A development of enterprise resource planning success model for accounting professionals

riances assumed .105 .746 .923 118 .358 .168 .182 -.192 .527 Equal variances not assumed .897 75.042 .373 .168 .187 -.205 .540 PAB16 Equal variances assumed 4.032 .047 2.030 118 .045 .384 .189 .010 .759 Equal variances not assumed 2.115 90.780 .057 .384 .182 .023 .745 PAB17 Equal variances assumed .542 .463 3.535 118 .051 .948 .268 .417 1.479 Equal variances not assumed 3.402 72.998 .061 .948 .279 .392 1.503 PAB18 Equal variances assumed .164 .686 2.547 118 .052 .634 .249 .141 1.127 Equal variances not assumed 2.506 77.564 .054 .634 .253 .130 1.138 USE1 Equal variances assumed 3.641 .059 3.280 118 .001 .636 .194 .252 1.020 Equal variances not assumed 3.079 68.381 .003 .636 .207 .224 1.048 USE2 Equal variances assumed .087 .769 2.154 118 .033 .509 .236 .041 .977 Equal variances not assumed 2.054 71.290 .044 .509 .248 .015 1.003 USE3 Equal variances assumed 2.648 .106 2.497 118 .054 .447 .179 .093 .802 Equal variances not assumed 2.347 68.583 .062 .447 .191 .067 .827 USE4 Equal variances assumed .000 .986 1.716 118 .089 .315 .184 -.049 .678 Equal variances not assumed 1.692 78.072 .095 .315 .186 -.056 .685 USE5 Equal variances assumed .370 .544 2.163 118 .063 .489 .226 .041 .937 Equal variances not assumed 2.032 68.528 .076 .489 .241 .009 .969 USE6 Equal variances assumed .127 .722 .770 118 .443 .185 .240 -.290 .659 Equal variances not assumed .769 80.754 .444 .185 .240 -.293 .662 USE7 Equal variances assumed .043 .835 1.246 118 .215 .234 .188 -.138 .605 Equal variances not assumed 1.260 83.670 .211 .234 .186 -.135 .603 USE8 Equal variances assumed .243 .623 1.967 118 .052 .467 .237 -.003 .936 222 Equal variances not assumed 1.885 72.216 .063 .467 .247 -.027 .960 USE9 Equal variances assumed .934 .336 .641 118 .523 .137 .213 -.286 .559 Equal variances not assumed .665 89.636 .508 .137 .206 -.272 .546 USE10 Equal variances assumed .039 .844 1.410 118 .161 .275 .195 -.111 .662 Equal variances not assumed 1.406 80.507 .164 .275 .196 -.114 .664 USE11 Equal variances assumed .311 .578 .733 118 .465 .196 .267 -.333 .726 Equal variances not assumed .756 88.171 .452 .196 .259 -.319 .712 SAT1 Equal variances assumed .028 .867 1.880 118 .063 .362 .192 -.019 .743 Equal variances not assumed 1.875 80.472 .064 .362 .193 -.022 .746 SAT2 Equal variances assumed 1.249 .266 1.976 118 .051 .377 .191 -.001 .755 Equal variances not assumed 1.899 72.836 .061 .377 .199 -.019 .773 SAT3 Equal variances assumed .400 .528 1.377 118 .171 .244 .177 -.107 .594 Equal variances not assumed 1.416 87.690 .160 .244 .172 -.098 .585 SAT4 Equal variances assumed .929 .337 1.876 118 .063 .355 .189 -.020 .729 Equal variances not assumed 1.824 75.115 .072 .355 .194 -.033 .742 AP1 Equal variances assumed .010 .919 2.980 118 .064 .394 .132 .132 .656 Equal variances not assumed 3.075 88.461 .063 .394 .128 .139 .649 AP2 Equal variances assumed 3.451 .066 3.209 118 .072 .454 .142 .174 .735 Equal variances not assumed 3.079 72.484 .073 .454 .148 .160 .749 AP3 Equal variances assumed .849 .359 2.717 118 .068 .379 .140 .103 .656 Equal variances not assumed 2.712 80.666 .068 .379 .140 .101 .658 AP4 Equal variances assumed .004 .951 2.505 118 .054 .321 .128 .067 .574 Equal variances not assumed 2.529 83.203 .053 .321 .127 .068 .573 OP1 Equal variances assumed .207 .650 1.682 118 .095 .267 .159 -.047 .582 Equal variances not assumed 1.665 78.738 .100 .267 .161 -.052 .587 OP2 Equal variances assumed 4.361 .039 .362 118 .718 .049 .135 -.218 .316 Equal variances not assumed .402 106.710 .688 .049 .121 -.192 .289 OP3 Equal variances assumed 1.446 .232 .639 118 .524 .087 .136 -.182 .356 Equal variances not assumed .624 76.058 .535 .087 .139 -.190 .364 223 OP4 Equal variances assumed .268 .606 .245 118 .807 .035 .144 -.249 .320 Equal variances not assumed .249 84.833 .804 .035 .141 -.246 .316 OP5 Equal variances assumed .572 .451 .519 118 .605 .070 .136 -.198 .339 Equal variances not assumed .552 95.915 .583 .070 .128 -.183 .324 OP6 Equal variances assumed .992 .321 2.399 118 .058 .344 .144 .060 .628 Equal variances not assumed 2.353 76.901 .061 .344 .146 .053 .636 OP7 Equal variances assumed 2.844 .094 1.520 118 .131 .174 .115 -.053 .402 Equal variances not assumed 1.617 96.098 .109 .174 .108 -.040 .389 OP8 Equal variances assumed .404 .526 .468 118 .641 .056 .120 -.181 .294 Equal variances not assumed .471 82.616 .639 .056 .119 -.181 .293 U1 Equal variances assumed 4.691 .032 4.421 118 .000 1.416 .320 .782 2.050 Equal variances not assumed 4.192 70.235 .000 1.416 .338 .742 2.090 U2 Equal variances assumed 7.042 .009 4.237 118 .000 1.203 .284 .641 1.766 Equal variances not assumed 3.888 64.568 .000 1.203 .309 .585 1.821 U3 Equal variances assumed 1.662 .200 3.587 118 .000 1.040 .290 .466 1.615 Equal variances not assumed 3.485 74.910 .001 1.040 .299 .446 1.635 224 Appendix 5.4 Total Variance Explained for Common Method Bias Test Component Initial Eigenvalues Extraction Sums of Squared Loadings Total % of Variance Cumulative % Total % of Variance Cumulative % 1 28.029 42.468 42.468 28.029 42.468 42.468 2 4.274 6.476 48.944 3 3.596 5.449 54.393 4 3.074 4.657 59.050 5 2.291 3.471 62.521 6 2.050 3.107 65.628 7 1.643 2.489 68.117 8 1.591 2.410 70.527 9 1.445 2.190 72.717 10 1.352 2.049 74.766 11 1.008 1.527 76.293 12 .965 1.462 77.755 13 .945 1.432 79.188 14 .878 1.331 80.519 15 .834 1.264 81.783 16 .755 1.144 82.927 17 .734 1.112 84.038 18 .657 .996 85.034 19 .636 .964 85.998 20 .544 .824 86.822 21 .520 .787 87.609 22 .503 .762 88.371 23 .480 .727 89.098 24 .465 .705 89.803 25 .452 .686 90.488 26 .404 .612 91.100 27 .384 .582 91.683 28 .368 .557 92.240 29 .350 .530 92.770 30 .328 .498 93.268 31 .323 .490 93.757 32 .304 .460 94.218 33 .285 .431 94.649 34 .275 .416 95.065 35 .262 .397 95.462 225 36 .241 .365 95.826 37 .214 .324 96.151 38 .208 .315 96.466 39 .186 .282 96.748 40 .178 .269 97.017 41 .169 .255 97.272 42 .154 .234 97.506 43 .145 .220 97.726 44 .139 .211 97.937 45 .138 .208 98.146 46 .120 .181 98.327 47 .113 .171 98.498 48 .110 .167 98.665 49 .106 .160 98.825 50 .091 .137 98.963 51 .084 .128 99.091 52 .078 .118 99.208 53 .072 .109 99.317 54 .067 .101 99.418 55 .058 .088 99.507 56 .049 .074 99.580 57 .046 .069 99.649 58 .040 .061 99.710 59 .037 .056 99.766 60 .033 .051 99.816 61 .030 .045 99.862 62 .026 .039 99.900 63 .022 .033 99.934 64 .019 .029 99.963 65 .013 .020 99.983 66 .011 .017 100.000 226 Appendix 5.5 Development of the final sample Number of potential respondents emailed 5110 Collected responses 569 Less: Screening and cleaning Organization not-adopting ERP system 177 Irrelevant length of ERP adoption 0 Irrelevant respondent position 78 Respondent without enough working experience year 50 Incomplete cases 102 Response time of less than 10 minutes 26 433 Straight lining 0 Outlier 16 Final sample 120 (Source: by author) 227 Appendix 5.6 Instrument reliability assessment Internal consistency is adopted to assess reliability through two statistical indicators: (1) item- total correlation and (2) Cronbach’s alpha. Each statistics examines different aspect of reliability issue  Item-total correlation indicates if a single item is reliable in representing the instrument. Specifically, low correlation between items indicates that the items do not represent the same instrument. A correlation value less than 0.3 indicates that the corresponding item does not correlate very well with the scale overall and thus may be dropped (Hair Jr. et al., 2010). High value (>0.95), however, are suspect as they indicate multi-collinearity and the possibility that respondents have not answered objectively (Straub et al., 2004).  Cronbach’s alpha is a widely used index to assess internal consistency reliability (Streiner, 2003). It indicates if the items as a whole represent the instrument. Values of Cronbach’s alpha range from 0 to 1, with higher values indicate greater reliability (Straub et al., 2004). DeVellis (2016) proposes a common accepted rule describing internal consistency using Cronbach’s alpha as follow: The ‘Cronbach’s alpha if item deleted’ presents the hypothetical Cronbach’s alpha coefficients that would be obtained if each item is deleted. Hence, by examining the results, a researcher can immediately determine whether deleting a particular item would increase or decrease overall reliability. Based on the following Table of results of Conbach’s Alpha Test, the item-total correlation value of each item to its designated instrument is greater than the cut-off value of 0.3. This indicates that each item has strong internal consistency with other items of the instrument. In meantime, the Cronbach’s alpha values of the instruments are almost greater 0.9. The figures confirm that all items significantly measure the same instrument. Therefore, reliability of each instrument is extremely excellent and all items are remained with 66 variables from nine instruments. 228 Results of Conbach’s Alpha Test (Source: by author) Var. Item It em -T o ta la C o rr el at io n A lp h a if i te m d el et ed Alphab Var. Item It em -T o ta l C o rr el at io n A lp h a if i te m d el et ed Alpha SQ SQ1 .659 .834 .858 USE USE1 .571 .927 .928 SQ2 .604 .840 USE2 .757 .919 SQ3 .584 .843 USE3 .663 .924 SQ4 .515 .852 USE4 .742 .921 SQ5 .638 .837 USE5 .795 .918 SQ6 .643 .837 USE6 .716 .921 SQ7 .644 .836 USE7 .660 .924 SQ8 .580 .846 USE8 .817 .916 IQ IQ1 .639 .903 .906 USE9 .695 .922 IQ2 .706 .895 USE10 .727 .921 IQ3 .785 .882 USE11 .676 .924 IQ4 .845 .873 SAT SAT1 .903 .946 .959 IQ5 .721 .892 SAT2 .910 .944 IQ6 .752 .887 SAT3 .864 .957 PAB PAB1 .662 .957 .959 SAT4 .927 .939 PAB2 .740 .956 AP AP1 .817 .885 .915 PAB3 .770 .956 AP2 .808 .889 PAB4 .768 .956 AP3 .833 .880 PAB5 .688 .957 AP4 .766 .903 PAB6 .798 .955 OP OP1 .704 .910 .918 PAB7 .808 .955 OP2 .793 .902 PAB8 .799 .955 OP3 .743 .906 PAB9 .833 .955 OP4 .682 .911 PAB10 .827 .955 OP5 .748 .905 PAB11 .781 .956 OP6 .737 .906 PAB12 .850 .955 OP7 .754 .906 PAB13 .806 .955 OP8 .697 .910 PAB14 .669 .957 Ex_U U1 .899 .917 .946 PAB15 .688 .957 U2 .894 .918 PAB16 .700 .957 U3 .880 .928 PAB17 .624 .959 ITU ITU1 .744 .881 .899 PAB18 .579 .959 ITU2 .877 .836 ITU3 .869 .837 ITU4 .649 .925 a 0.3=< All item-total correlation <=0.95 indicates Reliability (Hair Jr. et al., 2010) b All instruments’ alpha >0.7 indicates Reliability (DeVellis, 2016) 229 Appendix 5.7 Sample characteristics consideration Testing for Normality Normality is degree to which the distribution of the sample data corresponds to a normal distribution (Hair Jr. et al., 2010). There are two kinds of normality involving into univariate normality and multivariate normality. “A variable is multivariate normal, it is then univariate normal while the reverse is not necessarily true However, multivariate normality is more difficult to test In most cases, assessing and achieving univariate normality for all variables is sufficient, multivariate normality is addressed only when it is especially critical” (Hair Jr. et al., 2010, p. 70). The shape of any distribution can be described by two measures: kurtosis and skewness. Kurtosis refers to the “peakedness” or “flatness” of distribution compared with the normal distribution while skewness is used to describe the balance of the distribution (Hair Jr. et al., 2010, p. 70). If either zskewness and zkurtosis exceeds +/-1.96 at 0.05 significance level, it can be said that the degree to which the skewness and peakedness of the distribution vary from the normality distribution (Hair Jr. et al., 2010). Kurtosis and Skewness test is calculated on SPSS and shows that almost 66 variables except for ITU3 and ITU4, which are not main variables of the ESMAP, meet the proposed level (-1.96 <= either zskewness and zkurtosis <= 1.96), so that they are normally distributed. The results of the normality test are display in the following table. 230 Results of Normal Distribution Test Var. Mean Std. Dev. zskewness Zkurtosis Var. Mean Std. Dev. zskewness Zkurtosis SQ1 5.35 1.120 -0.109 -0.823 USE2 5.41 1.247 -0.713 0.495 SQ2 5.01 1.192 0.014 -0.812 USE3 5.73 0.950 -0.396 -0.182 SQ3 5.84 1.021 -0.881 0.463 USE4 5.50 0.961 0.000 -0.683 SQ4 5.02 1.283 -0.444 -0.030 USE5 5.20 1.192 -0.517 0.487 SQ5 5.43 1.083 -0.573 0.115 USE6 5.32 1.243 -0.732 1.022 SQ6 5.61 1.007 -0.453 -0.285 USE7 5.64 0.977 -0.709 0.304 SQ7 5.14 1.197 -0.278 -0.325 USE8 5.26 1.247 -0.637 0.403 SQ8 4.70 1.430 -0.631 0.314 USE9 5.36 1.106 -0.373 -0.175 IQ1 5.49 0.944 -0.403 -0.909 USE10 5.43 1.018 -0.109 -0.557 IQ2 5.37 1.122 -0.259 -0.616 USE11 4.96 1.387 -0.732 0.415 IQ3 5.30 1.034 -0.307 -0.045 SAT1 5.43 1.010 -0.561 0.386 IQ4 5.41 1.025 -0.510 0.204 SAT2 5.52 1.004 -0.882 1.299 IQ5 5.37 1.004 -0.590 0.228 SAT3 5.58 0.923 -0.420 -0.107 IQ6 5.21 1.092 -0.465 0.308 SAT4 5.58 0.993 -0.630 0.680 PAB1 5.68 1.022 -0.557 -0.241 AP1 3.97 0.709 -0.240 -0.204 PAB2 5.38 1.054 -0.628 0.336 AP2 3.93 0.764 -0.346 -0.190 PAB3 5.65 1.157 -1.002 1.149 AP3 3.91 0.745 -0.347 -0.045 PAB4 5.63 1.045 -0.519 -0.068 AP4 3.99 0.680 -0.153 -0.313 PAB5 5.29 1.233 -0.822 1.181 OP1 3.88 0.832 -0.400 0.108 PAB6 5.65 1.042 -0.748 0.639 OP2 3.98 0.698 -0.279 -0.054 PAB7 5.63 1.070 -0.625 -0.014 OP3 4.01 0.704 -0.305 -0.078 PAB8 5.63 1.020 -0.514 -0.106 OP4 3.95 0.743 -0.544 0.396 PAB9 5.55 1.068 -0.490 -0.328 OP5 3.90 0.703 -0.596 0.788 PAB10 5.59 1.024 -0.680 0.260 OP6 3.96 0.760 -0.513 0.199 PAB11 5.50 1.069 -0.860 0.878 OP7 4.04 0.600 -0.14 -0.165 PAB12 5.50 1.004 -0.532 0.404 OP8 4.18 0.622 -0.143 -0.498 PAB13 5.63 0.978 -0.738 0.991 U1 5.23 1.789 -0.918 -0.213 PAB14 5.80 0.913 -0.601 0.221 U2 5.48 1.577 -1.137 0.620 PAB15 5.53 0.943 -0.256 -0.315 U3 5.42 1.580 -1.027 0.261 PAB16 5.38 0.996 -0.348 -0.195 ITU1 6.18 0.976 -1.243 1.354 PAB17 4.92 1.459 -0.795 0.608 ITU2 6.14 0.938 -1.159 1.198 PAB18 4.78 1.323 -0.322 0.161 ITU3 6.18 0.979 -1.528 3.067 USE1 5.83 1.048 -0.684 -0.154 ITU4 5.92 1.142 -1.418 2.886 * (-1.96 <= either zskewness and zkurtosis <= 1.96) indicates that the data set is normal distribution (Hair Jr. et al., 2010) Testing for homoscedasticity “When the variance of the error terms ( e) appears constant over a range of predictor variable, the data are said to be homoscedastic” (Hair Jr. et al., 2010, p. 34). Homoscedasticity is evaluated for pairs of variables. To conduct the homoscedasticity test, the variance of the dependent variable values are compared at each value of the independent variable. If the variance values are relatively equal, it indicates that the dependence relationship between the dependent variable and the independent variable exist (Osterlind, Tabachnick, & 231 Fidell, 2001). The result of homoscedasticity is represented through either graphical or statistical methods. This study applies a graphical method. The examination of homoscedasticity shows that homoscedasticity of the data set is certain. Some examples of the homoscedasticity test are presented as follow: AP1 (independent variable) and OP1 (dependent variable). R2 linear = 0.815. Variables are homoscedastic. AP1 (independent variable) and OP8 (dependent variable). R2 linear = 0.759. Variables are homoscedastic. 232 Testing for linearity Linearity refers to a consistent slope of change that represents the relationship between the independent variable and the dependent variable. If a relationship is nonlinear, the statistics, which assume it is linear, will not only underestimate the strength of the relationship but also cause inaccurate predictions, especially when the result is generalized beyond the range of the sample (Osterlind et al., 2001). The linearity test is achieved through either graphical or statistical method. This study utilizes the statistical method. The statistical test for linearity is conducted by analyzing the correlation matrices for the dependent variables and the independent variables. If the correlation coefficient between an independent variable and a dependent variable is not statistically significant (p>0.05), the relationship is linear. This study chooses to test randomly linearity between independent variable AP4 and dependent variables OP1-OP8. The deviation for linearity test available in the ANOVA test on SPSS indicates that the significant values for deviation from linearity are greater than 0.05 (see the following table). Therefore, the assumption of linearity in the current data set is valid. Table of Test of linearity between independent variable AP4 and dependent variables OP1-OP8 ANOVA Table Sum of Squares df Mean Square F Sig. OP1 * AP4 Between Groups (Combined) 25.653 3 8.551 13.629 .000 Linearity 25.041 1 25.041 39.913 .000 Deviation from Linearity .612 2 .306 .488 .615 Within Groups 82.817 132 .627 Total 108.471 135 OP2 * AP4 Between Groups (Combined) 25.072 3 8.357 19.462 .000 Linearity 23.151 1 23.151 53.910 .000 Deviation from Linearity 1.921 2 .961 2.237 .111 Within Groups 56.685 132 .429 Total 81.757 135 OP3 * AP4 Between Groups (Combined) 26.096 3 8.699 17.095 .000 Linearity 25.790 1 25.790 50.682 .000 Deviation from Linearity .307 2 .153 .302 .740 Within Groups 67.168 132 .509 Total 93.265 135 OP4 * Between Groups (Combined) 24.000 3 8.000 14.554 .000 Linearity 23.066 1 23.066 41.962 .000 233 AP4 Deviation from Linearity .934 2 .467 .850 .430 Within Groups 72.559 132 .550 Total 96.559 135 OP5 * AP4 Between Groups (Combined) 26.009 3 8.670 18.533 .000 Linearity 24.812 1 24.812 53.041 .000 Deviation from Linearity 1.197 2 .599 1.280 .282 Within Groups 61.748 132 .468 Total 87.757 135 OP6 * AP4 Between Groups (Combined) 28.122 3 9.374 17.998 .000 Linearity 27.634 1 27.634 53.055 .000 Deviation from Linearity .488 2 .244 .469 .627 Within Groups 68.753 132 .521 Total 96.875 135 OP7 * AP4 Between Groups (Combined) 23.099 3 7.700 21.247 .000 Linearity 21.712 1 21.712 59.913 .000 Deviation from Linearity 1.387 2 .694 1.914 .152 Within Groups 47.835 132 .362 Total 70.934 135 OP8 * AP4 Between Groups (Combined) 20.404 3 6.801 19.542 .000 Linearity 20.191 1 20.191 58.014 .000 Deviation from Linearity .213 2 .106 .306 .737 Within Groups 45.942 132 .348 Total 66.346 135 234 Appendix 5.8 Process of interpreting the factors (1) Determine the number of factors through the percentage of variance (% Var) and Variance (Eigenvalue). The percentage of variance (% Var) indicates the amount of variance that the factors explain. The variance equals eigenvalue when principal components analysis is used to extract factors. Hair Jr. et al. (2010) and Fornell and Larcker (1981) suggest retaining the factors that explain an acceptable level of variance, which means that the percentage of variance is greater than 60% and eigenvalue is greater than 1. (2) Item consideration (Garver & Mentzer, 1999; Hair Jr. et al., 2010):  Retain items with high loadings ( ³0.5)  Consider items with medium loadings (0.4,0.5):  Loading difference ³0.3: keep item  Loading Difference <0.3: delete item  The deletion of items is undertaken with careful consideration. In terms of statistic aspect, the elimination is reasonable if it makes increase the percentage of variance of the scale. In terms of conceptualization aspect, the elimination will be able to lead losing the content validity of the corresponding instrument. Therefore, an item will be deleted if the remaining items on the scale cover all the important dimensions of the instrument.  Eliminate items with low loadings ( £0.4) (3) When deciding to delete any item of an instrument, EFA will be re-run, the process of interpreting the factors will be repeated. 235 Appendix 5.9 EFA Test Results - Scales without modifications Items Factor loadinga % Variance extractedb Eigen- valuec Instrument structure Information quality IQ4 IQ3 IQ6 IQ5 IQ2 IQ1 0.91 0.83 0.80 0.77 0.74 0.67 68.26 4.10 One factor extractedd Satisfaction SAT4 SAT2 SAT1 SAT3 0.96 0.94 0.93 0.88 89.21 3.57 One factor extracted Accountant performance AP3 AP1 AP2 AP4 0.89 0.87 0.86 0.81 79.71 3.19 One factor extracted Organizational performance OP2 OP7 OP5 OP6 OP3 OP1 OP8 OP4 0.83 0.80 0.79 0.78 0.77 0.74 0.73 0.71 64.22 5.14 One factor extracted Extent of Use U1 U2 U3 0.94 0.93 0.91 90.37 2.72 One factor extracted Intend to Use 78.23 3.13 One factor extracted ITU1 0.80 ITU2 0.96 ITU3 0.94 ITU4 0.67 a All item Factor loadings >=0.5 indicates related items are retained (Hair Jr. et al., 2010) b All % Var. >= 60% indicates the number of factors extracted (Fornell & Larcker, 1981) c All eigenvalue >1 indicates the number of factors extracted (Fornell & Larcker, 1981) d All instrument extracting one factor are first-order instruments (Hair Jr. et al., 2010) 236 Appendix 5.10 Summary after Factor Analysis Construct Factor Items Number of items Cronbach’s alpha* System quality (SQ) System related quality (SQ_system) SQ1, SQ2, SQ7 3 0.83 Task related quality (SQ_task) SQ5, SQ6, SQ8 3 0.80 Information quality (IQ) IQ IQ1, IQ2, IQ3, IQ4, IQ5, IQ6 6 0.90 Perceived Accounting Benefit (PAB) Operational accounting benefit (PAB_operation) PAB6, PAB7, PAB8, PAB9, PAB10 5 0.96 Organizational accounting benefit (PAB_organization) PAB11, PAB12, PAB13, PAB14, PAB15, PAB16, PAB17, PAB18 8 0.91 Use (USE) Decision support (USE_decision) USE1, USE3, USE4, USE7 4 0.85 Work integration (USE_work) USE5, USE6, USE8, USE11 4 0.89 Satisfaction (SAT) SAT SAT1, SAT2, SAT3, SAT4 4 0.95 Accountant performance (AP) AP AP1, AP2, AP3, AP4 4 0.91 Organizational performance (OP) OP OP1, OP2, OP3, OP4, OP5, OP6, OP7, OP8 8 0.92 Extent of Use (Ex_U) Ex_U U1, U2, U3 3 0.95 Intend to Use (ITU) ITU ITU1, ITU2, ITU3, ITU4 4 0.90 * All Cronbach’s alpha >= 0.7 indicate that all factors are reliability (DeVellis, 2016) 237 Appendix 5.11.1 Results of fit consideration of higher-factor instruments – PAB Amos model fit indicators (Hu & Bentler, 1999) Indicators Abbr. Best Construct: Perceived Accounting Benefit Model 1 Model 2 First-order construct Second-order construct Chi-square/df CMIN/DF <3 good; <5 sometimes permissible 7.452 3.382 p-value for the model p-value >0.05 0.00 0.00 Comparative fit index CFI >0.95 great >0.90 traditional >0.8 sometimes permissible 0.735 0.904 Goodness of Fit index GFI >0.95 0.50 0.782 Adjusted Goodness of Fit index AGFI >0.80 0.301 0.691 Root Mean Square RMR Close 0 0.143 0.098 Root Mean Squared error of approximation RMSEA <0.05 good 0.05-0.10 moderate >0.10 bad 0.233 0.141 PCLOSE PCLOSE >0.05 0.00 0.000 238 Appendix 5.11.2 Results of fit consideration of higher-factor instruments - USE Amos model fit indicators (Hu & Bentler, 1999) Construct: Use Indicators Abbr. Best Model 1 Model 2 First-order construct Second-order construct Chi-square/df CMIN/DF <3 good; <5 sometimes permissible 7.612 3.277 p-value for the model p-value >0.05 0.00 0.00 Comparative fit index CFI >0.95 great >0.90 traditional >0.8 sometimes permissible 0.784 0.929 Goodness of Fit index GFI >0.95 0.711 0.898 Adjusted Goodness of Fit index AGFI >0.80 0.480 0.806 Root Mean Square RMR Close 0 0.119 0.085 Root Mean Squared error of approximation RMSEA <0.05 good 0.05-0.10 moderate >0.10 bad 0.236 0.138 PCLOSE PCLOSE >0.05 0.00 0.02 239 Appendix 5.12 Cross-factor loadings SQ_t ask SQ_sys tem IQ PAB_oga nizational PAB_oper ational USE_ work USE_d ecision SAT AP OP SQ5 0.908 0.424 0.634 0.594 0.443 0.291 0.348 0.66 0.397 0.392 SQ6 0.892 0.47 0.686 0.606 0.507 0.32 0.402 0.667 0.425 0.463 SQ8 0.773 0.447 0.586 0.447 0.445 0.243 0.216 0.537 0.234 0.369 SQ2 0.386 0.863 0.503 0.433 0.501 0.413 0.42 0.389 0.346 0.213 SQ1 0.458 0.88 0.465 0.434 0.365 0.372 0.356 0.44 0.27 0.286 SQ7 0.498 0.847 0.527 0.416 0.452 0.179 0.285 0.42 0.343 0.295 IQ4 0.654 0.422 0.902 0.597 0.529 0.36 0.446 0.61 0.465 0.437 IQ3 0.628 0.499 0.857 0.564 0.493 0.375 0.406 0.572 0.428 0.396 IQ6 0.598 0.515 0.838 0.595 0.547 0.349 0.467 0.586 0.451 0.379 IQ5 0.598 0.425 0.81 0.615 0.589 0.283 0.382 0.538 0.376 0.395 IQ2 0.524 0.449 0.792 0.513 0.459 0.241 0.41 0.481 0.306 0.344 IQ1 0.654 0.544 0.748 0.555 0.524 0.331 0.46 0.563 0.396 0.407 PAB15 0.47 0.334 0.456 0.83 0.536 0.46 0.602 0.597 0.446 0.557 PAB16 0.567 0.272 0.544 0.799 0.534 0.522 0.494 0.653 0.512 0.54 PAB17 0.477 0.368 0.523 0.762 0.486 0.462 0.444 0.527 0.517 0.575 PAB14 0.473 0.376 0.493 0.781 0.538 0.284 0.593 0.574 0.512 0.495 PAB12 0.565 0.565 0.648 0.891 0.733 0.451 0.634 0.688 0.492 0.502 PAB13 0.539 0.468 0.645 0.867 0.695 0.374 0.605 0.68 0.484 0.526 PAB11 0.528 0.443 0.586 0.838 0.67 0.61 0.62 0.636 0.503 0.561 PAB18 0.515 0.309 0.566 0.637 0.477 0.437 0.446 0.516 0.456 0.538 PAB7 0.476 0.412 0.57 0.662 0.956 0.345 0.467 0.582 0.49 0.416 PAB6 0.48 0.445 0.566 0.646 0.951 0.356 0.444 0.54 0.474 0.388 PAB8 0.477 0.399 0.571 0.638 0.95 0.321 0.431 0.523 0.435 0.44 PAB9 0.492 0.498 0.573 0.718 0.929 0.434 0.554 0.644 0.585 0.465 PAB10 0.586 0.596 0.664 0.742 0.857 0.507 0.562 0.618 0.515 0.448 USE11 0.223 0.332 0.3 0.411 0.265 0.869 0.432 0.352 0.313 0.275 USE8 0.326 0.338 0.398 0.534 0.445 0.918 0.585 0.463 0.467 0.354 USE5 0.302 0.281 0.385 0.507 0.409 0.905 0.576 0.462 0.481 0.384 USE6 0.308 0.347 0.293 0.497 0.353 0.81 0.625 0.469 0.36 0.335 USE3 0.316 0.38 0.443 0.529 0.454 0.51 0.851 0.455 0.472 0.338 USE4 0.312 0.368 0.393 0.604 0.424 0.628 0.875 0.481 0.459 0.429 USE1 0.404 0.34 0.496 0.64 0.49 0.414 0.785 0.424 0.385 0.432 USE7 0.242 0.269 0.415 0.549 0.408 0.545 0.813 0.494 0.53 0.442 SAT1 0.704 0.463 0.65 0.708 0.552 0.488 0.507 0.945 0.634 0.651 SAT2 0.679 0.465 0.648 0.752 0.594 0.509 0.562 0.951 0.688 0.625 SAT3 0.667 0.429 0.635 0.73 0.639 0.453 0.537 0.922 0.586 0.586 SAT4 0.692 0.467 0.633 0.684 0.586 0.441 0.502 0.959 0.672 0.621 AP1 0.332 0.326 0.424 0.574 0.514 0.436 0.509 0.595 0.896 0.532 AP2 0.296 0.267 0.368 0.464 0.427 0.359 0.49 0.557 0.887 0.505 AP3 0.478 0.35 0.508 0.573 0.46 0.437 0.483 0.693 0.917 0.643 240 AP4 0.353 0.368 0.448 0.549 0.525 0.428 0.508 0.583 0.87 0.562 OP2 0.466 0.267 0.437 0.633 0.489 0.275 0.41 0.584 0.507 0.846 OP7 0.444 0.235 0.408 0.532 0.306 0.288 0.402 0.605 0.574 0.834 OP5 0.309 0.199 0.274 0.449 0.324 0.286 0.317 0.531 0.532 0.817 OP6 0.345 0.174 0.385 0.474 0.279 0.362 0.381 0.53 0.551 0.818 OP3 0.337 0.284 0.382 0.537 0.45 0.363 0.408 0.495 0.452 0.794 OP1 0.433 0.321 0.508 0.527 0.441 0.269 0.388 0.507 0.453 0.77 OP8 0.388 0.209 0.384 0.604 0.346 0.298 0.49 0.466 0.519 0.777 OP4 0.325 0.322 0.289 0.505 0.388 0.347 0.359 0.48 0.435 0.747 Appendix 5.13 Internal consistency and convergent validity results of the first order factor “PAB_operational” after eliminating indicator PAB18 PAB_operational with PAB18 (the original measurement model) PAB_operational without PAB18 Composite Reliability 0.935 0.939 AVE 0.646 0.688 241 Appendix 5.14 Decision-making process for keeping or deleting formative indicators basing on outer weight and outer loading Hair Jr et al. (2014) 242 Appendix 5.15 Results of alternative models analysis Table 1: Direct relationships for Hypothesis testing (alternative model related to Extent of Use) Hypothesis Relationship Std Beta Std Error [t - value]^ Decision H1: AP -> OP 0.6379 0.0667 9.4765*** Supported H2: Ex_U -> AP 0.0636 0.093 0.6202 No H3: SAT -> AP 0.6503 0.0643 10.2516*** Supported H4a: SQ -> Ex_U 0.4461 0.1483 3.0417** Supported H4b: SQ -> SAT 0.2909 0.0899 3.2691** Supported H5a: IQ -> Ex_U -0.1327 0.1622 0.8722 No H5b: IQ -> SAT 0.124 0.0979 1.2363 No H6a: PAB -> Ex_U 0.0343 0.1626 0.2278 No H6b: PAB -> SAT 0.4674 0.1012 4.6101*** Supported *p<0.05, **p<0.01, ***p<0.001 Table 2: Direct relationships for Hypothesis testing (alternative model related to Extended Use) H: Relationship Std Beta Std Error [t - value]^ Decision H1: AP -> OP 0.6369 0.0656 9.6359*** Supported H2: Extended_U -> AP 0.2008 0.08 2.3949* Supported H3: SAT -> AP 0.5533 0.0775 7.2462*** Supported H4a: SQ -> Extended_U 0.082 0.1085 0.822 No H4b: SQ -> SAT 0.2906 0.0874 3.3731*** Supported H5a: IQ -> Extended_U 0.0764 0.1089 0.5993 No H5b: IQ -> SAT 0.1285 0.1024 1.1609 No H6a: PAB -> Extended_U 0.4787 0.1081 4.4775*** Supported H6b: PAB -> SAT 0.4644 0.101 4.6328*** Supported *p<0.05, **p<0.01, ***p<0.001 243 Figure 1: Hypotheses Testing: Bootstrapping Direct Effect Results (alternative model related to Extent of Use) 244 Figure 5.8: Hypotheses Testing: Bootstrapping Direct Effect Results (alternative model related to Extended Use 245 Appendix 6.1 Demographic characteristics of surveyed companies Frequency % Type of ownership 100% foreign-owned enterprises 23 19.2 State-owned enterprises (>=51% government capital) 22 18.3 Private enterprises/ limited enterprises 54 45 Joint venture with foreign partners 14 11.7 Joint venture with domestic partners 7 5.8 Total 120 100.0 Type of industry sectors Manufacture 72 60.0 Commerce 44 36.7 Services 42 35.0 Total 120 Type of industry Bank, insurance, investment 2 1.7 Chemical & Pharmaceuticals 3 2.5 Dairy, food & meat products 28 23.3 Electrical & Electronics 7 5.8 Medical & healthcare 8 6.8 Information technology (IT) 10 8.3 Manufacturing 12 10.0 Retail/ Wholesale/ Distribution 25 20.8 Telecommunications 3 2.5 Transportation, logistics & courier 7 5.8 Construction 6 5.0 Others (beverages, fashion, design, FMCG,) 9 7.5 Total 120 100.0 Company size (paid-in capital) in VND billion < 10 3 2.5 10 – 50 6 5.0 >50 – 100 11 9.2 >100 – 200 12 10.0 >200 – 500 14 11.7 >500 – 1000 22 18.3 >1000 52 43.3 Total 120 100.0 Company size (number of employees) <=50 8 6.7 50 – 200 13 10.8 201 – 500 29 24.2 501 – 1000 23 19.2 1001 – 5000 32 26.7 5001 – 10000 9 7.5 >10000 6 5.0 Total 120 100.0 (Source: by author) 246 Appendix 6.2 Demography characteristics of ERP system Frequency % Type of ERP software Oracle 20 16.7 SAP 43 35.8 XMAN (ERP) 2 1.7 SalesUp ERP 2 1.7 Navision 3 2.5 Microsoft Dynamic 4 3.3 Lemon 3 2.5 FAST (ERP) 3 2.5 Others (AMIS – MISA, Bamboo, Bravo, Bross, IMAS, Maconomy, MMIS, Peoplesoft, PERP) 40 33.3 Total 120 100.0 Years that ERP is implemented and used at current company <1 year 0 0.0 1- 2 years 21 17.5 >2 – 4 years 16 13.3 >4 – 6 years 37 30.8 >6 – 8 years 13 10.8 >8 years 33 27.5 Total 120 100.0 (Source: by author) 247 Appendix 6.3 Demographic characteristics of informants Frequency % Min Max Mean Position in the firm (job title) Chief Finance Officer CFO 15 12.5 Chief accountant 39 32.5 Internal controller 45 37.5 Internal auditor 15 12.5 Management accountant 6 5.0 Total 120 100.0 Position in organization’s hierarchy Top management position 27 22.5 Mid-level personnel 51 42.5 Senior staff 39 32.5 Staff 3 2.5 Total 120 100.0 Gender Female 63 52.5 Male 57 47.5 Total 120 100.0 Education background College degree 0 0.0 University bachelor degree 101 84.2 Master degree 19 15.8 Total 120 100.0 Age < 25 3 2.5 25 – 34 66 55.0 35 – 44 51 42.5 >44 0 00.0 Total 120 100.0 Experience Years at current position (years) 1 20 6.5 Years using ERP at current position (years) 1 5 2.7 Extent of using ERP system (the degree users perceive the following statements according to 7-likert scale ranging from strongly disagree to strongly agree) Many hours per day at work 1 7 5.2 Many times per day at work 1 7 5.5 Overall, use ERP a lot 1 7 5.4 Intend to continuously use (the degree users perceive the following statements according to 7-likert scale ranging from strongly disagree to strongly agree) We intend to continue using the ERP in our job 3 7 6.2 We intend to use more functions of the ERP 3 7 6.1 We intend to continue using the ERP for processing more tasks 2 7 6.2 We intend to suggest that our company should continue to use the current ERP system 1 7 5.9 (Source: by author) 248 Appendix 6.4 Comparing sample means The Independent Samples T test is generally used for comparing sample means to indicate whether there is sufficient evidence to infer that the means of the two sample distributions significantly differ from each other. The two samples are measured on common some variables of interest but there is no overlap of memberships between two groups. These variables are the length of ERP adoption, organizational size and utilized ERP packages. Based on Trọng and Ngọc (2008), if the significance level of Levene’s Test is less than 0.05, the significance level of T-test is the value of hypothesis “Equal variances assumed”, in contrast, if the significant level of Levene’s Test is greater than 0.05, the significance level of T-test is the value of hypothesis “Equal variances not assumed”. And the significance level of T-test must be below 0.05 to assert that there exists the difference to population between two groups. Accordingly, in terms of p-values greater than 0.05, there are not enough proofs to prove that there exist the differences between each group set (see Table 1, 2 and 3). Table 1: Results of the Independent Samples T-test by the length of ERP adoption Levene's Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2-tailed) Mean Difference SQ Equal variances assumed .094 .760 -1.565 118 .120 -.32780 Equal variances not assumed -1.533 28.514 .136 -.32780 IQ Equal variances assumed 3.433 .066 .048 118 .962 .00986 Equal variances not assumed .039 24.545 .970 .00986 PAB Equal variances assumed 1.782 .184 -1.070 118 .287 -.21645 Equal variances not assumed -.907 25.344 .373 -.21645 USE Equal variances assumed .991 .322 -1.011 118 .314 -.21122 Equal variances not assumed -.899 26.242 .377 -.21122 249 Table 2: Results of the Independent Samples T-test by the organizational size Levene's Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2-tailed) Mean Difference SQ Equal variances assumed 1.549 .216 -.503 118 .616 -.10833 Equal variances not assumed -.554 30.120 .583 -.10833 IQ Equal variances assumed .571 .451 -.230 118 .819 -.04833 Equal variances not assumed -.241 28.532 .811 -.04833 PAB Equal variances assumed .030 .864 -1.089 118 .278 -.22462 Equal variances not assumed -1.050 26.259 .303 -.22462 USE Equal variances assumed 2.699 .103 -.240 118 .811 -.05125 Equal variances not assumed -.207 24.116 .838 -.05125 SAT Equal variances assumed .082 .775 -.131 118 .896 -.03000 Equal variances not assumed -.140 29.015 .889 -.03000 AP Equal variances assumed .111 .740 .661 118 .510 .10500 Equal variances not assumed .690 28.358 .496 .10500 OP Equal variances assumed .987 .322 .964 118 .337 .13375 Equal variances not assumed 1.097 31.339 .281 .13375 SAT Equal variances assumed .474 .493 -.717 118 .475 -.16017 Equal variances not assumed -.656 26.844 .518 -.16017 AP Equal variances assumed 15.533 .000 .481 118 .631 .07504 Equal variances not assumed .361 23.541 .721 .07504 OP Equal variances assumed .698 .405 -.639 118 .524 -.08712 Equal variances not assumed -.621 28.336 .540 -.08712 250 Table 3: Results of the Independent Samples T-test by the adopted ERP packages Levene's Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2-tailed) Mean Difference SQ Equal variances assumed .000 .997 .165 118 .869 .02659 Equal variances not assumed .165 114.399 .870 .02659 IQ Equal variances assumed .075 .785 -.281 118 .779 -.04414 Equal variances not assumed -.280 114.036 .780 -.04414 PAB Equal variances assumed .196 .658 1.575 118 .118 .24099 Equal variances not assumed 1.568 114.151 .120 .24099 USE Equal variances assumed 1.260 .264 .083 118 .934 .01326 Equal variances not assumed .082 109.653 .935 .01326 SAT Equal variances assumed .476 .491 .624 118 .534 .10610 Equal variances not assumed .619 111.277 .537 .10610 AP Equal variances assumed .441 .508 -.239 118 .811 -.02840 Equal variances not assumed -.238 113.786 .812 -.02840 OP Equal variances assumed 1.360 .246 .837 118 .404 .08678 Equal variances not assumed .828 107.842 .409 .08678 251 Appendix 6.5 Testing the D&M IS Success model (1992) without PAB Results of direct effects in the D&M IS success model (1992) H Relationship Std Beta Std Error [t - value]^ Decision 95% CI LL 95% CI UL H1: AP -> OP 0.6416 0.0613 10.3101*** Supported 0.4982 0.7124 H2: USE -> AP 0.2506 0.0715 3.5106*** Supported 0.124 0.3555 H3: SAT -> AP 0.536 0.0657 8.1686*** Supported 0.426 0.6462 H4a: SQ -> USE 0.2052 0.1193 1.7689 No 0.0039 0.3971 H4b: SQ -> SAT 0.4277 0.0849 5.0957*** Supported 0.2896 0.5708 H5a: IQ -> USE 0.3485 0.1186 2.9128** Supported 0.1428 0.532 H5b: IQ -> SAT 0.3575 0.0907 3.8697*** Supported 0.1939 0.4889 (Source: by author) 252 Appendix 7.1 Practical implications for ERP adoption organizatons’ stakeholders 7.3.2.1 For ERP vendors: The difference on the relationship between perceived accounting benefit and satisfaction by the ERP vendors recorded significantly indicates that at originations employing popular ERP packages such as SAP and Oracle, satisfaction of accounting professionals is increasingly higher when they perceive accounting-benefits that the ERP system brings for their work. This finding is meaningful for popular ERP vendors (SAP and Oracle). They understand their products better; determine their tools’ impacts when implementing them in practice; then provide valuable experiences for their customers in duration of ERP implementation and post- implementation. The more advantaged their customers receive from their products, the higher their reputation is. 7.3.2.2 For ERP-adopting organizations The ESMAP allows maximize the ERP’s impacts on accounting professionals, which in turn increasingly enhance the ERP’s impacts on organizations. To put it differently, organizations will obtain beneficial outcome when individuals (particular, accountant professional) use successfully the ERP system to add more value to organization. Therefore, the ERP-adopting organizations receive indirect benefits from the ESMAP. 7.3.2.3 For accounting professionals The ERP system is still company’s backbone for the planning, control, and execution of all business processes20, even in revolution of industry 4.0. However, the role of accounting professions, especially CFOs and senior controllers, has dramatically changed. They are no longer seen as passive manager, but partners actively involved in creating business strategies. The ESMAP guides them how to be come productive under ERP settings. In short, the ESMAP supports them to adjust their ERP adopting behavior in order to improve their work performance, and via this, enhance organizational performance. 20 According to Deloitte’s report “Industry 4.0: is your ERP system ready for the digital era?” quoted from website: https://www2.deloitte.com/content/dam/Deloitte/de/Documents/technology/Deloitte_ERP_Industrie- 4-0_Whitepaper.pdf 253 7.3.2.4 For organization management The results of this study support organization management to increasingly acknowledge the power of the ERP system for its usage of senior accounting experts. In other words, the study provides guidance on how to guide accountant professionals to effectively adopt the ERP system in organization management’s efforts to improve organizational performance. Besides, it allows management to better manage, control accounting experts and their work in ERP adoption and ongoing context. For example, organizational management should recognize that a high level of system quality and perceived accounting benefit are likely to result in improved organizational performance via increased effective use, satisfaction and accountant performance. More specifically, given discussions on findings in Chapter 6, which is modeled in Figure 7.1, the author identifies practical implications, summarizes and categories them into four instruction sets for organization management, namely: Black arrow line: direct effect Red arrow line: indirect effect Blue line: the relationship has difference by length of ERP system or by firm size or by ERP vendor Figure 7.1: The supported relationships useful for organization management A set of instructions on advancing organizational performance The following solutions are ordered in descend manner of impact extent of each construct on organizational performance. First, organization management should focus on finding how to improve accountant performance, since it is not only a direct driver, also a mediator controlling the impacts of use and satisfaction on organizational performance, and also a mediator managing the impacts of perceived accounting benefit on organizational performance. More specifically, management should frequently organize seminars for accounting department, or thoroughly communicate 254 to accounting-professionals on ERP system’s advantages for their work such as “they can learn much through the presence of the ERP system” (AP1), “the ERP system helps them to more enhance their awareness and recall of job-related information” (AP2), “the ERP system also helps them to more enhance their effectiveness in their job” (AP3), and “the ERP system allows them more increasing their productivity” (AP4). Noting that, large-sized companies’ management need to more focus on raising these perceptions of accounting professionals. Analogous to small-and-medium-sized firms, the higher accounting department perceives the ERP impacts on them; the more massive their organization performance is enhanced. Second, management should pay more attention on accounting department’s satisfaction of the ERP system, as it offers both direct and indirect impacts on organizational performance. In a little more detail, management needs to find out how to make accountant professional more satisfy of system quality (SAT1), information quality (SAT2), perceived accounting benefit (SAT3) and overall ERP system (SAT4). Third, the most different and dominant contribution of the ESMAP, management needs to find out how to improve accounting department’s perceptions of accounting-related benefits received from the ERP system as these perceptions influence directly and indirectly organizational performance. These perceptions include “the ERP system makes reduction of time for closure of monthly account” (PAB6), “the ERP system makes reduction of time for closure of quarterly account” (PAB7), “the ERP system makes reduction of time for closure of annual accounts” (PAB8), “the ERP system makes reduction of time for issuing of financial statements” (PAB9), “the ERP system increases flexibility in information generation” (PAB10), “the ERP system increase integration of accounting applications” (PAB11), “the ERP system improves decisions based on timely and reliable information” (PAB12, “the ERP system improve quality of reports- statements of account” (PAB13), “the ERP system improves internal audit function” (PAB14), “the ERP system improve working capital control” (PAB15), “the ERP system increase use of financial ratio analysis” (PAB16), “the ERP system make reduction of time for issuing payroll” (PAB17), “the ERP system makes reduction of personnel of accounting department” (PAB18). Overall, the more clarified the accountant professional perceive accounting-benefits obtained from the ERP system, the higher the organizational performance is. Simultaneously, the more clarified the accountant professional perceive accounting-related benefits received from the ERP system, the more effective they use the ERP system (when they are master of it), and in the meantime, the more satisfied they feel for the ERP system, then the more productive they are, and finally, the more gainful their organizations offer. 255 Fourth, management should concern about how to improve effective use because it has nearly full indirect impact on organizational performance via accountant performance. Based on the measures of effective use, the author suggests that making accounting professionals clarify how effectively the ERP system used for their decision support and work integration is appropriate solution to improve effective use. For example, effectiveness of using the ERP system means that using it enables them to improve the efficiency of the decision process (USE1), manage their work (USE3), make explicit the reasons for their decision (USE4), make sense out of data (USE7), communicate with people in other work groups (USE5), monitor their own performance (USE6), communicate with people in other departments (USE8), and keep people in other department informed (USE11). A set of instructions on enhancing accountant performance The following solutions are ordered in descend manner of impact extent of each construct on accountant performance. First, management should find out how to improve the satisfaction of accounting professionals, which is more necessary in large-sized companies. The more satisfied accounting professionals are with ERP system, the more strongly they agree that the ERP system supports them to perform well their routine work. Second, a different way to enhance accountant performance relates to increasing how effectively when using the ERP, especially in large-sized organizations. To put it differently, this study provides a strong proof for management to convince their inexperience and unconfident accounting professionals that the more extensively the ERP system is used for decision support and work integration, the more likely these experts will perceive that ERP system has considerably positive impacts on their work. Third, PAB has not direct impact on accountant performance, however PAB has indirect impact on accountant performance via use and satisfaction (parallel mediation model). Therefore, to advance accountant performance, management not only focuses on making use and satisfaction increased, also concentrate on making PAB improved. An instruction on improving effective use According to the ESMAP, to make accountant professionals use the EPR system more effectively, the only way that the management should conduct is to attempt to find how to increase PAB, as the more increasingly they perceive accounting-related benefits from the ERP system, the more effectively they use the ERP system. 256 A set of instructions on increasing accounting professionals’ satisfaction The following solutions are ordered in descend manner of impact extent of each construct on satisfaction First, organization management should find how to improve system quality, which in turn, absolutely, enhance satisfaction. For instance, information such as “the ERP system is easy to use” (SQ1), “the ERP system is easy to learn” (SQ2), “the ERP user interface can be easily adapted” (SQ7), “the ERP system meets users requirements” (SQ5), “the ERP system includes necessary features and functions for users job” (SQ6), and “the ERP can be easily modified or improved according to users needs” (SQ8) should be communicated commonly to accounting department. The earlier the accountant professionals realize system quality, the more satisfied they feel about the ERP system. Second, another solution of increasing satisfaction is making accountant professionals perceive accounting-related benefits, which the ERP system offers, more earlier and clarified. This is more meaningful in large-sized organizations or in firms employing popular ERP packages such as SAP and Oracle.