OEOI X
3
, with ROA dependent variable has a value of correlation
466 r 0.928, meaning that the correlation level are in strong correlation
Sugiyono, 2009. While the coefficient of determination analysis was used to determine the contribution effect of LDR X
1
, NPLs X
2
and OEOI X
3
with ROA as a dependent variable Y expressed as a percentage. Based on Table 3
that the R
2
value was 0.862 or 86.2 indicating that LDR X
1
, NPLs X
2
and OEOI X
3
, accounted for 86.2 of the ROA Y, while the remaining 13.8 influenced by other variables not examined.
Table 3 Test Results Correlation Coefficient and Coefficient of Determination
Model Summary
Model R
R square Adjusted R square Std. error of the estimate
1 .928
a
.862 .841
.12156 Source: Output SPSS 21.0.
a. Predictors: Constant, BOPO, LDR, NPLs b. Dependent Variable: ROA
4.4. Partial significance test t-test To examine hypotheses on the significance of the partial model used t-test.
It is intended to determine the effect of independent variables LDR, NPLs and OEOI partially to the dependent variable ROA. Partially, the
influence of the three independent variables to the ROA as an independent variables, shown in the Table 4 partial test results t-test, it
can be argued that: 1. Effect of LDR on the ROA
Partial test results between the LDR with an ROA shows the t-test value of 3.309 greater than t-table 2.074 with a significant value of 0.04
which is below 0.05. This means that the LDR has positive effects on ROA. Thus hypothesis H1 stating LDR positive effect on ROA is
acceptable. The test results are in line with previous research conducted by Nusantara 2009; Prasanjaya 2013; Artarina 2013;
Vong 2009; Widati 2012 and Restiyana 2011 which states that LDR has positive and significant impact on ROA in the banking firm.
However, contrary to the results of Arimi 2012 research result showed Loan to Deposit Ratio LDR hasn’t significant positive effect to Return
On Asset ROA. Similarly, the research result of Purwoko 2013 LDR whereas no significant effect on ROA.
467 2. Effect of NPLs on the ROA
Partial test results between NPLs to ROA shows the t-value of -2.422 greater than the t-table 2.064 with a significant value of 0.025 which
is below 0.05. This means that the NPLs has negative effect and significant effect to the return on assets ROA. Thus the hypothesis H2
which states NPLs negative effect to the ROA is acceptable. The test results are in line with the research results of Adeusi 2014; Sinha 2014;
Arimi 2012; Ahmad 2014; Puwoko 2013; Manikam 2013; Eng 2013; Nusantara 2009; Artarina 2013 and Restiyana 2011 which
states that NPL have significant negative effect on ROA. However, contrary to the research results according to Oktaviantari 2013 which
states that NPL have influence positive but it is not significant towards profitability ROA.
3. Effect of OEOI on the ROA Partial test results between the OEOI to ROA shows the t-test value of -
7.401 is greater than t-table 2.064 with a significant value of 0.000 which was below 0.05. This means that the OEOI has negative effects
and significant effect to the ROA. Thus hypothesis H3 which states OEOI negatively effect on ROA is acceptable. The test results are in line
with previous research conducted by Purwoko 2013; Oktaviantari 2013; Wibowo 2013; Chatarine 2014 and Restiyana 2011 which
states that OEOI had negative significant effect towards profitability ROA. However, contrary to the research results according to Sinha
2014; Nusantara 2009; Prasanjaya 2013; Francis 2013; Manikam 2013; Artarina 2013 and Widati 2012 which states that OEOI have
significant effect on ROA
Table 4 Partial Test Results T-Test
Model T-count
T-table Sig.
Description 1
Constant 16.078
.000 LDR
3.309 2.064
.004 Significant
NPLs -2.422
2.064 .025
Significant OEOI
-7.401 2.064
.000 Significant
Source: Output SPSS 21.0. Dependent variable: ROA
468 4.5. Simultaneous significant test F-test
F - test was conducted to determine the effect of independent variables LDR, NPLs and OEOI simultaneously to the dependent variable ROA.
Simultaneously influence of the three independent variables to the independent variables ROA is shown in Table 5. Based on the results of the
F-test calculations in Table 5, F-count was 41.580 greater than the F-table 3.028 with a significance value sig of 0.000 is below 0.05. This means
that the independent variables LDR, NPLs and OEOI simultaneously significant effect to dependent variable ROA. Thus the H4 hypothesis
which states LDR, NPLs and OEOI effect on ROA is acceptable. The test results are in line with previous research conducted by Prasanjaya 2013
that the CAR, OEOI, LDR and Firm Size with The F test influence the profitability. Furthermore, according to Purwoko 2013 coefficient of
determination, which shows the magnitude of the influence OEOI, NPL, NIM, CAR and LDR to ROA are 73.1, while the rest of 26.9 explained by
other causes outside the model. Then Manikam 2013, the research result shows OEOI, NIM and NPL has significant effect toward bank profitability.
And Eng 2013, the research result shows NIM, LDR and NPL have significant on ROA. Nusantara 2009, from the result of analyze indicate
that data NPL, CAR, LDR and OEOI in significant toward ROA bank go public at level of significant less than 5. Oktaviantari 2013, NPL influenced
positive but it is not significant towards profitability ROA. The OEOI influenced negative significant towards profitability ROA, and LDR
influenced positive significant towards profitability ROA. Prasanjaya 2013, from the F test result shows CAR, OEOI, LDR and Firm Size
significantly influence the profitability; Artarina 2013 the research result shows that simultaneously CAR, LDR, OEOI, and NPL have significant effect
on ROA. Widati 2012, the result of CAR and LDR are significant positive influence to the ROA, but PPAP have positive and not significant influence
to ROA, DER have positive significant to ROA, OEOI have positive and not significant to ROA. And Restiyana 2011, the results of this study indicate
that CAR, LDR, and the NIM has positive and significant impact on ROA in the banking firm. While the NPL and OEOI significantly and negatively
related to ROA banking.
469
Table 5 Simultaneous Test Results F-Test
Source: Output SPSS 21.0. ANOVA
Model Sum of squares
Df Mean square
F Sig.
1 Regression
1.843 3
.614 41.580
.000
b
Residual .296
20 .015
Total 2.139
23 a. Dependent variable ROA;
b. Predictors: constant, OEOI, LDR, NPLs 5. CONCLUSION
Based on the background, the formulation of the problem, hypotheses, methods and research results and discussion, some conclusions can be
drawn as follows: 1. In period 2010 -2015, the average value of ROA achieved by PT. Bank
Tabungan Negara Pesero Tbk. amounted 1.68 and until December 31, 2015 amounted to 1.61 lower than the national banks 2.32, and
other groups such as state owned banks 3.31, foreign owned banks 1.68, foreign exchange banks 1.75, non-foreign exchange
banks 1.65; regional development bank 2.40, but higher than joint venture banks 1.01. The lowest value of ROA at 1.02 is
achieved on III2014, while the highest value of ROA at 2.05 is achieved on IV2010.
2. Based on the results, it is concluded that the partial, LDR has positive and significant effect to ROA; The NPLs has negative and significant
effects to the ROA; and the OEOI has negative and significant effects to the ROA; Simultaneously that variable LDR, NPLs and OEOI significantly
influence to ROA variable.
3. The amount of the contribution or influence variable of NPLs and OEOI to the dependent variable of ROA is 86.2 while the remaining 13.8
influenced by other variables not examined in this study.