Communication Model Between Regional Institutions Towards Regional Development and Environment in North Sumatra Province

Regional planning experts have defined regions. Glasson (1974) in study of Talitha et al. [23] described the region as a continuous area located between the local level and the national level [9]. In Indonesia, the definition of the region has been defined in Law No. 24 of 1992 concerning spatial planning; the region is a space that is a geographical unit along with all related elements with boundaries and systems determined based on administrative aspects and functional aspects, this understanding can be understood that the region is not always physical but instead emphasizes the interaction between humans and the resources that exist in a particular geographical unit [10]. Conceptually, regions can be divided into four types, namely: (1) homogeneous areas; (2) nodal areas; (3) administrative areas; and (4) planning areas. Homogeneous areas are areas that are viewed from one [11].

According to kasikoen [12] based on the application of regional planning theory can be divided into four components:

a. Physical Planning [13]. Planning needs to be done to plan the region's development physically;

b. Macro-Economic Planning [14]. This planning deals with the economic planning of the region;

c. Social Planning [15]. Social planning discusses education, health, social integrity, living and working conditions, women, children, and criminal issues;

d. Development Planning. This concerns planning a comprehensive development program to achieve regional development [16].

Regional development is also an effort to empower three elements of stakeholders: The community [17], government, and entrepreneurs/private sector in a region, in utilizing natural resources and the environment by using technology [18]. According to the Director General of Spatial Planning (2003), the basic regional development principles are one as a growth center [19]. (1) Regional development is not only internal to the region but must be considered the spread or influence (spread effect) of growth that can be caused to the surrounding area, even nationally [20]. (2) Regional development requires cooperative development efforts between regions and is an essential requirement for the success of regional development [21]. (3) The pattern of regional development is integral, which is the integration of the areas covered in the region through an equitable approach [22]. (4) In regional development, market mechanisms must also be a prerequisite for regional development planning.

Communication models are idealized descriptions of what is needed for communication [23] (Figure 1).

1.png

4.1 Data quality testing validity test

Convergent validity is the correlation between the indicator score and its construct score. The PLS-SEM model meets convergent validity and can be valid if the outer loading value > 0.7 and the AVE value is > 0.5. The following results from the correlation between the indicator and its construct show the outer loading value > 0.7. The external loading value in the model can be seen in Table 2.

The variables in this study already have an AVE value of > 0.5. The AVE values in the model can be seen in Table 2 below:

Table 2. Average variance extracted (AVE) value

3.png

Figure 3. Outer loading

4.2 Discriminant validity

On the discriminant validity, used cross-loading values. An indicator is said to meet discriminant fact if the indicator's cross-loading value against its variables is the largest compared to other variables. The cross-loading value in the model can be seen in the Figure 3.

Based on Figure 4, three indicators are invalid because they have a loading factor of < 0.70; namely, in the interagency communication indicator on the dialogical dimension, there are two indicators, and in the deliberative size, one hand. Meanwhile, other indicators that make up each variable in this study have met the discriminant validity, namely the cross-loading value exceeding the value of 0.7. Therefore, it is necessary to dispose of invalid indicators. Here is a validity model after the disposal of a weak hand. Based on Figure 5. It can be known that all hands that make up each variable in this study have met the discriminant validity, namely, the cross-loading value exceeds the value of 0.7.

4.3 Reliable test

Composite reliability tests the reliability value of indicators on a variable. A variable is said to meet composite reliability if it has a combined reliability value of > 0.7. The composite reliability value of each variable can be seen in Table 3.

The table shows that the composite reliability value has qualified more than 0.7 [25]. The value of Cronbach’s Alpha above indicates a value above 0.6, proving that the measurements in this study are reliable.

Table 3. Composite reliability and Cronbach alpha values

4.png

Figure 4. Loading factor

4.4 Testing data analysis requirements (Inner model evaluation)

Evaluate the inner model or test the structural model to see the direct and indirect influences between the variables. Evaluation of the internal model with PLS-SEM begins by looking at the R-square value. Based on data processing with SmartPLS 3.0 Professional, the R-square value is generated in Table 4.

Table 4. R-square value

Based on the data above shows that the R-square value for the interagency communication variable is 0.944. This means that the percentage of the influence of communication development and communication effectiveness on communication between institutions is 94.4%, while other factors influence the remaining 5.6%. Furthermore, the R-square value for the regional planning variable is 0.699. This means that the percentage of the magnitude of the influence of communication development, the effectiveness of communication, and communication between institutions on regional development is 69.9%. In comparison, other factors influence the remaining 30.1%. Further, evaluate the inner model by looking at the path diagram, which shows how much influence the independent variable has on the dependent variable.

4.5 Data match test

For a model to meet the fit model criteria, the SMSR value must be less than 0.05 [26]. However, from the explanation on the SMART PLS website, the limitations or criteria of the FIT model include the following: RMS Teta or Root Mean Square Theta value (0.102 SRMR or Standardized Root Mean Square value <0.10 or < 0.08 NFI value > 0.9.

Table 5. FIT model fit test

Based on Table 5, it can be seen that the SRMR value is 0.012 < 0.05, while the NFI value is 0.942 > 0.90, so in conclusion, the model used can be said to be FIT.

4.6 Hypothesis testing

Simultaneous tests are used T-statistical tests intended to test the significance of the influence of Xi's exogenous variables as a whole on endogenous variables Y. This test is carried out by comparing the T values resulting from T-statistical calculations with T-table values. The null hypothesis will be accepted if the T-statistic value is smaller than the T-table value (T-statistic < T-table); this means that the alternative hypothesis is rejected. Conversely, the null hypothesis will be rejected if the T-statistic value is greater than or equal to the T-table value (T-statistic ≥ T-table); this means that the alternative view is accepted. The T-table value can be known based on a significance level of 0.05 with an observation number of 217, which is 1.25356. The following Table 6 shows the results of the t-statistical test.

The data shows t-statistics greater than 1.25356, proving that eight research hypotheses are accepted. The truth in the above hypothesis test is conducted by rejecting how the study's idea is acceptable. There is only one insignificant variable. Namely, the effectiveness of communication does not affect the region's development. Based on the UTAUT2 model [27], testing was carried out to determine whether public satisfaction could moderate the model in the study. Moderation testing in this study was carried out with two steps: negotiating effect analysis and multiple group analysis. The results of moderation testing using moderating effect analysis can be seen in Table 7.

Table 6. Path analysis results

Table 7. Public satisfaction moderation results

4.7 Communication development positively affects regional institutions in North Sumatra Province

The results of the SEM-PLS model prove that communication development positively affects communication between institutions (the proposed H1 can be supported/accepted). Civil Servants and Members of the North Sumatra Provincial Parliament feel the involvement of communication developed by the communication needs between institutions. So respondents should try to build and be able to have reliable (good) and superior communication and provide the highest benefit to the public. When political influence on institutions increases significantly, the need for a proper and practical framework for the development and analysis of public policy will be alarming; public policy is often inconsistent and unclear because the approaches of interagency communication for regional development adopted by many governments are often opposite.

Mc Call (2010), communication for development is realized as a strategy that promotes the social welfare of specific communities so that they can align the vision and mission of the region through interagency communication. The author concludes that communication for development is a social process based on promoting dialogue between society and decision-makers at the local, national, and regional levels. Its purpose is to promote, develop and implement policies and programs that improve the quality of life. 

5.png

Figure 5. Interagency communication model