The following are the stages that are taken into account while evaluating a PLS-SEM route model:
First, dependability, or the degree to which something can be relied upon, is assessed by looking at how well it conforms to expectations outside the system. The classic method for estimating reliability using intercorrelations across note indicator variables is Cronbach’s alpha, and it has been used for many years as a gold standard. All indicators are supposed to be equally dependable in Cronbach’s alpha, and all indications subject to extraneous loads are assumed to have the same structure. Priorities, however, are determined by indications from the PLS-SEM. In addition, Cronbach’s alpha tends to understate the dependability of external consistency, accounting for the sensitivity of the number of components in the table.
Cross Loading
Discriminant validity has been evaluated in two different ways. Taking a look at the indicators, cross loadings is one way to evaluate their discriminant validity. The outer loading of an indication should be greater on the linked construct than any other loadings. Discriminant validity issues arise when there are cross loadings that are larger than the outer loadings of the indicators. On the basis of [
39], these criteria are seen as somewhat lax when it comes to demonstrating discriminant validity. As a result, it is probable evidence that two or more constructs have discriminant validity (see
Table 6).
2: Assessment of the structural model (Inner Model)
After checking the construct phases for correctness and efficacy, the results of the structural model are evaluated by analyzing the dependent variables’ internal linkages. The prediction abilities and inter-construct relationships of the model are investigated. The next five phases, shown in
Figure 5, assess the structural model at this point.
Step 1: hypotheses testing (path coefficient)
By using the PLS method in Smart PLS, we were able to extract the structural model connections (i.e., the path coefficients) that represent the predicted links among the constructs. There is a predefined range for route coefficients, from 1 to +1, which is known as the limit. Strong positive connections (and the opposite for negative values) are always statistically significant when their estimated path coefficient is near to 1 (i.e., different from zero in the population). The estimated coefficients are most closely centered around 0. Weak associations and shallow values near to zero are often not significant (i.e., not substantially different from zero) [
35].
Bootstrapping may be used to acquire a coefficient’s standard error, which is then used to determine statistical significance. The next phase, following the guidelines laid forth in [
40], was using the bootstrapping technique to see whether a reflective indicator makes a significant contribution to the relevant construct. The observed t value may be calculated using the bootstrap standard error. With samples greater than 30, it is reasonable to make an approximation to the t distribution. Equally, the quantiles of the normal distribution may be used as cutoff points against which the observed t value can be evaluated. With a certain error probability, a coefficient is considered to be statistically significant when the empirical t value is bigger than the critical value (i.e., significance level). Critical values for two-tailed tests are typically 1.65 (significance level 5–10%), 1.96 (significance level 5–5%), and 2.57 (significance level 5–1%). Researchers in marketing often use a 5% significance level, although this is not always the case; nonetheless, consumer research studies often use a 1% significance level, especially when conducting trials. While performing an exploratory study, however, researchers often choose a 10% significance threshold. The significance level that should be used is ultimately determined by the nature of the research being conducted and its ultimate goal.
Construction projects in civil engineering in Malaysia are shown in
Figure 6 together with the evolution of key aspects including drivers and strategies. It features the most significant pieces of music and proven theories. Hypothesis 1: driver factors—DF—having a considerable impact on standard adoption is supported by SEM analysis (
Table 7). The same holds true for Hypothesis 2: strategy factors very heavily affect standard adoption.
Step 2: coefficient of determination (R2 value)
The R
2 value (or coefficient of determination) [
31,
39] is a key metric in PLS-SEM used to evaluate the strength of the structural model (R
2 value). To measure how well a model can predict the future, we may use the squared correlation between the observed and anticipated values of a single endogenous component. The coefficient of determination (R
2) is a statistical measure of the degree to which a set of independent variables can explain the observed pattern of the dependent variable [
41,
42]. While [
41,
43] determined that an R
2 of 0.10 is minimally acceptable, this value is context-dependent. Like PLS-SEM, [
29] suggests that R
2 values of 0.67, 0.33, and 0.19 are substantial, moderate, and weak, respectively, and that R
2 values below 0.19 are insufficient (see
Table 8).
The coefficient of determination (R-squared) is a statistical measure that represents the proportion of variance in the dependent variable that is explained by the independent variables in the model. In this case, the R-squared value of the SEM-PLS model is 0.40, which means that 40% of the variance in the dependent variable is explained by the independent variables in the model. A higher R-squared value indicates a stronger relationship between the independent and dependent variables and a better fit of the model to the data. However, the interpretation of the R-squared value may depend on the context of the study and the specific research question.
Step 3: measuring the effect size (f2)
According to [
35], the change in R
2 value when a specific exogenous construct is removed from the model may be used to assess whether the removed construct has a significant influence on the endogenous constructs, in addition to evaluating the R
2 values of all endogenous constructs. The f
2 effect size is a common term for this quantification. One such formula for expressing the extent of the impact is shown below [
44,
45,
46].
According to [
25,
47], f
2 measures how much each external latent construct influences the corresponding endogenous latent construct. The coefficient of determination (R
2) changes when an independent construct is removed from the path model, indicating whether the value of the latent exogenous construct had a substantial impact on the value of the latent endogenous construct before its removal. Based on [
44], the f
2 values were 0.35 (high impact), 0.15 (moderate effect), and 0.02 (low effect) (see
Table 9).
Step 4: blindfolding and predictive relevance (Q2)
Data points representing indicators of a chosen endogenous latent variable are omitted from the analysis, and the resulting variables are anticipated, as shown by [
35]. The blind approach may then evaluate how well the anticipated values match the actual ones. Predictive accuracy in a route model is high if the predicted value is relatively close to the actual value, while Q
2 values larger than 0 show that the model has predictive significance for a certain endogenous construct and are calculated by subtracting the observed value from the predicted value and adding a trivial prediction error (specified as the mean of the remaining data). On the other hand, negative numbers have no predictive value.
According to [
47], the blindfold approach is only used for endogenous single-item constructs and endogenous constructs with a reflecting measurement model specification. Blind calculations and cross-validation redundancy are used to determine Q
2 statistics as a quality metric for the PLS route model. Predicting the endogenous latent constructs is highly recommended by the Q
2 criteria (see
Table 10).
Predictive relevance (Q2) is a statistical measure used to evaluate the predictive power of a structural equation modeling (SEM) model. In this case, the SEM-PLS model has a Q2 value of 0.195, which means that the model can predict the endogenous latent components with an accuracy of 19.5%. A higher Q2 value indicates a better predictive power of the model. However, the interpretation of the Q2 value may depend on the context of the study and the specific research question. In general, a Q2 value greater than zero suggests that the model has predictive power, while a negative Q2 value indicates that the model has no predictive power.
Step 5: The goodness of fit of the model—GoF
To ensure that the model adequately describes the data, the goodness of fit (GoF) is used as a comprehensive measure for the model fit. In [
48], GoF was described as a global fit metric. It is the geometric mean of the average R
2 of the endogenous variables and the average variance extracted (AVE). GoF was developed to take into consideration the research model in its whole [
30,
49,
50], including its measurement and structural components. GoF can be calculated using the following formula:
According to [
51], a globally valid PLS model must have either a GoF value of zero, a small GoF value, a medium GoF value, or a high GoF value. The following table lists these factors (see
Table 11).
The goodness-of-fit (GoF) index is a statistical measure used to evaluate how well a structural equation modeling (SEM) model fits the data. In this case, the SEM-PLS model has a GoF value of 0.4569. A GoF value of 0.4569 suggests that the model has some degree of fit with the data, but the fit may not be optimal. However, the interpretation of the GoF value may depend on the context of the study and the specific research question. In general, a GoF value between 0.1 and 0.25 is considered weak, between 0.25 and 0.5 is moderate, and above 0.5 is strong. It is also important to note that the interpretation of the GoF index should be considered along with other statistical measures, such as the coefficient of determination (R-squared) and the predictive relevance (Q2), to provide a more comprehensive assessment of the model’s quality.
In order to investigate the structural model, it is important to grasp how the PLS-SEM adapts the model to empirical data in order to achieve optimal estimates of the parameters by maximizing the explained variance of the latent endogenous variable [
52]. The disadvantage of using goodness-of-fit measurements for the model, the structural model in PLS-SEM, is evaluated using heuristic criteria established by [
35]. This is because the model is seen to be well stated if it can accurately predict endogenous components [
53]. According to
Table 11, the effect of our model’s GoF result of 0.4569 on standard adoptions is minimal, falling between 0.1 and 0.25.
Notwithstanding the lack of resources in this field in poor nations, this research is one of the first to investigate the causes and methods for establishing a civil engineering standard measuring system in building projects in Malaysia. The study’s results also help in the implementation of best practices by identifying the most important drivers and tactics, enabling enterprises and governments interested in this topic to choose the most cost-effective civil engineering projects in the dynamic local construction sector. Analysis of the factor loading reveals that offering an efficient bill of quantities in better arrangement (factor loading = 0.774), and common knowledge in providing consistency, precision, and uniformity for measurement are key driving factors that contribute to standard acceptance (0.769). The third component that helps facilitate contract management is an improvement in construction project control and contract management (0.683). The last motivating aspect is the formulation’s simplification, which results in a competitive and reliable tender price and reduces the likelihood of needless disagreements (0.648).
When considering strategic considerations, however, standard-setting research and development is essential (0.771). Furthermore, it is important to have competent and proactive standard-method-of-measurement promotion teams and local authorities in place to ensure that the adopted standard is used in all cases (0.764) (0.755). Then, with a factor loading of 0.717 comes the implementation of established measuring rules and standards. Financial and additional market-based incentives for the adoption of the standard method of measurement (0.677), improved availability of information on the benefits of the standard method of measurement (0.663), and public awareness through workshops, seminars, and conferences (0.648) are the next three strategy factors. The stakeholders’ team may sort the many drivers and tactics according to the importance and rankings affecting the adoption of the standard, but it is impossible to build a single plan for all the aspects and give them the same attention, time, effort, and money (see
Figure 1).
The research aimed to investigate the relationship between the driver and strategy elements in the adoption of standardized measuring techniques in civil construction projects, particularly in emerging nations such as Malaysia. The study used structural equation modeling and the PLS-SEM technique to analyze the data collected through a questionnaire from quantity surveyors at quantity surveying consultancy companies using the standard measurement technique. The results of the study revealed that all classes significantly influence the adoption of the standard technique of assessment, but the barrier factors had the most impact. The adoption of a standardized technique of measuring was significantly impacted by the driver and strategy elements. The coefficient of determination (R-squared value) was found to be 0.4, indicating that 40% of the variance in the dependent variable is explained by the independent variables in the model. Moreover, the predictive relevance (Q2) was found to be 0.195, suggesting that the model can predict the endogenous latent components with an accuracy of 19.5%. Finally, the goodness-of-fit (GoF) index was found to be 0.4569, suggesting that the model fits the data to a moderate degree.
Based on the data, we can identify four drivers and seven strategic elements that influence the pace at which civil engineers adopt a standardized measuring technique. The findings of the structural equation model indicate that all items with factor loadings above the minimum value of 0.6 may explain the connection between driver and strategy factors and the adoption of the standard technique of assessment in construction projects. In contrast, an R-squared value of 0.400 indicates that 40% of the variance in the dependent variable(s) may be accounted for by the presence of one or more predictor factors. In addition, the Q2 is 0.195, which means that the conceptual model may make predictions about endogenous latent constructs. Hence, H1 and H2 are supported by a positive correlation. Our model received a high score on the goodness of fit of the model GoF. On the other hand, the path coefficient of the research hypothesis test, shown by the value of the beta coefficient, characterizes the strength of the relationship between the exogenous and endogenous latent constructs. The two most important determinants of the impact on the adoption of a standard technique of measurement in civil engineering are the driver factors (=0.302) and the strategy factors (=0.415).
The results of the study indicate that the driver and strategy factors have a significant impact on the adoption of standardized measuring techniques in civil construction projects. The finding that barrier factors have the most significant impact highlights the importance of identifying and addressing these barriers to increase the adoption of standardized measuring techniques in the industry. The relatively low R-squared and Q2 values suggest that there may be other factors not included in the model that also influence the adoption of standardized measuring techniques in civil construction projects.
The adoption of a standardized technique for measuring civil construction projects can be influenced by a range of factors, including drivers and strategies. In the context of emerging nations such as Malaysia, these factors may be particularly important due to the unique challenges and opportunities facing these countries. Drivers refer to the factors that motivate or incentivize the adoption of standardized measurement techniques. In the context of civil construction projects, some common drivers include the need for consistency and accuracy in measurement, compliance with industry regulations and standards, and the desire to improve communication and transparency among project stakeholders. Strategies refer to the approaches or methods used to facilitate the adoption of standardized measurement techniques. This may involve the development of training programs, the implementation of new measurement tools and technologies, or the creation of incentives or rewards for using standardized techniques. In emerging nations such as Malaysia, there may be additional drivers and strategies that are particularly relevant. For example, the need to attract foreign investment and compete with other countries in the region may be a driver for the adoption of standardized measurement techniques. Similarly, the use of government policies or regulations to encourage the adoption of standardized techniques may be an important strategy for promoting widespread adoption.
Nonetheless, there are still obstacles to using and implementing conventional measuring procedures when creating a bill of quantities. If the standard document is used, every scope of work will be clearly and frequently stated, ensuring that all contractors that participate in the project tender receive the same, accurate information. Costs associated with bidding uncertainty are reduced or eliminated thanks to this provision, and a more transparent and objective pool of qualified contractors is made possible. As a byproduct, it streamlines contract administration and removes budgetary uncertainty from projects. The standard measurement adaptation technique has been the subject of several investigations, but there has been little effort made to systematically analyze these or other relevant studies. To bridge the communication gap, the article discusses and defines the universally accepted measuring system in the building sector. This standardized way of measuring has not been widely used, however, and this is something that consultants, contractors, and clients in building projects are aware of. Quantifying building efforts is crucial for financial planning. A bill of quantities created from architectural and engineering drawings is the standard starting point for a quantity surveyor’s work in the construction and civil engineering industries [
5,
7,
20]. Disputes may be avoided with methodical bill of quantities preparation. As such, there has to be consistency in how uniformity is measured. The Malaysian Civil Engineering Standard Measurement Method (MyCESMM) was used for engineering projects, whereas the Malaysian Standard Method of Measurement of Building Works (SMM2) was used for construction [
7,
21,
22]. Disparities in measurement accuracy were a driving factor in deciding to conduct this investigation. According to Molloy, consultants employed conventional in-house measuring techniques, including several classifications and descriptions of the same thing (2007). According to [
21], RICS found 46 different standardized measuring methodologies throughout the world’s 27 nations. A literacy report notes, however, that some nations utilize a variety of non-conforming assessment techniques. In [
7], the authors provided more evidence of this issue by showing that no data from a global survey database (RICS 2003) because the members of RICS Malaysia did not investigate the reliability of the currently used measurement method. Key players in Malaysia’s construction sector referred to a wide range of measurement protocols for building and civil engineering despite a long gap in the research window (2003–2018). Although there have been empirical studies, [
7] noted, “there has been no coordinated effort to solve the issue that affects Malaysia’s building sector”.
Adopting standardized measurement methods in civil engineering construction projects can provide several benefits, including
Consistency: standardized measurement methods ensure consistency in the measurement and reporting of construction project quantities, which can help to reduce errors and discrepancies in the measurement process.
Improved accuracy: standardized measurement methods typically involve using more accurate and precise measurement tools and techniques, which can help to improve the accuracy of the measurements.
Efficiency: standardized measurement methods can help to streamline the measurement process, reducing the time and effort required to measure and report quantities. This can lead to increased efficiency and reduced costs.
Transparency: standardized measurement methods provide a transparent and objective way of measuring and reporting quantities, reducing the potential for disputes and conflicts between project stakeholders.
Improved communication: standardized measurement methods can help to improve communication between project stakeholders, providing a common language for discussing and reporting project quantities.
Compliance: standardized measurement methods are often required by industry regulations and standards, ensuring compliance with legal and regulatory requirements.