Structural Equation Models with Unobservable Variables and Measurement Error: Algebra and Statistics

In the citing paper

“Results revealed satisfactory reliability of all the measures with a threshold of 0.7, except for the measure of perceived physical conditions, which is greater than 0.6 [100].”

In the citing paper

“all discriminant validity requirements (Fornell and Larcker 1981) were met, since each scale ' s average variance extracted exceeded multiple squared correlations.”

In the citing paper

“The analysis showed that the square roots of the AVE of all constructs are higher than the correlation coefficients between constructs, indicating good discriminant validity.”

In the citing paper

“There was a higher discriminant validity in this study, and a path analysis of the structural equation model can be applied (Table 4).”

In the citing paper

“From the research by Fornell and Larcker (1981), for all the constructs, the CR should not be below 0.70 and the AVE should exceed 0.50, and the five constructs in this research met the two criteria (Table 3).”

In the citing paper

“The discriminant validity was also supported, considering that the inter-factor correlations were less than the square root of the factor's AVEs, as advocated by Fornell and Larcker and depicted in Table 2 [33].”

In the citing paper

“The reliability and the convergent validity were confirmed, as all five factors reached construct reliabilities (CRs) exceeding the generally recommended 0.70 threshold [31] and (b) the average variances extracted (AVE) exceeding the suggested 0.50 threshold [33], indicating that more than 50% of a factor's variance is due to its measurement items.”

In the citing paper

“As shown in Table 4, the squared root of the AVE for each construct (the value along the diagonal presented in bold) exceeded the correlation of that construct with other constructs, thereby establishing discriminant validity of all constructs.”

In the citing paper

“As per Fornell and Larcker's (1981) criteria, discriminant validity is established if the squared root of the AVE for each construct is larger than the correlation of that construct with other constructs.”

In the citing paper

“In the same vein, CR and rho_A were higher than the recommended thresholds 0.50 (Fornell and Larcker 1981) and 0.70 (Camilleri 2022;Ravand and Firoozi 2016) respectively, with the lowest CR equal to 0.889 and the lowest rho_A equal to 0.853.”

In the citing paper

“All the constructs exhibit AVE values higher than 0.5 (Fornell and Larcker 1981) with the lowest AVE equal to 0.618.”

In the citing paper

“all AVE values ranged from 0.604 to 0.698, surpassing the threshold of 0.50, which confirms adequate convergent validity”

In the citing paper

“the square root of the AVE for each construct exceeds its highest correlation with any other construct, providing additional evidence of discriminant validity at the construct level.”

In the citing paper

“Thus, the measurement models exhibited sufficient reliability, convergent validity and discriminant validity (Fornell & Larcker, 1981).”

In the citing paper

“In addition, we measured discriminant validity by comparing the AVE and correlation efficiency of the constructs, and more specifically, if the AVE of each construct was greater than the square of the correlation coefficient of two constructs (AVE > ∅2), the measurement model had discriminant validity (Fornell & Larcker, 1981).”

In the citing paper

“Almost all AVE values are above the suggested threshold (Fornell and Larcker, 1981); values above 0.7 are considered good, while a value of 0.5 is acceptable.”