The role of smart governance in ensuring the success of smart cities: a case of Thailand

Figures & data

Figure 1. PartLabel-upper Theoretical framework proposed.

Table 1. Construct and observable variables.

Transparency (TRAN)
TRAN1	Disclosure of information to citizens promotes greater transparency in the decision-making processes of government agencies.
TRAN2	The use of technology increases the transparency of government agencies’ operations.
TRAN3	Transparency in the information and actions of government agencies reduces the occurrence of corruption.
TRAN4	Transparency in the information and actions of government agencies empowers the civic sector.
Collaboration (COL)
COL1	The collaboration between citizens and the government agencies generates the effectiveness of smart cities.
COL2	Government agencies that consider public opinion on urban governance issues typically carry out their duties more effectively.
COL3	Public participation in the budget administration of state agencies fosters greater cooperation between citizens and state agencies.
COL4	Strong community leaders can represent the citizens in effective collaboration with state agencies.
Partnership (PAN)
PAN1	Partnership with educational institutions for technological development increases the technological skill for technology users.
PAN2	Partnership with educational institutions for technological development can develop the economy
PAN3	Partnerships with other organizations, such as non-governmental organizations, associations, cooperatives, religious institutions, and so on, are essential to the city’s technological development.
PAN4	The partnership between local, regional, and central governments is essential to effectively applying technology to city development.
Participation (PAT)
PAT1	Public participation promotes greater transparency in governmental processes.
PAT2	The use of technology enables greater citizen participation in government agencies’ operations.
PAT3	Public participation increases the effectiveness of government administration.
PAT4	Public participation enhances communication between the public and private sectors.
Communication (COM)
COM1	Communication between government agencies and citizens promotes transparency in decision-making.
COM2	The use of technology enhances the effeciency of communication between government agencies and citizen.
COM3	Tools that facilitate government-to-citizen communication should be considered as priority.
COM4	Social networks are practical and efficient tools for government-to-citizen communication.
Accountability (ACC)
ACC1	Transparency in the operation of government agencies helps prevent them from being held accountable for misconduct.
ACC2	Government agencies’ responsibility improves government administration.
ACC3	Accountability of government agencies reduces the occurrence of corruption.
ACC4	The accountability of government agencies will result in a stronger electoral base for executives, thereby ensuring the continuity of government project implementation.
Effectiveness of smart cities (EFF)
EFF1	The establishment of a network between the public and private sectors fosters innovation, which is essential to the effectiveness of a smart city.
EFF2	The technology, which includes security solutions and big data, is the instrument that will enable a smart city to operate more efficiently.
EFF3	Transparency of government agencies enable a smart city to operate more efficiently
EFF4	Collaboration between government agencies and the citizens through the use of relevant technology promotes more efficient smart city management.
EFF5	Public participation combined with effective communication promotes more efficient smart city management.

Table 2. Descriptive statistics.

Demographic Variable	Category	Segment 1 (KKC)		Segment 2 (CNX)		Total		Significant Chi-Square Test
		n	%	n	%	n	%
Segment Size		393	51.2	374	48.8	767	100
Gender	Male	149	19.4	73	9.5	222	28.9
	Female	244	31.8	301	39.2	545	71.1	<.001
Age	18–30 years old	126	16.4	255	33.2	381	49.7
	31–40 years old	129	16.8	36	4.7	165	21.5
	41–50 years old	79	10.3	36	4.7	115	15
	51–60 years old	40	5.2	27	3.5	67	8.7
	Above 60 years old	19	2.5	20	2.6	39	5.1	<.001
Level of Education	Elementary school	13	1.7	12	1.6	25	3.3
	Secondary education	45	5.9	19	2.5	64	8.3
	High vocational certificate	75	9.8	58	7.6	133	17.3
	B.A. (Bachelor of Arts)	238	31	258	33.6	496	64.7
	M.A. (Master of Arts) or above	22	2.9	27	3.5	49	6.4	<.009
Occupation	Student	73	9.5	199	25.9	272	35.4
	Business owner	115	15	75	9.8	190	24.8
	Government officer	78	10.2	25	3.3	103	13.5
	Office worker	98	12.8	41	5.3	139	18.1
	Retired	12	1.6	2	0.3	14	1.8
	Unemployed	17	2.2	32	4.2	49	6.4	<.001
Monthly Income	Less than 5,000 THB	29	3.8	107	14	136	17.7
	5,001–10,000 THB	59	6.9	134	17.5	193	24.4
	10,001–15,000 THB	78	10.2	40	5.2	118	15.4
	15,001–20,000 THB	75	9.8	40	5.2	115	15
	20,001–25,000 THB	56	7.3	30	3.9	86	11.2
	Above 25,000 THB	102	13.3	23	3	125	16.3	<.001
Years of Residence	0-5 years	76	9.9	226	29.5	302	39.5
	6-10 years	122	15.9	58	7.6	180	23.5
	Above 10 years	193	25.2	90	11.8	283	37	<.001
Area of Living	In Municipal Area	387	50.5	358	46.7	745	97.1
	Outside of Municipal Area	6	0.8	16	2.1	22	2.9	0.022

Source: data adapted from authors (2023).

Table 3. The goodness of the fit of the measurement model.

Fit index	Value	Threshold	Assessment
p-value	0.000		Acceptable for complex model
CMIN/df	2.988	<3.00	Passed
TLI	0.941	>0.90	Passed
CFI	0.948	>0.90	Passed
IFI	0.948	>0.90	Passed
RMSEA	0.051	<0.10	Passed

Note. that CMIN/df = Chi-square/degree of freedom; TLI = Tucker–Lewis index; CFI = comparative fit index; IFI = incremental fit index; RMSEA = root mean square error approximation.

Table 4. Convergent validity.

Construct	Indicator	Loading	Cronbach α	AVE	CR
Transparency	TRAN1	0.799	0.829	0.551	0.831
	TRAN2	0.750
	TRAN3	0.704
	TRAN4	0.713
Collaboration	COL1	0.764	0.828	0.549	0.839
	COL2	0.804
	COL3	0.709
	COL4	0.680
Partnership	PAN1	0.835	0.829	0.548	0.827
	PAN2	0.735
	PAN3	0.625
	PAN4	0.750
Participation	PAT1	0.748	0.843	0.575	0.844
	PAT2	0.770
	PAT3	0.807
	PAT4	0.704
Communication	COM1	0.757	0.855	0.594	0.854
	COM2	0.805
	COM3	0.760
	COM4	0.760
Accountability	ACC1	0.721	0.822	0.538	0.823
	ACC2	0.792
	ACC3	0.743
	ACC4	0.674
Effectiveness of smart cities	EFF1	0.740	0.885	0.606	0.885
	EFF2	0.777
	EFF3	0.803
	EFF4	0.790
	EFF5	0.781

Note. that AVE = average variance extracted; CR = composite validity.

Table 5. Discriminant validity.

Fornell & Larker Criterion
Construct	CR	AVE		EFF		ACC	COM		PAT		PAN	COL		TRAN
EFF	0.831	0.551		0.778
ACC	0.839	0.549		0.695		0.734
COM	0.827	0.548		0.907		0.683	0.771
PAT	0.844	0.575		0.901		0.728	0.914		0.758
PAN	0.854	0.594		0.86		0.636	0.85		0.844		0.740
COL	0.823	0.538		0.771		0.651	0.779		0.83		0.706	0.741
TRAN	0.885	0.606		0.778		0.784	0.802		0.828		0.766	0.829		0.742
HTMT Ratio Approach
EFF	-		-		-			-		-			-		-
ACC	0.54		-		-			-		-			-		-
COM	0.73		0.52		-			-		-			-		-
PAT	0.72		0.55		0.71			-		-			-		-
PAN	0.67		0.47		0.46			0.64		-			-		-
COL	0.64		0.49		0.60			0.63		0.53			-		-
TRAN	0.61		0.59		0.62			0.63		0.58			0.62		-

Source: data adapted from authors (2023).

Table 6. The goodness of fit of the structural model (SEM).

Fit index	Value	Threshold	Assessment
p-value	0.000		Acceptable for complex model
CMIN/df	2.988	<3.00	Passed
TLI	0.941	>0.90	Passed
CFI	0.948	>0.90	Passed
IFI	0.948	>0.90	Passed
RMSEA	0.051	<0.10	Passed

Note. that CMIN/df = Chi-square/degree of freedom, TLI = Tucker-Lewis Index, CFI = Comparative Fit Index, IFI = Incremental Fit Index, RMSEA = Root Mean Square Error of Approximation.

Table 7. Test results from the structural model.

Hypothesis	Endogenous variable	Exogenous variable	Model	Path coefficients (β)	P-value	Result
H1	Transparency	Effectiveness of Smart Cities	TRAN→ EFF	0.001	.233	Rejected
H2	Collaboration	Effectiveness of Smart Cities	COLL→ EFF	0.078	.257	Rejected
H3	Partnership	Effectiveness of Smart Cities	PAN → EFF	0.376	***	Supported
H4	Participation	Effectiveness of Smart Cities	PAT→ EFF	0.269	.01**	Supported
H5	Communication	Effectiveness of Smart Cities	COM → EFF	0.207	***	Supported
H6	Accountability	Effectiveness of Smart Cities	ACC → EFF	0.090	.07	Rejected

Noted: *** p-value < 0.001, ** p-value < 0.01.

Table 8. Measurement invariance.

Fit Index	Configural Invariance (Unconstrained Model)	Metric Invariance (Equal Factor Loading)	Scalar Invariance (Equal Intercept)	Threshold	Assessment
p-value	0.000
CMIN/df	2.058	2.089	2.707	<3.00	Acceptable
TLI	0.936	0.934	0.897	>0.90	(Partial)
CFI	0.944	0.94	0.903	>0.90	Acceptable
IFI	0.944	0.941	0.904	>0.90	Acceptable
RMSEA	0.037	0.038	0.047	<0.10	Acceptable

Source. Data adapted from authors (2023).

Table 9. Test results from loading differences.

Hypothesis	Endogenous variable	Exogenous variable	Model	Standardized Loading		Critical Ratio Difference	Threshold
				(KKC)	(CNX)
H1	Transparency	Effectiveness of Smart Cities	TRAN→ EFF	−0.035	−0.008	|0.117|	|1.96|
H2	Collaboration	Effectiveness of Smart Cities	COLL→ EFF	0.057	−0.006	|-0.207|	|1.96|
H3	Partnership	Effectiveness of Smart Cities	PAN → EFF	0.337***	0.311	|-0.039|	|1.96|
H4	Participation	Effectiveness of Smart Cities	PAT→ EFF	0.306**	0.308	|-0.098|	|1.96|
H5	Communication	Effectiveness of Smart Cities	COM → EFF	0.253**	−0.123	|-0.628|	|1.96|
H6	Accountability	Effectiveness of Smart Cities	ACC → EFF	0.064	0.548***	|2.895|*	|1.96|

Source: Data adapted from authors (2023). Note. *** p-value < 0.001, ** p-value < 0.01, * p-value < 0.05.

Data availability statement

The data will be available upon request. Authors agree to make available data and materials supporting the results or analyses in our paper.