Journal of Business Management and Economic Development E-ISSN 2986-9072 P-ISSN 3031-9269 Volume 4 Issue 01.
January 2026.
Pp.
DOI: https://doi.
org/10.
59653/jbmed.
Copyright by Author Business Research Methods and Methodology in Practice:
Understanding the Advanced SmartPls Path Models in Structural Equation Modeling Abdul-Kahar Adam University of Education Winneba.
Ghana Corresponding Email: akadam@uew.
Received: 01-01-2026 Reviewed: 03-02-2026 Accepted: 05-03-2026 Abstract The purpose of this paper is to advocate the use of SmartPLS in scientific research by adopting business methodologies and variables, namely discipline, good governance, and probity.
The aim is to present the findings and data analysis using SmartPLS 4.
4 components.
The study population comprised 187 students in the Level 300 HRM class, and a sample size of 126 was determined using Krejcie and Morgan's .
The results were based on factor loadings, discriminant validity, composite reliability, and collinearity statistics.
The findings indicate that the outer loadings with appropriate p-values indicate that the majority of the factors are significantly loaded.
But it is not a surprise that 1a.
<- Good Governance, 1c.
<Good Governance, and 2a.
<- Control of Corruption is not significantly loaded because it means that there are certain effects that affect Good Governance and the control of corruption.
Responsibility and Discipline failed the discriminant validity test, indicating that the Follow Responsibility factor and the Discipline variable were not in agreement with the measurement, implying that certain factors affect this variable-factor relationship.
SmartPLS 4.
4 App was used to run the data.
Keywords: Factor Loadings.
Construct Reliability and Validity.
Discriminant Validity.
Collinearity statistics (VIF).
Pre-testing.
Pilot Testing.
Discipline.
Good Governance.
Probity Introduction The main purpose of this paper is to present a practical scientific overview of the methodology for analyzing SmartPLS results and their effects, and to encourage future researchers to adopt the best quantitative approaches.
SmartPLS is a research tool or app used to scientifically define path models among variables.
The aim is to elaborate on the technicalities and styles of presenting SmartPLS findings, with components that underscore the research's significance.
The objective is to serve as an enhancement and resource to help young Journal of Business Management and Economic Development researchers better understand and adopt SmartPLS in their methodological design, and to build knowledge as part of this objective.
Literature Review Business Research Methods Vs Methodology Business research is a type of business activity that involves preparing data collection, analyzing the information received, interpreting it, and forming data findings (Kramarenko.
At the preparation stage for population and sample size determination and data collection, tasks are set, and research methods are selected.
Mexon & Kumar .
discussed business research as the systematic collection and analysis of data to answer problems facing management or executives.
Business research can also be used to solve business-related problems, for example, determining how to reduce excess inventory in the organization.
In these views.
Mexon & Kumar .
wrote that research methods encompass all approaches, techniques, and procedures used to conduct research.
Thus, research techniques or methods are the approaches researchers use to conduct research.
The two main research approaches to consider are quantitative and qualitative.
On the other hand, research methodology refers to the systematic methods used to solve research problems.
it is the science of conducting research.
These involve the various steps generally adopted by the researcher to study the research problems and address existing gaps.
Therefore, in simple terms, research methods are the techniques or approaches used in the research process, whereas research methodology refers to the scientific approaches the researcher uses to apply those methods.
Partial Least Squares Structural Equation Modeling (PLS-SEM) is regarded by some as an advanced statistical technique for analyzing and interpreting complex causal relationships among observed and latent variables.
This was originally conceptualized and introduced by Herman Wold in the early 1980s as a statistical method tailored for econometric modeling (Chinnaraju, 2.
Moreover.
Partial Least Squares Structural Equation Modeling (PLSSEM) serves as a comprehensive methodological framework that critically addresses theoretical underpinnings through rigorous analytical approaches and state-of-the-art modeling techniques vital to contemporary business research (Chinnaraju, 2.
This research presents empirical evidence on adapted variables, namely: Good Governance.
Probity, and Discipline, used in the data gathering and analysis.
Discipline is an independent variable.
Good Governance is a dependent variable, and Probity is a Mediator The various empirical literature on these variables are as follows:
Discipline and Good Governance Research conducted by Mira and Hammadache, .
about the relationship between good governance and economic growth .
, indicated that several econometric research studies proved that there is a positive relationship between good governance in the sense of discipline and probity policies within market-enhancing governance (Kauffman & Business Research Methods and Methodology in Practice: Understanding the Advanced SmartPls Path Models in Structural Equation Modeling Kray, 2003.
Knack & Keefer, 1.
Even though good governance policy allows developing countries like Ghana to achieve only minimal economic advancement through political reforms, with the hope of reaching developed-state status.
Additionally, following previous empirical studies on neo-institutionalism, the relationship among economic growth, societal discipline, and good governance is examined.
This proved two divergent theories as state failure in the developing countries (Kauffman, et al.
,1999, 2005.
Knack & Keefer, 1995.
Hall, and Jones,1999.
Barro, 1996.
Robert, 2002.
Rodrik, 1995, 1996, 2004, 2006, 2.
that is components of good governance like corruption, instability of rights to property and lack of democracy, the other is the weak institutions are the cause to good governance.
Hence, the remedies are to foster discipline and probity for the integrity of governance (Khan, 2019, 2005, 2.
There is always a mismatch between policies and institutions for good governance practices (Mira & Hammadache, 2.
They assumed that good governance policies would only be relevant if developing countries reached an appreciable level of social and economic development, at which point institutions responsible for good governance could impose discipline to bolster good governance within their economies.
Williams .
, in his research on governance and the discipline of development, stated that good governance, as presented by the World Bank, is best understood and practiced in terms of discipline, as his studies showed.
This means that the relationship between good governance and discipline is Moreover, good governance presents an ideal situation in practice, where it interlocks with disciplinary practices: when the state is disciplined by society, society is also disciplined by the state.
Hence, good governance and discipline are like birds with the same This article further suggested that the practice of theory and good governance provides insights into working ideology and the development of discipline.
The study affirmed a clear association between good governance and efficient, effective administrative systems in a democracy (Wani, 2014.
Martin, 2.
Discipline and Probity Whitton .
conducted research on the effective implementation of ethics in civil service and government, based on integrity and good governance among public officers.
This was conducted across New Zealand.
Australia, the UK.
Korea.
Canada.
Ethiopia, and Morocco.
The research stated that specific strategies to consider include identifying risks to integrity/probity, maintaining discipline, among other factors.
It concluded that officers responsible for maladministration can be disciplined by their employers to improve integrity and probity.
Research conducted by Mohd et al.
argued that the values of honesty, hard work, discipline, and loyalty can predict Muslim Owner Managers through entrepreneurial The findings indicated that discipline and hard work, which are integrity and probity, are positively related to innovative orientations.
This also explains that a Muslim owner-manager has religious values such as discipline, integrity, and probity.
Shih et al.
examined differences among teachers and their backgrounds within the school environment.
one of the research findings was that teachers aged 41 to 50 with master's or doctoral degrees demonstrated strong integrity and discipline at that level.
Both positive discipline and integrity have a significant impact on the delivery of instruction in public and private schools.
Journal of Business Management and Economic Development Central Bank of Ireland .
established a policy on guidance on fitness and probity standards, in which it stated that probity is a character acquired from a personAos past behavior.
That is, if a person is found to not possess probity qualities or features, such as a lack of integrity, honesty, and ethical judgment, then that person cannot be engaged in any serious Therefore, it is necessary to always consider due diligence first, which means discipline and probity are associated with one another.
The Institute of Internal Auditors .
compiled a fact sheet on Probity, affirming that ethical behavior is evidence of the probity process and that it is complete uprightness, integrity, and honesty.
Probity and Good Governance Shimawua .
conducted research on probity, ethics, and good governance in the public sector within Government and Administration and argued that current studies have shown that awareness among the general public regarding probity and matters of integrity in the public sector is very high.
This study was conducted from different perspectives, such as organizational and individual, and explored ways to help government agencies meet expected The best way to understand probity is to first think of the principles of ethical conduct, honesty, and transparent processes (Ombudsman, 2.
And that there is a clear association between the concepts of ethics and probity, whereby ethics is about the rights, fairness, and justice, and all these operate with Good Governance.
Probity is about setting standards and values within organizations and ensuring that those standards/values are implemented through the enactment of policies and codes of conduct for Good Governance practice (Shimawua, 2.
It further stated that achieving effective governance depends on maintaining and developing appropriate and acceptable governance structures and frameworks, and on applying the governance systems the government chooses, with a commitment to operationalizing effective and efficient functioning.
By and large, in recent times, there has been government scrutiny of their procurement and governance processes.
therefore, there would be consequences for ignoring probity.
Hence, the right culture must always be established and demonstrated, showing discipline and good governance.
The study of Good Governance is broad and therefore interacts with many factors within an organizational setting.
Ibori .
examined the collapse of probity and good governance and found consensus that the lack of probity and accountability in governance is the main factor affecting the performance of African governments.
This statement indicates a relationship between good governance and probity in the management of public-sector businesses (Ibori, 2.
This was based on the researcher's research findings and analyses.
Researched on probity principles with good governance and culture, and in their discussion, they stated that documented policies and procedures in governance do not guarantee that the workers will understand what they are, but rather through active promotion of Probity and Good Governance principles, which is a clear affirmative findings statement (Ibori, 2.
According to General Knowledge Today, .
, the position on probity in governance where it stated that the constitutions and laws serve as a legal framework for governance practice, together with probity, which is the uprightness, integrity, and the rules of conduct within the legal framework for the governance institutions and the relationship of government and the people to flourish.
This simply means that government organizations are set up to effect Good Business Research Methods and Methodology in Practice: Understanding the Advanced SmartPls Path Models in Structural Equation Modeling Governance and must comply with the constitution and the legal framework through Probity The importance of applying the principles of good governance to achieve a brighter future for a nation or a company can be realized through a framework of accountability for such practices (Zirman & Oktari, 2.
Probity Probity is important for both public sector governance and private corporations.
Shimawua .
conducted research on public perception of Probity in public service and its impact on sustainable development and confirmed that acceptable conduct (Disciplin.
and standards established by the state (Good Governanc.
cannot be compromised, and therefore, all public service personnel must demonstrate Probity at all levels for sustainable development.
Hence, this will bring Good Governance to bear.
Definition of Terms Discipline: In a behavioral context, discipline is, in practice, the control of a personAos behavior through the regulation of actions in accordance with established expectations, rules, and goals, guided by self-control and consequences for noncompliance.
Good Governance: This is derived from the term "governance," and the opposite is "bad governance.
" It is a process by which institutions or corporate entities ensure that all systems, policies, and processes designed and established to achieve the objectives are strictly followed and complied with, without any deviation from the original plan.
In other words, good governance has 8 major characteristics according to the United Nations: consensus-oriented, participatory, transparent, accountable, effective and efficient, responsive, rule of law, and equitable and inclusive.
Probity: This is the act of being equitable and fair by showing evidence of ethical behavior in practice by following or adopting well-designed procedures and processes to ensure that decisions are transparent, fair, and defensible.
It also relies on transparency and accountability in decision-making within the procurement process and in behavioral attitudes towards decision-making.
Research Method This is a scientific design using quantitative methods to demonstrate the display of SmartPLS App in defining variable paths.
SmartPLS is a software that shows a graphical user interface for variance-based structural equation modeling (SEM) using the Partial Least Squares (PLS) path modeling method (Hair et.
, 2.
The target population was the University of Education.
Winneba, students at level 300 of the Department of Management Sciences.
The population was 187, and the sample size of 126 was deduced using Krejcie and Morgan's .
A questionnaire was designed in Google Forms with Likert-scale items and administered to collect data.
Journal of Business Management and Economic Development Instrument Development The questionnaire for this research was adapted from previously established research and data collection methods across various theories.
It has been carefully adopted, and a few portions have been edited to suit the purpose of this research.
It has been grouped and categorized into three different sections.
Section A is about Good Governance (Dependent Variabl.
Section B is about Discipline (Independent Variabl.
, and Section C is about Probity (Mediator Variabl.
These are all subdivided into subheadings for clarity of data presentation.
Tronvoll .
argued that their questionnaire included components of the baseline conditions or services of governance and corruption, and had an experimental component assessing both negative and positive changes in the quality of government in terms of dealing with government, trust, and future satisfaction.
Table 1 below was concluded after the pre-testing and piloting of the questionnaire.
Table 1: Instruments in the Questionnaires as adopted Section Sources Variables Number of Items Good Governance:
GBS .
GBS .
*Quality of Good Governance.
*Corruption Control.
Discipline:
Tronvoll .
* Follow Responsibilities - Satisfaction with 4 government service.
* Follow Responsibilities - Satisfaction with 6 government service.
*Fairness Ae Transparency.
Probity:
*Integrity.
*Integrity.
GBS .
GBS .
GBS .
OECD .
Otoghile et *Integrity.
TOTAL QUESTIONS
Section A.
Ae Good Governance The researcher used and adopted three instruments in this section, namely the Euromed Survey .
and the GBS .
IEMed represents the European Institute of the Mediterranean.
GBS represents the Global Barometer Survey.
and USAID represents the United States Agency for International Development.
IEMed is a consortium comprising the Catalan Government, the Barcelona City Council, and the Spanish Ministry of Foreign Affairs and Cooperation.
Global Barometer Surveys is a collaborative research project on six barometers regionally, including Africa, which is known as Afrobarometer.
East and Southeast Asia (Asian Baromete.
South Asia (South Asia Baromete.
Central and South America (Latino Barometr.
, the Middle East (Arab Baromete.
, and countries of the Soviet Union (Eurasia Baromete.
It measures, at a mass level, the current social, political, and economic Business Research Methods and Methodology in Practice: Understanding the Advanced SmartPls Path Models in Structural Equation Modeling climate worldwide, offering an independent, nonpartisan, scientific, and multidisciplinary view of public opinion on a broad range of government policies and other relevant issues.
The GBS network is over 70% of the worldAos population, and itAos still expanding.
USAID is an American international organization that leads humanitarian efforts to save lives, strengthen democratic governance, reduce poverty, and more.
The researcher chose these instruments for this section because the questions are directly relevant to and address the thesis's objective.
Also, it is easy to administer to the target population.
This instrument measures the areas of general perspective, democratic governance process.
Control of Corruption, and quality of governance.
Table 2: Distribution of Good Governance Items Elements Quality of Governance (QoG) Control of Corruption (CoC) Section B.
Question No.
Number of Items 1a to 1e 2a to 2b Discipline The research combines several sub-instruments to assess how discipline governance systems and policies should or must be structured.
These are in the areas of trustworthiness, fairness, adherence to responsibilities and procedures, and cultural norms.
Here, the instruments are adopted from Tronvoll .
Court .
, and GBS .
as appropriate.
GBS, as discussed in section A, is more relevant to the thesis's objectives.
Tronvoll .
research survey was on citizensAo attitudes to good governance and corruption.
It has all the indications of discipline, policies, and people's behaviors towards governance.
Hence, it is ideal for the thesis's purpose and objective.
This was conducted under ILPI (International Law and Policy Institut.
in Zanzibar.
Court .
is a co-director of the World Governance Survey (WGS) project together with Goran Hyden.
They conducted a comprehensive national assessment of governance in India.
The researcher found it necessary to adopt portions of this questionnaire to fulfil some of the objectives of this thesis, thereby improving the presentation of data and analysis.
Table 3: Distribution of Discipline Items Elements Follow Responsibilities Ae satisfaction with government service Follow Responsibilities Ae satisfaction with government service Fairness Section C.
Question No.
Number of Items 3a to 3d 4a to 4f 5a to 5h Probity The researcher adopted instruments such as Integrity from GBS .
OECD .
, and Otoghile et al.
OECD stands for the Organization for Economic Co-operation and Development.
The OECD's mission is to promote policies that improve the economic and social well-being of people worldwide.
And thereby, it provides a platform for governments to work together by sharing ideas and experiences and seeking solutions to their economic Otoghile et al.
wrote a research paper on the quest for good governance in Nigeria: a survey of peopleAos perception in Benin City.
It includes guidance on government Journal of Business Management and Economic Development policy assessment in the module, which has been adopted because it aligns with the general objective of this thesis.
GBS ensures that data are comparable and reliable by sharing a common research agenda that explores citizensAo attitudes towards democracy, political reforms, and Hence.
GBS has, over the years, established standardized questionnaire modules across all the regions it operates, while also allowing the regions to retain their own specific items that reflect particular local concerns.
The questionnaire was designed using a mixedmethods approach, including Likert-scale items and descriptive questions to collect both qualitative and quantitative data for analysis.
Table 4: Distribution of Probity Items Elements Question No.
Probity Ae Integrity general Probity Integrity in the Government Financial Sector for Projects and Investment Number Items 6a to 6f 8a to 8g Result and Discussion Pre-test of the Questionnaires Results After the questionnaire was designed based on various sources, a pre-test was conducted with 10 respondents.
The respondents were part of the target population for this research, namely Level 300 Students.
First of all, the pre-test was done to ascertain its relevance for efficiency and effectiveness, accuracy, and content validity.
Pre-testing was also conducted to determine the strengths and weaknesses of the adopted questionnaires as There are two main types of pre-testing, namely participating and undeclared.
Participating means participants were told the questionnaire was for a practice run and, in this case, were asked to explain their reactions to its form, order, and wording.
The purpose was to enable the researcher to determine whether they fully understood the questionnaire.
On the other hand, in the undeclared pre-testing, the respondents were not told it was a pre-test, so they answered it as their real opinions.
Hence, it helped the researcher deduce the type of analysis and the standardization of the questions to be adopted.
Henceforth, the two types of pre-tests were conducted.
According to Converse & Presser .
, it is important and good practice for researchers to conduct participatory research, followed by an undeclared pre-test.
With this, the variation of the questions, meaning, difficulty, respondentsAo interest, and attention will all be known and captured.
Above all, it means that the questionnaire's reliability and validity can be assessed by comparing one pre-test with another from respondents/participants (Weisberg et al.
, 1.
The following were the ratings method used for the pretesting of the predesigned questionnaires which had in total of 47 questions at first for Quantitative survey and 50 questions for Qualitative interview, sub-questions inclusive in all, which subsequently reduced to a number of 44 questions for Quantitative survey and 45 questions for Qualitative interview Business Research Methods and Methodology in Practice: Understanding the Advanced SmartPls Path Models in Structural Equation Modeling after complying with the pre-test comments, meaning some questions were deleted Relevance rating:
A1- Not Relevant A2- Item needs some revision A3- Relevant but need minor revision A4- Very relevant Simplicity rating:
B1- Not simple B2- Item needs some revision B3-Simple but need minor revision B4- Very simple Figure 1: Pre-test of Questionnaire Ratings Five participants were engaged as members of the class to rate the predesigned questionnaire using a participatory method, in which they rated the questions according to the above options.
The results per observation and analysis of the pre-tested questionnaires are shown in the tables below:
Table 5: Pre-test Results for Relevance Rating Quantitative survey Qualitative survey Total Responses Expected *Note: 10 questionnaires were pre-tested in each case.
Hence, 10 x 47 = 470.
10 x 50 = 500 Table 6: Pre-test Results for Simplicity Rating Quantitative survey Qualitative survey Total Responses Expected The above tables show that 361 and 402 ratings represented very relevant and very simple, respectively, for the Quantitative survey questionnaire.
Meanwhile, the 418 and 440 ratings represent, respectively, very relevant and very simple for the Qualitative interview The tables above show that 3 Quantitative survey questions and 5 Qualitative interview questions were deleted as appropriate.
General remarks by participants after the exercise were that Ausome items need to be revised.
good work done.
It is involving,Ay and another one stated.
Augenerally, questions are relevant and simple to be rated.
Not much difficulty will be encountered by any graduate or undergraduate who works on this Ay In view of these observations, the researcher decided to review the questionnaire by editing it to the level of piloting.
Many more questions were deleted, and some were edited accordingly before piloting.
Moreover, another pre-test was conducted using the predesigned questionnaires, administered via an undeclared method, with respondents filling in the answers.
In fact, the questionnaires were almost 100% completed without difficulty.
With the results from these two pretested sessions, the final questionnaire, informed by other theories and relevant sources, can stand the test of Good Governance.
Discipline, and Probity and, more importantly, serve the purpose of qualitative and quantitative analysis within a mixed-methods This means the content is better suited to testing at the pilot stage, given the overwhelmingly high ratings for relevance and simplicity.
Journal of Business Management and Economic Development Pilot Study Table 7: Results of the Pilot Study for reliability and validity of Constructs Variables Discipline:
Fairness Follow Responsibilities Good Governance Probity Ae Integrity in the Government Financial Sector Probity Number of Items Conbach Alpha 882= 0.
289= 0.
707= 0.
695= 0.
681= 0.
Findings and Analysis Table 8: Outer Loadings Mean.
STDEV.
T values, p
<- Good
Governance <- Good
Governance <- Good
Governance <- Good
Governance <- Good
Governance <- Control
of Corruption <- Control
of Corruption <Discipline <- Follow
Responsibilitie <- Follow
Responsibilitie <- Follow
Responsibilitie <Discipline Origina
(O)
Sample mean (M) Standard
(STDEV)
T statistics (|O/STDEV|) Implication not significantly not significantly not significantly not significantly Business Research Methods and Methodology in Practice: Understanding the Advanced SmartPls Path Models in Structural Equation Modeling <- Follow Responsibilitie <- Follow Responsibilitie <- Follow Responsibilitie 4e <- Follow Responsibilitie <- Follow Responsibilitie <- Fairness <- Fairness <- Fairness <- Fairness <- Fairness <- Fairness <- Fairness <- Fairness <- Fairness <- Fairness <- Probity <- Probity <- Probity <- Probity <- Probity <- Probity <- Probity <- Probity <- Probity not significantly Journal of Business Management and Economic Development <- Probity <- Probity <- Probity <- Probity <- Probity From Table 8 above shows the outer loadings in bootstrap significance which indicates that the constructs paths to the individual factors or variables that have p-value of 0.
000 and above are all significantly loaded.
Table 9: Outer Mean.
STDEV.
T values, p values Original sample (O) Sample mean (M) Standard
(STDEV)
T statistics (|O/STDEV|) <- Discipline <- Follow Responsibilities <- Follow Responsibilities <- Follow Responsibilities <- Discipline <- Follow Responsibilities <- Follow Responsibilities <- Follow Responsibilities 4e <- Follow Responsibilities <- Good Governance <- Good Governance <- Good Governance <- Good Governance <- Good Governance <- Control of Corruption <- Control of Corruption Implication not significantly not significantly not significantly not significantly Business Research Methods and Methodology in Practice: Understanding the Advanced SmartPls Path Models in Structural Equation Modeling <- Follow Responsibilities <- Fairness <- Fairness <- Fairness <- Fairness <- Fairness <- Fairness <- Fairness <- Fairness <- Fairness <- Fairness <- Probity <- Probity <- Probity <- Probity <- Probity <- Probity <- Probity <- Probity <- Probity <- Probity <- Probity <- Probity <- Probity <- Probity not significantly not significantly not significantly not significantly not significantly Journal of Business Management and Economic Development From the above table 9 indicates that all the constructs or indicators to their corresponding factors or variables have outer weights that have p-value of 0.
000 and above means that all are significantly weighted.
Table 10: Construct reliability and validity Cronbach's Composite reliability .
Composite reliability .
Control of Corruption Discipline Fairness Follow Responsibilities Good Governance Probity Average variance extracted (AVE) The above Table 10 shows the construct reliability and validity of constructs such as control of corruption, with Composite Reliability .
and Average Variance Extracted (AVE) values of 0.
918 and 0.
565, respectively, indicating reliability and validity.
The discipline construct has a composite reliability .
730 and an Average Variance Extracted (AVE) 585, both indicating reliability and validity.
Fairness constructs have CronbachAos Alpha of 764.
Composite Reliability .
771, and Composite Reliability .
meaning that Fairness is most reliable and valid in this context.
Follow Responsibility, which also has CronbachAos Alpha of 0.
Composite Reliability .
798, and Composite Reliability .
809, indicating reliability and validity.
On the other hand.
Probity has CronbachAos Alpha of 0.
Composite Reliability .
799, and Composite Reliability .
815, which were all reliable and valid.
The only variable constructs that failed reliability and validity were Good Governance constructs.
Discriminant validity Table 11: Heterotrait-monotrait ratio (HTMT) - Matrix Control of Corruption Discipline Fairness Follow Responsibilities Good Governance Probity Control of Corruption Discipline Fairness Follow Responsibilities Good Governance Probity Business Research Methods and Methodology in Practice: Understanding the Advanced SmartPls Path Models in Structural Equation Modeling Table 12: Heterotrait-monotrait ratio (HTMT) - List Discipline <-> Control of Corruption Fairness <-> Control of Corruption Fairness <-> Discipline Follow Responsibilities <-> Control of Corruption Follow Responsibilities <-> Discipline Follow Responsibilities <-> Fairness Good Governance <-> Control of Corruption Good Governance <-> Discipline Good Governance <-> Fairness Good Governance <-> Follow Responsibilities Probity <-> Control of Corruption Probity <-> Discipline Probity <-> Fairness Probity <-> Follow Responsibilities Probity <-> Good Governance Heterotrait-monotrait ratio (HTMT) Tables 11 and 12 above indicate that Heterotrait-Monotrait Ratio (HTMT), which Henseler et al.
developed, simulates and demonstrates the lack of discriminant validity in the Fornell-Larcker criterion and cross-loading shortcomings by using the HTMT ratio.
This HTMT ratio is the geometric mean of the heterotrait-heteromethod correlations of indicators across constructs measuring different phenomena, divided by the average of the monotraitheteromethod correlations of indicators within the same construct.
With a well-fitted model, heterotrait correlations are expected to be smaller than monotrait correlations because the HTMT ratio must be below 1.
0 in the model table.
Henseler et al.
established that if the HTMT ratio is less than 0.
90, there is discriminant validity between the pair of reflective this cutoff point has also been used by Gold et al.
and Teo et al.
But Clark and Watson .
and Kline .
established a very stringent cutoff of 0.
Therefore, the tables above show that apart from Follow Responsibility and Discipline, which 050, which fails the discriminant validity, all the rest of the variable constructs proved and qualified with the discriminant validity.
Collinearity statistics (VIF) Table 13: Outer model - List
VIF
Journal of Business Management and Economic Development Table 14: Inner model - Matrix Control of Corruption Discipl Control of Corruption Discipline Fairness Fairn Follow Responsibilities Good Governance Probi Business Research Methods and Methodology in Practice: Understanding the Advanced SmartPls Path Models in Structural Equation Modeling Follow Responsibilities Good Governance Probity Table 15: Inner model - List Control of Corruption -> Good Governance Discipline -> Good Governance Discipline -> Probity Fairness -> Discipline Follow Responsibilities -> Discipline Probity -> Good Governance VIF Tables 13, 14, and 15 above present Collinearity Statistics, also known as Variance Inflation Factors (VIF.
, for inner values.
Hair et al.
argued that researchers should evaluate the data and results for issues related to influential outliers.
Collinearity occurs when two indicators are highly correlated.
when more than two indicators are involved, it is called Multicollinearity.
Variance Inflation Factor (VIF) is a related measure of Collinearity, defined as the reciprocal of the tolerance.
The term VIF is derived from the square root of the VIF .
OoVIF) as the degree to which the standard error has been increased due to the occurrence of Collinearity.
Multicollinearity in OLS regression occurs when two or more independent variables are highly intercorrelated.
It inflates standard errors and makes significance tests of the independent variables unreliable by preventing researchers from seeing the relative importance of one independent variable compared to another.
Now the rule is that problematic Multicollinearity may occur when the VIF is greater than 4.
0, though some use a cut-off of 5.
Another rule for the VIF is when the tolerance is smaller than 0.
25, and some use a cut-off of In reflective models such as this, the variables are modelled as single predictors of the indicator variables, which are the dependent variables.
Hence, in the reflective model measurement.
Multicollinearity is not a problem, though in SmartPLS the outcome will show the VIF statistic for the outer measurement model, whether the measurement model is formative or reflective (Garson, 2.
But in either formative or reflective model, there is always the likelihood that multicollinearity occurs at the structural level, that is, the variables modelled cause the endogenous variable to be Multicollinear.
Structural multicollinearity is an issue and problem in either formative or reflective models for the same reason as it is in the Ordinary Least Squares (OLS) regression models.
Therefore, the above tables show that there is VIF which has passed the collinearity statistics test.
Journal of Business Management and Economic Development Model fit Table 16: Fit summary SRMR d_ULS NFI Saturated model Estimated model The table above shows that d_ULS and d_G have p-values greater than 0.
05, indicating a good fit.
But with SRMR and NFI the model could not meet the condition that SRMR is less 08 and NFI is greater than 0.
Conclusion The outer loadings of the factors, along with their corresponding p-values, indicate that most factors are significantly loaded.
But it is not a surprise that 1a.
<- Good Governance, 1c.
<- Good Governance, and 2a.
<- Control of Corruption is not significantly loaded because it means that there are certain effects that affect Good Governance and control of corruption.
This means that the agreement that the measurement obtained was very low to support these The outer weights also prove that 1a.
<- Good Governance, 1c.
<- Good Governance, and 2a.
<- Control of Corruption is not significantly weighted, while the other factors have significant weights.
It is also obvious that Good Governance constructs failed their reliability and validity test, meaning that the measurements are not in good agreement or do not support the variable constructs.
Responsibility and Discipline failed the discriminant validity test, indicating that the Follow Responsibility factor and the Discipline variable were not in agreement with the measurement, implying that certain factors affect this variable-factor relationship.
The collinearity statistics show that the VIF has passed the collinearity test.
The model fit statistics satisfy some components but not all.
The d_ULS and d_G satisfied the Model Fit, while SRMR and NFI did not.
Research Implication The major implication of this research is knowledge-based, as it uses SmartPLS in scientific research.
It will pre-inform researchers on how to approach the findings and analysis of SmartPLS results.
It also adds value to the narration of the variable factors and their theoretical measurement conclusions, in terms of p-values indicating significance for various theories, by the establishment.
Business Research Methods and Methodology in Practice: Understanding the Advanced SmartPls Path Models in Structural Equation Modeling References