**4. Discussion**

The EFQM Excellence Model is based on the logical assumption that excellence in enablers will lead to superior results, and thus leadership drives policy and strategy, people management, and partnerships and resources, and these three elements influence the results through suitable processes [45]. As a quality model, the EFQM Excellence Model explains, through its enabler criteria, the areas that the organization should consider as input to improve its results, as well as the result indicators that must be achieved. In this regard, the EFQM Excellence model provides a pattern of relationships both between enablers and results, and between the criteria. Empirical evidence shows that significant relationships exist between the result elements, where results on one level contribute to outcomes on others [46,47]. The excellence model assumes that customer results, people results, and society results will, together, ultimately infer organizational performance. Research on the

relationship between enablers and results indicates that weaknesses in leadership can affect people, customer, societal, and key organizational results [48,49].

Based on the findings of this research, reliability of the empirical data was tested through Cronbach's alpha (0.883) and passed the required level of significance, which is 0.7. This result can be compared with the result by Carlos [50], Cronbach's alpha of 0.71. Hence, the result confirmed that the items (questions) have relatively high internal consistency. The selected regression model (ordinal logistics regression model) was proved to be a good fit with our data and passed the required level of significance (see Table 2). In addition to this, there was an assumption for the ordinal regression model to be met in relation to correlation, which states that correlation among independent variables should not be highly correlated. On the basis of this assumption, it can be observed from the Table 4 that there is a positive and negative correlation between independent variables. If the correlation is greater than 0.7, it is said to be highly correlated (multicollinearity). Accordingly, the assumption is met on the basis of this benchmark. The maximum correlation among independent variables is 0.662, for partnership and process variables, while the lowest correlation is −0.003 for partnership and customer results. These statistics indicate that the ordinal regression is an appropriate model to analyze and interpret the data.

We also considered the results of model fitting, goodness of fit, Pseudo R2 and test of parallel line. Nagelkerke's (R2) statistics showed that the independent variables explain about 71.9% of the variations in the outcomes. All those statistics confirm that the model is a good fit to explain the outcome. Based on these test results, estimates (coefficients) presented in Table 5 are calculated. Those estimates or coefficients of the independent variables (Strategy, Leadership, People, Partnership, Process, People Result, Customer Result, and Societal Result) determine the dependent variable (organizational performance). Coefficients are the change in the response associated with a one-unit change of the independent, all other independents being held constant. In short, these parameters are values for the regression equation to predict the dependent variable.

The performance of an urban cadastral system is measured through 8 independent variables, each with a satisfaction score between 1 (not at all satisfied) and 5 (completely satisfied), thus the sum of the overall performance of the cadastral organization could achieve a minimum of 8 and a maximum of 40. Since the performance of an urban cadastral system is measured only through the eight independent variables, there is no intercept (β0) that can be added with the rest of the estimated coefficients. Hence, based on the results obtained in Table 6, each independent variable predicted the dependent variable in different weights. In the case of the urban cadastral system of Addis Ababa, people result and customer result were estimated with the highest scores among all the independent variables, which are 1.724 and 1.657, respectively, while strategy and leadership were estimated with the lowest values, which are 0.354 and 0.489, respectively. Accordingly, people result was affected the most compared to the rest of the variables, and on the other hand, policy and strategy has less effect on the overall organizational result. Thus, the high value of the regression estimation parameter implies that there is strong causal relationship between the independent and dependent variables. Therefore, the overall performance of the organization is evaluated out of 40. Based on the findings of this research, the urban cadastral system organization scored 24.196 out of 40. In percentage, the Addis Ababa urban cadastral system organization has an overall performance of 62.3%.

With regard to interviews, ten cadastral sub-city directors were interviewed to respond to questions related to the performance of their organization in relation to EFQM criteria. They responded that the most bottlenecks for the achievement of their organization are strategic planning, quality of leadership, bureaucratic processes, and supply of resources. In this regard, most of the respondents (80%) agreed with this idea, in that it was not designed in a way that responds to the existing circumstances of the city.

In addition to this, results from the focus group discussion confirm that most problems emanate from the strategic plan, supply of resources, and leadership skill. Comparing these responses with the interview and questionnaire, it measures the reality of the organization.
